diff options
author | hiddenpath <hiddenpath@yandex-team.com> | 2024-04-02 23:50:23 +0300 |
---|---|---|
committer | hiddenpath <hiddenpath@yandex-team.com> | 2024-04-03 00:02:31 +0300 |
commit | 8923c6d2c438e0aeed2e06b8b0275e1864eeee33 (patch) | |
tree | 6b5e476699fc0be5091cb650654ef5f602c8afff /contrib/go/_std_1.22/src/encoding | |
parent | d18afd09df2a08cd023012593b46109b77713a6c (diff) | |
download | ydb-8923c6d2c438e0aeed2e06b8b0275e1864eeee33.tar.gz |
Update golang to 1.22.1
2967d19c907adf59101a1f47b4208bd0b04a6186
Diffstat (limited to 'contrib/go/_std_1.22/src/encoding')
43 files changed, 18536 insertions, 0 deletions
diff --git a/contrib/go/_std_1.22/src/encoding/asn1/asn1.go b/contrib/go/_std_1.22/src/encoding/asn1/asn1.go new file mode 100644 index 0000000000..781ab87691 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/asn1/asn1.go @@ -0,0 +1,1124 @@ +// Copyright 2009 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +// Package asn1 implements parsing of DER-encoded ASN.1 data structures, +// as defined in ITU-T Rec X.690. +// +// See also “A Layman's Guide to a Subset of ASN.1, BER, and DER,” +// http://luca.ntop.org/Teaching/Appunti/asn1.html. +package asn1 + +// ASN.1 is a syntax for specifying abstract objects and BER, DER, PER, XER etc +// are different encoding formats for those objects. Here, we'll be dealing +// with DER, the Distinguished Encoding Rules. DER is used in X.509 because +// it's fast to parse and, unlike BER, has a unique encoding for every object. +// When calculating hashes over objects, it's important that the resulting +// bytes be the same at both ends and DER removes this margin of error. +// +// ASN.1 is very complex and this package doesn't attempt to implement +// everything by any means. + +import ( + "errors" + "fmt" + "math" + "math/big" + "reflect" + "strconv" + "strings" + "time" + "unicode/utf16" + "unicode/utf8" +) + +// A StructuralError suggests that the ASN.1 data is valid, but the Go type +// which is receiving it doesn't match. +type StructuralError struct { + Msg string +} + +func (e StructuralError) Error() string { return "asn1: structure error: " + e.Msg } + +// A SyntaxError suggests that the ASN.1 data is invalid. +type SyntaxError struct { + Msg string +} + +func (e SyntaxError) Error() string { return "asn1: syntax error: " + e.Msg } + +// We start by dealing with each of the primitive types in turn. + +// BOOLEAN + +func parseBool(bytes []byte) (ret bool, err error) { + if len(bytes) != 1 { + err = SyntaxError{"invalid boolean"} + return + } + + // DER demands that "If the encoding represents the boolean value TRUE, + // its single contents octet shall have all eight bits set to one." + // Thus only 0 and 255 are valid encoded values. + switch bytes[0] { + case 0: + ret = false + case 0xff: + ret = true + default: + err = SyntaxError{"invalid boolean"} + } + + return +} + +// INTEGER + +// checkInteger returns nil if the given bytes are a valid DER-encoded +// INTEGER and an error otherwise. +func checkInteger(bytes []byte) error { + if len(bytes) == 0 { + return StructuralError{"empty integer"} + } + if len(bytes) == 1 { + return nil + } + if (bytes[0] == 0 && bytes[1]&0x80 == 0) || (bytes[0] == 0xff && bytes[1]&0x80 == 0x80) { + return StructuralError{"integer not minimally-encoded"} + } + return nil +} + +// parseInt64 treats the given bytes as a big-endian, signed integer and +// returns the result. +func parseInt64(bytes []byte) (ret int64, err error) { + err = checkInteger(bytes) + if err != nil { + return + } + if len(bytes) > 8 { + // We'll overflow an int64 in this case. + err = StructuralError{"integer too large"} + return + } + for bytesRead := 0; bytesRead < len(bytes); bytesRead++ { + ret <<= 8 + ret |= int64(bytes[bytesRead]) + } + + // Shift up and down in order to sign extend the result. + ret <<= 64 - uint8(len(bytes))*8 + ret >>= 64 - uint8(len(bytes))*8 + return +} + +// parseInt32 treats the given bytes as a big-endian, signed integer and returns +// the result. +func parseInt32(bytes []byte) (int32, error) { + if err := checkInteger(bytes); err != nil { + return 0, err + } + ret64, err := parseInt64(bytes) + if err != nil { + return 0, err + } + if ret64 != int64(int32(ret64)) { + return 0, StructuralError{"integer too large"} + } + return int32(ret64), nil +} + +var bigOne = big.NewInt(1) + +// parseBigInt treats the given bytes as a big-endian, signed integer and returns +// the result. +func parseBigInt(bytes []byte) (*big.Int, error) { + if err := checkInteger(bytes); err != nil { + return nil, err + } + ret := new(big.Int) + if len(bytes) > 0 && bytes[0]&0x80 == 0x80 { + // This is a negative number. + notBytes := make([]byte, len(bytes)) + for i := range notBytes { + notBytes[i] = ^bytes[i] + } + ret.SetBytes(notBytes) + ret.Add(ret, bigOne) + ret.Neg(ret) + return ret, nil + } + ret.SetBytes(bytes) + return ret, nil +} + +// BIT STRING + +// BitString is the structure to use when you want an ASN.1 BIT STRING type. A +// bit string is padded up to the nearest byte in memory and the number of +// valid bits is recorded. Padding bits will be zero. +type BitString struct { + Bytes []byte // bits packed into bytes. + BitLength int // length in bits. +} + +// At returns the bit at the given index. If the index is out of range it +// returns 0. +func (b BitString) At(i int) int { + if i < 0 || i >= b.BitLength { + return 0 + } + x := i / 8 + y := 7 - uint(i%8) + return int(b.Bytes[x]>>y) & 1 +} + +// RightAlign returns a slice where the padding bits are at the beginning. The +// slice may share memory with the BitString. +func (b BitString) RightAlign() []byte { + shift := uint(8 - (b.BitLength % 8)) + if shift == 8 || len(b.Bytes) == 0 { + return b.Bytes + } + + a := make([]byte, len(b.Bytes)) + a[0] = b.Bytes[0] >> shift + for i := 1; i < len(b.Bytes); i++ { + a[i] = b.Bytes[i-1] << (8 - shift) + a[i] |= b.Bytes[i] >> shift + } + + return a +} + +// parseBitString parses an ASN.1 bit string from the given byte slice and returns it. +func parseBitString(bytes []byte) (ret BitString, err error) { + if len(bytes) == 0 { + err = SyntaxError{"zero length BIT STRING"} + return + } + paddingBits := int(bytes[0]) + if paddingBits > 7 || + len(bytes) == 1 && paddingBits > 0 || + bytes[len(bytes)-1]&((1<<bytes[0])-1) != 0 { + err = SyntaxError{"invalid padding bits in BIT STRING"} + return + } + ret.BitLength = (len(bytes)-1)*8 - paddingBits + ret.Bytes = bytes[1:] + return +} + +// NULL + +// NullRawValue is a [RawValue] with its Tag set to the ASN.1 NULL type tag (5). +var NullRawValue = RawValue{Tag: TagNull} + +// NullBytes contains bytes representing the DER-encoded ASN.1 NULL type. +var NullBytes = []byte{TagNull, 0} + +// OBJECT IDENTIFIER + +// An ObjectIdentifier represents an ASN.1 OBJECT IDENTIFIER. +type ObjectIdentifier []int + +// Equal reports whether oi and other represent the same identifier. +func (oi ObjectIdentifier) Equal(other ObjectIdentifier) bool { + if len(oi) != len(other) { + return false + } + for i := 0; i < len(oi); i++ { + if oi[i] != other[i] { + return false + } + } + + return true +} + +func (oi ObjectIdentifier) String() string { + var s strings.Builder + s.Grow(32) + + buf := make([]byte, 0, 19) + for i, v := range oi { + if i > 0 { + s.WriteByte('.') + } + s.Write(strconv.AppendInt(buf, int64(v), 10)) + } + + return s.String() +} + +// parseObjectIdentifier parses an OBJECT IDENTIFIER from the given bytes and +// returns it. An object identifier is a sequence of variable length integers +// that are assigned in a hierarchy. +func parseObjectIdentifier(bytes []byte) (s ObjectIdentifier, err error) { + if len(bytes) == 0 { + err = SyntaxError{"zero length OBJECT IDENTIFIER"} + return + } + + // In the worst case, we get two elements from the first byte (which is + // encoded differently) and then every varint is a single byte long. + s = make([]int, len(bytes)+1) + + // The first varint is 40*value1 + value2: + // According to this packing, value1 can take the values 0, 1 and 2 only. + // When value1 = 0 or value1 = 1, then value2 is <= 39. When value1 = 2, + // then there are no restrictions on value2. + v, offset, err := parseBase128Int(bytes, 0) + if err != nil { + return + } + if v < 80 { + s[0] = v / 40 + s[1] = v % 40 + } else { + s[0] = 2 + s[1] = v - 80 + } + + i := 2 + for ; offset < len(bytes); i++ { + v, offset, err = parseBase128Int(bytes, offset) + if err != nil { + return + } + s[i] = v + } + s = s[0:i] + return +} + +// ENUMERATED + +// An Enumerated is represented as a plain int. +type Enumerated int + +// FLAG + +// A Flag accepts any data and is set to true if present. +type Flag bool + +// parseBase128Int parses a base-128 encoded int from the given offset in the +// given byte slice. It returns the value and the new offset. +func parseBase128Int(bytes []byte, initOffset int) (ret, offset int, err error) { + offset = initOffset + var ret64 int64 + for shifted := 0; offset < len(bytes); shifted++ { + // 5 * 7 bits per byte == 35 bits of data + // Thus the representation is either non-minimal or too large for an int32 + if shifted == 5 { + err = StructuralError{"base 128 integer too large"} + return + } + ret64 <<= 7 + b := bytes[offset] + // integers should be minimally encoded, so the leading octet should + // never be 0x80 + if shifted == 0 && b == 0x80 { + err = SyntaxError{"integer is not minimally encoded"} + return + } + ret64 |= int64(b & 0x7f) + offset++ + if b&0x80 == 0 { + ret = int(ret64) + // Ensure that the returned value fits in an int on all platforms + if ret64 > math.MaxInt32 { + err = StructuralError{"base 128 integer too large"} + } + return + } + } + err = SyntaxError{"truncated base 128 integer"} + return +} + +// UTCTime + +func parseUTCTime(bytes []byte) (ret time.Time, err error) { + s := string(bytes) + + formatStr := "0601021504Z0700" + ret, err = time.Parse(formatStr, s) + if err != nil { + formatStr = "060102150405Z0700" + ret, err = time.Parse(formatStr, s) + } + if err != nil { + return + } + + if serialized := ret.Format(formatStr); serialized != s { + err = fmt.Errorf("asn1: time did not serialize back to the original value and may be invalid: given %q, but serialized as %q", s, serialized) + return + } + + if ret.Year() >= 2050 { + // UTCTime only encodes times prior to 2050. See https://tools.ietf.org/html/rfc5280#section-4.1.2.5.1 + ret = ret.AddDate(-100, 0, 0) + } + + return +} + +// parseGeneralizedTime parses the GeneralizedTime from the given byte slice +// and returns the resulting time. +func parseGeneralizedTime(bytes []byte) (ret time.Time, err error) { + const formatStr = "20060102150405.999999999Z0700" + s := string(bytes) + + if ret, err = time.Parse(formatStr, s); err != nil { + return + } + + if serialized := ret.Format(formatStr); serialized != s { + err = fmt.Errorf("asn1: time did not serialize back to the original value and may be invalid: given %q, but serialized as %q", s, serialized) + } + + return +} + +// NumericString + +// parseNumericString parses an ASN.1 NumericString from the given byte array +// and returns it. +func parseNumericString(bytes []byte) (ret string, err error) { + for _, b := range bytes { + if !isNumeric(b) { + return "", SyntaxError{"NumericString contains invalid character"} + } + } + return string(bytes), nil +} + +// isNumeric reports whether the given b is in the ASN.1 NumericString set. +func isNumeric(b byte) bool { + return '0' <= b && b <= '9' || + b == ' ' +} + +// PrintableString + +// parsePrintableString parses an ASN.1 PrintableString from the given byte +// array and returns it. +func parsePrintableString(bytes []byte) (ret string, err error) { + for _, b := range bytes { + if !isPrintable(b, allowAsterisk, allowAmpersand) { + err = SyntaxError{"PrintableString contains invalid character"} + return + } + } + ret = string(bytes) + return +} + +type asteriskFlag bool +type ampersandFlag bool + +const ( + allowAsterisk asteriskFlag = true + rejectAsterisk asteriskFlag = false + + allowAmpersand ampersandFlag = true + rejectAmpersand ampersandFlag = false +) + +// isPrintable reports whether the given b is in the ASN.1 PrintableString set. +// If asterisk is allowAsterisk then '*' is also allowed, reflecting existing +// practice. If ampersand is allowAmpersand then '&' is allowed as well. +func isPrintable(b byte, asterisk asteriskFlag, ampersand ampersandFlag) bool { + return 'a' <= b && b <= 'z' || + 'A' <= b && b <= 'Z' || + '0' <= b && b <= '9' || + '\'' <= b && b <= ')' || + '+' <= b && b <= '/' || + b == ' ' || + b == ':' || + b == '=' || + b == '?' || + // This is technically not allowed in a PrintableString. + // However, x509 certificates with wildcard strings don't + // always use the correct string type so we permit it. + (bool(asterisk) && b == '*') || + // This is not technically allowed either. However, not + // only is it relatively common, but there are also a + // handful of CA certificates that contain it. At least + // one of which will not expire until 2027. + (bool(ampersand) && b == '&') +} + +// IA5String + +// parseIA5String parses an ASN.1 IA5String (ASCII string) from the given +// byte slice and returns it. +func parseIA5String(bytes []byte) (ret string, err error) { + for _, b := range bytes { + if b >= utf8.RuneSelf { + err = SyntaxError{"IA5String contains invalid character"} + return + } + } + ret = string(bytes) + return +} + +// T61String + +// parseT61String parses an ASN.1 T61String (8-bit clean string) from the given +// byte slice and returns it. +func parseT61String(bytes []byte) (ret string, err error) { + return string(bytes), nil +} + +// UTF8String + +// parseUTF8String parses an ASN.1 UTF8String (raw UTF-8) from the given byte +// array and returns it. +func parseUTF8String(bytes []byte) (ret string, err error) { + if !utf8.Valid(bytes) { + return "", errors.New("asn1: invalid UTF-8 string") + } + return string(bytes), nil +} + +// BMPString + +// parseBMPString parses an ASN.1 BMPString (Basic Multilingual Plane of +// ISO/IEC/ITU 10646-1) from the given byte slice and returns it. +func parseBMPString(bmpString []byte) (string, error) { + if len(bmpString)%2 != 0 { + return "", errors.New("pkcs12: odd-length BMP string") + } + + // Strip terminator if present. + if l := len(bmpString); l >= 2 && bmpString[l-1] == 0 && bmpString[l-2] == 0 { + bmpString = bmpString[:l-2] + } + + s := make([]uint16, 0, len(bmpString)/2) + for len(bmpString) > 0 { + s = append(s, uint16(bmpString[0])<<8+uint16(bmpString[1])) + bmpString = bmpString[2:] + } + + return string(utf16.Decode(s)), nil +} + +// A RawValue represents an undecoded ASN.1 object. +type RawValue struct { + Class, Tag int + IsCompound bool + Bytes []byte + FullBytes []byte // includes the tag and length +} + +// RawContent is used to signal that the undecoded, DER data needs to be +// preserved for a struct. To use it, the first field of the struct must have +// this type. It's an error for any of the other fields to have this type. +type RawContent []byte + +// Tagging + +// parseTagAndLength parses an ASN.1 tag and length pair from the given offset +// into a byte slice. It returns the parsed data and the new offset. SET and +// SET OF (tag 17) are mapped to SEQUENCE and SEQUENCE OF (tag 16) since we +// don't distinguish between ordered and unordered objects in this code. +func parseTagAndLength(bytes []byte, initOffset int) (ret tagAndLength, offset int, err error) { + offset = initOffset + // parseTagAndLength should not be called without at least a single + // byte to read. Thus this check is for robustness: + if offset >= len(bytes) { + err = errors.New("asn1: internal error in parseTagAndLength") + return + } + b := bytes[offset] + offset++ + ret.class = int(b >> 6) + ret.isCompound = b&0x20 == 0x20 + ret.tag = int(b & 0x1f) + + // If the bottom five bits are set, then the tag number is actually base 128 + // encoded afterwards + if ret.tag == 0x1f { + ret.tag, offset, err = parseBase128Int(bytes, offset) + if err != nil { + return + } + // Tags should be encoded in minimal form. + if ret.tag < 0x1f { + err = SyntaxError{"non-minimal tag"} + return + } + } + if offset >= len(bytes) { + err = SyntaxError{"truncated tag or length"} + return + } + b = bytes[offset] + offset++ + if b&0x80 == 0 { + // The length is encoded in the bottom 7 bits. + ret.length = int(b & 0x7f) + } else { + // Bottom 7 bits give the number of length bytes to follow. + numBytes := int(b & 0x7f) + if numBytes == 0 { + err = SyntaxError{"indefinite length found (not DER)"} + return + } + ret.length = 0 + for i := 0; i < numBytes; i++ { + if offset >= len(bytes) { + err = SyntaxError{"truncated tag or length"} + return + } + b = bytes[offset] + offset++ + if ret.length >= 1<<23 { + // We can't shift ret.length up without + // overflowing. + err = StructuralError{"length too large"} + return + } + ret.length <<= 8 + ret.length |= int(b) + if ret.length == 0 { + // DER requires that lengths be minimal. + err = StructuralError{"superfluous leading zeros in length"} + return + } + } + // Short lengths must be encoded in short form. + if ret.length < 0x80 { + err = StructuralError{"non-minimal length"} + return + } + } + + return +} + +// parseSequenceOf is used for SEQUENCE OF and SET OF values. It tries to parse +// a number of ASN.1 values from the given byte slice and returns them as a +// slice of Go values of the given type. +func parseSequenceOf(bytes []byte, sliceType reflect.Type, elemType reflect.Type) (ret reflect.Value, err error) { + matchAny, expectedTag, compoundType, ok := getUniversalType(elemType) + if !ok { + err = StructuralError{"unknown Go type for slice"} + return + } + + // First we iterate over the input and count the number of elements, + // checking that the types are correct in each case. + numElements := 0 + for offset := 0; offset < len(bytes); { + var t tagAndLength + t, offset, err = parseTagAndLength(bytes, offset) + if err != nil { + return + } + switch t.tag { + case TagIA5String, TagGeneralString, TagT61String, TagUTF8String, TagNumericString, TagBMPString: + // We pretend that various other string types are + // PRINTABLE STRINGs so that a sequence of them can be + // parsed into a []string. + t.tag = TagPrintableString + case TagGeneralizedTime, TagUTCTime: + // Likewise, both time types are treated the same. + t.tag = TagUTCTime + } + + if !matchAny && (t.class != ClassUniversal || t.isCompound != compoundType || t.tag != expectedTag) { + err = StructuralError{"sequence tag mismatch"} + return + } + if invalidLength(offset, t.length, len(bytes)) { + err = SyntaxError{"truncated sequence"} + return + } + offset += t.length + numElements++ + } + ret = reflect.MakeSlice(sliceType, numElements, numElements) + params := fieldParameters{} + offset := 0 + for i := 0; i < numElements; i++ { + offset, err = parseField(ret.Index(i), bytes, offset, params) + if err != nil { + return + } + } + return +} + +var ( + bitStringType = reflect.TypeFor[BitString]() + objectIdentifierType = reflect.TypeFor[ObjectIdentifier]() + enumeratedType = reflect.TypeFor[Enumerated]() + flagType = reflect.TypeFor[Flag]() + timeType = reflect.TypeFor[time.Time]() + rawValueType = reflect.TypeFor[RawValue]() + rawContentsType = reflect.TypeFor[RawContent]() + bigIntType = reflect.TypeFor[*big.Int]() +) + +// invalidLength reports whether offset + length > sliceLength, or if the +// addition would overflow. +func invalidLength(offset, length, sliceLength int) bool { + return offset+length < offset || offset+length > sliceLength +} + +// parseField is the main parsing function. Given a byte slice and an offset +// into the array, it will try to parse a suitable ASN.1 value out and store it +// in the given Value. +func parseField(v reflect.Value, bytes []byte, initOffset int, params fieldParameters) (offset int, err error) { + offset = initOffset + fieldType := v.Type() + + // If we have run out of data, it may be that there are optional elements at the end. + if offset == len(bytes) { + if !setDefaultValue(v, params) { + err = SyntaxError{"sequence truncated"} + } + return + } + + // Deal with the ANY type. + if ifaceType := fieldType; ifaceType.Kind() == reflect.Interface && ifaceType.NumMethod() == 0 { + var t tagAndLength + t, offset, err = parseTagAndLength(bytes, offset) + if err != nil { + return + } + if invalidLength(offset, t.length, len(bytes)) { + err = SyntaxError{"data truncated"} + return + } + var result any + if !t.isCompound && t.class == ClassUniversal { + innerBytes := bytes[offset : offset+t.length] + switch t.tag { + case TagPrintableString: + result, err = parsePrintableString(innerBytes) + case TagNumericString: + result, err = parseNumericString(innerBytes) + case TagIA5String: + result, err = parseIA5String(innerBytes) + case TagT61String: + result, err = parseT61String(innerBytes) + case TagUTF8String: + result, err = parseUTF8String(innerBytes) + case TagInteger: + result, err = parseInt64(innerBytes) + case TagBitString: + result, err = parseBitString(innerBytes) + case TagOID: + result, err = parseObjectIdentifier(innerBytes) + case TagUTCTime: + result, err = parseUTCTime(innerBytes) + case TagGeneralizedTime: + result, err = parseGeneralizedTime(innerBytes) + case TagOctetString: + result = innerBytes + case TagBMPString: + result, err = parseBMPString(innerBytes) + default: + // If we don't know how to handle the type, we just leave Value as nil. + } + } + offset += t.length + if err != nil { + return + } + if result != nil { + v.Set(reflect.ValueOf(result)) + } + return + } + + t, offset, err := parseTagAndLength(bytes, offset) + if err != nil { + return + } + if params.explicit { + expectedClass := ClassContextSpecific + if params.application { + expectedClass = ClassApplication + } + if offset == len(bytes) { + err = StructuralError{"explicit tag has no child"} + return + } + if t.class == expectedClass && t.tag == *params.tag && (t.length == 0 || t.isCompound) { + if fieldType == rawValueType { + // The inner element should not be parsed for RawValues. + } else if t.length > 0 { + t, offset, err = parseTagAndLength(bytes, offset) + if err != nil { + return + } + } else { + if fieldType != flagType { + err = StructuralError{"zero length explicit tag was not an asn1.Flag"} + return + } + v.SetBool(true) + return + } + } else { + // The tags didn't match, it might be an optional element. + ok := setDefaultValue(v, params) + if ok { + offset = initOffset + } else { + err = StructuralError{"explicitly tagged member didn't match"} + } + return + } + } + + matchAny, universalTag, compoundType, ok1 := getUniversalType(fieldType) + if !ok1 { + err = StructuralError{fmt.Sprintf("unknown Go type: %v", fieldType)} + return + } + + // Special case for strings: all the ASN.1 string types map to the Go + // type string. getUniversalType returns the tag for PrintableString + // when it sees a string, so if we see a different string type on the + // wire, we change the universal type to match. + if universalTag == TagPrintableString { + if t.class == ClassUniversal { + switch t.tag { + case TagIA5String, TagGeneralString, TagT61String, TagUTF8String, TagNumericString, TagBMPString: + universalTag = t.tag + } + } else if params.stringType != 0 { + universalTag = params.stringType + } + } + + // Special case for time: UTCTime and GeneralizedTime both map to the + // Go type time.Time. + if universalTag == TagUTCTime && t.tag == TagGeneralizedTime && t.class == ClassUniversal { + universalTag = TagGeneralizedTime + } + + if params.set { + universalTag = TagSet + } + + matchAnyClassAndTag := matchAny + expectedClass := ClassUniversal + expectedTag := universalTag + + if !params.explicit && params.tag != nil { + expectedClass = ClassContextSpecific + expectedTag = *params.tag + matchAnyClassAndTag = false + } + + if !params.explicit && params.application && params.tag != nil { + expectedClass = ClassApplication + expectedTag = *params.tag + matchAnyClassAndTag = false + } + + if !params.explicit && params.private && params.tag != nil { + expectedClass = ClassPrivate + expectedTag = *params.tag + matchAnyClassAndTag = false + } + + // We have unwrapped any explicit tagging at this point. + if !matchAnyClassAndTag && (t.class != expectedClass || t.tag != expectedTag) || + (!matchAny && t.isCompound != compoundType) { + // Tags don't match. Again, it could be an optional element. + ok := setDefaultValue(v, params) + if ok { + offset = initOffset + } else { + err = StructuralError{fmt.Sprintf("tags don't match (%d vs %+v) %+v %s @%d", expectedTag, t, params, fieldType.Name(), offset)} + } + return + } + if invalidLength(offset, t.length, len(bytes)) { + err = SyntaxError{"data truncated"} + return + } + innerBytes := bytes[offset : offset+t.length] + offset += t.length + + // We deal with the structures defined in this package first. + switch v := v.Addr().Interface().(type) { + case *RawValue: + *v = RawValue{t.class, t.tag, t.isCompound, innerBytes, bytes[initOffset:offset]} + return + case *ObjectIdentifier: + *v, err = parseObjectIdentifier(innerBytes) + return + case *BitString: + *v, err = parseBitString(innerBytes) + return + case *time.Time: + if universalTag == TagUTCTime { + *v, err = parseUTCTime(innerBytes) + return + } + *v, err = parseGeneralizedTime(innerBytes) + return + case *Enumerated: + parsedInt, err1 := parseInt32(innerBytes) + if err1 == nil { + *v = Enumerated(parsedInt) + } + err = err1 + return + case *Flag: + *v = true + return + case **big.Int: + parsedInt, err1 := parseBigInt(innerBytes) + if err1 == nil { + *v = parsedInt + } + err = err1 + return + } + switch val := v; val.Kind() { + case reflect.Bool: + parsedBool, err1 := parseBool(innerBytes) + if err1 == nil { + val.SetBool(parsedBool) + } + err = err1 + return + case reflect.Int, reflect.Int32, reflect.Int64: + if val.Type().Size() == 4 { + parsedInt, err1 := parseInt32(innerBytes) + if err1 == nil { + val.SetInt(int64(parsedInt)) + } + err = err1 + } else { + parsedInt, err1 := parseInt64(innerBytes) + if err1 == nil { + val.SetInt(parsedInt) + } + err = err1 + } + return + // TODO(dfc) Add support for the remaining integer types + case reflect.Struct: + structType := fieldType + + for i := 0; i < structType.NumField(); i++ { + if !structType.Field(i).IsExported() { + err = StructuralError{"struct contains unexported fields"} + return + } + } + + if structType.NumField() > 0 && + structType.Field(0).Type == rawContentsType { + bytes := bytes[initOffset:offset] + val.Field(0).Set(reflect.ValueOf(RawContent(bytes))) + } + + innerOffset := 0 + for i := 0; i < structType.NumField(); i++ { + field := structType.Field(i) + if i == 0 && field.Type == rawContentsType { + continue + } + innerOffset, err = parseField(val.Field(i), innerBytes, innerOffset, parseFieldParameters(field.Tag.Get("asn1"))) + if err != nil { + return + } + } + // We allow extra bytes at the end of the SEQUENCE because + // adding elements to the end has been used in X.509 as the + // version numbers have increased. + return + case reflect.Slice: + sliceType := fieldType + if sliceType.Elem().Kind() == reflect.Uint8 { + val.Set(reflect.MakeSlice(sliceType, len(innerBytes), len(innerBytes))) + reflect.Copy(val, reflect.ValueOf(innerBytes)) + return + } + newSlice, err1 := parseSequenceOf(innerBytes, sliceType, sliceType.Elem()) + if err1 == nil { + val.Set(newSlice) + } + err = err1 + return + case reflect.String: + var v string + switch universalTag { + case TagPrintableString: + v, err = parsePrintableString(innerBytes) + case TagNumericString: + v, err = parseNumericString(innerBytes) + case TagIA5String: + v, err = parseIA5String(innerBytes) + case TagT61String: + v, err = parseT61String(innerBytes) + case TagUTF8String: + v, err = parseUTF8String(innerBytes) + case TagGeneralString: + // GeneralString is specified in ISO-2022/ECMA-35, + // A brief review suggests that it includes structures + // that allow the encoding to change midstring and + // such. We give up and pass it as an 8-bit string. + v, err = parseT61String(innerBytes) + case TagBMPString: + v, err = parseBMPString(innerBytes) + + default: + err = SyntaxError{fmt.Sprintf("internal error: unknown string type %d", universalTag)} + } + if err == nil { + val.SetString(v) + } + return + } + err = StructuralError{"unsupported: " + v.Type().String()} + return +} + +// canHaveDefaultValue reports whether k is a Kind that we will set a default +// value for. (A signed integer, essentially.) +func canHaveDefaultValue(k reflect.Kind) bool { + switch k { + case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64: + return true + } + + return false +} + +// setDefaultValue is used to install a default value, from a tag string, into +// a Value. It is successful if the field was optional, even if a default value +// wasn't provided or it failed to install it into the Value. +func setDefaultValue(v reflect.Value, params fieldParameters) (ok bool) { + if !params.optional { + return + } + ok = true + if params.defaultValue == nil { + return + } + if canHaveDefaultValue(v.Kind()) { + v.SetInt(*params.defaultValue) + } + return +} + +// Unmarshal parses the DER-encoded ASN.1 data structure b +// and uses the reflect package to fill in an arbitrary value pointed at by val. +// Because Unmarshal uses the reflect package, the structs +// being written to must use upper case field names. If val +// is nil or not a pointer, Unmarshal returns an error. +// +// After parsing b, any bytes that were leftover and not used to fill +// val will be returned in rest. When parsing a SEQUENCE into a struct, +// any trailing elements of the SEQUENCE that do not have matching +// fields in val will not be included in rest, as these are considered +// valid elements of the SEQUENCE and not trailing data. +// +// - An ASN.1 INTEGER can be written to an int, int32, int64, +// or *[big.Int]. +// If the encoded value does not fit in the Go type, +// Unmarshal returns a parse error. +// +// - An ASN.1 BIT STRING can be written to a [BitString]. +// +// - An ASN.1 OCTET STRING can be written to a []byte. +// +// - An ASN.1 OBJECT IDENTIFIER can be written to an [ObjectIdentifier]. +// +// - An ASN.1 ENUMERATED can be written to an [Enumerated]. +// +// - An ASN.1 UTCTIME or GENERALIZEDTIME can be written to a [time.Time]. +// +// - An ASN.1 PrintableString, IA5String, or NumericString can be written to a string. +// +// - Any of the above ASN.1 values can be written to an interface{}. +// The value stored in the interface has the corresponding Go type. +// For integers, that type is int64. +// +// - An ASN.1 SEQUENCE OF x or SET OF x can be written +// to a slice if an x can be written to the slice's element type. +// +// - An ASN.1 SEQUENCE or SET can be written to a struct +// if each of the elements in the sequence can be +// written to the corresponding element in the struct. +// +// The following tags on struct fields have special meaning to Unmarshal: +// +// application specifies that an APPLICATION tag is used +// private specifies that a PRIVATE tag is used +// default:x sets the default value for optional integer fields (only used if optional is also present) +// explicit specifies that an additional, explicit tag wraps the implicit one +// optional marks the field as ASN.1 OPTIONAL +// set causes a SET, rather than a SEQUENCE type to be expected +// tag:x specifies the ASN.1 tag number; implies ASN.1 CONTEXT SPECIFIC +// +// When decoding an ASN.1 value with an IMPLICIT tag into a string field, +// Unmarshal will default to a PrintableString, which doesn't support +// characters such as '@' and '&'. To force other encodings, use the following +// tags: +// +// ia5 causes strings to be unmarshaled as ASN.1 IA5String values +// numeric causes strings to be unmarshaled as ASN.1 NumericString values +// utf8 causes strings to be unmarshaled as ASN.1 UTF8String values +// +// If the type of the first field of a structure is RawContent then the raw +// ASN1 contents of the struct will be stored in it. +// +// If the name of a slice type ends with "SET" then it's treated as if +// the "set" tag was set on it. This results in interpreting the type as a +// SET OF x rather than a SEQUENCE OF x. This can be used with nested slices +// where a struct tag cannot be given. +// +// Other ASN.1 types are not supported; if it encounters them, +// Unmarshal returns a parse error. +func Unmarshal(b []byte, val any) (rest []byte, err error) { + return UnmarshalWithParams(b, val, "") +} + +// An invalidUnmarshalError describes an invalid argument passed to Unmarshal. +// (The argument to Unmarshal must be a non-nil pointer.) +type invalidUnmarshalError struct { + Type reflect.Type +} + +func (e *invalidUnmarshalError) Error() string { + if e.Type == nil { + return "asn1: Unmarshal recipient value is nil" + } + + if e.Type.Kind() != reflect.Pointer { + return "asn1: Unmarshal recipient value is non-pointer " + e.Type.String() + } + return "asn1: Unmarshal recipient value is nil " + e.Type.String() +} + +// UnmarshalWithParams allows field parameters to be specified for the +// top-level element. The form of the params is the same as the field tags. +func UnmarshalWithParams(b []byte, val any, params string) (rest []byte, err error) { + v := reflect.ValueOf(val) + if v.Kind() != reflect.Pointer || v.IsNil() { + return nil, &invalidUnmarshalError{reflect.TypeOf(val)} + } + offset, err := parseField(v.Elem(), b, 0, parseFieldParameters(params)) + if err != nil { + return nil, err + } + return b[offset:], nil +} diff --git a/contrib/go/_std_1.22/src/encoding/asn1/common.go b/contrib/go/_std_1.22/src/encoding/asn1/common.go new file mode 100644 index 0000000000..40115df8b4 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/asn1/common.go @@ -0,0 +1,185 @@ +// Copyright 2009 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package asn1 + +import ( + "reflect" + "strconv" + "strings" +) + +// ASN.1 objects have metadata preceding them: +// the tag: the type of the object +// a flag denoting if this object is compound or not +// the class type: the namespace of the tag +// the length of the object, in bytes + +// Here are some standard tags and classes + +// ASN.1 tags represent the type of the following object. +const ( + TagBoolean = 1 + TagInteger = 2 + TagBitString = 3 + TagOctetString = 4 + TagNull = 5 + TagOID = 6 + TagEnum = 10 + TagUTF8String = 12 + TagSequence = 16 + TagSet = 17 + TagNumericString = 18 + TagPrintableString = 19 + TagT61String = 20 + TagIA5String = 22 + TagUTCTime = 23 + TagGeneralizedTime = 24 + TagGeneralString = 27 + TagBMPString = 30 +) + +// ASN.1 class types represent the namespace of the tag. +const ( + ClassUniversal = 0 + ClassApplication = 1 + ClassContextSpecific = 2 + ClassPrivate = 3 +) + +type tagAndLength struct { + class, tag, length int + isCompound bool +} + +// ASN.1 has IMPLICIT and EXPLICIT tags, which can be translated as "instead +// of" and "in addition to". When not specified, every primitive type has a +// default tag in the UNIVERSAL class. +// +// For example: a BIT STRING is tagged [UNIVERSAL 3] by default (although ASN.1 +// doesn't actually have a UNIVERSAL keyword). However, by saying [IMPLICIT +// CONTEXT-SPECIFIC 42], that means that the tag is replaced by another. +// +// On the other hand, if it said [EXPLICIT CONTEXT-SPECIFIC 10], then an +// /additional/ tag would wrap the default tag. This explicit tag will have the +// compound flag set. +// +// (This is used in order to remove ambiguity with optional elements.) +// +// You can layer EXPLICIT and IMPLICIT tags to an arbitrary depth, however we +// don't support that here. We support a single layer of EXPLICIT or IMPLICIT +// tagging with tag strings on the fields of a structure. + +// fieldParameters is the parsed representation of tag string from a structure field. +type fieldParameters struct { + optional bool // true iff the field is OPTIONAL + explicit bool // true iff an EXPLICIT tag is in use. + application bool // true iff an APPLICATION tag is in use. + private bool // true iff a PRIVATE tag is in use. + defaultValue *int64 // a default value for INTEGER typed fields (maybe nil). + tag *int // the EXPLICIT or IMPLICIT tag (maybe nil). + stringType int // the string tag to use when marshaling. + timeType int // the time tag to use when marshaling. + set bool // true iff this should be encoded as a SET + omitEmpty bool // true iff this should be omitted if empty when marshaling. + + // Invariants: + // if explicit is set, tag is non-nil. +} + +// Given a tag string with the format specified in the package comment, +// parseFieldParameters will parse it into a fieldParameters structure, +// ignoring unknown parts of the string. +func parseFieldParameters(str string) (ret fieldParameters) { + var part string + for len(str) > 0 { + part, str, _ = strings.Cut(str, ",") + switch { + case part == "optional": + ret.optional = true + case part == "explicit": + ret.explicit = true + if ret.tag == nil { + ret.tag = new(int) + } + case part == "generalized": + ret.timeType = TagGeneralizedTime + case part == "utc": + ret.timeType = TagUTCTime + case part == "ia5": + ret.stringType = TagIA5String + case part == "printable": + ret.stringType = TagPrintableString + case part == "numeric": + ret.stringType = TagNumericString + case part == "utf8": + ret.stringType = TagUTF8String + case strings.HasPrefix(part, "default:"): + i, err := strconv.ParseInt(part[8:], 10, 64) + if err == nil { + ret.defaultValue = new(int64) + *ret.defaultValue = i + } + case strings.HasPrefix(part, "tag:"): + i, err := strconv.Atoi(part[4:]) + if err == nil { + ret.tag = new(int) + *ret.tag = i + } + case part == "set": + ret.set = true + case part == "application": + ret.application = true + if ret.tag == nil { + ret.tag = new(int) + } + case part == "private": + ret.private = true + if ret.tag == nil { + ret.tag = new(int) + } + case part == "omitempty": + ret.omitEmpty = true + } + } + return +} + +// Given a reflected Go type, getUniversalType returns the default tag number +// and expected compound flag. +func getUniversalType(t reflect.Type) (matchAny bool, tagNumber int, isCompound, ok bool) { + switch t { + case rawValueType: + return true, -1, false, true + case objectIdentifierType: + return false, TagOID, false, true + case bitStringType: + return false, TagBitString, false, true + case timeType: + return false, TagUTCTime, false, true + case enumeratedType: + return false, TagEnum, false, true + case bigIntType: + return false, TagInteger, false, true + } + switch t.Kind() { + case reflect.Bool: + return false, TagBoolean, false, true + case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64: + return false, TagInteger, false, true + case reflect.Struct: + return false, TagSequence, true, true + case reflect.Slice: + if t.Elem().Kind() == reflect.Uint8 { + return false, TagOctetString, false, true + } + if strings.HasSuffix(t.Name(), "SET") { + return false, TagSet, true, true + } + return false, TagSequence, true, true + case reflect.String: + return false, TagPrintableString, false, true + } + return false, 0, false, false +} diff --git a/contrib/go/_std_1.22/src/encoding/asn1/marshal.go b/contrib/go/_std_1.22/src/encoding/asn1/marshal.go new file mode 100644 index 0000000000..d8c8fe17b3 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/asn1/marshal.go @@ -0,0 +1,747 @@ +// Copyright 2009 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package asn1 + +import ( + "bytes" + "errors" + "fmt" + "math/big" + "reflect" + "sort" + "time" + "unicode/utf8" +) + +var ( + byte00Encoder encoder = byteEncoder(0x00) + byteFFEncoder encoder = byteEncoder(0xff) +) + +// encoder represents an ASN.1 element that is waiting to be marshaled. +type encoder interface { + // Len returns the number of bytes needed to marshal this element. + Len() int + // Encode encodes this element by writing Len() bytes to dst. + Encode(dst []byte) +} + +type byteEncoder byte + +func (c byteEncoder) Len() int { + return 1 +} + +func (c byteEncoder) Encode(dst []byte) { + dst[0] = byte(c) +} + +type bytesEncoder []byte + +func (b bytesEncoder) Len() int { + return len(b) +} + +func (b bytesEncoder) Encode(dst []byte) { + if copy(dst, b) != len(b) { + panic("internal error") + } +} + +type stringEncoder string + +func (s stringEncoder) Len() int { + return len(s) +} + +func (s stringEncoder) Encode(dst []byte) { + if copy(dst, s) != len(s) { + panic("internal error") + } +} + +type multiEncoder []encoder + +func (m multiEncoder) Len() int { + var size int + for _, e := range m { + size += e.Len() + } + return size +} + +func (m multiEncoder) Encode(dst []byte) { + var off int + for _, e := range m { + e.Encode(dst[off:]) + off += e.Len() + } +} + +type setEncoder []encoder + +func (s setEncoder) Len() int { + var size int + for _, e := range s { + size += e.Len() + } + return size +} + +func (s setEncoder) Encode(dst []byte) { + // Per X690 Section 11.6: The encodings of the component values of a + // set-of value shall appear in ascending order, the encodings being + // compared as octet strings with the shorter components being padded + // at their trailing end with 0-octets. + // + // First we encode each element to its TLV encoding and then use + // octetSort to get the ordering expected by X690 DER rules before + // writing the sorted encodings out to dst. + l := make([][]byte, len(s)) + for i, e := range s { + l[i] = make([]byte, e.Len()) + e.Encode(l[i]) + } + + sort.Slice(l, func(i, j int) bool { + // Since we are using bytes.Compare to compare TLV encodings we + // don't need to right pad s[i] and s[j] to the same length as + // suggested in X690. If len(s[i]) < len(s[j]) the length octet of + // s[i], which is the first determining byte, will inherently be + // smaller than the length octet of s[j]. This lets us skip the + // padding step. + return bytes.Compare(l[i], l[j]) < 0 + }) + + var off int + for _, b := range l { + copy(dst[off:], b) + off += len(b) + } +} + +type taggedEncoder struct { + // scratch contains temporary space for encoding the tag and length of + // an element in order to avoid extra allocations. + scratch [8]byte + tag encoder + body encoder +} + +func (t *taggedEncoder) Len() int { + return t.tag.Len() + t.body.Len() +} + +func (t *taggedEncoder) Encode(dst []byte) { + t.tag.Encode(dst) + t.body.Encode(dst[t.tag.Len():]) +} + +type int64Encoder int64 + +func (i int64Encoder) Len() int { + n := 1 + + for i > 127 { + n++ + i >>= 8 + } + + for i < -128 { + n++ + i >>= 8 + } + + return n +} + +func (i int64Encoder) Encode(dst []byte) { + n := i.Len() + + for j := 0; j < n; j++ { + dst[j] = byte(i >> uint((n-1-j)*8)) + } +} + +func base128IntLength(n int64) int { + if n == 0 { + return 1 + } + + l := 0 + for i := n; i > 0; i >>= 7 { + l++ + } + + return l +} + +func appendBase128Int(dst []byte, n int64) []byte { + l := base128IntLength(n) + + for i := l - 1; i >= 0; i-- { + o := byte(n >> uint(i*7)) + o &= 0x7f + if i != 0 { + o |= 0x80 + } + + dst = append(dst, o) + } + + return dst +} + +func makeBigInt(n *big.Int) (encoder, error) { + if n == nil { + return nil, StructuralError{"empty integer"} + } + + if n.Sign() < 0 { + // A negative number has to be converted to two's-complement + // form. So we'll invert and subtract 1. If the + // most-significant-bit isn't set then we'll need to pad the + // beginning with 0xff in order to keep the number negative. + nMinus1 := new(big.Int).Neg(n) + nMinus1.Sub(nMinus1, bigOne) + bytes := nMinus1.Bytes() + for i := range bytes { + bytes[i] ^= 0xff + } + if len(bytes) == 0 || bytes[0]&0x80 == 0 { + return multiEncoder([]encoder{byteFFEncoder, bytesEncoder(bytes)}), nil + } + return bytesEncoder(bytes), nil + } else if n.Sign() == 0 { + // Zero is written as a single 0 zero rather than no bytes. + return byte00Encoder, nil + } else { + bytes := n.Bytes() + if len(bytes) > 0 && bytes[0]&0x80 != 0 { + // We'll have to pad this with 0x00 in order to stop it + // looking like a negative number. + return multiEncoder([]encoder{byte00Encoder, bytesEncoder(bytes)}), nil + } + return bytesEncoder(bytes), nil + } +} + +func appendLength(dst []byte, i int) []byte { + n := lengthLength(i) + + for ; n > 0; n-- { + dst = append(dst, byte(i>>uint((n-1)*8))) + } + + return dst +} + +func lengthLength(i int) (numBytes int) { + numBytes = 1 + for i > 255 { + numBytes++ + i >>= 8 + } + return +} + +func appendTagAndLength(dst []byte, t tagAndLength) []byte { + b := uint8(t.class) << 6 + if t.isCompound { + b |= 0x20 + } + if t.tag >= 31 { + b |= 0x1f + dst = append(dst, b) + dst = appendBase128Int(dst, int64(t.tag)) + } else { + b |= uint8(t.tag) + dst = append(dst, b) + } + + if t.length >= 128 { + l := lengthLength(t.length) + dst = append(dst, 0x80|byte(l)) + dst = appendLength(dst, t.length) + } else { + dst = append(dst, byte(t.length)) + } + + return dst +} + +type bitStringEncoder BitString + +func (b bitStringEncoder) Len() int { + return len(b.Bytes) + 1 +} + +func (b bitStringEncoder) Encode(dst []byte) { + dst[0] = byte((8 - b.BitLength%8) % 8) + if copy(dst[1:], b.Bytes) != len(b.Bytes) { + panic("internal error") + } +} + +type oidEncoder []int + +func (oid oidEncoder) Len() int { + l := base128IntLength(int64(oid[0]*40 + oid[1])) + for i := 2; i < len(oid); i++ { + l += base128IntLength(int64(oid[i])) + } + return l +} + +func (oid oidEncoder) Encode(dst []byte) { + dst = appendBase128Int(dst[:0], int64(oid[0]*40+oid[1])) + for i := 2; i < len(oid); i++ { + dst = appendBase128Int(dst, int64(oid[i])) + } +} + +func makeObjectIdentifier(oid []int) (e encoder, err error) { + if len(oid) < 2 || oid[0] > 2 || (oid[0] < 2 && oid[1] >= 40) { + return nil, StructuralError{"invalid object identifier"} + } + + return oidEncoder(oid), nil +} + +func makePrintableString(s string) (e encoder, err error) { + for i := 0; i < len(s); i++ { + // The asterisk is often used in PrintableString, even though + // it is invalid. If a PrintableString was specifically + // requested then the asterisk is permitted by this code. + // Ampersand is allowed in parsing due a handful of CA + // certificates, however when making new certificates + // it is rejected. + if !isPrintable(s[i], allowAsterisk, rejectAmpersand) { + return nil, StructuralError{"PrintableString contains invalid character"} + } + } + + return stringEncoder(s), nil +} + +func makeIA5String(s string) (e encoder, err error) { + for i := 0; i < len(s); i++ { + if s[i] > 127 { + return nil, StructuralError{"IA5String contains invalid character"} + } + } + + return stringEncoder(s), nil +} + +func makeNumericString(s string) (e encoder, err error) { + for i := 0; i < len(s); i++ { + if !isNumeric(s[i]) { + return nil, StructuralError{"NumericString contains invalid character"} + } + } + + return stringEncoder(s), nil +} + +func makeUTF8String(s string) encoder { + return stringEncoder(s) +} + +func appendTwoDigits(dst []byte, v int) []byte { + return append(dst, byte('0'+(v/10)%10), byte('0'+v%10)) +} + +func appendFourDigits(dst []byte, v int) []byte { + var bytes [4]byte + for i := range bytes { + bytes[3-i] = '0' + byte(v%10) + v /= 10 + } + return append(dst, bytes[:]...) +} + +func outsideUTCRange(t time.Time) bool { + year := t.Year() + return year < 1950 || year >= 2050 +} + +func makeUTCTime(t time.Time) (e encoder, err error) { + dst := make([]byte, 0, 18) + + dst, err = appendUTCTime(dst, t) + if err != nil { + return nil, err + } + + return bytesEncoder(dst), nil +} + +func makeGeneralizedTime(t time.Time) (e encoder, err error) { + dst := make([]byte, 0, 20) + + dst, err = appendGeneralizedTime(dst, t) + if err != nil { + return nil, err + } + + return bytesEncoder(dst), nil +} + +func appendUTCTime(dst []byte, t time.Time) (ret []byte, err error) { + year := t.Year() + + switch { + case 1950 <= year && year < 2000: + dst = appendTwoDigits(dst, year-1900) + case 2000 <= year && year < 2050: + dst = appendTwoDigits(dst, year-2000) + default: + return nil, StructuralError{"cannot represent time as UTCTime"} + } + + return appendTimeCommon(dst, t), nil +} + +func appendGeneralizedTime(dst []byte, t time.Time) (ret []byte, err error) { + year := t.Year() + if year < 0 || year > 9999 { + return nil, StructuralError{"cannot represent time as GeneralizedTime"} + } + + dst = appendFourDigits(dst, year) + + return appendTimeCommon(dst, t), nil +} + +func appendTimeCommon(dst []byte, t time.Time) []byte { + _, month, day := t.Date() + + dst = appendTwoDigits(dst, int(month)) + dst = appendTwoDigits(dst, day) + + hour, min, sec := t.Clock() + + dst = appendTwoDigits(dst, hour) + dst = appendTwoDigits(dst, min) + dst = appendTwoDigits(dst, sec) + + _, offset := t.Zone() + + switch { + case offset/60 == 0: + return append(dst, 'Z') + case offset > 0: + dst = append(dst, '+') + case offset < 0: + dst = append(dst, '-') + } + + offsetMinutes := offset / 60 + if offsetMinutes < 0 { + offsetMinutes = -offsetMinutes + } + + dst = appendTwoDigits(dst, offsetMinutes/60) + dst = appendTwoDigits(dst, offsetMinutes%60) + + return dst +} + +func stripTagAndLength(in []byte) []byte { + _, offset, err := parseTagAndLength(in, 0) + if err != nil { + return in + } + return in[offset:] +} + +func makeBody(value reflect.Value, params fieldParameters) (e encoder, err error) { + switch value.Type() { + case flagType: + return bytesEncoder(nil), nil + case timeType: + t := value.Interface().(time.Time) + if params.timeType == TagGeneralizedTime || outsideUTCRange(t) { + return makeGeneralizedTime(t) + } + return makeUTCTime(t) + case bitStringType: + return bitStringEncoder(value.Interface().(BitString)), nil + case objectIdentifierType: + return makeObjectIdentifier(value.Interface().(ObjectIdentifier)) + case bigIntType: + return makeBigInt(value.Interface().(*big.Int)) + } + + switch v := value; v.Kind() { + case reflect.Bool: + if v.Bool() { + return byteFFEncoder, nil + } + return byte00Encoder, nil + case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64: + return int64Encoder(v.Int()), nil + case reflect.Struct: + t := v.Type() + + for i := 0; i < t.NumField(); i++ { + if !t.Field(i).IsExported() { + return nil, StructuralError{"struct contains unexported fields"} + } + } + + startingField := 0 + + n := t.NumField() + if n == 0 { + return bytesEncoder(nil), nil + } + + // If the first element of the structure is a non-empty + // RawContents, then we don't bother serializing the rest. + if t.Field(0).Type == rawContentsType { + s := v.Field(0) + if s.Len() > 0 { + bytes := s.Bytes() + /* The RawContents will contain the tag and + * length fields but we'll also be writing + * those ourselves, so we strip them out of + * bytes */ + return bytesEncoder(stripTagAndLength(bytes)), nil + } + + startingField = 1 + } + + switch n1 := n - startingField; n1 { + case 0: + return bytesEncoder(nil), nil + case 1: + return makeField(v.Field(startingField), parseFieldParameters(t.Field(startingField).Tag.Get("asn1"))) + default: + m := make([]encoder, n1) + for i := 0; i < n1; i++ { + m[i], err = makeField(v.Field(i+startingField), parseFieldParameters(t.Field(i+startingField).Tag.Get("asn1"))) + if err != nil { + return nil, err + } + } + + return multiEncoder(m), nil + } + case reflect.Slice: + sliceType := v.Type() + if sliceType.Elem().Kind() == reflect.Uint8 { + return bytesEncoder(v.Bytes()), nil + } + + var fp fieldParameters + + switch l := v.Len(); l { + case 0: + return bytesEncoder(nil), nil + case 1: + return makeField(v.Index(0), fp) + default: + m := make([]encoder, l) + + for i := 0; i < l; i++ { + m[i], err = makeField(v.Index(i), fp) + if err != nil { + return nil, err + } + } + + if params.set { + return setEncoder(m), nil + } + return multiEncoder(m), nil + } + case reflect.String: + switch params.stringType { + case TagIA5String: + return makeIA5String(v.String()) + case TagPrintableString: + return makePrintableString(v.String()) + case TagNumericString: + return makeNumericString(v.String()) + default: + return makeUTF8String(v.String()), nil + } + } + + return nil, StructuralError{"unknown Go type"} +} + +func makeField(v reflect.Value, params fieldParameters) (e encoder, err error) { + if !v.IsValid() { + return nil, fmt.Errorf("asn1: cannot marshal nil value") + } + // If the field is an interface{} then recurse into it. + if v.Kind() == reflect.Interface && v.Type().NumMethod() == 0 { + return makeField(v.Elem(), params) + } + + if v.Kind() == reflect.Slice && v.Len() == 0 && params.omitEmpty { + return bytesEncoder(nil), nil + } + + if params.optional && params.defaultValue != nil && canHaveDefaultValue(v.Kind()) { + defaultValue := reflect.New(v.Type()).Elem() + defaultValue.SetInt(*params.defaultValue) + + if reflect.DeepEqual(v.Interface(), defaultValue.Interface()) { + return bytesEncoder(nil), nil + } + } + + // If no default value is given then the zero value for the type is + // assumed to be the default value. This isn't obviously the correct + // behavior, but it's what Go has traditionally done. + if params.optional && params.defaultValue == nil { + if reflect.DeepEqual(v.Interface(), reflect.Zero(v.Type()).Interface()) { + return bytesEncoder(nil), nil + } + } + + if v.Type() == rawValueType { + rv := v.Interface().(RawValue) + if len(rv.FullBytes) != 0 { + return bytesEncoder(rv.FullBytes), nil + } + + t := new(taggedEncoder) + + t.tag = bytesEncoder(appendTagAndLength(t.scratch[:0], tagAndLength{rv.Class, rv.Tag, len(rv.Bytes), rv.IsCompound})) + t.body = bytesEncoder(rv.Bytes) + + return t, nil + } + + matchAny, tag, isCompound, ok := getUniversalType(v.Type()) + if !ok || matchAny { + return nil, StructuralError{fmt.Sprintf("unknown Go type: %v", v.Type())} + } + + if params.timeType != 0 && tag != TagUTCTime { + return nil, StructuralError{"explicit time type given to non-time member"} + } + + if params.stringType != 0 && tag != TagPrintableString { + return nil, StructuralError{"explicit string type given to non-string member"} + } + + switch tag { + case TagPrintableString: + if params.stringType == 0 { + // This is a string without an explicit string type. We'll use + // a PrintableString if the character set in the string is + // sufficiently limited, otherwise we'll use a UTF8String. + for _, r := range v.String() { + if r >= utf8.RuneSelf || !isPrintable(byte(r), rejectAsterisk, rejectAmpersand) { + if !utf8.ValidString(v.String()) { + return nil, errors.New("asn1: string not valid UTF-8") + } + tag = TagUTF8String + break + } + } + } else { + tag = params.stringType + } + case TagUTCTime: + if params.timeType == TagGeneralizedTime || outsideUTCRange(v.Interface().(time.Time)) { + tag = TagGeneralizedTime + } + } + + if params.set { + if tag != TagSequence { + return nil, StructuralError{"non sequence tagged as set"} + } + tag = TagSet + } + + // makeField can be called for a slice that should be treated as a SET + // but doesn't have params.set set, for instance when using a slice + // with the SET type name suffix. In this case getUniversalType returns + // TagSet, but makeBody doesn't know about that so will treat the slice + // as a sequence. To work around this we set params.set. + if tag == TagSet && !params.set { + params.set = true + } + + t := new(taggedEncoder) + + t.body, err = makeBody(v, params) + if err != nil { + return nil, err + } + + bodyLen := t.body.Len() + + class := ClassUniversal + if params.tag != nil { + if params.application { + class = ClassApplication + } else if params.private { + class = ClassPrivate + } else { + class = ClassContextSpecific + } + + if params.explicit { + t.tag = bytesEncoder(appendTagAndLength(t.scratch[:0], tagAndLength{ClassUniversal, tag, bodyLen, isCompound})) + + tt := new(taggedEncoder) + + tt.body = t + + tt.tag = bytesEncoder(appendTagAndLength(tt.scratch[:0], tagAndLength{ + class: class, + tag: *params.tag, + length: bodyLen + t.tag.Len(), + isCompound: true, + })) + + return tt, nil + } + + // implicit tag. + tag = *params.tag + } + + t.tag = bytesEncoder(appendTagAndLength(t.scratch[:0], tagAndLength{class, tag, bodyLen, isCompound})) + + return t, nil +} + +// Marshal returns the ASN.1 encoding of val. +// +// In addition to the struct tags recognized by Unmarshal, the following can be +// used: +// +// ia5: causes strings to be marshaled as ASN.1, IA5String values +// omitempty: causes empty slices to be skipped +// printable: causes strings to be marshaled as ASN.1, PrintableString values +// utf8: causes strings to be marshaled as ASN.1, UTF8String values +// utc: causes time.Time to be marshaled as ASN.1, UTCTime values +// generalized: causes time.Time to be marshaled as ASN.1, GeneralizedTime values +func Marshal(val any) ([]byte, error) { + return MarshalWithParams(val, "") +} + +// MarshalWithParams allows field parameters to be specified for the +// top-level element. The form of the params is the same as the field tags. +func MarshalWithParams(val any, params string) ([]byte, error) { + e, err := makeField(reflect.ValueOf(val), parseFieldParameters(params)) + if err != nil { + return nil, err + } + b := make([]byte, e.Len()) + e.Encode(b) + return b, nil +} diff --git a/contrib/go/_std_1.22/src/encoding/asn1/ya.make b/contrib/go/_std_1.22/src/encoding/asn1/ya.make new file mode 100644 index 0000000000..f9dc003f73 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/asn1/ya.make @@ -0,0 +1,9 @@ +GO_LIBRARY() +IF (TRUE) + SRCS( + asn1.go + common.go + marshal.go + ) +ENDIF() +END() diff --git a/contrib/go/_std_1.22/src/encoding/base64/base64.go b/contrib/go/_std_1.22/src/encoding/base64/base64.go new file mode 100644 index 0000000000..87f45863bd --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/base64/base64.go @@ -0,0 +1,661 @@ +// Copyright 2009 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +// Package base64 implements base64 encoding as specified by RFC 4648. +package base64 + +import ( + "encoding/binary" + "io" + "slices" + "strconv" +) + +/* + * Encodings + */ + +// An Encoding is a radix 64 encoding/decoding scheme, defined by a +// 64-character alphabet. The most common encoding is the "base64" +// encoding defined in RFC 4648 and used in MIME (RFC 2045) and PEM +// (RFC 1421). RFC 4648 also defines an alternate encoding, which is +// the standard encoding with - and _ substituted for + and /. +type Encoding struct { + encode [64]byte // mapping of symbol index to symbol byte value + decodeMap [256]uint8 // mapping of symbol byte value to symbol index + padChar rune + strict bool +} + +const ( + StdPadding rune = '=' // Standard padding character + NoPadding rune = -1 // No padding +) + +const ( + decodeMapInitialize = "" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + invalidIndex = '\xff' +) + +// NewEncoding returns a new padded Encoding defined by the given alphabet, +// which must be a 64-byte string that contains unique byte values and +// does not contain the padding character or CR / LF ('\r', '\n'). +// The alphabet is treated as a sequence of byte values +// without any special treatment for multi-byte UTF-8. +// The resulting Encoding uses the default padding character ('='), +// which may be changed or disabled via [Encoding.WithPadding]. +func NewEncoding(encoder string) *Encoding { + if len(encoder) != 64 { + panic("encoding alphabet is not 64-bytes long") + } + + e := new(Encoding) + e.padChar = StdPadding + copy(e.encode[:], encoder) + copy(e.decodeMap[:], decodeMapInitialize) + + for i := 0; i < len(encoder); i++ { + // Note: While we document that the alphabet cannot contain + // the padding character, we do not enforce it since we do not know + // if the caller intends to switch the padding from StdPadding later. + switch { + case encoder[i] == '\n' || encoder[i] == '\r': + panic("encoding alphabet contains newline character") + case e.decodeMap[encoder[i]] != invalidIndex: + panic("encoding alphabet includes duplicate symbols") + } + e.decodeMap[encoder[i]] = uint8(i) + } + return e +} + +// WithPadding creates a new encoding identical to enc except +// with a specified padding character, or [NoPadding] to disable padding. +// The padding character must not be '\r' or '\n', +// must not be contained in the encoding's alphabet, +// must not be negative, and must be a rune equal or below '\xff'. +// Padding characters above '\x7f' are encoded as their exact byte value +// rather than using the UTF-8 representation of the codepoint. +func (enc Encoding) WithPadding(padding rune) *Encoding { + switch { + case padding < NoPadding || padding == '\r' || padding == '\n' || padding > 0xff: + panic("invalid padding") + case padding != NoPadding && enc.decodeMap[byte(padding)] != invalidIndex: + panic("padding contained in alphabet") + } + enc.padChar = padding + return &enc +} + +// Strict creates a new encoding identical to enc except with +// strict decoding enabled. In this mode, the decoder requires that +// trailing padding bits are zero, as described in RFC 4648 section 3.5. +// +// Note that the input is still malleable, as new line characters +// (CR and LF) are still ignored. +func (enc Encoding) Strict() *Encoding { + enc.strict = true + return &enc +} + +// StdEncoding is the standard base64 encoding, as defined in RFC 4648. +var StdEncoding = NewEncoding("ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/") + +// URLEncoding is the alternate base64 encoding defined in RFC 4648. +// It is typically used in URLs and file names. +var URLEncoding = NewEncoding("ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-_") + +// RawStdEncoding is the standard raw, unpadded base64 encoding, +// as defined in RFC 4648 section 3.2. +// This is the same as [StdEncoding] but omits padding characters. +var RawStdEncoding = StdEncoding.WithPadding(NoPadding) + +// RawURLEncoding is the unpadded alternate base64 encoding defined in RFC 4648. +// It is typically used in URLs and file names. +// This is the same as [URLEncoding] but omits padding characters. +var RawURLEncoding = URLEncoding.WithPadding(NoPadding) + +/* + * Encoder + */ + +// Encode encodes src using the encoding enc, +// writing [Encoding.EncodedLen](len(src)) bytes to dst. +// +// The encoding pads the output to a multiple of 4 bytes, +// so Encode is not appropriate for use on individual blocks +// of a large data stream. Use [NewEncoder] instead. +func (enc *Encoding) Encode(dst, src []byte) { + if len(src) == 0 { + return + } + // enc is a pointer receiver, so the use of enc.encode within the hot + // loop below means a nil check at every operation. Lift that nil check + // outside of the loop to speed up the encoder. + _ = enc.encode + + di, si := 0, 0 + n := (len(src) / 3) * 3 + for si < n { + // Convert 3x 8bit source bytes into 4 bytes + val := uint(src[si+0])<<16 | uint(src[si+1])<<8 | uint(src[si+2]) + + dst[di+0] = enc.encode[val>>18&0x3F] + dst[di+1] = enc.encode[val>>12&0x3F] + dst[di+2] = enc.encode[val>>6&0x3F] + dst[di+3] = enc.encode[val&0x3F] + + si += 3 + di += 4 + } + + remain := len(src) - si + if remain == 0 { + return + } + // Add the remaining small block + val := uint(src[si+0]) << 16 + if remain == 2 { + val |= uint(src[si+1]) << 8 + } + + dst[di+0] = enc.encode[val>>18&0x3F] + dst[di+1] = enc.encode[val>>12&0x3F] + + switch remain { + case 2: + dst[di+2] = enc.encode[val>>6&0x3F] + if enc.padChar != NoPadding { + dst[di+3] = byte(enc.padChar) + } + case 1: + if enc.padChar != NoPadding { + dst[di+2] = byte(enc.padChar) + dst[di+3] = byte(enc.padChar) + } + } +} + +// AppendEncode appends the base64 encoded src to dst +// and returns the extended buffer. +func (enc *Encoding) AppendEncode(dst, src []byte) []byte { + n := enc.EncodedLen(len(src)) + dst = slices.Grow(dst, n) + enc.Encode(dst[len(dst):][:n], src) + return dst[:len(dst)+n] +} + +// EncodeToString returns the base64 encoding of src. +func (enc *Encoding) EncodeToString(src []byte) string { + buf := make([]byte, enc.EncodedLen(len(src))) + enc.Encode(buf, src) + return string(buf) +} + +type encoder struct { + err error + enc *Encoding + w io.Writer + buf [3]byte // buffered data waiting to be encoded + nbuf int // number of bytes in buf + out [1024]byte // output buffer +} + +func (e *encoder) Write(p []byte) (n int, err error) { + if e.err != nil { + return 0, e.err + } + + // Leading fringe. + if e.nbuf > 0 { + var i int + for i = 0; i < len(p) && e.nbuf < 3; i++ { + e.buf[e.nbuf] = p[i] + e.nbuf++ + } + n += i + p = p[i:] + if e.nbuf < 3 { + return + } + e.enc.Encode(e.out[:], e.buf[:]) + if _, e.err = e.w.Write(e.out[:4]); e.err != nil { + return n, e.err + } + e.nbuf = 0 + } + + // Large interior chunks. + for len(p) >= 3 { + nn := len(e.out) / 4 * 3 + if nn > len(p) { + nn = len(p) + nn -= nn % 3 + } + e.enc.Encode(e.out[:], p[:nn]) + if _, e.err = e.w.Write(e.out[0 : nn/3*4]); e.err != nil { + return n, e.err + } + n += nn + p = p[nn:] + } + + // Trailing fringe. + copy(e.buf[:], p) + e.nbuf = len(p) + n += len(p) + return +} + +// Close flushes any pending output from the encoder. +// It is an error to call Write after calling Close. +func (e *encoder) Close() error { + // If there's anything left in the buffer, flush it out + if e.err == nil && e.nbuf > 0 { + e.enc.Encode(e.out[:], e.buf[:e.nbuf]) + _, e.err = e.w.Write(e.out[:e.enc.EncodedLen(e.nbuf)]) + e.nbuf = 0 + } + return e.err +} + +// NewEncoder returns a new base64 stream encoder. Data written to +// the returned writer will be encoded using enc and then written to w. +// Base64 encodings operate in 4-byte blocks; when finished +// writing, the caller must Close the returned encoder to flush any +// partially written blocks. +func NewEncoder(enc *Encoding, w io.Writer) io.WriteCloser { + return &encoder{enc: enc, w: w} +} + +// EncodedLen returns the length in bytes of the base64 encoding +// of an input buffer of length n. +func (enc *Encoding) EncodedLen(n int) int { + if enc.padChar == NoPadding { + return n/3*4 + (n%3*8+5)/6 // minimum # chars at 6 bits per char + } + return (n + 2) / 3 * 4 // minimum # 4-char quanta, 3 bytes each +} + +/* + * Decoder + */ + +type CorruptInputError int64 + +func (e CorruptInputError) Error() string { + return "illegal base64 data at input byte " + strconv.FormatInt(int64(e), 10) +} + +// decodeQuantum decodes up to 4 base64 bytes. The received parameters are +// the destination buffer dst, the source buffer src and an index in the +// source buffer si. +// It returns the number of bytes read from src, the number of bytes written +// to dst, and an error, if any. +func (enc *Encoding) decodeQuantum(dst, src []byte, si int) (nsi, n int, err error) { + // Decode quantum using the base64 alphabet + var dbuf [4]byte + dlen := 4 + + // Lift the nil check outside of the loop. + _ = enc.decodeMap + + for j := 0; j < len(dbuf); j++ { + if len(src) == si { + switch { + case j == 0: + return si, 0, nil + case j == 1, enc.padChar != NoPadding: + return si, 0, CorruptInputError(si - j) + } + dlen = j + break + } + in := src[si] + si++ + + out := enc.decodeMap[in] + if out != 0xff { + dbuf[j] = out + continue + } + + if in == '\n' || in == '\r' { + j-- + continue + } + + if rune(in) != enc.padChar { + return si, 0, CorruptInputError(si - 1) + } + + // We've reached the end and there's padding + switch j { + case 0, 1: + // incorrect padding + return si, 0, CorruptInputError(si - 1) + case 2: + // "==" is expected, the first "=" is already consumed. + // skip over newlines + for si < len(src) && (src[si] == '\n' || src[si] == '\r') { + si++ + } + if si == len(src) { + // not enough padding + return si, 0, CorruptInputError(len(src)) + } + if rune(src[si]) != enc.padChar { + // incorrect padding + return si, 0, CorruptInputError(si - 1) + } + + si++ + } + + // skip over newlines + for si < len(src) && (src[si] == '\n' || src[si] == '\r') { + si++ + } + if si < len(src) { + // trailing garbage + err = CorruptInputError(si) + } + dlen = j + break + } + + // Convert 4x 6bit source bytes into 3 bytes + val := uint(dbuf[0])<<18 | uint(dbuf[1])<<12 | uint(dbuf[2])<<6 | uint(dbuf[3]) + dbuf[2], dbuf[1], dbuf[0] = byte(val>>0), byte(val>>8), byte(val>>16) + switch dlen { + case 4: + dst[2] = dbuf[2] + dbuf[2] = 0 + fallthrough + case 3: + dst[1] = dbuf[1] + if enc.strict && dbuf[2] != 0 { + return si, 0, CorruptInputError(si - 1) + } + dbuf[1] = 0 + fallthrough + case 2: + dst[0] = dbuf[0] + if enc.strict && (dbuf[1] != 0 || dbuf[2] != 0) { + return si, 0, CorruptInputError(si - 2) + } + } + + return si, dlen - 1, err +} + +// AppendDecode appends the base64 decoded src to dst +// and returns the extended buffer. +// If the input is malformed, it returns the partially decoded src and an error. +func (enc *Encoding) AppendDecode(dst, src []byte) ([]byte, error) { + // Compute the output size without padding to avoid over allocating. + n := len(src) + for n > 0 && rune(src[n-1]) == enc.padChar { + n-- + } + n = decodedLen(n, NoPadding) + + dst = slices.Grow(dst, n) + n, err := enc.Decode(dst[len(dst):][:n], src) + return dst[:len(dst)+n], err +} + +// DecodeString returns the bytes represented by the base64 string s. +func (enc *Encoding) DecodeString(s string) ([]byte, error) { + dbuf := make([]byte, enc.DecodedLen(len(s))) + n, err := enc.Decode(dbuf, []byte(s)) + return dbuf[:n], err +} + +type decoder struct { + err error + readErr error // error from r.Read + enc *Encoding + r io.Reader + buf [1024]byte // leftover input + nbuf int + out []byte // leftover decoded output + outbuf [1024 / 4 * 3]byte +} + +func (d *decoder) Read(p []byte) (n int, err error) { + // Use leftover decoded output from last read. + if len(d.out) > 0 { + n = copy(p, d.out) + d.out = d.out[n:] + return n, nil + } + + if d.err != nil { + return 0, d.err + } + + // This code assumes that d.r strips supported whitespace ('\r' and '\n'). + + // Refill buffer. + for d.nbuf < 4 && d.readErr == nil { + nn := len(p) / 3 * 4 + if nn < 4 { + nn = 4 + } + if nn > len(d.buf) { + nn = len(d.buf) + } + nn, d.readErr = d.r.Read(d.buf[d.nbuf:nn]) + d.nbuf += nn + } + + if d.nbuf < 4 { + if d.enc.padChar == NoPadding && d.nbuf > 0 { + // Decode final fragment, without padding. + var nw int + nw, d.err = d.enc.Decode(d.outbuf[:], d.buf[:d.nbuf]) + d.nbuf = 0 + d.out = d.outbuf[:nw] + n = copy(p, d.out) + d.out = d.out[n:] + if n > 0 || len(p) == 0 && len(d.out) > 0 { + return n, nil + } + if d.err != nil { + return 0, d.err + } + } + d.err = d.readErr + if d.err == io.EOF && d.nbuf > 0 { + d.err = io.ErrUnexpectedEOF + } + return 0, d.err + } + + // Decode chunk into p, or d.out and then p if p is too small. + nr := d.nbuf / 4 * 4 + nw := d.nbuf / 4 * 3 + if nw > len(p) { + nw, d.err = d.enc.Decode(d.outbuf[:], d.buf[:nr]) + d.out = d.outbuf[:nw] + n = copy(p, d.out) + d.out = d.out[n:] + } else { + n, d.err = d.enc.Decode(p, d.buf[:nr]) + } + d.nbuf -= nr + copy(d.buf[:d.nbuf], d.buf[nr:]) + return n, d.err +} + +// Decode decodes src using the encoding enc. It writes at most +// [Encoding.DecodedLen](len(src)) bytes to dst and returns the number of bytes +// written. If src contains invalid base64 data, it will return the +// number of bytes successfully written and [CorruptInputError]. +// New line characters (\r and \n) are ignored. +func (enc *Encoding) Decode(dst, src []byte) (n int, err error) { + if len(src) == 0 { + return 0, nil + } + + // Lift the nil check outside of the loop. enc.decodeMap is directly + // used later in this function, to let the compiler know that the + // receiver can't be nil. + _ = enc.decodeMap + + si := 0 + for strconv.IntSize >= 64 && len(src)-si >= 8 && len(dst)-n >= 8 { + src2 := src[si : si+8] + if dn, ok := assemble64( + enc.decodeMap[src2[0]], + enc.decodeMap[src2[1]], + enc.decodeMap[src2[2]], + enc.decodeMap[src2[3]], + enc.decodeMap[src2[4]], + enc.decodeMap[src2[5]], + enc.decodeMap[src2[6]], + enc.decodeMap[src2[7]], + ); ok { + binary.BigEndian.PutUint64(dst[n:], dn) + n += 6 + si += 8 + } else { + var ninc int + si, ninc, err = enc.decodeQuantum(dst[n:], src, si) + n += ninc + if err != nil { + return n, err + } + } + } + + for len(src)-si >= 4 && len(dst)-n >= 4 { + src2 := src[si : si+4] + if dn, ok := assemble32( + enc.decodeMap[src2[0]], + enc.decodeMap[src2[1]], + enc.decodeMap[src2[2]], + enc.decodeMap[src2[3]], + ); ok { + binary.BigEndian.PutUint32(dst[n:], dn) + n += 3 + si += 4 + } else { + var ninc int + si, ninc, err = enc.decodeQuantum(dst[n:], src, si) + n += ninc + if err != nil { + return n, err + } + } + } + + for si < len(src) { + var ninc int + si, ninc, err = enc.decodeQuantum(dst[n:], src, si) + n += ninc + if err != nil { + return n, err + } + } + return n, err +} + +// assemble32 assembles 4 base64 digits into 3 bytes. +// Each digit comes from the decode map, and will be 0xff +// if it came from an invalid character. +func assemble32(n1, n2, n3, n4 byte) (dn uint32, ok bool) { + // Check that all the digits are valid. If any of them was 0xff, their + // bitwise OR will be 0xff. + if n1|n2|n3|n4 == 0xff { + return 0, false + } + return uint32(n1)<<26 | + uint32(n2)<<20 | + uint32(n3)<<14 | + uint32(n4)<<8, + true +} + +// assemble64 assembles 8 base64 digits into 6 bytes. +// Each digit comes from the decode map, and will be 0xff +// if it came from an invalid character. +func assemble64(n1, n2, n3, n4, n5, n6, n7, n8 byte) (dn uint64, ok bool) { + // Check that all the digits are valid. If any of them was 0xff, their + // bitwise OR will be 0xff. + if n1|n2|n3|n4|n5|n6|n7|n8 == 0xff { + return 0, false + } + return uint64(n1)<<58 | + uint64(n2)<<52 | + uint64(n3)<<46 | + uint64(n4)<<40 | + uint64(n5)<<34 | + uint64(n6)<<28 | + uint64(n7)<<22 | + uint64(n8)<<16, + true +} + +type newlineFilteringReader struct { + wrapped io.Reader +} + +func (r *newlineFilteringReader) Read(p []byte) (int, error) { + n, err := r.wrapped.Read(p) + for n > 0 { + offset := 0 + for i, b := range p[:n] { + if b != '\r' && b != '\n' { + if i != offset { + p[offset] = b + } + offset++ + } + } + if offset > 0 { + return offset, err + } + // Previous buffer entirely whitespace, read again + n, err = r.wrapped.Read(p) + } + return n, err +} + +// NewDecoder constructs a new base64 stream decoder. +func NewDecoder(enc *Encoding, r io.Reader) io.Reader { + return &decoder{enc: enc, r: &newlineFilteringReader{r}} +} + +// DecodedLen returns the maximum length in bytes of the decoded data +// corresponding to n bytes of base64-encoded data. +func (enc *Encoding) DecodedLen(n int) int { + return decodedLen(n, enc.padChar) +} + +func decodedLen(n int, padChar rune) int { + if padChar == NoPadding { + // Unpadded data may end with partial block of 2-3 characters. + return n/4*3 + n%4*6/8 + } + // Padded base64 should always be a multiple of 4 characters in length. + return n / 4 * 3 +} diff --git a/contrib/go/_std_1.22/src/encoding/base64/ya.make b/contrib/go/_std_1.22/src/encoding/base64/ya.make new file mode 100644 index 0000000000..282e7a5a03 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/base64/ya.make @@ -0,0 +1,7 @@ +GO_LIBRARY() +IF (TRUE) + SRCS( + base64.go + ) +ENDIF() +END() diff --git a/contrib/go/_std_1.22/src/encoding/binary/binary.go b/contrib/go/_std_1.22/src/encoding/binary/binary.go new file mode 100644 index 0000000000..f001be8386 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/binary/binary.go @@ -0,0 +1,815 @@ +// Copyright 2009 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +// Package binary implements simple translation between numbers and byte +// sequences and encoding and decoding of varints. +// +// Numbers are translated by reading and writing fixed-size values. +// A fixed-size value is either a fixed-size arithmetic +// type (bool, int8, uint8, int16, float32, complex64, ...) +// or an array or struct containing only fixed-size values. +// +// The varint functions encode and decode single integer values using +// a variable-length encoding; smaller values require fewer bytes. +// For a specification, see +// https://developers.google.com/protocol-buffers/docs/encoding. +// +// This package favors simplicity over efficiency. Clients that require +// high-performance serialization, especially for large data structures, +// should look at more advanced solutions such as the [encoding/gob] +// package or [google.golang.org/protobuf] for protocol buffers. +package binary + +import ( + "errors" + "io" + "math" + "reflect" + "sync" +) + +// A ByteOrder specifies how to convert byte slices into +// 16-, 32-, or 64-bit unsigned integers. +// +// It is implemented by [LittleEndian], [BigEndian], and [NativeEndian]. +type ByteOrder interface { + Uint16([]byte) uint16 + Uint32([]byte) uint32 + Uint64([]byte) uint64 + PutUint16([]byte, uint16) + PutUint32([]byte, uint32) + PutUint64([]byte, uint64) + String() string +} + +// AppendByteOrder specifies how to append 16-, 32-, or 64-bit unsigned integers +// into a byte slice. +// +// It is implemented by [LittleEndian], [BigEndian], and [NativeEndian]. +type AppendByteOrder interface { + AppendUint16([]byte, uint16) []byte + AppendUint32([]byte, uint32) []byte + AppendUint64([]byte, uint64) []byte + String() string +} + +// LittleEndian is the little-endian implementation of [ByteOrder] and [AppendByteOrder]. +var LittleEndian littleEndian + +// BigEndian is the big-endian implementation of [ByteOrder] and [AppendByteOrder]. +var BigEndian bigEndian + +type littleEndian struct{} + +func (littleEndian) Uint16(b []byte) uint16 { + _ = b[1] // bounds check hint to compiler; see golang.org/issue/14808 + return uint16(b[0]) | uint16(b[1])<<8 +} + +func (littleEndian) PutUint16(b []byte, v uint16) { + _ = b[1] // early bounds check to guarantee safety of writes below + b[0] = byte(v) + b[1] = byte(v >> 8) +} + +func (littleEndian) AppendUint16(b []byte, v uint16) []byte { + return append(b, + byte(v), + byte(v>>8), + ) +} + +func (littleEndian) Uint32(b []byte) uint32 { + _ = b[3] // bounds check hint to compiler; see golang.org/issue/14808 + return uint32(b[0]) | uint32(b[1])<<8 | uint32(b[2])<<16 | uint32(b[3])<<24 +} + +func (littleEndian) PutUint32(b []byte, v uint32) { + _ = b[3] // early bounds check to guarantee safety of writes below + b[0] = byte(v) + b[1] = byte(v >> 8) + b[2] = byte(v >> 16) + b[3] = byte(v >> 24) +} + +func (littleEndian) AppendUint32(b []byte, v uint32) []byte { + return append(b, + byte(v), + byte(v>>8), + byte(v>>16), + byte(v>>24), + ) +} + +func (littleEndian) Uint64(b []byte) uint64 { + _ = b[7] // bounds check hint to compiler; see golang.org/issue/14808 + return uint64(b[0]) | uint64(b[1])<<8 | uint64(b[2])<<16 | uint64(b[3])<<24 | + uint64(b[4])<<32 | uint64(b[5])<<40 | uint64(b[6])<<48 | uint64(b[7])<<56 +} + +func (littleEndian) PutUint64(b []byte, v uint64) { + _ = b[7] // early bounds check to guarantee safety of writes below + b[0] = byte(v) + b[1] = byte(v >> 8) + b[2] = byte(v >> 16) + b[3] = byte(v >> 24) + b[4] = byte(v >> 32) + b[5] = byte(v >> 40) + b[6] = byte(v >> 48) + b[7] = byte(v >> 56) +} + +func (littleEndian) AppendUint64(b []byte, v uint64) []byte { + return append(b, + byte(v), + byte(v>>8), + byte(v>>16), + byte(v>>24), + byte(v>>32), + byte(v>>40), + byte(v>>48), + byte(v>>56), + ) +} + +func (littleEndian) String() string { return "LittleEndian" } + +func (littleEndian) GoString() string { return "binary.LittleEndian" } + +type bigEndian struct{} + +func (bigEndian) Uint16(b []byte) uint16 { + _ = b[1] // bounds check hint to compiler; see golang.org/issue/14808 + return uint16(b[1]) | uint16(b[0])<<8 +} + +func (bigEndian) PutUint16(b []byte, v uint16) { + _ = b[1] // early bounds check to guarantee safety of writes below + b[0] = byte(v >> 8) + b[1] = byte(v) +} + +func (bigEndian) AppendUint16(b []byte, v uint16) []byte { + return append(b, + byte(v>>8), + byte(v), + ) +} + +func (bigEndian) Uint32(b []byte) uint32 { + _ = b[3] // bounds check hint to compiler; see golang.org/issue/14808 + return uint32(b[3]) | uint32(b[2])<<8 | uint32(b[1])<<16 | uint32(b[0])<<24 +} + +func (bigEndian) PutUint32(b []byte, v uint32) { + _ = b[3] // early bounds check to guarantee safety of writes below + b[0] = byte(v >> 24) + b[1] = byte(v >> 16) + b[2] = byte(v >> 8) + b[3] = byte(v) +} + +func (bigEndian) AppendUint32(b []byte, v uint32) []byte { + return append(b, + byte(v>>24), + byte(v>>16), + byte(v>>8), + byte(v), + ) +} + +func (bigEndian) Uint64(b []byte) uint64 { + _ = b[7] // bounds check hint to compiler; see golang.org/issue/14808 + return uint64(b[7]) | uint64(b[6])<<8 | uint64(b[5])<<16 | uint64(b[4])<<24 | + uint64(b[3])<<32 | uint64(b[2])<<40 | uint64(b[1])<<48 | uint64(b[0])<<56 +} + +func (bigEndian) PutUint64(b []byte, v uint64) { + _ = b[7] // early bounds check to guarantee safety of writes below + b[0] = byte(v >> 56) + b[1] = byte(v >> 48) + b[2] = byte(v >> 40) + b[3] = byte(v >> 32) + b[4] = byte(v >> 24) + b[5] = byte(v >> 16) + b[6] = byte(v >> 8) + b[7] = byte(v) +} + +func (bigEndian) AppendUint64(b []byte, v uint64) []byte { + return append(b, + byte(v>>56), + byte(v>>48), + byte(v>>40), + byte(v>>32), + byte(v>>24), + byte(v>>16), + byte(v>>8), + byte(v), + ) +} + +func (bigEndian) String() string { return "BigEndian" } + +func (bigEndian) GoString() string { return "binary.BigEndian" } + +func (nativeEndian) String() string { return "NativeEndian" } + +func (nativeEndian) GoString() string { return "binary.NativeEndian" } + +// Read reads structured binary data from r into data. +// Data must be a pointer to a fixed-size value or a slice +// of fixed-size values. +// Bytes read from r are decoded using the specified byte order +// and written to successive fields of the data. +// When decoding boolean values, a zero byte is decoded as false, and +// any other non-zero byte is decoded as true. +// When reading into structs, the field data for fields with +// blank (_) field names is skipped; i.e., blank field names +// may be used for padding. +// When reading into a struct, all non-blank fields must be exported +// or Read may panic. +// +// The error is [io.EOF] only if no bytes were read. +// If an [io.EOF] happens after reading some but not all the bytes, +// Read returns [io.ErrUnexpectedEOF]. +func Read(r io.Reader, order ByteOrder, data any) error { + // Fast path for basic types and slices. + if n := intDataSize(data); n != 0 { + bs := make([]byte, n) + if _, err := io.ReadFull(r, bs); err != nil { + return err + } + switch data := data.(type) { + case *bool: + *data = bs[0] != 0 + case *int8: + *data = int8(bs[0]) + case *uint8: + *data = bs[0] + case *int16: + *data = int16(order.Uint16(bs)) + case *uint16: + *data = order.Uint16(bs) + case *int32: + *data = int32(order.Uint32(bs)) + case *uint32: + *data = order.Uint32(bs) + case *int64: + *data = int64(order.Uint64(bs)) + case *uint64: + *data = order.Uint64(bs) + case *float32: + *data = math.Float32frombits(order.Uint32(bs)) + case *float64: + *data = math.Float64frombits(order.Uint64(bs)) + case []bool: + for i, x := range bs { // Easier to loop over the input for 8-bit values. + data[i] = x != 0 + } + case []int8: + for i, x := range bs { + data[i] = int8(x) + } + case []uint8: + copy(data, bs) + case []int16: + for i := range data { + data[i] = int16(order.Uint16(bs[2*i:])) + } + case []uint16: + for i := range data { + data[i] = order.Uint16(bs[2*i:]) + } + case []int32: + for i := range data { + data[i] = int32(order.Uint32(bs[4*i:])) + } + case []uint32: + for i := range data { + data[i] = order.Uint32(bs[4*i:]) + } + case []int64: + for i := range data { + data[i] = int64(order.Uint64(bs[8*i:])) + } + case []uint64: + for i := range data { + data[i] = order.Uint64(bs[8*i:]) + } + case []float32: + for i := range data { + data[i] = math.Float32frombits(order.Uint32(bs[4*i:])) + } + case []float64: + for i := range data { + data[i] = math.Float64frombits(order.Uint64(bs[8*i:])) + } + default: + n = 0 // fast path doesn't apply + } + if n != 0 { + return nil + } + } + + // Fallback to reflect-based decoding. + v := reflect.ValueOf(data) + size := -1 + switch v.Kind() { + case reflect.Pointer: + v = v.Elem() + size = dataSize(v) + case reflect.Slice: + size = dataSize(v) + } + if size < 0 { + return errors.New("binary.Read: invalid type " + reflect.TypeOf(data).String()) + } + d := &decoder{order: order, buf: make([]byte, size)} + if _, err := io.ReadFull(r, d.buf); err != nil { + return err + } + d.value(v) + return nil +} + +// Write writes the binary representation of data into w. +// Data must be a fixed-size value or a slice of fixed-size +// values, or a pointer to such data. +// Boolean values encode as one byte: 1 for true, and 0 for false. +// Bytes written to w are encoded using the specified byte order +// and read from successive fields of the data. +// When writing structs, zero values are written for fields +// with blank (_) field names. +func Write(w io.Writer, order ByteOrder, data any) error { + // Fast path for basic types and slices. + if n := intDataSize(data); n != 0 { + bs := make([]byte, n) + switch v := data.(type) { + case *bool: + if *v { + bs[0] = 1 + } else { + bs[0] = 0 + } + case bool: + if v { + bs[0] = 1 + } else { + bs[0] = 0 + } + case []bool: + for i, x := range v { + if x { + bs[i] = 1 + } else { + bs[i] = 0 + } + } + case *int8: + bs[0] = byte(*v) + case int8: + bs[0] = byte(v) + case []int8: + for i, x := range v { + bs[i] = byte(x) + } + case *uint8: + bs[0] = *v + case uint8: + bs[0] = v + case []uint8: + bs = v + case *int16: + order.PutUint16(bs, uint16(*v)) + case int16: + order.PutUint16(bs, uint16(v)) + case []int16: + for i, x := range v { + order.PutUint16(bs[2*i:], uint16(x)) + } + case *uint16: + order.PutUint16(bs, *v) + case uint16: + order.PutUint16(bs, v) + case []uint16: + for i, x := range v { + order.PutUint16(bs[2*i:], x) + } + case *int32: + order.PutUint32(bs, uint32(*v)) + case int32: + order.PutUint32(bs, uint32(v)) + case []int32: + for i, x := range v { + order.PutUint32(bs[4*i:], uint32(x)) + } + case *uint32: + order.PutUint32(bs, *v) + case uint32: + order.PutUint32(bs, v) + case []uint32: + for i, x := range v { + order.PutUint32(bs[4*i:], x) + } + case *int64: + order.PutUint64(bs, uint64(*v)) + case int64: + order.PutUint64(bs, uint64(v)) + case []int64: + for i, x := range v { + order.PutUint64(bs[8*i:], uint64(x)) + } + case *uint64: + order.PutUint64(bs, *v) + case uint64: + order.PutUint64(bs, v) + case []uint64: + for i, x := range v { + order.PutUint64(bs[8*i:], x) + } + case *float32: + order.PutUint32(bs, math.Float32bits(*v)) + case float32: + order.PutUint32(bs, math.Float32bits(v)) + case []float32: + for i, x := range v { + order.PutUint32(bs[4*i:], math.Float32bits(x)) + } + case *float64: + order.PutUint64(bs, math.Float64bits(*v)) + case float64: + order.PutUint64(bs, math.Float64bits(v)) + case []float64: + for i, x := range v { + order.PutUint64(bs[8*i:], math.Float64bits(x)) + } + } + _, err := w.Write(bs) + return err + } + + // Fallback to reflect-based encoding. + v := reflect.Indirect(reflect.ValueOf(data)) + size := dataSize(v) + if size < 0 { + return errors.New("binary.Write: some values are not fixed-sized in type " + reflect.TypeOf(data).String()) + } + buf := make([]byte, size) + e := &encoder{order: order, buf: buf} + e.value(v) + _, err := w.Write(buf) + return err +} + +// Size returns how many bytes [Write] would generate to encode the value v, which +// must be a fixed-size value or a slice of fixed-size values, or a pointer to such data. +// If v is neither of these, Size returns -1. +func Size(v any) int { + return dataSize(reflect.Indirect(reflect.ValueOf(v))) +} + +var structSize sync.Map // map[reflect.Type]int + +// dataSize returns the number of bytes the actual data represented by v occupies in memory. +// For compound structures, it sums the sizes of the elements. Thus, for instance, for a slice +// it returns the length of the slice times the element size and does not count the memory +// occupied by the header. If the type of v is not acceptable, dataSize returns -1. +func dataSize(v reflect.Value) int { + switch v.Kind() { + case reflect.Slice: + if s := sizeof(v.Type().Elem()); s >= 0 { + return s * v.Len() + } + + case reflect.Struct: + t := v.Type() + if size, ok := structSize.Load(t); ok { + return size.(int) + } + size := sizeof(t) + structSize.Store(t, size) + return size + + default: + if v.IsValid() { + return sizeof(v.Type()) + } + } + + return -1 +} + +// sizeof returns the size >= 0 of variables for the given type or -1 if the type is not acceptable. +func sizeof(t reflect.Type) int { + switch t.Kind() { + case reflect.Array: + if s := sizeof(t.Elem()); s >= 0 { + return s * t.Len() + } + + case reflect.Struct: + sum := 0 + for i, n := 0, t.NumField(); i < n; i++ { + s := sizeof(t.Field(i).Type) + if s < 0 { + return -1 + } + sum += s + } + return sum + + case reflect.Bool, + reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, + reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64, + reflect.Float32, reflect.Float64, reflect.Complex64, reflect.Complex128: + return int(t.Size()) + } + + return -1 +} + +type coder struct { + order ByteOrder + buf []byte + offset int +} + +type decoder coder +type encoder coder + +func (d *decoder) bool() bool { + x := d.buf[d.offset] + d.offset++ + return x != 0 +} + +func (e *encoder) bool(x bool) { + if x { + e.buf[e.offset] = 1 + } else { + e.buf[e.offset] = 0 + } + e.offset++ +} + +func (d *decoder) uint8() uint8 { + x := d.buf[d.offset] + d.offset++ + return x +} + +func (e *encoder) uint8(x uint8) { + e.buf[e.offset] = x + e.offset++ +} + +func (d *decoder) uint16() uint16 { + x := d.order.Uint16(d.buf[d.offset : d.offset+2]) + d.offset += 2 + return x +} + +func (e *encoder) uint16(x uint16) { + e.order.PutUint16(e.buf[e.offset:e.offset+2], x) + e.offset += 2 +} + +func (d *decoder) uint32() uint32 { + x := d.order.Uint32(d.buf[d.offset : d.offset+4]) + d.offset += 4 + return x +} + +func (e *encoder) uint32(x uint32) { + e.order.PutUint32(e.buf[e.offset:e.offset+4], x) + e.offset += 4 +} + +func (d *decoder) uint64() uint64 { + x := d.order.Uint64(d.buf[d.offset : d.offset+8]) + d.offset += 8 + return x +} + +func (e *encoder) uint64(x uint64) { + e.order.PutUint64(e.buf[e.offset:e.offset+8], x) + e.offset += 8 +} + +func (d *decoder) int8() int8 { return int8(d.uint8()) } + +func (e *encoder) int8(x int8) { e.uint8(uint8(x)) } + +func (d *decoder) int16() int16 { return int16(d.uint16()) } + +func (e *encoder) int16(x int16) { e.uint16(uint16(x)) } + +func (d *decoder) int32() int32 { return int32(d.uint32()) } + +func (e *encoder) int32(x int32) { e.uint32(uint32(x)) } + +func (d *decoder) int64() int64 { return int64(d.uint64()) } + +func (e *encoder) int64(x int64) { e.uint64(uint64(x)) } + +func (d *decoder) value(v reflect.Value) { + switch v.Kind() { + case reflect.Array: + l := v.Len() + for i := 0; i < l; i++ { + d.value(v.Index(i)) + } + + case reflect.Struct: + t := v.Type() + l := v.NumField() + for i := 0; i < l; i++ { + // Note: Calling v.CanSet() below is an optimization. + // It would be sufficient to check the field name, + // but creating the StructField info for each field is + // costly (run "go test -bench=ReadStruct" and compare + // results when making changes to this code). + if v := v.Field(i); v.CanSet() || t.Field(i).Name != "_" { + d.value(v) + } else { + d.skip(v) + } + } + + case reflect.Slice: + l := v.Len() + for i := 0; i < l; i++ { + d.value(v.Index(i)) + } + + case reflect.Bool: + v.SetBool(d.bool()) + + case reflect.Int8: + v.SetInt(int64(d.int8())) + case reflect.Int16: + v.SetInt(int64(d.int16())) + case reflect.Int32: + v.SetInt(int64(d.int32())) + case reflect.Int64: + v.SetInt(d.int64()) + + case reflect.Uint8: + v.SetUint(uint64(d.uint8())) + case reflect.Uint16: + v.SetUint(uint64(d.uint16())) + case reflect.Uint32: + v.SetUint(uint64(d.uint32())) + case reflect.Uint64: + v.SetUint(d.uint64()) + + case reflect.Float32: + v.SetFloat(float64(math.Float32frombits(d.uint32()))) + case reflect.Float64: + v.SetFloat(math.Float64frombits(d.uint64())) + + case reflect.Complex64: + v.SetComplex(complex( + float64(math.Float32frombits(d.uint32())), + float64(math.Float32frombits(d.uint32())), + )) + case reflect.Complex128: + v.SetComplex(complex( + math.Float64frombits(d.uint64()), + math.Float64frombits(d.uint64()), + )) + } +} + +func (e *encoder) value(v reflect.Value) { + switch v.Kind() { + case reflect.Array: + l := v.Len() + for i := 0; i < l; i++ { + e.value(v.Index(i)) + } + + case reflect.Struct: + t := v.Type() + l := v.NumField() + for i := 0; i < l; i++ { + // see comment for corresponding code in decoder.value() + if v := v.Field(i); v.CanSet() || t.Field(i).Name != "_" { + e.value(v) + } else { + e.skip(v) + } + } + + case reflect.Slice: + l := v.Len() + for i := 0; i < l; i++ { + e.value(v.Index(i)) + } + + case reflect.Bool: + e.bool(v.Bool()) + + case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64: + switch v.Type().Kind() { + case reflect.Int8: + e.int8(int8(v.Int())) + case reflect.Int16: + e.int16(int16(v.Int())) + case reflect.Int32: + e.int32(int32(v.Int())) + case reflect.Int64: + e.int64(v.Int()) + } + + case reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr: + switch v.Type().Kind() { + case reflect.Uint8: + e.uint8(uint8(v.Uint())) + case reflect.Uint16: + e.uint16(uint16(v.Uint())) + case reflect.Uint32: + e.uint32(uint32(v.Uint())) + case reflect.Uint64: + e.uint64(v.Uint()) + } + + case reflect.Float32, reflect.Float64: + switch v.Type().Kind() { + case reflect.Float32: + e.uint32(math.Float32bits(float32(v.Float()))) + case reflect.Float64: + e.uint64(math.Float64bits(v.Float())) + } + + case reflect.Complex64, reflect.Complex128: + switch v.Type().Kind() { + case reflect.Complex64: + x := v.Complex() + e.uint32(math.Float32bits(float32(real(x)))) + e.uint32(math.Float32bits(float32(imag(x)))) + case reflect.Complex128: + x := v.Complex() + e.uint64(math.Float64bits(real(x))) + e.uint64(math.Float64bits(imag(x))) + } + } +} + +func (d *decoder) skip(v reflect.Value) { + d.offset += dataSize(v) +} + +func (e *encoder) skip(v reflect.Value) { + n := dataSize(v) + zero := e.buf[e.offset : e.offset+n] + for i := range zero { + zero[i] = 0 + } + e.offset += n +} + +// intDataSize returns the size of the data required to represent the data when encoded. +// It returns zero if the type cannot be implemented by the fast path in Read or Write. +func intDataSize(data any) int { + switch data := data.(type) { + case bool, int8, uint8, *bool, *int8, *uint8: + return 1 + case []bool: + return len(data) + case []int8: + return len(data) + case []uint8: + return len(data) + case int16, uint16, *int16, *uint16: + return 2 + case []int16: + return 2 * len(data) + case []uint16: + return 2 * len(data) + case int32, uint32, *int32, *uint32: + return 4 + case []int32: + return 4 * len(data) + case []uint32: + return 4 * len(data) + case int64, uint64, *int64, *uint64: + return 8 + case []int64: + return 8 * len(data) + case []uint64: + return 8 * len(data) + case float32, *float32: + return 4 + case float64, *float64: + return 8 + case []float32: + return 4 * len(data) + case []float64: + return 8 * len(data) + } + return 0 +} diff --git a/contrib/go/_std_1.22/src/encoding/binary/native_endian_little.go b/contrib/go/_std_1.22/src/encoding/binary/native_endian_little.go new file mode 100644 index 0000000000..38d3e9b695 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/binary/native_endian_little.go @@ -0,0 +1,14 @@ +// Copyright 2023 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +//go:build 386 || amd64 || amd64p32 || alpha || arm || arm64 || loong64 || mipsle || mips64le || mips64p32le || nios2 || ppc64le || riscv || riscv64 || sh || wasm + +package binary + +type nativeEndian struct { + littleEndian +} + +// NativeEndian is the native-endian implementation of [ByteOrder] and [AppendByteOrder]. +var NativeEndian nativeEndian diff --git a/contrib/go/_std_1.22/src/encoding/binary/varint.go b/contrib/go/_std_1.22/src/encoding/binary/varint.go new file mode 100644 index 0000000000..64dd9d61b4 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/binary/varint.go @@ -0,0 +1,166 @@ +// Copyright 2011 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package binary + +// This file implements "varint" encoding of 64-bit integers. +// The encoding is: +// - unsigned integers are serialized 7 bits at a time, starting with the +// least significant bits +// - the most significant bit (msb) in each output byte indicates if there +// is a continuation byte (msb = 1) +// - signed integers are mapped to unsigned integers using "zig-zag" +// encoding: Positive values x are written as 2*x + 0, negative values +// are written as 2*(^x) + 1; that is, negative numbers are complemented +// and whether to complement is encoded in bit 0. +// +// Design note: +// At most 10 bytes are needed for 64-bit values. The encoding could +// be more dense: a full 64-bit value needs an extra byte just to hold bit 63. +// Instead, the msb of the previous byte could be used to hold bit 63 since we +// know there can't be more than 64 bits. This is a trivial improvement and +// would reduce the maximum encoding length to 9 bytes. However, it breaks the +// invariant that the msb is always the "continuation bit" and thus makes the +// format incompatible with a varint encoding for larger numbers (say 128-bit). + +import ( + "errors" + "io" +) + +// MaxVarintLenN is the maximum length of a varint-encoded N-bit integer. +const ( + MaxVarintLen16 = 3 + MaxVarintLen32 = 5 + MaxVarintLen64 = 10 +) + +// AppendUvarint appends the varint-encoded form of x, +// as generated by [PutUvarint], to buf and returns the extended buffer. +func AppendUvarint(buf []byte, x uint64) []byte { + for x >= 0x80 { + buf = append(buf, byte(x)|0x80) + x >>= 7 + } + return append(buf, byte(x)) +} + +// PutUvarint encodes a uint64 into buf and returns the number of bytes written. +// If the buffer is too small, PutUvarint will panic. +func PutUvarint(buf []byte, x uint64) int { + i := 0 + for x >= 0x80 { + buf[i] = byte(x) | 0x80 + x >>= 7 + i++ + } + buf[i] = byte(x) + return i + 1 +} + +// Uvarint decodes a uint64 from buf and returns that value and the +// number of bytes read (> 0). If an error occurred, the value is 0 +// and the number of bytes n is <= 0 meaning: +// +// n == 0: buf too small +// n < 0: value larger than 64 bits (overflow) +// and -n is the number of bytes read +func Uvarint(buf []byte) (uint64, int) { + var x uint64 + var s uint + for i, b := range buf { + if i == MaxVarintLen64 { + // Catch byte reads past MaxVarintLen64. + // See issue https://golang.org/issues/41185 + return 0, -(i + 1) // overflow + } + if b < 0x80 { + if i == MaxVarintLen64-1 && b > 1 { + return 0, -(i + 1) // overflow + } + return x | uint64(b)<<s, i + 1 + } + x |= uint64(b&0x7f) << s + s += 7 + } + return 0, 0 +} + +// AppendVarint appends the varint-encoded form of x, +// as generated by [PutVarint], to buf and returns the extended buffer. +func AppendVarint(buf []byte, x int64) []byte { + ux := uint64(x) << 1 + if x < 0 { + ux = ^ux + } + return AppendUvarint(buf, ux) +} + +// PutVarint encodes an int64 into buf and returns the number of bytes written. +// If the buffer is too small, PutVarint will panic. +func PutVarint(buf []byte, x int64) int { + ux := uint64(x) << 1 + if x < 0 { + ux = ^ux + } + return PutUvarint(buf, ux) +} + +// Varint decodes an int64 from buf and returns that value and the +// number of bytes read (> 0). If an error occurred, the value is 0 +// and the number of bytes n is <= 0 with the following meaning: +// +// n == 0: buf too small +// n < 0: value larger than 64 bits (overflow) +// and -n is the number of bytes read +func Varint(buf []byte) (int64, int) { + ux, n := Uvarint(buf) // ok to continue in presence of error + x := int64(ux >> 1) + if ux&1 != 0 { + x = ^x + } + return x, n +} + +var errOverflow = errors.New("binary: varint overflows a 64-bit integer") + +// ReadUvarint reads an encoded unsigned integer from r and returns it as a uint64. +// The error is [io.EOF] only if no bytes were read. +// If an [io.EOF] happens after reading some but not all the bytes, +// ReadUvarint returns [io.ErrUnexpectedEOF]. +func ReadUvarint(r io.ByteReader) (uint64, error) { + var x uint64 + var s uint + for i := 0; i < MaxVarintLen64; i++ { + b, err := r.ReadByte() + if err != nil { + if i > 0 && err == io.EOF { + err = io.ErrUnexpectedEOF + } + return x, err + } + if b < 0x80 { + if i == MaxVarintLen64-1 && b > 1 { + return x, errOverflow + } + return x | uint64(b)<<s, nil + } + x |= uint64(b&0x7f) << s + s += 7 + } + return x, errOverflow +} + +// ReadVarint reads an encoded signed integer from r and returns it as an int64. +// The error is [io.EOF] only if no bytes were read. +// If an [io.EOF] happens after reading some but not all the bytes, +// ReadVarint returns [io.ErrUnexpectedEOF]. +func ReadVarint(r io.ByteReader) (int64, error) { + ux, err := ReadUvarint(r) // ok to continue in presence of error + x := int64(ux >> 1) + if ux&1 != 0 { + x = ^x + } + return x, err +} diff --git a/contrib/go/_std_1.22/src/encoding/binary/ya.make b/contrib/go/_std_1.22/src/encoding/binary/ya.make new file mode 100644 index 0000000000..79775ce9f1 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/binary/ya.make @@ -0,0 +1,9 @@ +GO_LIBRARY() +IF (TRUE) + SRCS( + binary.go + native_endian_little.go + varint.go + ) +ENDIF() +END() diff --git a/contrib/go/_std_1.22/src/encoding/csv/reader.go b/contrib/go/_std_1.22/src/encoding/csv/reader.go new file mode 100644 index 0000000000..d9cab86572 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/csv/reader.go @@ -0,0 +1,466 @@ +// Copyright 2011 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +// Package csv reads and writes comma-separated values (CSV) files. +// There are many kinds of CSV files; this package supports the format +// described in RFC 4180. +// +// A csv file contains zero or more records of one or more fields per record. +// Each record is separated by the newline character. The final record may +// optionally be followed by a newline character. +// +// field1,field2,field3 +// +// White space is considered part of a field. +// +// Carriage returns before newline characters are silently removed. +// +// Blank lines are ignored. A line with only whitespace characters (excluding +// the ending newline character) is not considered a blank line. +// +// Fields which start and stop with the quote character " are called +// quoted-fields. The beginning and ending quote are not part of the +// field. +// +// The source: +// +// normal string,"quoted-field" +// +// results in the fields +// +// {`normal string`, `quoted-field`} +// +// Within a quoted-field a quote character followed by a second quote +// character is considered a single quote. +// +// "the ""word"" is true","a ""quoted-field""" +// +// results in +// +// {`the "word" is true`, `a "quoted-field"`} +// +// Newlines and commas may be included in a quoted-field +// +// "Multi-line +// field","comma is ," +// +// results in +// +// {`Multi-line +// field`, `comma is ,`} +package csv + +import ( + "bufio" + "bytes" + "errors" + "fmt" + "io" + "unicode" + "unicode/utf8" +) + +// A ParseError is returned for parsing errors. +// Line and column numbers are 1-indexed. +type ParseError struct { + StartLine int // Line where the record starts + Line int // Line where the error occurred + Column int // Column (1-based byte index) where the error occurred + Err error // The actual error +} + +func (e *ParseError) Error() string { + if e.Err == ErrFieldCount { + return fmt.Sprintf("record on line %d: %v", e.Line, e.Err) + } + if e.StartLine != e.Line { + return fmt.Sprintf("record on line %d; parse error on line %d, column %d: %v", e.StartLine, e.Line, e.Column, e.Err) + } + return fmt.Sprintf("parse error on line %d, column %d: %v", e.Line, e.Column, e.Err) +} + +func (e *ParseError) Unwrap() error { return e.Err } + +// These are the errors that can be returned in [ParseError.Err]. +var ( + ErrBareQuote = errors.New("bare \" in non-quoted-field") + ErrQuote = errors.New("extraneous or missing \" in quoted-field") + ErrFieldCount = errors.New("wrong number of fields") + + // Deprecated: ErrTrailingComma is no longer used. + ErrTrailingComma = errors.New("extra delimiter at end of line") +) + +var errInvalidDelim = errors.New("csv: invalid field or comment delimiter") + +func validDelim(r rune) bool { + return r != 0 && r != '"' && r != '\r' && r != '\n' && utf8.ValidRune(r) && r != utf8.RuneError +} + +// A Reader reads records from a CSV-encoded file. +// +// As returned by [NewReader], a Reader expects input conforming to RFC 4180. +// The exported fields can be changed to customize the details before the +// first call to [Reader.Read] or [Reader.ReadAll]. +// +// The Reader converts all \r\n sequences in its input to plain \n, +// including in multiline field values, so that the returned data does +// not depend on which line-ending convention an input file uses. +type Reader struct { + // Comma is the field delimiter. + // It is set to comma (',') by NewReader. + // Comma must be a valid rune and must not be \r, \n, + // or the Unicode replacement character (0xFFFD). + Comma rune + + // Comment, if not 0, is the comment character. Lines beginning with the + // Comment character without preceding whitespace are ignored. + // With leading whitespace the Comment character becomes part of the + // field, even if TrimLeadingSpace is true. + // Comment must be a valid rune and must not be \r, \n, + // or the Unicode replacement character (0xFFFD). + // It must also not be equal to Comma. + Comment rune + + // FieldsPerRecord is the number of expected fields per record. + // If FieldsPerRecord is positive, Read requires each record to + // have the given number of fields. If FieldsPerRecord is 0, Read sets it to + // the number of fields in the first record, so that future records must + // have the same field count. If FieldsPerRecord is negative, no check is + // made and records may have a variable number of fields. + FieldsPerRecord int + + // If LazyQuotes is true, a quote may appear in an unquoted field and a + // non-doubled quote may appear in a quoted field. + LazyQuotes bool + + // If TrimLeadingSpace is true, leading white space in a field is ignored. + // This is done even if the field delimiter, Comma, is white space. + TrimLeadingSpace bool + + // ReuseRecord controls whether calls to Read may return a slice sharing + // the backing array of the previous call's returned slice for performance. + // By default, each call to Read returns newly allocated memory owned by the caller. + ReuseRecord bool + + // Deprecated: TrailingComma is no longer used. + TrailingComma bool + + r *bufio.Reader + + // numLine is the current line being read in the CSV file. + numLine int + + // offset is the input stream byte offset of the current reader position. + offset int64 + + // rawBuffer is a line buffer only used by the readLine method. + rawBuffer []byte + + // recordBuffer holds the unescaped fields, one after another. + // The fields can be accessed by using the indexes in fieldIndexes. + // E.g., For the row `a,"b","c""d",e`, recordBuffer will contain `abc"de` + // and fieldIndexes will contain the indexes [1, 2, 5, 6]. + recordBuffer []byte + + // fieldIndexes is an index of fields inside recordBuffer. + // The i'th field ends at offset fieldIndexes[i] in recordBuffer. + fieldIndexes []int + + // fieldPositions is an index of field positions for the + // last record returned by Read. + fieldPositions []position + + // lastRecord is a record cache and only used when ReuseRecord == true. + lastRecord []string +} + +// NewReader returns a new Reader that reads from r. +func NewReader(r io.Reader) *Reader { + return &Reader{ + Comma: ',', + r: bufio.NewReader(r), + } +} + +// Read reads one record (a slice of fields) from r. +// If the record has an unexpected number of fields, +// Read returns the record along with the error [ErrFieldCount]. +// If the record contains a field that cannot be parsed, +// Read returns a partial record along with the parse error. +// The partial record contains all fields read before the error. +// If there is no data left to be read, Read returns nil, [io.EOF]. +// If [Reader.ReuseRecord] is true, the returned slice may be shared +// between multiple calls to Read. +func (r *Reader) Read() (record []string, err error) { + if r.ReuseRecord { + record, err = r.readRecord(r.lastRecord) + r.lastRecord = record + } else { + record, err = r.readRecord(nil) + } + return record, err +} + +// FieldPos returns the line and column corresponding to +// the start of the field with the given index in the slice most recently +// returned by [Reader.Read]. Numbering of lines and columns starts at 1; +// columns are counted in bytes, not runes. +// +// If this is called with an out-of-bounds index, it panics. +func (r *Reader) FieldPos(field int) (line, column int) { + if field < 0 || field >= len(r.fieldPositions) { + panic("out of range index passed to FieldPos") + } + p := &r.fieldPositions[field] + return p.line, p.col +} + +// InputOffset returns the input stream byte offset of the current reader +// position. The offset gives the location of the end of the most recently +// read row and the beginning of the next row. +func (r *Reader) InputOffset() int64 { + return r.offset +} + +// pos holds the position of a field in the current line. +type position struct { + line, col int +} + +// ReadAll reads all the remaining records from r. +// Each record is a slice of fields. +// A successful call returns err == nil, not err == [io.EOF]. Because ReadAll is +// defined to read until EOF, it does not treat end of file as an error to be +// reported. +func (r *Reader) ReadAll() (records [][]string, err error) { + for { + record, err := r.readRecord(nil) + if err == io.EOF { + return records, nil + } + if err != nil { + return nil, err + } + records = append(records, record) + } +} + +// readLine reads the next line (with the trailing endline). +// If EOF is hit without a trailing endline, it will be omitted. +// If some bytes were read, then the error is never [io.EOF]. +// The result is only valid until the next call to readLine. +func (r *Reader) readLine() ([]byte, error) { + line, err := r.r.ReadSlice('\n') + if err == bufio.ErrBufferFull { + r.rawBuffer = append(r.rawBuffer[:0], line...) + for err == bufio.ErrBufferFull { + line, err = r.r.ReadSlice('\n') + r.rawBuffer = append(r.rawBuffer, line...) + } + line = r.rawBuffer + } + readSize := len(line) + if readSize > 0 && err == io.EOF { + err = nil + // For backwards compatibility, drop trailing \r before EOF. + if line[readSize-1] == '\r' { + line = line[:readSize-1] + } + } + r.numLine++ + r.offset += int64(readSize) + // Normalize \r\n to \n on all input lines. + if n := len(line); n >= 2 && line[n-2] == '\r' && line[n-1] == '\n' { + line[n-2] = '\n' + line = line[:n-1] + } + return line, err +} + +// lengthNL reports the number of bytes for the trailing \n. +func lengthNL(b []byte) int { + if len(b) > 0 && b[len(b)-1] == '\n' { + return 1 + } + return 0 +} + +// nextRune returns the next rune in b or utf8.RuneError. +func nextRune(b []byte) rune { + r, _ := utf8.DecodeRune(b) + return r +} + +func (r *Reader) readRecord(dst []string) ([]string, error) { + if r.Comma == r.Comment || !validDelim(r.Comma) || (r.Comment != 0 && !validDelim(r.Comment)) { + return nil, errInvalidDelim + } + + // Read line (automatically skipping past empty lines and any comments). + var line []byte + var errRead error + for errRead == nil { + line, errRead = r.readLine() + if r.Comment != 0 && nextRune(line) == r.Comment { + line = nil + continue // Skip comment lines + } + if errRead == nil && len(line) == lengthNL(line) { + line = nil + continue // Skip empty lines + } + break + } + if errRead == io.EOF { + return nil, errRead + } + + // Parse each field in the record. + var err error + const quoteLen = len(`"`) + commaLen := utf8.RuneLen(r.Comma) + recLine := r.numLine // Starting line for record + r.recordBuffer = r.recordBuffer[:0] + r.fieldIndexes = r.fieldIndexes[:0] + r.fieldPositions = r.fieldPositions[:0] + pos := position{line: r.numLine, col: 1} +parseField: + for { + if r.TrimLeadingSpace { + i := bytes.IndexFunc(line, func(r rune) bool { + return !unicode.IsSpace(r) + }) + if i < 0 { + i = len(line) + pos.col -= lengthNL(line) + } + line = line[i:] + pos.col += i + } + if len(line) == 0 || line[0] != '"' { + // Non-quoted string field + i := bytes.IndexRune(line, r.Comma) + field := line + if i >= 0 { + field = field[:i] + } else { + field = field[:len(field)-lengthNL(field)] + } + // Check to make sure a quote does not appear in field. + if !r.LazyQuotes { + if j := bytes.IndexByte(field, '"'); j >= 0 { + col := pos.col + j + err = &ParseError{StartLine: recLine, Line: r.numLine, Column: col, Err: ErrBareQuote} + break parseField + } + } + r.recordBuffer = append(r.recordBuffer, field...) + r.fieldIndexes = append(r.fieldIndexes, len(r.recordBuffer)) + r.fieldPositions = append(r.fieldPositions, pos) + if i >= 0 { + line = line[i+commaLen:] + pos.col += i + commaLen + continue parseField + } + break parseField + } else { + // Quoted string field + fieldPos := pos + line = line[quoteLen:] + pos.col += quoteLen + for { + i := bytes.IndexByte(line, '"') + if i >= 0 { + // Hit next quote. + r.recordBuffer = append(r.recordBuffer, line[:i]...) + line = line[i+quoteLen:] + pos.col += i + quoteLen + switch rn := nextRune(line); { + case rn == '"': + // `""` sequence (append quote). + r.recordBuffer = append(r.recordBuffer, '"') + line = line[quoteLen:] + pos.col += quoteLen + case rn == r.Comma: + // `",` sequence (end of field). + line = line[commaLen:] + pos.col += commaLen + r.fieldIndexes = append(r.fieldIndexes, len(r.recordBuffer)) + r.fieldPositions = append(r.fieldPositions, fieldPos) + continue parseField + case lengthNL(line) == len(line): + // `"\n` sequence (end of line). + r.fieldIndexes = append(r.fieldIndexes, len(r.recordBuffer)) + r.fieldPositions = append(r.fieldPositions, fieldPos) + break parseField + case r.LazyQuotes: + // `"` sequence (bare quote). + r.recordBuffer = append(r.recordBuffer, '"') + default: + // `"*` sequence (invalid non-escaped quote). + err = &ParseError{StartLine: recLine, Line: r.numLine, Column: pos.col - quoteLen, Err: ErrQuote} + break parseField + } + } else if len(line) > 0 { + // Hit end of line (copy all data so far). + r.recordBuffer = append(r.recordBuffer, line...) + if errRead != nil { + break parseField + } + pos.col += len(line) + line, errRead = r.readLine() + if len(line) > 0 { + pos.line++ + pos.col = 1 + } + if errRead == io.EOF { + errRead = nil + } + } else { + // Abrupt end of file (EOF or error). + if !r.LazyQuotes && errRead == nil { + err = &ParseError{StartLine: recLine, Line: pos.line, Column: pos.col, Err: ErrQuote} + break parseField + } + r.fieldIndexes = append(r.fieldIndexes, len(r.recordBuffer)) + r.fieldPositions = append(r.fieldPositions, fieldPos) + break parseField + } + } + } + } + if err == nil { + err = errRead + } + + // Create a single string and create slices out of it. + // This pins the memory of the fields together, but allocates once. + str := string(r.recordBuffer) // Convert to string once to batch allocations + dst = dst[:0] + if cap(dst) < len(r.fieldIndexes) { + dst = make([]string, len(r.fieldIndexes)) + } + dst = dst[:len(r.fieldIndexes)] + var preIdx int + for i, idx := range r.fieldIndexes { + dst[i] = str[preIdx:idx] + preIdx = idx + } + + // Check or update the expected fields per record. + if r.FieldsPerRecord > 0 { + if len(dst) != r.FieldsPerRecord && err == nil { + err = &ParseError{ + StartLine: recLine, + Line: recLine, + Column: 1, + Err: ErrFieldCount, + } + } + } else if r.FieldsPerRecord == 0 { + r.FieldsPerRecord = len(dst) + } + return dst, err +} diff --git a/contrib/go/_std_1.22/src/encoding/csv/writer.go b/contrib/go/_std_1.22/src/encoding/csv/writer.go new file mode 100644 index 0000000000..ff3142f0bb --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/csv/writer.go @@ -0,0 +1,184 @@ +// Copyright 2011 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package csv + +import ( + "bufio" + "io" + "strings" + "unicode" + "unicode/utf8" +) + +// A Writer writes records using CSV encoding. +// +// As returned by [NewWriter], a Writer writes records terminated by a +// newline and uses ',' as the field delimiter. The exported fields can be +// changed to customize the details before +// the first call to [Writer.Write] or [Writer.WriteAll]. +// +// [Writer.Comma] is the field delimiter. +// +// If [Writer.UseCRLF] is true, +// the Writer ends each output line with \r\n instead of \n. +// +// The writes of individual records are buffered. +// After all data has been written, the client should call the +// [Writer.Flush] method to guarantee all data has been forwarded to +// the underlying [io.Writer]. Any errors that occurred should +// be checked by calling the [Writer.Error] method. +type Writer struct { + Comma rune // Field delimiter (set to ',' by NewWriter) + UseCRLF bool // True to use \r\n as the line terminator + w *bufio.Writer +} + +// NewWriter returns a new Writer that writes to w. +func NewWriter(w io.Writer) *Writer { + return &Writer{ + Comma: ',', + w: bufio.NewWriter(w), + } +} + +// Write writes a single CSV record to w along with any necessary quoting. +// A record is a slice of strings with each string being one field. +// Writes are buffered, so [Writer.Flush] must eventually be called to ensure +// that the record is written to the underlying [io.Writer]. +func (w *Writer) Write(record []string) error { + if !validDelim(w.Comma) { + return errInvalidDelim + } + + for n, field := range record { + if n > 0 { + if _, err := w.w.WriteRune(w.Comma); err != nil { + return err + } + } + + // If we don't have to have a quoted field then just + // write out the field and continue to the next field. + if !w.fieldNeedsQuotes(field) { + if _, err := w.w.WriteString(field); err != nil { + return err + } + continue + } + + if err := w.w.WriteByte('"'); err != nil { + return err + } + for len(field) > 0 { + // Search for special characters. + i := strings.IndexAny(field, "\"\r\n") + if i < 0 { + i = len(field) + } + + // Copy verbatim everything before the special character. + if _, err := w.w.WriteString(field[:i]); err != nil { + return err + } + field = field[i:] + + // Encode the special character. + if len(field) > 0 { + var err error + switch field[0] { + case '"': + _, err = w.w.WriteString(`""`) + case '\r': + if !w.UseCRLF { + err = w.w.WriteByte('\r') + } + case '\n': + if w.UseCRLF { + _, err = w.w.WriteString("\r\n") + } else { + err = w.w.WriteByte('\n') + } + } + field = field[1:] + if err != nil { + return err + } + } + } + if err := w.w.WriteByte('"'); err != nil { + return err + } + } + var err error + if w.UseCRLF { + _, err = w.w.WriteString("\r\n") + } else { + err = w.w.WriteByte('\n') + } + return err +} + +// Flush writes any buffered data to the underlying [io.Writer]. +// To check if an error occurred during Flush, call [Writer.Error]. +func (w *Writer) Flush() { + w.w.Flush() +} + +// Error reports any error that has occurred during +// a previous [Writer.Write] or [Writer.Flush]. +func (w *Writer) Error() error { + _, err := w.w.Write(nil) + return err +} + +// WriteAll writes multiple CSV records to w using [Writer.Write] and +// then calls [Writer.Flush], returning any error from the Flush. +func (w *Writer) WriteAll(records [][]string) error { + for _, record := range records { + err := w.Write(record) + if err != nil { + return err + } + } + return w.w.Flush() +} + +// fieldNeedsQuotes reports whether our field must be enclosed in quotes. +// Fields with a Comma, fields with a quote or newline, and +// fields which start with a space must be enclosed in quotes. +// We used to quote empty strings, but we do not anymore (as of Go 1.4). +// The two representations should be equivalent, but Postgres distinguishes +// quoted vs non-quoted empty string during database imports, and it has +// an option to force the quoted behavior for non-quoted CSV but it has +// no option to force the non-quoted behavior for quoted CSV, making +// CSV with quoted empty strings strictly less useful. +// Not quoting the empty string also makes this package match the behavior +// of Microsoft Excel and Google Drive. +// For Postgres, quote the data terminating string `\.`. +func (w *Writer) fieldNeedsQuotes(field string) bool { + if field == "" { + return false + } + + if field == `\.` { + return true + } + + if w.Comma < utf8.RuneSelf { + for i := 0; i < len(field); i++ { + c := field[i] + if c == '\n' || c == '\r' || c == '"' || c == byte(w.Comma) { + return true + } + } + } else { + if strings.ContainsRune(field, w.Comma) || strings.ContainsAny(field, "\"\r\n") { + return true + } + } + + r1, _ := utf8.DecodeRuneInString(field) + return unicode.IsSpace(r1) +} diff --git a/contrib/go/_std_1.22/src/encoding/csv/ya.make b/contrib/go/_std_1.22/src/encoding/csv/ya.make new file mode 100644 index 0000000000..b3fe9b7d97 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/csv/ya.make @@ -0,0 +1,8 @@ +GO_LIBRARY() +IF (TRUE) + SRCS( + reader.go + writer.go + ) +ENDIF() +END() diff --git a/contrib/go/_std_1.22/src/encoding/encoding.go b/contrib/go/_std_1.22/src/encoding/encoding.go new file mode 100644 index 0000000000..50acf3c23a --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/encoding.go @@ -0,0 +1,54 @@ +// Copyright 2013 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +// Package encoding defines interfaces shared by other packages that +// convert data to and from byte-level and textual representations. +// Packages that check for these interfaces include encoding/gob, +// encoding/json, and encoding/xml. As a result, implementing an +// interface once can make a type useful in multiple encodings. +// Standard types that implement these interfaces include time.Time and net.IP. +// The interfaces come in pairs that produce and consume encoded data. +// +// Adding encoding/decoding methods to existing types may constitute a breaking change, +// as they can be used for serialization in communicating with programs +// written with different library versions. +// The policy for packages maintained by the Go project is to only allow +// the addition of marshaling functions if no existing, reasonable marshaling exists. +package encoding + +// BinaryMarshaler is the interface implemented by an object that can +// marshal itself into a binary form. +// +// MarshalBinary encodes the receiver into a binary form and returns the result. +type BinaryMarshaler interface { + MarshalBinary() (data []byte, err error) +} + +// BinaryUnmarshaler is the interface implemented by an object that can +// unmarshal a binary representation of itself. +// +// UnmarshalBinary must be able to decode the form generated by MarshalBinary. +// UnmarshalBinary must copy the data if it wishes to retain the data +// after returning. +type BinaryUnmarshaler interface { + UnmarshalBinary(data []byte) error +} + +// TextMarshaler is the interface implemented by an object that can +// marshal itself into a textual form. +// +// MarshalText encodes the receiver into UTF-8-encoded text and returns the result. +type TextMarshaler interface { + MarshalText() (text []byte, err error) +} + +// TextUnmarshaler is the interface implemented by an object that can +// unmarshal a textual representation of itself. +// +// UnmarshalText must be able to decode the form generated by MarshalText. +// UnmarshalText must copy the text if it wishes to retain the text +// after returning. +type TextUnmarshaler interface { + UnmarshalText(text []byte) error +} diff --git a/contrib/go/_std_1.22/src/encoding/gob/dec_helpers.go b/contrib/go/_std_1.22/src/encoding/gob/dec_helpers.go new file mode 100644 index 0000000000..44a74e2442 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/gob/dec_helpers.go @@ -0,0 +1,548 @@ +// Code generated by go run decgen.go -output dec_helpers.go; DO NOT EDIT. + +// Copyright 2014 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package gob + +import ( + "math" + "reflect" +) + +var decArrayHelper = map[reflect.Kind]decHelper{ + reflect.Bool: decBoolArray, + reflect.Complex64: decComplex64Array, + reflect.Complex128: decComplex128Array, + reflect.Float32: decFloat32Array, + reflect.Float64: decFloat64Array, + reflect.Int: decIntArray, + reflect.Int16: decInt16Array, + reflect.Int32: decInt32Array, + reflect.Int64: decInt64Array, + reflect.Int8: decInt8Array, + reflect.String: decStringArray, + reflect.Uint: decUintArray, + reflect.Uint16: decUint16Array, + reflect.Uint32: decUint32Array, + reflect.Uint64: decUint64Array, + reflect.Uintptr: decUintptrArray, +} + +var decSliceHelper = map[reflect.Kind]decHelper{ + reflect.Bool: decBoolSlice, + reflect.Complex64: decComplex64Slice, + reflect.Complex128: decComplex128Slice, + reflect.Float32: decFloat32Slice, + reflect.Float64: decFloat64Slice, + reflect.Int: decIntSlice, + reflect.Int16: decInt16Slice, + reflect.Int32: decInt32Slice, + reflect.Int64: decInt64Slice, + reflect.Int8: decInt8Slice, + reflect.String: decStringSlice, + reflect.Uint: decUintSlice, + reflect.Uint16: decUint16Slice, + reflect.Uint32: decUint32Slice, + reflect.Uint64: decUint64Slice, + reflect.Uintptr: decUintptrSlice, +} + +func decBoolArray(state *decoderState, v reflect.Value, length int, ovfl error) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return decBoolSlice(state, v.Slice(0, v.Len()), length, ovfl) +} + +func decBoolSlice(state *decoderState, v reflect.Value, length int, ovfl error) bool { + slice, ok := v.Interface().([]bool) + if !ok { + // It is kind bool but not type bool. TODO: We can handle this unsafely. + return false + } + for i := 0; i < length; i++ { + if state.b.Len() == 0 { + errorf("decoding bool array or slice: length exceeds input size (%d elements)", length) + } + if i >= len(slice) { + // This is a slice that we only partially allocated. + growSlice(v, &slice, length) + } + slice[i] = state.decodeUint() != 0 + } + return true +} + +func decComplex64Array(state *decoderState, v reflect.Value, length int, ovfl error) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return decComplex64Slice(state, v.Slice(0, v.Len()), length, ovfl) +} + +func decComplex64Slice(state *decoderState, v reflect.Value, length int, ovfl error) bool { + slice, ok := v.Interface().([]complex64) + if !ok { + // It is kind complex64 but not type complex64. TODO: We can handle this unsafely. + return false + } + for i := 0; i < length; i++ { + if state.b.Len() == 0 { + errorf("decoding complex64 array or slice: length exceeds input size (%d elements)", length) + } + if i >= len(slice) { + // This is a slice that we only partially allocated. + growSlice(v, &slice, length) + } + real := float32FromBits(state.decodeUint(), ovfl) + imag := float32FromBits(state.decodeUint(), ovfl) + slice[i] = complex(float32(real), float32(imag)) + } + return true +} + +func decComplex128Array(state *decoderState, v reflect.Value, length int, ovfl error) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return decComplex128Slice(state, v.Slice(0, v.Len()), length, ovfl) +} + +func decComplex128Slice(state *decoderState, v reflect.Value, length int, ovfl error) bool { + slice, ok := v.Interface().([]complex128) + if !ok { + // It is kind complex128 but not type complex128. TODO: We can handle this unsafely. + return false + } + for i := 0; i < length; i++ { + if state.b.Len() == 0 { + errorf("decoding complex128 array or slice: length exceeds input size (%d elements)", length) + } + if i >= len(slice) { + // This is a slice that we only partially allocated. + growSlice(v, &slice, length) + } + real := float64FromBits(state.decodeUint()) + imag := float64FromBits(state.decodeUint()) + slice[i] = complex(real, imag) + } + return true +} + +func decFloat32Array(state *decoderState, v reflect.Value, length int, ovfl error) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return decFloat32Slice(state, v.Slice(0, v.Len()), length, ovfl) +} + +func decFloat32Slice(state *decoderState, v reflect.Value, length int, ovfl error) bool { + slice, ok := v.Interface().([]float32) + if !ok { + // It is kind float32 but not type float32. TODO: We can handle this unsafely. + return false + } + for i := 0; i < length; i++ { + if state.b.Len() == 0 { + errorf("decoding float32 array or slice: length exceeds input size (%d elements)", length) + } + if i >= len(slice) { + // This is a slice that we only partially allocated. + growSlice(v, &slice, length) + } + slice[i] = float32(float32FromBits(state.decodeUint(), ovfl)) + } + return true +} + +func decFloat64Array(state *decoderState, v reflect.Value, length int, ovfl error) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return decFloat64Slice(state, v.Slice(0, v.Len()), length, ovfl) +} + +func decFloat64Slice(state *decoderState, v reflect.Value, length int, ovfl error) bool { + slice, ok := v.Interface().([]float64) + if !ok { + // It is kind float64 but not type float64. TODO: We can handle this unsafely. + return false + } + for i := 0; i < length; i++ { + if state.b.Len() == 0 { + errorf("decoding float64 array or slice: length exceeds input size (%d elements)", length) + } + if i >= len(slice) { + // This is a slice that we only partially allocated. + growSlice(v, &slice, length) + } + slice[i] = float64FromBits(state.decodeUint()) + } + return true +} + +func decIntArray(state *decoderState, v reflect.Value, length int, ovfl error) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return decIntSlice(state, v.Slice(0, v.Len()), length, ovfl) +} + +func decIntSlice(state *decoderState, v reflect.Value, length int, ovfl error) bool { + slice, ok := v.Interface().([]int) + if !ok { + // It is kind int but not type int. TODO: We can handle this unsafely. + return false + } + for i := 0; i < length; i++ { + if state.b.Len() == 0 { + errorf("decoding int array or slice: length exceeds input size (%d elements)", length) + } + if i >= len(slice) { + // This is a slice that we only partially allocated. + growSlice(v, &slice, length) + } + x := state.decodeInt() + // MinInt and MaxInt + if x < ^int64(^uint(0)>>1) || int64(^uint(0)>>1) < x { + error_(ovfl) + } + slice[i] = int(x) + } + return true +} + +func decInt16Array(state *decoderState, v reflect.Value, length int, ovfl error) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return decInt16Slice(state, v.Slice(0, v.Len()), length, ovfl) +} + +func decInt16Slice(state *decoderState, v reflect.Value, length int, ovfl error) bool { + slice, ok := v.Interface().([]int16) + if !ok { + // It is kind int16 but not type int16. TODO: We can handle this unsafely. + return false + } + for i := 0; i < length; i++ { + if state.b.Len() == 0 { + errorf("decoding int16 array or slice: length exceeds input size (%d elements)", length) + } + if i >= len(slice) { + // This is a slice that we only partially allocated. + growSlice(v, &slice, length) + } + x := state.decodeInt() + if x < math.MinInt16 || math.MaxInt16 < x { + error_(ovfl) + } + slice[i] = int16(x) + } + return true +} + +func decInt32Array(state *decoderState, v reflect.Value, length int, ovfl error) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return decInt32Slice(state, v.Slice(0, v.Len()), length, ovfl) +} + +func decInt32Slice(state *decoderState, v reflect.Value, length int, ovfl error) bool { + slice, ok := v.Interface().([]int32) + if !ok { + // It is kind int32 but not type int32. TODO: We can handle this unsafely. + return false + } + for i := 0; i < length; i++ { + if state.b.Len() == 0 { + errorf("decoding int32 array or slice: length exceeds input size (%d elements)", length) + } + if i >= len(slice) { + // This is a slice that we only partially allocated. + growSlice(v, &slice, length) + } + x := state.decodeInt() + if x < math.MinInt32 || math.MaxInt32 < x { + error_(ovfl) + } + slice[i] = int32(x) + } + return true +} + +func decInt64Array(state *decoderState, v reflect.Value, length int, ovfl error) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return decInt64Slice(state, v.Slice(0, v.Len()), length, ovfl) +} + +func decInt64Slice(state *decoderState, v reflect.Value, length int, ovfl error) bool { + slice, ok := v.Interface().([]int64) + if !ok { + // It is kind int64 but not type int64. TODO: We can handle this unsafely. + return false + } + for i := 0; i < length; i++ { + if state.b.Len() == 0 { + errorf("decoding int64 array or slice: length exceeds input size (%d elements)", length) + } + if i >= len(slice) { + // This is a slice that we only partially allocated. + growSlice(v, &slice, length) + } + slice[i] = state.decodeInt() + } + return true +} + +func decInt8Array(state *decoderState, v reflect.Value, length int, ovfl error) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return decInt8Slice(state, v.Slice(0, v.Len()), length, ovfl) +} + +func decInt8Slice(state *decoderState, v reflect.Value, length int, ovfl error) bool { + slice, ok := v.Interface().([]int8) + if !ok { + // It is kind int8 but not type int8. TODO: We can handle this unsafely. + return false + } + for i := 0; i < length; i++ { + if state.b.Len() == 0 { + errorf("decoding int8 array or slice: length exceeds input size (%d elements)", length) + } + if i >= len(slice) { + // This is a slice that we only partially allocated. + growSlice(v, &slice, length) + } + x := state.decodeInt() + if x < math.MinInt8 || math.MaxInt8 < x { + error_(ovfl) + } + slice[i] = int8(x) + } + return true +} + +func decStringArray(state *decoderState, v reflect.Value, length int, ovfl error) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return decStringSlice(state, v.Slice(0, v.Len()), length, ovfl) +} + +func decStringSlice(state *decoderState, v reflect.Value, length int, ovfl error) bool { + slice, ok := v.Interface().([]string) + if !ok { + // It is kind string but not type string. TODO: We can handle this unsafely. + return false + } + for i := 0; i < length; i++ { + if state.b.Len() == 0 { + errorf("decoding string array or slice: length exceeds input size (%d elements)", length) + } + if i >= len(slice) { + // This is a slice that we only partially allocated. + growSlice(v, &slice, length) + } + u := state.decodeUint() + n := int(u) + if n < 0 || uint64(n) != u || n > state.b.Len() { + errorf("length of string exceeds input size (%d bytes)", u) + } + if n > state.b.Len() { + errorf("string data too long for buffer: %d", n) + } + // Read the data. + data := state.b.Bytes() + if len(data) < n { + errorf("invalid string length %d: exceeds input size %d", n, len(data)) + } + slice[i] = string(data[:n]) + state.b.Drop(n) + } + return true +} + +func decUintArray(state *decoderState, v reflect.Value, length int, ovfl error) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return decUintSlice(state, v.Slice(0, v.Len()), length, ovfl) +} + +func decUintSlice(state *decoderState, v reflect.Value, length int, ovfl error) bool { + slice, ok := v.Interface().([]uint) + if !ok { + // It is kind uint but not type uint. TODO: We can handle this unsafely. + return false + } + for i := 0; i < length; i++ { + if state.b.Len() == 0 { + errorf("decoding uint array or slice: length exceeds input size (%d elements)", length) + } + if i >= len(slice) { + // This is a slice that we only partially allocated. + growSlice(v, &slice, length) + } + x := state.decodeUint() + /*TODO if math.MaxUint32 < x { + error_(ovfl) + }*/ + slice[i] = uint(x) + } + return true +} + +func decUint16Array(state *decoderState, v reflect.Value, length int, ovfl error) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return decUint16Slice(state, v.Slice(0, v.Len()), length, ovfl) +} + +func decUint16Slice(state *decoderState, v reflect.Value, length int, ovfl error) bool { + slice, ok := v.Interface().([]uint16) + if !ok { + // It is kind uint16 but not type uint16. TODO: We can handle this unsafely. + return false + } + for i := 0; i < length; i++ { + if state.b.Len() == 0 { + errorf("decoding uint16 array or slice: length exceeds input size (%d elements)", length) + } + if i >= len(slice) { + // This is a slice that we only partially allocated. + growSlice(v, &slice, length) + } + x := state.decodeUint() + if math.MaxUint16 < x { + error_(ovfl) + } + slice[i] = uint16(x) + } + return true +} + +func decUint32Array(state *decoderState, v reflect.Value, length int, ovfl error) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return decUint32Slice(state, v.Slice(0, v.Len()), length, ovfl) +} + +func decUint32Slice(state *decoderState, v reflect.Value, length int, ovfl error) bool { + slice, ok := v.Interface().([]uint32) + if !ok { + // It is kind uint32 but not type uint32. TODO: We can handle this unsafely. + return false + } + for i := 0; i < length; i++ { + if state.b.Len() == 0 { + errorf("decoding uint32 array or slice: length exceeds input size (%d elements)", length) + } + if i >= len(slice) { + // This is a slice that we only partially allocated. + growSlice(v, &slice, length) + } + x := state.decodeUint() + if math.MaxUint32 < x { + error_(ovfl) + } + slice[i] = uint32(x) + } + return true +} + +func decUint64Array(state *decoderState, v reflect.Value, length int, ovfl error) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return decUint64Slice(state, v.Slice(0, v.Len()), length, ovfl) +} + +func decUint64Slice(state *decoderState, v reflect.Value, length int, ovfl error) bool { + slice, ok := v.Interface().([]uint64) + if !ok { + // It is kind uint64 but not type uint64. TODO: We can handle this unsafely. + return false + } + for i := 0; i < length; i++ { + if state.b.Len() == 0 { + errorf("decoding uint64 array or slice: length exceeds input size (%d elements)", length) + } + if i >= len(slice) { + // This is a slice that we only partially allocated. + growSlice(v, &slice, length) + } + slice[i] = state.decodeUint() + } + return true +} + +func decUintptrArray(state *decoderState, v reflect.Value, length int, ovfl error) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return decUintptrSlice(state, v.Slice(0, v.Len()), length, ovfl) +} + +func decUintptrSlice(state *decoderState, v reflect.Value, length int, ovfl error) bool { + slice, ok := v.Interface().([]uintptr) + if !ok { + // It is kind uintptr but not type uintptr. TODO: We can handle this unsafely. + return false + } + for i := 0; i < length; i++ { + if state.b.Len() == 0 { + errorf("decoding uintptr array or slice: length exceeds input size (%d elements)", length) + } + if i >= len(slice) { + // This is a slice that we only partially allocated. + growSlice(v, &slice, length) + } + x := state.decodeUint() + if uint64(^uintptr(0)) < x { + error_(ovfl) + } + slice[i] = uintptr(x) + } + return true +} + +// growSlice is called for a slice that we only partially allocated, +// to grow it up to length. +func growSlice[E any](v reflect.Value, ps *[]E, length int) { + var zero E + s := *ps + s = append(s, zero) + cp := cap(s) + if cp > length { + cp = length + } + s = s[:cp] + v.Set(reflect.ValueOf(s)) + *ps = s +} diff --git a/contrib/go/_std_1.22/src/encoding/gob/decode.go b/contrib/go/_std_1.22/src/encoding/gob/decode.go new file mode 100644 index 0000000000..d178b2b2fb --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/gob/decode.go @@ -0,0 +1,1306 @@ +// Copyright 2009 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +//go:generate go run decgen.go -output dec_helpers.go + +package gob + +import ( + "encoding" + "errors" + "internal/saferio" + "io" + "math" + "math/bits" + "reflect" +) + +var ( + errBadUint = errors.New("gob: encoded unsigned integer out of range") + errBadType = errors.New("gob: unknown type id or corrupted data") + errRange = errors.New("gob: bad data: field numbers out of bounds") +) + +type decHelper func(state *decoderState, v reflect.Value, length int, ovfl error) bool + +// decoderState is the execution state of an instance of the decoder. A new state +// is created for nested objects. +type decoderState struct { + dec *Decoder + // The buffer is stored with an extra indirection because it may be replaced + // if we load a type during decode (when reading an interface value). + b *decBuffer + fieldnum int // the last field number read. + next *decoderState // for free list +} + +// decBuffer is an extremely simple, fast implementation of a read-only byte buffer. +// It is initialized by calling Size and then copying the data into the slice returned by Bytes(). +type decBuffer struct { + data []byte + offset int // Read offset. +} + +func (d *decBuffer) Read(p []byte) (int, error) { + n := copy(p, d.data[d.offset:]) + if n == 0 && len(p) != 0 { + return 0, io.EOF + } + d.offset += n + return n, nil +} + +func (d *decBuffer) Drop(n int) { + if n > d.Len() { + panic("drop") + } + d.offset += n +} + +func (d *decBuffer) ReadByte() (byte, error) { + if d.offset >= len(d.data) { + return 0, io.EOF + } + c := d.data[d.offset] + d.offset++ + return c, nil +} + +func (d *decBuffer) Len() int { + return len(d.data) - d.offset +} + +func (d *decBuffer) Bytes() []byte { + return d.data[d.offset:] +} + +// SetBytes sets the buffer to the bytes, discarding any existing data. +func (d *decBuffer) SetBytes(data []byte) { + d.data = data + d.offset = 0 +} + +func (d *decBuffer) Reset() { + d.data = d.data[0:0] + d.offset = 0 +} + +// We pass the bytes.Buffer separately for easier testing of the infrastructure +// without requiring a full Decoder. +func (dec *Decoder) newDecoderState(buf *decBuffer) *decoderState { + d := dec.freeList + if d == nil { + d = new(decoderState) + d.dec = dec + } else { + dec.freeList = d.next + } + d.b = buf + return d +} + +func (dec *Decoder) freeDecoderState(d *decoderState) { + d.next = dec.freeList + dec.freeList = d +} + +func overflow(name string) error { + return errors.New(`value for "` + name + `" out of range`) +} + +// decodeUintReader reads an encoded unsigned integer from an io.Reader. +// Used only by the Decoder to read the message length. +func decodeUintReader(r io.Reader, buf []byte) (x uint64, width int, err error) { + width = 1 + n, err := io.ReadFull(r, buf[0:width]) + if n == 0 { + return + } + b := buf[0] + if b <= 0x7f { + return uint64(b), width, nil + } + n = -int(int8(b)) + if n > uint64Size { + err = errBadUint + return + } + width, err = io.ReadFull(r, buf[0:n]) + if err != nil { + if err == io.EOF { + err = io.ErrUnexpectedEOF + } + return + } + // Could check that the high byte is zero but it's not worth it. + for _, b := range buf[0:width] { + x = x<<8 | uint64(b) + } + width++ // +1 for length byte + return +} + +// decodeUint reads an encoded unsigned integer from state.r. +// Does not check for overflow. +func (state *decoderState) decodeUint() (x uint64) { + b, err := state.b.ReadByte() + if err != nil { + error_(err) + } + if b <= 0x7f { + return uint64(b) + } + n := -int(int8(b)) + if n > uint64Size { + error_(errBadUint) + } + buf := state.b.Bytes() + if len(buf) < n { + errorf("invalid uint data length %d: exceeds input size %d", n, len(buf)) + } + // Don't need to check error; it's safe to loop regardless. + // Could check that the high byte is zero but it's not worth it. + for _, b := range buf[0:n] { + x = x<<8 | uint64(b) + } + state.b.Drop(n) + return x +} + +// decodeInt reads an encoded signed integer from state.r. +// Does not check for overflow. +func (state *decoderState) decodeInt() int64 { + x := state.decodeUint() + if x&1 != 0 { + return ^int64(x >> 1) + } + return int64(x >> 1) +} + +// getLength decodes the next uint and makes sure it is a possible +// size for a data item that follows, which means it must fit in a +// non-negative int and fit in the buffer. +func (state *decoderState) getLength() (int, bool) { + n := int(state.decodeUint()) + if n < 0 || state.b.Len() < n || tooBig <= n { + return 0, false + } + return n, true +} + +// decOp is the signature of a decoding operator for a given type. +type decOp func(i *decInstr, state *decoderState, v reflect.Value) + +// The 'instructions' of the decoding machine +type decInstr struct { + op decOp + field int // field number of the wire type + index []int // field access indices for destination type + ovfl error // error message for overflow/underflow (for arrays, of the elements) +} + +// ignoreUint discards a uint value with no destination. +func ignoreUint(i *decInstr, state *decoderState, v reflect.Value) { + state.decodeUint() +} + +// ignoreTwoUints discards a uint value with no destination. It's used to skip +// complex values. +func ignoreTwoUints(i *decInstr, state *decoderState, v reflect.Value) { + state.decodeUint() + state.decodeUint() +} + +// Since the encoder writes no zeros, if we arrive at a decoder we have +// a value to extract and store. The field number has already been read +// (it's how we knew to call this decoder). +// Each decoder is responsible for handling any indirections associated +// with the data structure. If any pointer so reached is nil, allocation must +// be done. + +// decAlloc takes a value and returns a settable value that can +// be assigned to. If the value is a pointer, decAlloc guarantees it points to storage. +// The callers to the individual decoders are expected to have used decAlloc. +// The individual decoders don't need it. +func decAlloc(v reflect.Value) reflect.Value { + for v.Kind() == reflect.Pointer { + if v.IsNil() { + v.Set(reflect.New(v.Type().Elem())) + } + v = v.Elem() + } + return v +} + +// decBool decodes a uint and stores it as a boolean in value. +func decBool(i *decInstr, state *decoderState, value reflect.Value) { + value.SetBool(state.decodeUint() != 0) +} + +// decInt8 decodes an integer and stores it as an int8 in value. +func decInt8(i *decInstr, state *decoderState, value reflect.Value) { + v := state.decodeInt() + if v < math.MinInt8 || math.MaxInt8 < v { + error_(i.ovfl) + } + value.SetInt(v) +} + +// decUint8 decodes an unsigned integer and stores it as a uint8 in value. +func decUint8(i *decInstr, state *decoderState, value reflect.Value) { + v := state.decodeUint() + if math.MaxUint8 < v { + error_(i.ovfl) + } + value.SetUint(v) +} + +// decInt16 decodes an integer and stores it as an int16 in value. +func decInt16(i *decInstr, state *decoderState, value reflect.Value) { + v := state.decodeInt() + if v < math.MinInt16 || math.MaxInt16 < v { + error_(i.ovfl) + } + value.SetInt(v) +} + +// decUint16 decodes an unsigned integer and stores it as a uint16 in value. +func decUint16(i *decInstr, state *decoderState, value reflect.Value) { + v := state.decodeUint() + if math.MaxUint16 < v { + error_(i.ovfl) + } + value.SetUint(v) +} + +// decInt32 decodes an integer and stores it as an int32 in value. +func decInt32(i *decInstr, state *decoderState, value reflect.Value) { + v := state.decodeInt() + if v < math.MinInt32 || math.MaxInt32 < v { + error_(i.ovfl) + } + value.SetInt(v) +} + +// decUint32 decodes an unsigned integer and stores it as a uint32 in value. +func decUint32(i *decInstr, state *decoderState, value reflect.Value) { + v := state.decodeUint() + if math.MaxUint32 < v { + error_(i.ovfl) + } + value.SetUint(v) +} + +// decInt64 decodes an integer and stores it as an int64 in value. +func decInt64(i *decInstr, state *decoderState, value reflect.Value) { + v := state.decodeInt() + value.SetInt(v) +} + +// decUint64 decodes an unsigned integer and stores it as a uint64 in value. +func decUint64(i *decInstr, state *decoderState, value reflect.Value) { + v := state.decodeUint() + value.SetUint(v) +} + +// Floating-point numbers are transmitted as uint64s holding the bits +// of the underlying representation. They are sent byte-reversed, with +// the exponent end coming out first, so integer floating point numbers +// (for example) transmit more compactly. This routine does the +// unswizzling. +func float64FromBits(u uint64) float64 { + v := bits.ReverseBytes64(u) + return math.Float64frombits(v) +} + +// float32FromBits decodes an unsigned integer, treats it as a 32-bit floating-point +// number, and returns it. It's a helper function for float32 and complex64. +// It returns a float64 because that's what reflection needs, but its return +// value is known to be accurately representable in a float32. +func float32FromBits(u uint64, ovfl error) float64 { + v := float64FromBits(u) + av := v + if av < 0 { + av = -av + } + // +Inf is OK in both 32- and 64-bit floats. Underflow is always OK. + if math.MaxFloat32 < av && av <= math.MaxFloat64 { + error_(ovfl) + } + return v +} + +// decFloat32 decodes an unsigned integer, treats it as a 32-bit floating-point +// number, and stores it in value. +func decFloat32(i *decInstr, state *decoderState, value reflect.Value) { + value.SetFloat(float32FromBits(state.decodeUint(), i.ovfl)) +} + +// decFloat64 decodes an unsigned integer, treats it as a 64-bit floating-point +// number, and stores it in value. +func decFloat64(i *decInstr, state *decoderState, value reflect.Value) { + value.SetFloat(float64FromBits(state.decodeUint())) +} + +// decComplex64 decodes a pair of unsigned integers, treats them as a +// pair of floating point numbers, and stores them as a complex64 in value. +// The real part comes first. +func decComplex64(i *decInstr, state *decoderState, value reflect.Value) { + real := float32FromBits(state.decodeUint(), i.ovfl) + imag := float32FromBits(state.decodeUint(), i.ovfl) + value.SetComplex(complex(real, imag)) +} + +// decComplex128 decodes a pair of unsigned integers, treats them as a +// pair of floating point numbers, and stores them as a complex128 in value. +// The real part comes first. +func decComplex128(i *decInstr, state *decoderState, value reflect.Value) { + real := float64FromBits(state.decodeUint()) + imag := float64FromBits(state.decodeUint()) + value.SetComplex(complex(real, imag)) +} + +// decUint8Slice decodes a byte slice and stores in value a slice header +// describing the data. +// uint8 slices are encoded as an unsigned count followed by the raw bytes. +func decUint8Slice(i *decInstr, state *decoderState, value reflect.Value) { + n, ok := state.getLength() + if !ok { + errorf("bad %s slice length: %d", value.Type(), n) + } + if value.Cap() < n { + safe := saferio.SliceCap[byte](uint64(n)) + if safe < 0 { + errorf("%s slice too big: %d elements", value.Type(), n) + } + value.Set(reflect.MakeSlice(value.Type(), safe, safe)) + ln := safe + i := 0 + for i < n { + if i >= ln { + // We didn't allocate the entire slice, + // due to using saferio.SliceCap. + // Grow the slice for one more element. + // The slice is full, so this should + // bump up the capacity. + value.Grow(1) + } + // Copy into s up to the capacity or n, + // whichever is less. + ln = value.Cap() + if ln > n { + ln = n + } + value.SetLen(ln) + sub := value.Slice(i, ln) + if _, err := state.b.Read(sub.Bytes()); err != nil { + errorf("error decoding []byte at %d: %s", i, err) + } + i = ln + } + } else { + value.SetLen(n) + if _, err := state.b.Read(value.Bytes()); err != nil { + errorf("error decoding []byte: %s", err) + } + } +} + +// decString decodes byte array and stores in value a string header +// describing the data. +// Strings are encoded as an unsigned count followed by the raw bytes. +func decString(i *decInstr, state *decoderState, value reflect.Value) { + n, ok := state.getLength() + if !ok { + errorf("bad %s slice length: %d", value.Type(), n) + } + // Read the data. + data := state.b.Bytes() + if len(data) < n { + errorf("invalid string length %d: exceeds input size %d", n, len(data)) + } + s := string(data[:n]) + state.b.Drop(n) + value.SetString(s) +} + +// ignoreUint8Array skips over the data for a byte slice value with no destination. +func ignoreUint8Array(i *decInstr, state *decoderState, value reflect.Value) { + n, ok := state.getLength() + if !ok { + errorf("slice length too large") + } + bn := state.b.Len() + if bn < n { + errorf("invalid slice length %d: exceeds input size %d", n, bn) + } + state.b.Drop(n) +} + +// Execution engine + +// The encoder engine is an array of instructions indexed by field number of the incoming +// decoder. It is executed with random access according to field number. +type decEngine struct { + instr []decInstr + numInstr int // the number of active instructions +} + +// decodeSingle decodes a top-level value that is not a struct and stores it in value. +// Such values are preceded by a zero, making them have the memory layout of a +// struct field (although with an illegal field number). +func (dec *Decoder) decodeSingle(engine *decEngine, value reflect.Value) { + state := dec.newDecoderState(&dec.buf) + defer dec.freeDecoderState(state) + state.fieldnum = singletonField + if state.decodeUint() != 0 { + errorf("decode: corrupted data: non-zero delta for singleton") + } + instr := &engine.instr[singletonField] + instr.op(instr, state, value) +} + +// decodeStruct decodes a top-level struct and stores it in value. +// Indir is for the value, not the type. At the time of the call it may +// differ from ut.indir, which was computed when the engine was built. +// This state cannot arise for decodeSingle, which is called directly +// from the user's value, not from the innards of an engine. +func (dec *Decoder) decodeStruct(engine *decEngine, value reflect.Value) { + state := dec.newDecoderState(&dec.buf) + defer dec.freeDecoderState(state) + state.fieldnum = -1 + for state.b.Len() > 0 { + delta := int(state.decodeUint()) + if delta < 0 { + errorf("decode: corrupted data: negative delta") + } + if delta == 0 { // struct terminator is zero delta fieldnum + break + } + if state.fieldnum >= len(engine.instr)-delta { // subtract to compare without overflow + error_(errRange) + } + fieldnum := state.fieldnum + delta + instr := &engine.instr[fieldnum] + var field reflect.Value + if instr.index != nil { + // Otherwise the field is unknown to us and instr.op is an ignore op. + field = value.FieldByIndex(instr.index) + if field.Kind() == reflect.Pointer { + field = decAlloc(field) + } + } + instr.op(instr, state, field) + state.fieldnum = fieldnum + } +} + +var noValue reflect.Value + +// ignoreStruct discards the data for a struct with no destination. +func (dec *Decoder) ignoreStruct(engine *decEngine) { + state := dec.newDecoderState(&dec.buf) + defer dec.freeDecoderState(state) + state.fieldnum = -1 + for state.b.Len() > 0 { + delta := int(state.decodeUint()) + if delta < 0 { + errorf("ignore decode: corrupted data: negative delta") + } + if delta == 0 { // struct terminator is zero delta fieldnum + break + } + fieldnum := state.fieldnum + delta + if fieldnum >= len(engine.instr) { + error_(errRange) + } + instr := &engine.instr[fieldnum] + instr.op(instr, state, noValue) + state.fieldnum = fieldnum + } +} + +// ignoreSingle discards the data for a top-level non-struct value with no +// destination. It's used when calling Decode with a nil value. +func (dec *Decoder) ignoreSingle(engine *decEngine) { + state := dec.newDecoderState(&dec.buf) + defer dec.freeDecoderState(state) + state.fieldnum = singletonField + delta := int(state.decodeUint()) + if delta != 0 { + errorf("decode: corrupted data: non-zero delta for singleton") + } + instr := &engine.instr[singletonField] + instr.op(instr, state, noValue) +} + +// decodeArrayHelper does the work for decoding arrays and slices. +func (dec *Decoder) decodeArrayHelper(state *decoderState, value reflect.Value, elemOp decOp, length int, ovfl error, helper decHelper) { + if helper != nil && helper(state, value, length, ovfl) { + return + } + instr := &decInstr{elemOp, 0, nil, ovfl} + isPtr := value.Type().Elem().Kind() == reflect.Pointer + ln := value.Len() + for i := 0; i < length; i++ { + if state.b.Len() == 0 { + errorf("decoding array or slice: length exceeds input size (%d elements)", length) + } + if i >= ln { + // This is a slice that we only partially allocated. + // Grow it up to length. + value.Grow(1) + cp := value.Cap() + if cp > length { + cp = length + } + value.SetLen(cp) + ln = cp + } + v := value.Index(i) + if isPtr { + v = decAlloc(v) + } + elemOp(instr, state, v) + } +} + +// decodeArray decodes an array and stores it in value. +// The length is an unsigned integer preceding the elements. Even though the length is redundant +// (it's part of the type), it's a useful check and is included in the encoding. +func (dec *Decoder) decodeArray(state *decoderState, value reflect.Value, elemOp decOp, length int, ovfl error, helper decHelper) { + if n := state.decodeUint(); n != uint64(length) { + errorf("length mismatch in decodeArray") + } + dec.decodeArrayHelper(state, value, elemOp, length, ovfl, helper) +} + +// decodeIntoValue is a helper for map decoding. +func decodeIntoValue(state *decoderState, op decOp, isPtr bool, value reflect.Value, instr *decInstr) reflect.Value { + v := value + if isPtr { + v = decAlloc(value) + } + + op(instr, state, v) + return value +} + +// decodeMap decodes a map and stores it in value. +// Maps are encoded as a length followed by key:value pairs. +// Because the internals of maps are not visible to us, we must +// use reflection rather than pointer magic. +func (dec *Decoder) decodeMap(mtyp reflect.Type, state *decoderState, value reflect.Value, keyOp, elemOp decOp, ovfl error) { + n := int(state.decodeUint()) + if value.IsNil() { + value.Set(reflect.MakeMapWithSize(mtyp, n)) + } + keyIsPtr := mtyp.Key().Kind() == reflect.Pointer + elemIsPtr := mtyp.Elem().Kind() == reflect.Pointer + keyInstr := &decInstr{keyOp, 0, nil, ovfl} + elemInstr := &decInstr{elemOp, 0, nil, ovfl} + keyP := reflect.New(mtyp.Key()) + elemP := reflect.New(mtyp.Elem()) + for i := 0; i < n; i++ { + key := decodeIntoValue(state, keyOp, keyIsPtr, keyP.Elem(), keyInstr) + elem := decodeIntoValue(state, elemOp, elemIsPtr, elemP.Elem(), elemInstr) + value.SetMapIndex(key, elem) + keyP.Elem().SetZero() + elemP.Elem().SetZero() + } +} + +// ignoreArrayHelper does the work for discarding arrays and slices. +func (dec *Decoder) ignoreArrayHelper(state *decoderState, elemOp decOp, length int) { + instr := &decInstr{elemOp, 0, nil, errors.New("no error")} + for i := 0; i < length; i++ { + if state.b.Len() == 0 { + errorf("decoding array or slice: length exceeds input size (%d elements)", length) + } + elemOp(instr, state, noValue) + } +} + +// ignoreArray discards the data for an array value with no destination. +func (dec *Decoder) ignoreArray(state *decoderState, elemOp decOp, length int) { + if n := state.decodeUint(); n != uint64(length) { + errorf("length mismatch in ignoreArray") + } + dec.ignoreArrayHelper(state, elemOp, length) +} + +// ignoreMap discards the data for a map value with no destination. +func (dec *Decoder) ignoreMap(state *decoderState, keyOp, elemOp decOp) { + n := int(state.decodeUint()) + keyInstr := &decInstr{keyOp, 0, nil, errors.New("no error")} + elemInstr := &decInstr{elemOp, 0, nil, errors.New("no error")} + for i := 0; i < n; i++ { + keyOp(keyInstr, state, noValue) + elemOp(elemInstr, state, noValue) + } +} + +// decodeSlice decodes a slice and stores it in value. +// Slices are encoded as an unsigned length followed by the elements. +func (dec *Decoder) decodeSlice(state *decoderState, value reflect.Value, elemOp decOp, ovfl error, helper decHelper) { + u := state.decodeUint() + typ := value.Type() + size := uint64(typ.Elem().Size()) + nBytes := u * size + n := int(u) + // Take care with overflow in this calculation. + if n < 0 || uint64(n) != u || nBytes > tooBig || (size > 0 && nBytes/size != u) { + // We don't check n against buffer length here because if it's a slice + // of interfaces, there will be buffer reloads. + errorf("%s slice too big: %d elements of %d bytes", typ.Elem(), u, size) + } + if value.Cap() < n { + safe := saferio.SliceCapWithSize(size, uint64(n)) + if safe < 0 { + errorf("%s slice too big: %d elements of %d bytes", typ.Elem(), u, size) + } + value.Set(reflect.MakeSlice(typ, safe, safe)) + } else { + value.SetLen(n) + } + dec.decodeArrayHelper(state, value, elemOp, n, ovfl, helper) +} + +// ignoreSlice skips over the data for a slice value with no destination. +func (dec *Decoder) ignoreSlice(state *decoderState, elemOp decOp) { + dec.ignoreArrayHelper(state, elemOp, int(state.decodeUint())) +} + +// decodeInterface decodes an interface value and stores it in value. +// Interfaces are encoded as the name of a concrete type followed by a value. +// If the name is empty, the value is nil and no value is sent. +func (dec *Decoder) decodeInterface(ityp reflect.Type, state *decoderState, value reflect.Value) { + // Read the name of the concrete type. + nr := state.decodeUint() + if nr > 1<<31 { // zero is permissible for anonymous types + errorf("invalid type name length %d", nr) + } + if nr > uint64(state.b.Len()) { + errorf("invalid type name length %d: exceeds input size", nr) + } + n := int(nr) + name := state.b.Bytes()[:n] + state.b.Drop(n) + // Allocate the destination interface value. + if len(name) == 0 { + // Copy the nil interface value to the target. + value.SetZero() + return + } + if len(name) > 1024 { + errorf("name too long (%d bytes): %.20q...", len(name), name) + } + // The concrete type must be registered. + typi, ok := nameToConcreteType.Load(string(name)) + if !ok { + errorf("name not registered for interface: %q", name) + } + typ := typi.(reflect.Type) + + // Read the type id of the concrete value. + concreteId := dec.decodeTypeSequence(true) + if concreteId < 0 { + error_(dec.err) + } + // Byte count of value is next; we don't care what it is (it's there + // in case we want to ignore the value by skipping it completely). + state.decodeUint() + // Read the concrete value. + v := allocValue(typ) + dec.decodeValue(concreteId, v) + if dec.err != nil { + error_(dec.err) + } + // Assign the concrete value to the interface. + // Tread carefully; it might not satisfy the interface. + if !typ.AssignableTo(ityp) { + errorf("%s is not assignable to type %s", typ, ityp) + } + // Copy the interface value to the target. + value.Set(v) +} + +// ignoreInterface discards the data for an interface value with no destination. +func (dec *Decoder) ignoreInterface(state *decoderState) { + // Read the name of the concrete type. + n, ok := state.getLength() + if !ok { + errorf("bad interface encoding: name too large for buffer") + } + bn := state.b.Len() + if bn < n { + errorf("invalid interface value length %d: exceeds input size %d", n, bn) + } + state.b.Drop(n) + id := dec.decodeTypeSequence(true) + if id < 0 { + error_(dec.err) + } + // At this point, the decoder buffer contains a delimited value. Just toss it. + n, ok = state.getLength() + if !ok { + errorf("bad interface encoding: data length too large for buffer") + } + state.b.Drop(n) +} + +// decodeGobDecoder decodes something implementing the GobDecoder interface. +// The data is encoded as a byte slice. +func (dec *Decoder) decodeGobDecoder(ut *userTypeInfo, state *decoderState, value reflect.Value) { + // Read the bytes for the value. + n, ok := state.getLength() + if !ok { + errorf("GobDecoder: length too large for buffer") + } + b := state.b.Bytes() + if len(b) < n { + errorf("GobDecoder: invalid data length %d: exceeds input size %d", n, len(b)) + } + b = b[:n] + state.b.Drop(n) + var err error + // We know it's one of these. + switch ut.externalDec { + case xGob: + err = value.Interface().(GobDecoder).GobDecode(b) + case xBinary: + err = value.Interface().(encoding.BinaryUnmarshaler).UnmarshalBinary(b) + case xText: + err = value.Interface().(encoding.TextUnmarshaler).UnmarshalText(b) + } + if err != nil { + error_(err) + } +} + +// ignoreGobDecoder discards the data for a GobDecoder value with no destination. +func (dec *Decoder) ignoreGobDecoder(state *decoderState) { + // Read the bytes for the value. + n, ok := state.getLength() + if !ok { + errorf("GobDecoder: length too large for buffer") + } + bn := state.b.Len() + if bn < n { + errorf("GobDecoder: invalid data length %d: exceeds input size %d", n, bn) + } + state.b.Drop(n) +} + +// Index by Go types. +var decOpTable = [...]decOp{ + reflect.Bool: decBool, + reflect.Int8: decInt8, + reflect.Int16: decInt16, + reflect.Int32: decInt32, + reflect.Int64: decInt64, + reflect.Uint8: decUint8, + reflect.Uint16: decUint16, + reflect.Uint32: decUint32, + reflect.Uint64: decUint64, + reflect.Float32: decFloat32, + reflect.Float64: decFloat64, + reflect.Complex64: decComplex64, + reflect.Complex128: decComplex128, + reflect.String: decString, +} + +// Indexed by gob types. tComplex will be added during type.init(). +var decIgnoreOpMap = map[typeId]decOp{ + tBool: ignoreUint, + tInt: ignoreUint, + tUint: ignoreUint, + tFloat: ignoreUint, + tBytes: ignoreUint8Array, + tString: ignoreUint8Array, + tComplex: ignoreTwoUints, +} + +// decOpFor returns the decoding op for the base type under rt and +// the indirection count to reach it. +func (dec *Decoder) decOpFor(wireId typeId, rt reflect.Type, name string, inProgress map[reflect.Type]*decOp) *decOp { + ut := userType(rt) + // If the type implements GobEncoder, we handle it without further processing. + if ut.externalDec != 0 { + return dec.gobDecodeOpFor(ut) + } + + // If this type is already in progress, it's a recursive type (e.g. map[string]*T). + // Return the pointer to the op we're already building. + if opPtr := inProgress[rt]; opPtr != nil { + return opPtr + } + typ := ut.base + var op decOp + k := typ.Kind() + if int(k) < len(decOpTable) { + op = decOpTable[k] + } + if op == nil { + inProgress[rt] = &op + // Special cases + switch t := typ; t.Kind() { + case reflect.Array: + name = "element of " + name + elemId := dec.wireType[wireId].ArrayT.Elem + elemOp := dec.decOpFor(elemId, t.Elem(), name, inProgress) + ovfl := overflow(name) + helper := decArrayHelper[t.Elem().Kind()] + op = func(i *decInstr, state *decoderState, value reflect.Value) { + state.dec.decodeArray(state, value, *elemOp, t.Len(), ovfl, helper) + } + + case reflect.Map: + keyId := dec.wireType[wireId].MapT.Key + elemId := dec.wireType[wireId].MapT.Elem + keyOp := dec.decOpFor(keyId, t.Key(), "key of "+name, inProgress) + elemOp := dec.decOpFor(elemId, t.Elem(), "element of "+name, inProgress) + ovfl := overflow(name) + op = func(i *decInstr, state *decoderState, value reflect.Value) { + state.dec.decodeMap(t, state, value, *keyOp, *elemOp, ovfl) + } + + case reflect.Slice: + name = "element of " + name + if t.Elem().Kind() == reflect.Uint8 { + op = decUint8Slice + break + } + var elemId typeId + if tt := builtinIdToType(wireId); tt != nil { + elemId = tt.(*sliceType).Elem + } else { + elemId = dec.wireType[wireId].SliceT.Elem + } + elemOp := dec.decOpFor(elemId, t.Elem(), name, inProgress) + ovfl := overflow(name) + helper := decSliceHelper[t.Elem().Kind()] + op = func(i *decInstr, state *decoderState, value reflect.Value) { + state.dec.decodeSlice(state, value, *elemOp, ovfl, helper) + } + + case reflect.Struct: + // Generate a closure that calls out to the engine for the nested type. + ut := userType(typ) + enginePtr, err := dec.getDecEnginePtr(wireId, ut) + if err != nil { + error_(err) + } + op = func(i *decInstr, state *decoderState, value reflect.Value) { + // indirect through enginePtr to delay evaluation for recursive structs. + dec.decodeStruct(*enginePtr, value) + } + case reflect.Interface: + op = func(i *decInstr, state *decoderState, value reflect.Value) { + state.dec.decodeInterface(t, state, value) + } + } + } + if op == nil { + errorf("decode can't handle type %s", rt) + } + return &op +} + +var maxIgnoreNestingDepth = 10000 + +// decIgnoreOpFor returns the decoding op for a field that has no destination. +func (dec *Decoder) decIgnoreOpFor(wireId typeId, inProgress map[typeId]*decOp, depth int) *decOp { + if depth > maxIgnoreNestingDepth { + error_(errors.New("invalid nesting depth")) + } + // If this type is already in progress, it's a recursive type (e.g. map[string]*T). + // Return the pointer to the op we're already building. + if opPtr := inProgress[wireId]; opPtr != nil { + return opPtr + } + op, ok := decIgnoreOpMap[wireId] + if !ok { + inProgress[wireId] = &op + if wireId == tInterface { + // Special case because it's a method: the ignored item might + // define types and we need to record their state in the decoder. + op = func(i *decInstr, state *decoderState, value reflect.Value) { + state.dec.ignoreInterface(state) + } + return &op + } + // Special cases + wire := dec.wireType[wireId] + switch { + case wire == nil: + errorf("bad data: undefined type %s", wireId.string()) + case wire.ArrayT != nil: + elemId := wire.ArrayT.Elem + elemOp := dec.decIgnoreOpFor(elemId, inProgress, depth+1) + op = func(i *decInstr, state *decoderState, value reflect.Value) { + state.dec.ignoreArray(state, *elemOp, wire.ArrayT.Len) + } + + case wire.MapT != nil: + keyId := dec.wireType[wireId].MapT.Key + elemId := dec.wireType[wireId].MapT.Elem + keyOp := dec.decIgnoreOpFor(keyId, inProgress, depth+1) + elemOp := dec.decIgnoreOpFor(elemId, inProgress, depth+1) + op = func(i *decInstr, state *decoderState, value reflect.Value) { + state.dec.ignoreMap(state, *keyOp, *elemOp) + } + + case wire.SliceT != nil: + elemId := wire.SliceT.Elem + elemOp := dec.decIgnoreOpFor(elemId, inProgress, depth+1) + op = func(i *decInstr, state *decoderState, value reflect.Value) { + state.dec.ignoreSlice(state, *elemOp) + } + + case wire.StructT != nil: + // Generate a closure that calls out to the engine for the nested type. + enginePtr, err := dec.getIgnoreEnginePtr(wireId) + if err != nil { + error_(err) + } + op = func(i *decInstr, state *decoderState, value reflect.Value) { + // indirect through enginePtr to delay evaluation for recursive structs + state.dec.ignoreStruct(*enginePtr) + } + + case wire.GobEncoderT != nil, wire.BinaryMarshalerT != nil, wire.TextMarshalerT != nil: + op = func(i *decInstr, state *decoderState, value reflect.Value) { + state.dec.ignoreGobDecoder(state) + } + } + } + if op == nil { + errorf("bad data: ignore can't handle type %s", wireId.string()) + } + return &op +} + +// gobDecodeOpFor returns the op for a type that is known to implement +// GobDecoder. +func (dec *Decoder) gobDecodeOpFor(ut *userTypeInfo) *decOp { + rcvrType := ut.user + if ut.decIndir == -1 { + rcvrType = reflect.PointerTo(rcvrType) + } else if ut.decIndir > 0 { + for i := int8(0); i < ut.decIndir; i++ { + rcvrType = rcvrType.Elem() + } + } + var op decOp + op = func(i *decInstr, state *decoderState, value reflect.Value) { + // We now have the base type. We need its address if the receiver is a pointer. + if value.Kind() != reflect.Pointer && rcvrType.Kind() == reflect.Pointer { + value = value.Addr() + } + state.dec.decodeGobDecoder(ut, state, value) + } + return &op +} + +// compatibleType asks: Are these two gob Types compatible? +// Answers the question for basic types, arrays, maps and slices, plus +// GobEncoder/Decoder pairs. +// Structs are considered ok; fields will be checked later. +func (dec *Decoder) compatibleType(fr reflect.Type, fw typeId, inProgress map[reflect.Type]typeId) bool { + if rhs, ok := inProgress[fr]; ok { + return rhs == fw + } + inProgress[fr] = fw + ut := userType(fr) + wire, ok := dec.wireType[fw] + // If wire was encoded with an encoding method, fr must have that method. + // And if not, it must not. + // At most one of the booleans in ut is set. + // We could possibly relax this constraint in the future in order to + // choose the decoding method using the data in the wireType. + // The parentheses look odd but are correct. + if (ut.externalDec == xGob) != (ok && wire.GobEncoderT != nil) || + (ut.externalDec == xBinary) != (ok && wire.BinaryMarshalerT != nil) || + (ut.externalDec == xText) != (ok && wire.TextMarshalerT != nil) { + return false + } + if ut.externalDec != 0 { // This test trumps all others. + return true + } + switch t := ut.base; t.Kind() { + default: + // chan, etc: cannot handle. + return false + case reflect.Bool: + return fw == tBool + case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64: + return fw == tInt + case reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr: + return fw == tUint + case reflect.Float32, reflect.Float64: + return fw == tFloat + case reflect.Complex64, reflect.Complex128: + return fw == tComplex + case reflect.String: + return fw == tString + case reflect.Interface: + return fw == tInterface + case reflect.Array: + if !ok || wire.ArrayT == nil { + return false + } + array := wire.ArrayT + return t.Len() == array.Len && dec.compatibleType(t.Elem(), array.Elem, inProgress) + case reflect.Map: + if !ok || wire.MapT == nil { + return false + } + MapType := wire.MapT + return dec.compatibleType(t.Key(), MapType.Key, inProgress) && dec.compatibleType(t.Elem(), MapType.Elem, inProgress) + case reflect.Slice: + // Is it an array of bytes? + if t.Elem().Kind() == reflect.Uint8 { + return fw == tBytes + } + // Extract and compare element types. + var sw *sliceType + if tt := builtinIdToType(fw); tt != nil { + sw, _ = tt.(*sliceType) + } else if wire != nil { + sw = wire.SliceT + } + elem := userType(t.Elem()).base + return sw != nil && dec.compatibleType(elem, sw.Elem, inProgress) + case reflect.Struct: + return true + } +} + +// typeString returns a human-readable description of the type identified by remoteId. +func (dec *Decoder) typeString(remoteId typeId) string { + typeLock.Lock() + defer typeLock.Unlock() + if t := idToType(remoteId); t != nil { + // globally known type. + return t.string() + } + return dec.wireType[remoteId].string() +} + +// compileSingle compiles the decoder engine for a non-struct top-level value, including +// GobDecoders. +func (dec *Decoder) compileSingle(remoteId typeId, ut *userTypeInfo) (engine *decEngine, err error) { + rt := ut.user + engine = new(decEngine) + engine.instr = make([]decInstr, 1) // one item + name := rt.String() // best we can do + if !dec.compatibleType(rt, remoteId, make(map[reflect.Type]typeId)) { + remoteType := dec.typeString(remoteId) + // Common confusing case: local interface type, remote concrete type. + if ut.base.Kind() == reflect.Interface && remoteId != tInterface { + return nil, errors.New("gob: local interface type " + name + " can only be decoded from remote interface type; received concrete type " + remoteType) + } + return nil, errors.New("gob: decoding into local type " + name + ", received remote type " + remoteType) + } + op := dec.decOpFor(remoteId, rt, name, make(map[reflect.Type]*decOp)) + ovfl := errors.New(`value for "` + name + `" out of range`) + engine.instr[singletonField] = decInstr{*op, singletonField, nil, ovfl} + engine.numInstr = 1 + return +} + +// compileIgnoreSingle compiles the decoder engine for a non-struct top-level value that will be discarded. +func (dec *Decoder) compileIgnoreSingle(remoteId typeId) *decEngine { + engine := new(decEngine) + engine.instr = make([]decInstr, 1) // one item + op := dec.decIgnoreOpFor(remoteId, make(map[typeId]*decOp), 0) + ovfl := overflow(dec.typeString(remoteId)) + engine.instr[0] = decInstr{*op, 0, nil, ovfl} + engine.numInstr = 1 + return engine +} + +// compileDec compiles the decoder engine for a value. If the value is not a struct, +// it calls out to compileSingle. +func (dec *Decoder) compileDec(remoteId typeId, ut *userTypeInfo) (engine *decEngine, err error) { + defer catchError(&err) + rt := ut.base + srt := rt + if srt.Kind() != reflect.Struct || ut.externalDec != 0 { + return dec.compileSingle(remoteId, ut) + } + var wireStruct *structType + // Builtin types can come from global pool; the rest must be defined by the decoder. + // Also we know we're decoding a struct now, so the client must have sent one. + if t := builtinIdToType(remoteId); t != nil { + wireStruct, _ = t.(*structType) + } else { + wire := dec.wireType[remoteId] + if wire == nil { + error_(errBadType) + } + wireStruct = wire.StructT + } + if wireStruct == nil { + errorf("type mismatch in decoder: want struct type %s; got non-struct", rt) + } + engine = new(decEngine) + engine.instr = make([]decInstr, len(wireStruct.Field)) + seen := make(map[reflect.Type]*decOp) + // Loop over the fields of the wire type. + for fieldnum := 0; fieldnum < len(wireStruct.Field); fieldnum++ { + wireField := wireStruct.Field[fieldnum] + if wireField.Name == "" { + errorf("empty name for remote field of type %s", wireStruct.Name) + } + ovfl := overflow(wireField.Name) + // Find the field of the local type with the same name. + localField, present := srt.FieldByName(wireField.Name) + // TODO(r): anonymous names + if !present || !isExported(wireField.Name) { + op := dec.decIgnoreOpFor(wireField.Id, make(map[typeId]*decOp), 0) + engine.instr[fieldnum] = decInstr{*op, fieldnum, nil, ovfl} + continue + } + if !dec.compatibleType(localField.Type, wireField.Id, make(map[reflect.Type]typeId)) { + errorf("wrong type (%s) for received field %s.%s", localField.Type, wireStruct.Name, wireField.Name) + } + op := dec.decOpFor(wireField.Id, localField.Type, localField.Name, seen) + engine.instr[fieldnum] = decInstr{*op, fieldnum, localField.Index, ovfl} + engine.numInstr++ + } + return +} + +// getDecEnginePtr returns the engine for the specified type. +func (dec *Decoder) getDecEnginePtr(remoteId typeId, ut *userTypeInfo) (enginePtr **decEngine, err error) { + rt := ut.user + decoderMap, ok := dec.decoderCache[rt] + if !ok { + decoderMap = make(map[typeId]**decEngine) + dec.decoderCache[rt] = decoderMap + } + if enginePtr, ok = decoderMap[remoteId]; !ok { + // To handle recursive types, mark this engine as underway before compiling. + enginePtr = new(*decEngine) + decoderMap[remoteId] = enginePtr + *enginePtr, err = dec.compileDec(remoteId, ut) + if err != nil { + delete(decoderMap, remoteId) + } + } + return +} + +// emptyStruct is the type we compile into when ignoring a struct value. +type emptyStruct struct{} + +var emptyStructType = reflect.TypeFor[emptyStruct]() + +// getIgnoreEnginePtr returns the engine for the specified type when the value is to be discarded. +func (dec *Decoder) getIgnoreEnginePtr(wireId typeId) (enginePtr **decEngine, err error) { + var ok bool + if enginePtr, ok = dec.ignorerCache[wireId]; !ok { + // To handle recursive types, mark this engine as underway before compiling. + enginePtr = new(*decEngine) + dec.ignorerCache[wireId] = enginePtr + wire := dec.wireType[wireId] + if wire != nil && wire.StructT != nil { + *enginePtr, err = dec.compileDec(wireId, userType(emptyStructType)) + } else { + *enginePtr = dec.compileIgnoreSingle(wireId) + } + if err != nil { + delete(dec.ignorerCache, wireId) + } + } + return +} + +// decodeValue decodes the data stream representing a value and stores it in value. +func (dec *Decoder) decodeValue(wireId typeId, value reflect.Value) { + defer catchError(&dec.err) + // If the value is nil, it means we should just ignore this item. + if !value.IsValid() { + dec.decodeIgnoredValue(wireId) + return + } + // Dereference down to the underlying type. + ut := userType(value.Type()) + base := ut.base + var enginePtr **decEngine + enginePtr, dec.err = dec.getDecEnginePtr(wireId, ut) + if dec.err != nil { + return + } + value = decAlloc(value) + engine := *enginePtr + if st := base; st.Kind() == reflect.Struct && ut.externalDec == 0 { + wt := dec.wireType[wireId] + if engine.numInstr == 0 && st.NumField() > 0 && + wt != nil && len(wt.StructT.Field) > 0 { + name := base.Name() + errorf("type mismatch: no fields matched compiling decoder for %s", name) + } + dec.decodeStruct(engine, value) + } else { + dec.decodeSingle(engine, value) + } +} + +// decodeIgnoredValue decodes the data stream representing a value of the specified type and discards it. +func (dec *Decoder) decodeIgnoredValue(wireId typeId) { + var enginePtr **decEngine + enginePtr, dec.err = dec.getIgnoreEnginePtr(wireId) + if dec.err != nil { + return + } + wire := dec.wireType[wireId] + if wire != nil && wire.StructT != nil { + dec.ignoreStruct(*enginePtr) + } else { + dec.ignoreSingle(*enginePtr) + } +} + +const ( + intBits = 32 << (^uint(0) >> 63) + uintptrBits = 32 << (^uintptr(0) >> 63) +) + +func init() { + var iop, uop decOp + switch intBits { + case 32: + iop = decInt32 + uop = decUint32 + case 64: + iop = decInt64 + uop = decUint64 + default: + panic("gob: unknown size of int/uint") + } + decOpTable[reflect.Int] = iop + decOpTable[reflect.Uint] = uop + + // Finally uintptr + switch uintptrBits { + case 32: + uop = decUint32 + case 64: + uop = decUint64 + default: + panic("gob: unknown size of uintptr") + } + decOpTable[reflect.Uintptr] = uop +} + +// Gob depends on being able to take the address +// of zeroed Values it creates, so use this wrapper instead +// of the standard reflect.Zero. +// Each call allocates once. +func allocValue(t reflect.Type) reflect.Value { + return reflect.New(t).Elem() +} diff --git a/contrib/go/_std_1.22/src/encoding/gob/decoder.go b/contrib/go/_std_1.22/src/encoding/gob/decoder.go new file mode 100644 index 0000000000..c4b6088013 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/gob/decoder.go @@ -0,0 +1,237 @@ +// Copyright 2009 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package gob + +import ( + "bufio" + "errors" + "internal/saferio" + "io" + "reflect" + "sync" +) + +// tooBig provides a sanity check for sizes; used in several places. Upper limit +// of is 1GB on 32-bit systems, 8GB on 64-bit, allowing room to grow a little +// without overflow. +const tooBig = (1 << 30) << (^uint(0) >> 62) + +// A Decoder manages the receipt of type and data information read from the +// remote side of a connection. It is safe for concurrent use by multiple +// goroutines. +// +// The Decoder does only basic sanity checking on decoded input sizes, +// and its limits are not configurable. Take caution when decoding gob data +// from untrusted sources. +type Decoder struct { + mutex sync.Mutex // each item must be received atomically + r io.Reader // source of the data + buf decBuffer // buffer for more efficient i/o from r + wireType map[typeId]*wireType // map from remote ID to local description + decoderCache map[reflect.Type]map[typeId]**decEngine // cache of compiled engines + ignorerCache map[typeId]**decEngine // ditto for ignored objects + freeList *decoderState // list of free decoderStates; avoids reallocation + countBuf []byte // used for decoding integers while parsing messages + err error +} + +// NewDecoder returns a new decoder that reads from the [io.Reader]. +// If r does not also implement [io.ByteReader], it will be wrapped in a +// [bufio.Reader]. +func NewDecoder(r io.Reader) *Decoder { + dec := new(Decoder) + // We use the ability to read bytes as a plausible surrogate for buffering. + if _, ok := r.(io.ByteReader); !ok { + r = bufio.NewReader(r) + } + dec.r = r + dec.wireType = make(map[typeId]*wireType) + dec.decoderCache = make(map[reflect.Type]map[typeId]**decEngine) + dec.ignorerCache = make(map[typeId]**decEngine) + dec.countBuf = make([]byte, 9) // counts may be uint64s (unlikely!), require 9 bytes + + return dec +} + +// recvType loads the definition of a type. +func (dec *Decoder) recvType(id typeId) { + // Have we already seen this type? That's an error + if id < firstUserId || dec.wireType[id] != nil { + dec.err = errors.New("gob: duplicate type received") + return + } + + // Type: + wire := new(wireType) + dec.decodeValue(tWireType, reflect.ValueOf(wire)) + if dec.err != nil { + return + } + // Remember we've seen this type. + dec.wireType[id] = wire +} + +var errBadCount = errors.New("invalid message length") + +// recvMessage reads the next count-delimited item from the input. It is the converse +// of Encoder.writeMessage. It returns false on EOF or other error reading the message. +func (dec *Decoder) recvMessage() bool { + // Read a count. + nbytes, _, err := decodeUintReader(dec.r, dec.countBuf) + if err != nil { + dec.err = err + return false + } + if nbytes >= tooBig { + dec.err = errBadCount + return false + } + dec.readMessage(int(nbytes)) + return dec.err == nil +} + +// readMessage reads the next nbytes bytes from the input. +func (dec *Decoder) readMessage(nbytes int) { + if dec.buf.Len() != 0 { + // The buffer should always be empty now. + panic("non-empty decoder buffer") + } + // Read the data + var buf []byte + buf, dec.err = saferio.ReadData(dec.r, uint64(nbytes)) + dec.buf.SetBytes(buf) + if dec.err == io.EOF { + dec.err = io.ErrUnexpectedEOF + } +} + +// toInt turns an encoded uint64 into an int, according to the marshaling rules. +func toInt(x uint64) int64 { + i := int64(x >> 1) + if x&1 != 0 { + i = ^i + } + return i +} + +func (dec *Decoder) nextInt() int64 { + n, _, err := decodeUintReader(&dec.buf, dec.countBuf) + if err != nil { + dec.err = err + } + return toInt(n) +} + +func (dec *Decoder) nextUint() uint64 { + n, _, err := decodeUintReader(&dec.buf, dec.countBuf) + if err != nil { + dec.err = err + } + return n +} + +// decodeTypeSequence parses: +// TypeSequence +// +// (TypeDefinition DelimitedTypeDefinition*)? +// +// and returns the type id of the next value. It returns -1 at +// EOF. Upon return, the remainder of dec.buf is the value to be +// decoded. If this is an interface value, it can be ignored by +// resetting that buffer. +func (dec *Decoder) decodeTypeSequence(isInterface bool) typeId { + firstMessage := true + for dec.err == nil { + if dec.buf.Len() == 0 { + if !dec.recvMessage() { + // We can only return io.EOF if the input was empty. + // If we read one or more type spec messages, + // require a data item message to follow. + // If we hit an EOF before that, then give ErrUnexpectedEOF. + if !firstMessage && dec.err == io.EOF { + dec.err = io.ErrUnexpectedEOF + } + break + } + } + // Receive a type id. + id := typeId(dec.nextInt()) + if id >= 0 { + // Value follows. + return id + } + // Type definition for (-id) follows. + dec.recvType(-id) + if dec.err != nil { + break + } + // When decoding an interface, after a type there may be a + // DelimitedValue still in the buffer. Skip its count. + // (Alternatively, the buffer is empty and the byte count + // will be absorbed by recvMessage.) + if dec.buf.Len() > 0 { + if !isInterface { + dec.err = errors.New("extra data in buffer") + break + } + dec.nextUint() + } + firstMessage = false + } + return -1 +} + +// Decode reads the next value from the input stream and stores +// it in the data represented by the empty interface value. +// If e is nil, the value will be discarded. Otherwise, +// the value underlying e must be a pointer to the +// correct type for the next data item received. +// If the input is at EOF, Decode returns [io.EOF] and +// does not modify e. +func (dec *Decoder) Decode(e any) error { + if e == nil { + return dec.DecodeValue(reflect.Value{}) + } + value := reflect.ValueOf(e) + // If e represents a value as opposed to a pointer, the answer won't + // get back to the caller. Make sure it's a pointer. + if value.Type().Kind() != reflect.Pointer { + dec.err = errors.New("gob: attempt to decode into a non-pointer") + return dec.err + } + return dec.DecodeValue(value) +} + +// DecodeValue reads the next value from the input stream. +// If v is the zero reflect.Value (v.Kind() == Invalid), DecodeValue discards the value. +// Otherwise, it stores the value into v. In that case, v must represent +// a non-nil pointer to data or be an assignable reflect.Value (v.CanSet()) +// If the input is at EOF, DecodeValue returns [io.EOF] and +// does not modify v. +func (dec *Decoder) DecodeValue(v reflect.Value) error { + if v.IsValid() { + if v.Kind() == reflect.Pointer && !v.IsNil() { + // That's okay, we'll store through the pointer. + } else if !v.CanSet() { + return errors.New("gob: DecodeValue of unassignable value") + } + } + // Make sure we're single-threaded through here. + dec.mutex.Lock() + defer dec.mutex.Unlock() + + dec.buf.Reset() // In case data lingers from previous invocation. + dec.err = nil + id := dec.decodeTypeSequence(false) + if dec.err == nil { + dec.decodeValue(id, v) + } + return dec.err +} + +// If debug.go is compiled into the program, debugFunc prints a human-readable +// representation of the gob data read from r by calling that file's Debug function. +// Otherwise it is nil. +var debugFunc func(io.Reader) diff --git a/contrib/go/_std_1.22/src/encoding/gob/doc.go b/contrib/go/_std_1.22/src/encoding/gob/doc.go new file mode 100644 index 0000000000..3f26ed8591 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/gob/doc.go @@ -0,0 +1,423 @@ +// Copyright 2009 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +/* +Package gob manages streams of gobs - binary values exchanged between an +[Encoder] (transmitter) and a [Decoder] (receiver). A typical use is transporting +arguments and results of remote procedure calls (RPCs) such as those provided by +[net/rpc]. + +The implementation compiles a custom codec for each data type in the stream and +is most efficient when a single [Encoder] is used to transmit a stream of values, +amortizing the cost of compilation. + +# Basics + +A stream of gobs is self-describing. Each data item in the stream is preceded by +a specification of its type, expressed in terms of a small set of predefined +types. Pointers are not transmitted, but the things they point to are +transmitted; that is, the values are flattened. Nil pointers are not permitted, +as they have no value. Recursive types work fine, but +recursive values (data with cycles) are problematic. This may change. + +To use gobs, create an [Encoder] and present it with a series of data items as +values or addresses that can be dereferenced to values. The [Encoder] makes sure +all type information is sent before it is needed. At the receive side, a +[Decoder] retrieves values from the encoded stream and unpacks them into local +variables. + +# Types and Values + +The source and destination values/types need not correspond exactly. For structs, +fields (identified by name) that are in the source but absent from the receiving +variable will be ignored. Fields that are in the receiving variable but missing +from the transmitted type or value will be ignored in the destination. If a field +with the same name is present in both, their types must be compatible. Both the +receiver and transmitter will do all necessary indirection and dereferencing to +convert between gobs and actual Go values. For instance, a gob type that is +schematically, + + struct { A, B int } + +can be sent from or received into any of these Go types: + + struct { A, B int } // the same + *struct { A, B int } // extra indirection of the struct + struct { *A, **B int } // extra indirection of the fields + struct { A, B int64 } // different concrete value type; see below + +It may also be received into any of these: + + struct { A, B int } // the same + struct { B, A int } // ordering doesn't matter; matching is by name + struct { A, B, C int } // extra field (C) ignored + struct { B int } // missing field (A) ignored; data will be dropped + struct { B, C int } // missing field (A) ignored; extra field (C) ignored. + +Attempting to receive into these types will draw a decode error: + + struct { A int; B uint } // change of signedness for B + struct { A int; B float } // change of type for B + struct { } // no field names in common + struct { C, D int } // no field names in common + +Integers are transmitted two ways: arbitrary precision signed integers or +arbitrary precision unsigned integers. There is no int8, int16 etc. +discrimination in the gob format; there are only signed and unsigned integers. As +described below, the transmitter sends the value in a variable-length encoding; +the receiver accepts the value and stores it in the destination variable. +Floating-point numbers are always sent using IEEE-754 64-bit precision (see +below). + +Signed integers may be received into any signed integer variable: int, int16, etc.; +unsigned integers may be received into any unsigned integer variable; and floating +point values may be received into any floating point variable. However, +the destination variable must be able to represent the value or the decode +operation will fail. + +Structs, arrays and slices are also supported. Structs encode and decode only +exported fields. Strings and arrays of bytes are supported with a special, +efficient representation (see below). When a slice is decoded, if the existing +slice has capacity the slice will be extended in place; if not, a new array is +allocated. Regardless, the length of the resulting slice reports the number of +elements decoded. + +In general, if allocation is required, the decoder will allocate memory. If not, +it will update the destination variables with values read from the stream. It does +not initialize them first, so if the destination is a compound value such as a +map, struct, or slice, the decoded values will be merged elementwise into the +existing variables. + +Functions and channels will not be sent in a gob. Attempting to encode such a value +at the top level will fail. A struct field of chan or func type is treated exactly +like an unexported field and is ignored. + +Gob can encode a value of any type implementing the [GobEncoder] or +[encoding.BinaryMarshaler] interfaces by calling the corresponding method, +in that order of preference. + +Gob can decode a value of any type implementing the [GobDecoder] or +[encoding.BinaryUnmarshaler] interfaces by calling the corresponding method, +again in that order of preference. + +# Encoding Details + +This section documents the encoding, details that are not important for most +users. Details are presented bottom-up. + +An unsigned integer is sent one of two ways. If it is less than 128, it is sent +as a byte with that value. Otherwise it is sent as a minimal-length big-endian +(high byte first) byte stream holding the value, preceded by one byte holding the +byte count, negated. Thus 0 is transmitted as (00), 7 is transmitted as (07) and +256 is transmitted as (FE 01 00). + +A boolean is encoded within an unsigned integer: 0 for false, 1 for true. + +A signed integer, i, is encoded within an unsigned integer, u. Within u, bits 1 +upward contain the value; bit 0 says whether they should be complemented upon +receipt. The encode algorithm looks like this: + + var u uint + if i < 0 { + u = (^uint(i) << 1) | 1 // complement i, bit 0 is 1 + } else { + u = (uint(i) << 1) // do not complement i, bit 0 is 0 + } + encodeUnsigned(u) + +The low bit is therefore analogous to a sign bit, but making it the complement bit +instead guarantees that the largest negative integer is not a special case. For +example, -129=^128=(^256>>1) encodes as (FE 01 01). + +Floating-point numbers are always sent as a representation of a float64 value. +That value is converted to a uint64 using [math.Float64bits]. The uint64 is then +byte-reversed and sent as a regular unsigned integer. The byte-reversal means the +exponent and high-precision part of the mantissa go first. Since the low bits are +often zero, this can save encoding bytes. For instance, 17.0 is encoded in only +three bytes (FE 31 40). + +Strings and slices of bytes are sent as an unsigned count followed by that many +uninterpreted bytes of the value. + +All other slices and arrays are sent as an unsigned count followed by that many +elements using the standard gob encoding for their type, recursively. + +Maps are sent as an unsigned count followed by that many key, element +pairs. Empty but non-nil maps are sent, so if the receiver has not allocated +one already, one will always be allocated on receipt unless the transmitted map +is nil and not at the top level. + +In slices and arrays, as well as maps, all elements, even zero-valued elements, +are transmitted, even if all the elements are zero. + +Structs are sent as a sequence of (field number, field value) pairs. The field +value is sent using the standard gob encoding for its type, recursively. If a +field has the zero value for its type (except for arrays; see above), it is omitted +from the transmission. The field number is defined by the type of the encoded +struct: the first field of the encoded type is field 0, the second is field 1, +etc. When encoding a value, the field numbers are delta encoded for efficiency +and the fields are always sent in order of increasing field number; the deltas are +therefore unsigned. The initialization for the delta encoding sets the field +number to -1, so an unsigned integer field 0 with value 7 is transmitted as unsigned +delta = 1, unsigned value = 7 or (01 07). Finally, after all the fields have been +sent a terminating mark denotes the end of the struct. That mark is a delta=0 +value, which has representation (00). + +Interface types are not checked for compatibility; all interface types are +treated, for transmission, as members of a single "interface" type, analogous to +int or []byte - in effect they're all treated as interface{}. Interface values +are transmitted as a string identifying the concrete type being sent (a name +that must be pre-defined by calling [Register]), followed by a byte count of the +length of the following data (so the value can be skipped if it cannot be +stored), followed by the usual encoding of concrete (dynamic) value stored in +the interface value. (A nil interface value is identified by the empty string +and transmits no value.) Upon receipt, the decoder verifies that the unpacked +concrete item satisfies the interface of the receiving variable. + +If a value is passed to [Encoder.Encode] and the type is not a struct (or pointer to struct, +etc.), for simplicity of processing it is represented as a struct of one field. +The only visible effect of this is to encode a zero byte after the value, just as +after the last field of an encoded struct, so that the decode algorithm knows when +the top-level value is complete. + +The representation of types is described below. When a type is defined on a given +connection between an [Encoder] and [Decoder], it is assigned a signed integer type +id. When [Encoder.Encode](v) is called, it makes sure there is an id assigned for +the type of v and all its elements and then it sends the pair (typeid, encoded-v) +where typeid is the type id of the encoded type of v and encoded-v is the gob +encoding of the value v. + +To define a type, the encoder chooses an unused, positive type id and sends the +pair (-type id, encoded-type) where encoded-type is the gob encoding of a wireType +description, constructed from these types: + + type wireType struct { + ArrayT *ArrayType + SliceT *SliceType + StructT *StructType + MapT *MapType + GobEncoderT *gobEncoderType + BinaryMarshalerT *gobEncoderType + TextMarshalerT *gobEncoderType + + } + type arrayType struct { + CommonType + Elem typeId + Len int + } + type CommonType struct { + Name string // the name of the struct type + Id int // the id of the type, repeated so it's inside the type + } + type sliceType struct { + CommonType + Elem typeId + } + type structType struct { + CommonType + Field []*fieldType // the fields of the struct. + } + type fieldType struct { + Name string // the name of the field. + Id int // the type id of the field, which must be already defined + } + type mapType struct { + CommonType + Key typeId + Elem typeId + } + type gobEncoderType struct { + CommonType + } + +If there are nested type ids, the types for all inner type ids must be defined +before the top-level type id is used to describe an encoded-v. + +For simplicity in setup, the connection is defined to understand these types a +priori, as well as the basic gob types int, uint, etc. Their ids are: + + bool 1 + int 2 + uint 3 + float 4 + []byte 5 + string 6 + complex 7 + interface 8 + // gap for reserved ids. + WireType 16 + ArrayType 17 + CommonType 18 + SliceType 19 + StructType 20 + FieldType 21 + // 22 is slice of fieldType. + MapType 23 + +Finally, each message created by a call to Encode is preceded by an encoded +unsigned integer count of the number of bytes remaining in the message. After +the initial type name, interface values are wrapped the same way; in effect, the +interface value acts like a recursive invocation of Encode. + +In summary, a gob stream looks like + + (byteCount (-type id, encoding of a wireType)* (type id, encoding of a value))* + +where * signifies zero or more repetitions and the type id of a value must +be predefined or be defined before the value in the stream. + +Compatibility: Any future changes to the package will endeavor to maintain +compatibility with streams encoded using previous versions. That is, any released +version of this package should be able to decode data written with any previously +released version, subject to issues such as security fixes. See the Go compatibility +document for background: https://golang.org/doc/go1compat + +See "Gobs of data" for a design discussion of the gob wire format: +https://blog.golang.org/gobs-of-data + +# Security + +This package is not designed to be hardened against adversarial inputs, and is +outside the scope of https://go.dev/security/policy. In particular, the [Decoder] +does only basic sanity checking on decoded input sizes, and its limits are not +configurable. Care should be taken when decoding gob data from untrusted +sources, which may consume significant resources. +*/ +package gob + +/* +Grammar: + +Tokens starting with a lower case letter are terminals; int(n) +and uint(n) represent the signed/unsigned encodings of the value n. + +GobStream: + DelimitedMessage* +DelimitedMessage: + uint(lengthOfMessage) Message +Message: + TypeSequence TypedValue +TypeSequence + (TypeDefinition DelimitedTypeDefinition*)? +DelimitedTypeDefinition: + uint(lengthOfTypeDefinition) TypeDefinition +TypedValue: + int(typeId) Value +TypeDefinition: + int(-typeId) encodingOfWireType +Value: + SingletonValue | StructValue +SingletonValue: + uint(0) FieldValue +FieldValue: + builtinValue | ArrayValue | MapValue | SliceValue | StructValue | InterfaceValue +InterfaceValue: + NilInterfaceValue | NonNilInterfaceValue +NilInterfaceValue: + uint(0) +NonNilInterfaceValue: + ConcreteTypeName TypeSequence InterfaceContents +ConcreteTypeName: + uint(lengthOfName) [already read=n] name +InterfaceContents: + int(concreteTypeId) DelimitedValue +DelimitedValue: + uint(length) Value +ArrayValue: + uint(n) FieldValue*n [n elements] +MapValue: + uint(n) (FieldValue FieldValue)*n [n (key, value) pairs] +SliceValue: + uint(n) FieldValue*n [n elements] +StructValue: + (uint(fieldDelta) FieldValue)* +*/ + +/* +For implementers and the curious, here is an encoded example. Given + type Point struct {X, Y int} +and the value + p := Point{22, 33} +the bytes transmitted that encode p will be: + 1f ff 81 03 01 01 05 50 6f 69 6e 74 01 ff 82 00 + 01 02 01 01 58 01 04 00 01 01 59 01 04 00 00 00 + 07 ff 82 01 2c 01 42 00 +They are determined as follows. + +Since this is the first transmission of type Point, the type descriptor +for Point itself must be sent before the value. This is the first type +we've sent on this Encoder, so it has type id 65 (0 through 64 are +reserved). + + 1f // This item (a type descriptor) is 31 bytes long. + ff 81 // The negative of the id for the type we're defining, -65. + // This is one byte (indicated by FF = -1) followed by + // ^-65<<1 | 1. The low 1 bit signals to complement the + // rest upon receipt. + + // Now we send a type descriptor, which is itself a struct (wireType). + // The type of wireType itself is known (it's built in, as is the type of + // all its components), so we just need to send a *value* of type wireType + // that represents type "Point". + // Here starts the encoding of that value. + // Set the field number implicitly to -1; this is done at the beginning + // of every struct, including nested structs. + 03 // Add 3 to field number; now 2 (wireType.structType; this is a struct). + // structType starts with an embedded CommonType, which appears + // as a regular structure here too. + 01 // add 1 to field number (now 0); start of embedded CommonType. + 01 // add 1 to field number (now 0, the name of the type) + 05 // string is (unsigned) 5 bytes long + 50 6f 69 6e 74 // wireType.structType.CommonType.name = "Point" + 01 // add 1 to field number (now 1, the id of the type) + ff 82 // wireType.structType.CommonType._id = 65 + 00 // end of embedded wiretype.structType.CommonType struct + 01 // add 1 to field number (now 1, the field array in wireType.structType) + 02 // There are two fields in the type (len(structType.field)) + 01 // Start of first field structure; add 1 to get field number 0: field[0].name + 01 // 1 byte + 58 // structType.field[0].name = "X" + 01 // Add 1 to get field number 1: field[0].id + 04 // structType.field[0].typeId is 2 (signed int). + 00 // End of structType.field[0]; start structType.field[1]; set field number to -1. + 01 // Add 1 to get field number 0: field[1].name + 01 // 1 byte + 59 // structType.field[1].name = "Y" + 01 // Add 1 to get field number 1: field[1].id + 04 // struct.Type.field[1].typeId is 2 (signed int). + 00 // End of structType.field[1]; end of structType.field. + 00 // end of wireType.structType structure + 00 // end of wireType structure + +Now we can send the Point value. Again the field number resets to -1: + + 07 // this value is 7 bytes long + ff 82 // the type number, 65 (1 byte (-FF) followed by 65<<1) + 01 // add one to field number, yielding field 0 + 2c // encoding of signed "22" (0x2c = 44 = 22<<1); Point.x = 22 + 01 // add one to field number, yielding field 1 + 42 // encoding of signed "33" (0x42 = 66 = 33<<1); Point.y = 33 + 00 // end of structure + +The type encoding is long and fairly intricate but we send it only once. +If p is transmitted a second time, the type is already known so the +output will be just: + + 07 ff 82 01 2c 01 42 00 + +A single non-struct value at top level is transmitted like a field with +delta tag 0. For instance, a signed integer with value 3 presented as +the argument to Encode will emit: + + 03 04 00 06 + +Which represents: + + 03 // this value is 3 bytes long + 04 // the type number, 2, represents an integer + 00 // tag delta 0 + 06 // value 3 + +*/ diff --git a/contrib/go/_std_1.22/src/encoding/gob/enc_helpers.go b/contrib/go/_std_1.22/src/encoding/gob/enc_helpers.go new file mode 100644 index 0000000000..c3b4ca8972 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/gob/enc_helpers.go @@ -0,0 +1,414 @@ +// Code generated by go run encgen.go -output enc_helpers.go; DO NOT EDIT. + +// Copyright 2014 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package gob + +import ( + "reflect" +) + +var encArrayHelper = map[reflect.Kind]encHelper{ + reflect.Bool: encBoolArray, + reflect.Complex64: encComplex64Array, + reflect.Complex128: encComplex128Array, + reflect.Float32: encFloat32Array, + reflect.Float64: encFloat64Array, + reflect.Int: encIntArray, + reflect.Int16: encInt16Array, + reflect.Int32: encInt32Array, + reflect.Int64: encInt64Array, + reflect.Int8: encInt8Array, + reflect.String: encStringArray, + reflect.Uint: encUintArray, + reflect.Uint16: encUint16Array, + reflect.Uint32: encUint32Array, + reflect.Uint64: encUint64Array, + reflect.Uintptr: encUintptrArray, +} + +var encSliceHelper = map[reflect.Kind]encHelper{ + reflect.Bool: encBoolSlice, + reflect.Complex64: encComplex64Slice, + reflect.Complex128: encComplex128Slice, + reflect.Float32: encFloat32Slice, + reflect.Float64: encFloat64Slice, + reflect.Int: encIntSlice, + reflect.Int16: encInt16Slice, + reflect.Int32: encInt32Slice, + reflect.Int64: encInt64Slice, + reflect.Int8: encInt8Slice, + reflect.String: encStringSlice, + reflect.Uint: encUintSlice, + reflect.Uint16: encUint16Slice, + reflect.Uint32: encUint32Slice, + reflect.Uint64: encUint64Slice, + reflect.Uintptr: encUintptrSlice, +} + +func encBoolArray(state *encoderState, v reflect.Value) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return encBoolSlice(state, v.Slice(0, v.Len())) +} + +func encBoolSlice(state *encoderState, v reflect.Value) bool { + slice, ok := v.Interface().([]bool) + if !ok { + // It is kind bool but not type bool. TODO: We can handle this unsafely. + return false + } + for _, x := range slice { + if x != false || state.sendZero { + if x { + state.encodeUint(1) + } else { + state.encodeUint(0) + } + } + } + return true +} + +func encComplex64Array(state *encoderState, v reflect.Value) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return encComplex64Slice(state, v.Slice(0, v.Len())) +} + +func encComplex64Slice(state *encoderState, v reflect.Value) bool { + slice, ok := v.Interface().([]complex64) + if !ok { + // It is kind complex64 but not type complex64. TODO: We can handle this unsafely. + return false + } + for _, x := range slice { + if x != 0+0i || state.sendZero { + rpart := floatBits(float64(real(x))) + ipart := floatBits(float64(imag(x))) + state.encodeUint(rpart) + state.encodeUint(ipart) + } + } + return true +} + +func encComplex128Array(state *encoderState, v reflect.Value) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return encComplex128Slice(state, v.Slice(0, v.Len())) +} + +func encComplex128Slice(state *encoderState, v reflect.Value) bool { + slice, ok := v.Interface().([]complex128) + if !ok { + // It is kind complex128 but not type complex128. TODO: We can handle this unsafely. + return false + } + for _, x := range slice { + if x != 0+0i || state.sendZero { + rpart := floatBits(real(x)) + ipart := floatBits(imag(x)) + state.encodeUint(rpart) + state.encodeUint(ipart) + } + } + return true +} + +func encFloat32Array(state *encoderState, v reflect.Value) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return encFloat32Slice(state, v.Slice(0, v.Len())) +} + +func encFloat32Slice(state *encoderState, v reflect.Value) bool { + slice, ok := v.Interface().([]float32) + if !ok { + // It is kind float32 but not type float32. TODO: We can handle this unsafely. + return false + } + for _, x := range slice { + if x != 0 || state.sendZero { + bits := floatBits(float64(x)) + state.encodeUint(bits) + } + } + return true +} + +func encFloat64Array(state *encoderState, v reflect.Value) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return encFloat64Slice(state, v.Slice(0, v.Len())) +} + +func encFloat64Slice(state *encoderState, v reflect.Value) bool { + slice, ok := v.Interface().([]float64) + if !ok { + // It is kind float64 but not type float64. TODO: We can handle this unsafely. + return false + } + for _, x := range slice { + if x != 0 || state.sendZero { + bits := floatBits(x) + state.encodeUint(bits) + } + } + return true +} + +func encIntArray(state *encoderState, v reflect.Value) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return encIntSlice(state, v.Slice(0, v.Len())) +} + +func encIntSlice(state *encoderState, v reflect.Value) bool { + slice, ok := v.Interface().([]int) + if !ok { + // It is kind int but not type int. TODO: We can handle this unsafely. + return false + } + for _, x := range slice { + if x != 0 || state.sendZero { + state.encodeInt(int64(x)) + } + } + return true +} + +func encInt16Array(state *encoderState, v reflect.Value) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return encInt16Slice(state, v.Slice(0, v.Len())) +} + +func encInt16Slice(state *encoderState, v reflect.Value) bool { + slice, ok := v.Interface().([]int16) + if !ok { + // It is kind int16 but not type int16. TODO: We can handle this unsafely. + return false + } + for _, x := range slice { + if x != 0 || state.sendZero { + state.encodeInt(int64(x)) + } + } + return true +} + +func encInt32Array(state *encoderState, v reflect.Value) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return encInt32Slice(state, v.Slice(0, v.Len())) +} + +func encInt32Slice(state *encoderState, v reflect.Value) bool { + slice, ok := v.Interface().([]int32) + if !ok { + // It is kind int32 but not type int32. TODO: We can handle this unsafely. + return false + } + for _, x := range slice { + if x != 0 || state.sendZero { + state.encodeInt(int64(x)) + } + } + return true +} + +func encInt64Array(state *encoderState, v reflect.Value) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return encInt64Slice(state, v.Slice(0, v.Len())) +} + +func encInt64Slice(state *encoderState, v reflect.Value) bool { + slice, ok := v.Interface().([]int64) + if !ok { + // It is kind int64 but not type int64. TODO: We can handle this unsafely. + return false + } + for _, x := range slice { + if x != 0 || state.sendZero { + state.encodeInt(x) + } + } + return true +} + +func encInt8Array(state *encoderState, v reflect.Value) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return encInt8Slice(state, v.Slice(0, v.Len())) +} + +func encInt8Slice(state *encoderState, v reflect.Value) bool { + slice, ok := v.Interface().([]int8) + if !ok { + // It is kind int8 but not type int8. TODO: We can handle this unsafely. + return false + } + for _, x := range slice { + if x != 0 || state.sendZero { + state.encodeInt(int64(x)) + } + } + return true +} + +func encStringArray(state *encoderState, v reflect.Value) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return encStringSlice(state, v.Slice(0, v.Len())) +} + +func encStringSlice(state *encoderState, v reflect.Value) bool { + slice, ok := v.Interface().([]string) + if !ok { + // It is kind string but not type string. TODO: We can handle this unsafely. + return false + } + for _, x := range slice { + if x != "" || state.sendZero { + state.encodeUint(uint64(len(x))) + state.b.WriteString(x) + } + } + return true +} + +func encUintArray(state *encoderState, v reflect.Value) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return encUintSlice(state, v.Slice(0, v.Len())) +} + +func encUintSlice(state *encoderState, v reflect.Value) bool { + slice, ok := v.Interface().([]uint) + if !ok { + // It is kind uint but not type uint. TODO: We can handle this unsafely. + return false + } + for _, x := range slice { + if x != 0 || state.sendZero { + state.encodeUint(uint64(x)) + } + } + return true +} + +func encUint16Array(state *encoderState, v reflect.Value) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return encUint16Slice(state, v.Slice(0, v.Len())) +} + +func encUint16Slice(state *encoderState, v reflect.Value) bool { + slice, ok := v.Interface().([]uint16) + if !ok { + // It is kind uint16 but not type uint16. TODO: We can handle this unsafely. + return false + } + for _, x := range slice { + if x != 0 || state.sendZero { + state.encodeUint(uint64(x)) + } + } + return true +} + +func encUint32Array(state *encoderState, v reflect.Value) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return encUint32Slice(state, v.Slice(0, v.Len())) +} + +func encUint32Slice(state *encoderState, v reflect.Value) bool { + slice, ok := v.Interface().([]uint32) + if !ok { + // It is kind uint32 but not type uint32. TODO: We can handle this unsafely. + return false + } + for _, x := range slice { + if x != 0 || state.sendZero { + state.encodeUint(uint64(x)) + } + } + return true +} + +func encUint64Array(state *encoderState, v reflect.Value) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return encUint64Slice(state, v.Slice(0, v.Len())) +} + +func encUint64Slice(state *encoderState, v reflect.Value) bool { + slice, ok := v.Interface().([]uint64) + if !ok { + // It is kind uint64 but not type uint64. TODO: We can handle this unsafely. + return false + } + for _, x := range slice { + if x != 0 || state.sendZero { + state.encodeUint(x) + } + } + return true +} + +func encUintptrArray(state *encoderState, v reflect.Value) bool { + // Can only slice if it is addressable. + if !v.CanAddr() { + return false + } + return encUintptrSlice(state, v.Slice(0, v.Len())) +} + +func encUintptrSlice(state *encoderState, v reflect.Value) bool { + slice, ok := v.Interface().([]uintptr) + if !ok { + // It is kind uintptr but not type uintptr. TODO: We can handle this unsafely. + return false + } + for _, x := range slice { + if x != 0 || state.sendZero { + state.encodeUint(uint64(x)) + } + } + return true +} diff --git a/contrib/go/_std_1.22/src/encoding/gob/encode.go b/contrib/go/_std_1.22/src/encoding/gob/encode.go new file mode 100644 index 0000000000..5f4d2539fa --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/gob/encode.go @@ -0,0 +1,670 @@ +// Copyright 2009 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +//go:generate go run encgen.go -output enc_helpers.go + +package gob + +import ( + "encoding" + "encoding/binary" + "math" + "math/bits" + "reflect" + "sync" +) + +const uint64Size = 8 + +type encHelper func(state *encoderState, v reflect.Value) bool + +// encoderState is the global execution state of an instance of the encoder. +// Field numbers are delta encoded and always increase. The field +// number is initialized to -1 so 0 comes out as delta(1). A delta of +// 0 terminates the structure. +type encoderState struct { + enc *Encoder + b *encBuffer + sendZero bool // encoding an array element or map key/value pair; send zero values + fieldnum int // the last field number written. + buf [1 + uint64Size]byte // buffer used by the encoder; here to avoid allocation. + next *encoderState // for free list +} + +// encBuffer is an extremely simple, fast implementation of a write-only byte buffer. +// It never returns a non-nil error, but Write returns an error value so it matches io.Writer. +type encBuffer struct { + data []byte + scratch [64]byte +} + +var encBufferPool = sync.Pool{ + New: func() any { + e := new(encBuffer) + e.data = e.scratch[0:0] + return e + }, +} + +func (e *encBuffer) writeByte(c byte) { + e.data = append(e.data, c) +} + +func (e *encBuffer) Write(p []byte) (int, error) { + e.data = append(e.data, p...) + return len(p), nil +} + +func (e *encBuffer) WriteString(s string) { + e.data = append(e.data, s...) +} + +func (e *encBuffer) Len() int { + return len(e.data) +} + +func (e *encBuffer) Bytes() []byte { + return e.data +} + +func (e *encBuffer) Reset() { + if len(e.data) >= tooBig { + e.data = e.scratch[0:0] + } else { + e.data = e.data[0:0] + } +} + +func (enc *Encoder) newEncoderState(b *encBuffer) *encoderState { + e := enc.freeList + if e == nil { + e = new(encoderState) + e.enc = enc + } else { + enc.freeList = e.next + } + e.sendZero = false + e.fieldnum = 0 + e.b = b + if len(b.data) == 0 { + b.data = b.scratch[0:0] + } + return e +} + +func (enc *Encoder) freeEncoderState(e *encoderState) { + e.next = enc.freeList + enc.freeList = e +} + +// Unsigned integers have a two-state encoding. If the number is less +// than 128 (0 through 0x7F), its value is written directly. +// Otherwise the value is written in big-endian byte order preceded +// by the byte length, negated. + +// encodeUint writes an encoded unsigned integer to state.b. +func (state *encoderState) encodeUint(x uint64) { + if x <= 0x7F { + state.b.writeByte(uint8(x)) + return + } + + binary.BigEndian.PutUint64(state.buf[1:], x) + bc := bits.LeadingZeros64(x) >> 3 // 8 - bytelen(x) + state.buf[bc] = uint8(bc - uint64Size) // and then we subtract 8 to get -bytelen(x) + + state.b.Write(state.buf[bc : uint64Size+1]) +} + +// encodeInt writes an encoded signed integer to state.w. +// The low bit of the encoding says whether to bit complement the (other bits of the) +// uint to recover the int. +func (state *encoderState) encodeInt(i int64) { + var x uint64 + if i < 0 { + x = uint64(^i<<1) | 1 + } else { + x = uint64(i << 1) + } + state.encodeUint(x) +} + +// encOp is the signature of an encoding operator for a given type. +type encOp func(i *encInstr, state *encoderState, v reflect.Value) + +// The 'instructions' of the encoding machine +type encInstr struct { + op encOp + field int // field number in input + index []int // struct index + indir int // how many pointer indirections to reach the value in the struct +} + +// update emits a field number and updates the state to record its value for delta encoding. +// If the instruction pointer is nil, it does nothing +func (state *encoderState) update(instr *encInstr) { + if instr != nil { + state.encodeUint(uint64(instr.field - state.fieldnum)) + state.fieldnum = instr.field + } +} + +// Each encoder for a composite is responsible for handling any +// indirections associated with the elements of the data structure. +// If any pointer so reached is nil, no bytes are written. If the +// data item is zero, no bytes are written. Single values - ints, +// strings etc. - are indirected before calling their encoders. +// Otherwise, the output (for a scalar) is the field number, as an +// encoded integer, followed by the field data in its appropriate +// format. + +// encIndirect dereferences pv indir times and returns the result. +func encIndirect(pv reflect.Value, indir int) reflect.Value { + for ; indir > 0; indir-- { + if pv.IsNil() { + break + } + pv = pv.Elem() + } + return pv +} + +// encBool encodes the bool referenced by v as an unsigned 0 or 1. +func encBool(i *encInstr, state *encoderState, v reflect.Value) { + b := v.Bool() + if b || state.sendZero { + state.update(i) + if b { + state.encodeUint(1) + } else { + state.encodeUint(0) + } + } +} + +// encInt encodes the signed integer (int int8 int16 int32 int64) referenced by v. +func encInt(i *encInstr, state *encoderState, v reflect.Value) { + value := v.Int() + if value != 0 || state.sendZero { + state.update(i) + state.encodeInt(value) + } +} + +// encUint encodes the unsigned integer (uint uint8 uint16 uint32 uint64 uintptr) referenced by v. +func encUint(i *encInstr, state *encoderState, v reflect.Value) { + value := v.Uint() + if value != 0 || state.sendZero { + state.update(i) + state.encodeUint(value) + } +} + +// floatBits returns a uint64 holding the bits of a floating-point number. +// Floating-point numbers are transmitted as uint64s holding the bits +// of the underlying representation. They are sent byte-reversed, with +// the exponent end coming out first, so integer floating point numbers +// (for example) transmit more compactly. This routine does the +// swizzling. +func floatBits(f float64) uint64 { + u := math.Float64bits(f) + return bits.ReverseBytes64(u) +} + +// encFloat encodes the floating point value (float32 float64) referenced by v. +func encFloat(i *encInstr, state *encoderState, v reflect.Value) { + f := v.Float() + if f != 0 || state.sendZero { + bits := floatBits(f) + state.update(i) + state.encodeUint(bits) + } +} + +// encComplex encodes the complex value (complex64 complex128) referenced by v. +// Complex numbers are just a pair of floating-point numbers, real part first. +func encComplex(i *encInstr, state *encoderState, v reflect.Value) { + c := v.Complex() + if c != 0+0i || state.sendZero { + rpart := floatBits(real(c)) + ipart := floatBits(imag(c)) + state.update(i) + state.encodeUint(rpart) + state.encodeUint(ipart) + } +} + +// encUint8Array encodes the byte array referenced by v. +// Byte arrays are encoded as an unsigned count followed by the raw bytes. +func encUint8Array(i *encInstr, state *encoderState, v reflect.Value) { + b := v.Bytes() + if len(b) > 0 || state.sendZero { + state.update(i) + state.encodeUint(uint64(len(b))) + state.b.Write(b) + } +} + +// encString encodes the string referenced by v. +// Strings are encoded as an unsigned count followed by the raw bytes. +func encString(i *encInstr, state *encoderState, v reflect.Value) { + s := v.String() + if len(s) > 0 || state.sendZero { + state.update(i) + state.encodeUint(uint64(len(s))) + state.b.WriteString(s) + } +} + +// encStructTerminator encodes the end of an encoded struct +// as delta field number of 0. +func encStructTerminator(i *encInstr, state *encoderState, v reflect.Value) { + state.encodeUint(0) +} + +// Execution engine + +// encEngine an array of instructions indexed by field number of the encoding +// data, typically a struct. It is executed top to bottom, walking the struct. +type encEngine struct { + instr []encInstr +} + +const singletonField = 0 + +// valid reports whether the value is valid and a non-nil pointer. +// (Slices, maps, and chans take care of themselves.) +func valid(v reflect.Value) bool { + switch v.Kind() { + case reflect.Invalid: + return false + case reflect.Pointer: + return !v.IsNil() + } + return true +} + +// encodeSingle encodes a single top-level non-struct value. +func (enc *Encoder) encodeSingle(b *encBuffer, engine *encEngine, value reflect.Value) { + state := enc.newEncoderState(b) + defer enc.freeEncoderState(state) + state.fieldnum = singletonField + // There is no surrounding struct to frame the transmission, so we must + // generate data even if the item is zero. To do this, set sendZero. + state.sendZero = true + instr := &engine.instr[singletonField] + if instr.indir > 0 { + value = encIndirect(value, instr.indir) + } + if valid(value) { + instr.op(instr, state, value) + } +} + +// encodeStruct encodes a single struct value. +func (enc *Encoder) encodeStruct(b *encBuffer, engine *encEngine, value reflect.Value) { + if !valid(value) { + return + } + state := enc.newEncoderState(b) + defer enc.freeEncoderState(state) + state.fieldnum = -1 + for i := 0; i < len(engine.instr); i++ { + instr := &engine.instr[i] + if i >= value.NumField() { + // encStructTerminator + instr.op(instr, state, reflect.Value{}) + break + } + field := value.FieldByIndex(instr.index) + if instr.indir > 0 { + field = encIndirect(field, instr.indir) + // TODO: Is field guaranteed valid? If so we could avoid this check. + if !valid(field) { + continue + } + } + instr.op(instr, state, field) + } +} + +// encodeArray encodes an array. +func (enc *Encoder) encodeArray(b *encBuffer, value reflect.Value, op encOp, elemIndir int, length int, helper encHelper) { + state := enc.newEncoderState(b) + defer enc.freeEncoderState(state) + state.fieldnum = -1 + state.sendZero = true + state.encodeUint(uint64(length)) + if helper != nil && helper(state, value) { + return + } + for i := 0; i < length; i++ { + elem := value.Index(i) + if elemIndir > 0 { + elem = encIndirect(elem, elemIndir) + // TODO: Is elem guaranteed valid? If so we could avoid this check. + if !valid(elem) { + errorf("encodeArray: nil element") + } + } + op(nil, state, elem) + } +} + +// encodeReflectValue is a helper for maps. It encodes the value v. +func encodeReflectValue(state *encoderState, v reflect.Value, op encOp, indir int) { + for i := 0; i < indir && v.IsValid(); i++ { + v = reflect.Indirect(v) + } + if !v.IsValid() { + errorf("encodeReflectValue: nil element") + } + op(nil, state, v) +} + +// encodeMap encodes a map as unsigned count followed by key:value pairs. +func (enc *Encoder) encodeMap(b *encBuffer, mv reflect.Value, keyOp, elemOp encOp, keyIndir, elemIndir int) { + state := enc.newEncoderState(b) + state.fieldnum = -1 + state.sendZero = true + state.encodeUint(uint64(mv.Len())) + mi := mv.MapRange() + for mi.Next() { + encodeReflectValue(state, mi.Key(), keyOp, keyIndir) + encodeReflectValue(state, mi.Value(), elemOp, elemIndir) + } + enc.freeEncoderState(state) +} + +// encodeInterface encodes the interface value iv. +// To send an interface, we send a string identifying the concrete type, followed +// by the type identifier (which might require defining that type right now), followed +// by the concrete value. A nil value gets sent as the empty string for the name, +// followed by no value. +func (enc *Encoder) encodeInterface(b *encBuffer, iv reflect.Value) { + // Gobs can encode nil interface values but not typed interface + // values holding nil pointers, since nil pointers point to no value. + elem := iv.Elem() + if elem.Kind() == reflect.Pointer && elem.IsNil() { + errorf("gob: cannot encode nil pointer of type %s inside interface", iv.Elem().Type()) + } + state := enc.newEncoderState(b) + state.fieldnum = -1 + state.sendZero = true + if iv.IsNil() { + state.encodeUint(0) + return + } + + ut := userType(iv.Elem().Type()) + namei, ok := concreteTypeToName.Load(ut.base) + if !ok { + errorf("type not registered for interface: %s", ut.base) + } + name := namei.(string) + + // Send the name. + state.encodeUint(uint64(len(name))) + state.b.WriteString(name) + // Define the type id if necessary. + enc.sendTypeDescriptor(enc.writer(), state, ut) + // Send the type id. + enc.sendTypeId(state, ut) + // Encode the value into a new buffer. Any nested type definitions + // should be written to b, before the encoded value. + enc.pushWriter(b) + data := encBufferPool.Get().(*encBuffer) + data.Write(spaceForLength) + enc.encode(data, elem, ut) + if enc.err != nil { + error_(enc.err) + } + enc.popWriter() + enc.writeMessage(b, data) + data.Reset() + encBufferPool.Put(data) + if enc.err != nil { + error_(enc.err) + } + enc.freeEncoderState(state) +} + +// encodeGobEncoder encodes a value that implements the GobEncoder interface. +// The data is sent as a byte array. +func (enc *Encoder) encodeGobEncoder(b *encBuffer, ut *userTypeInfo, v reflect.Value) { + // TODO: should we catch panics from the called method? + + var data []byte + var err error + // We know it's one of these. + switch ut.externalEnc { + case xGob: + data, err = v.Interface().(GobEncoder).GobEncode() + case xBinary: + data, err = v.Interface().(encoding.BinaryMarshaler).MarshalBinary() + case xText: + data, err = v.Interface().(encoding.TextMarshaler).MarshalText() + } + if err != nil { + error_(err) + } + state := enc.newEncoderState(b) + state.fieldnum = -1 + state.encodeUint(uint64(len(data))) + state.b.Write(data) + enc.freeEncoderState(state) +} + +var encOpTable = [...]encOp{ + reflect.Bool: encBool, + reflect.Int: encInt, + reflect.Int8: encInt, + reflect.Int16: encInt, + reflect.Int32: encInt, + reflect.Int64: encInt, + reflect.Uint: encUint, + reflect.Uint8: encUint, + reflect.Uint16: encUint, + reflect.Uint32: encUint, + reflect.Uint64: encUint, + reflect.Uintptr: encUint, + reflect.Float32: encFloat, + reflect.Float64: encFloat, + reflect.Complex64: encComplex, + reflect.Complex128: encComplex, + reflect.String: encString, +} + +// encOpFor returns (a pointer to) the encoding op for the base type under rt and +// the indirection count to reach it. +func encOpFor(rt reflect.Type, inProgress map[reflect.Type]*encOp, building map[*typeInfo]bool) (*encOp, int) { + ut := userType(rt) + // If the type implements GobEncoder, we handle it without further processing. + if ut.externalEnc != 0 { + return gobEncodeOpFor(ut) + } + // If this type is already in progress, it's a recursive type (e.g. map[string]*T). + // Return the pointer to the op we're already building. + if opPtr := inProgress[rt]; opPtr != nil { + return opPtr, ut.indir + } + typ := ut.base + indir := ut.indir + k := typ.Kind() + var op encOp + if int(k) < len(encOpTable) { + op = encOpTable[k] + } + if op == nil { + inProgress[rt] = &op + // Special cases + switch t := typ; t.Kind() { + case reflect.Slice: + if t.Elem().Kind() == reflect.Uint8 { + op = encUint8Array + break + } + // Slices have a header; we decode it to find the underlying array. + elemOp, elemIndir := encOpFor(t.Elem(), inProgress, building) + helper := encSliceHelper[t.Elem().Kind()] + op = func(i *encInstr, state *encoderState, slice reflect.Value) { + if !state.sendZero && slice.Len() == 0 { + return + } + state.update(i) + state.enc.encodeArray(state.b, slice, *elemOp, elemIndir, slice.Len(), helper) + } + case reflect.Array: + // True arrays have size in the type. + elemOp, elemIndir := encOpFor(t.Elem(), inProgress, building) + helper := encArrayHelper[t.Elem().Kind()] + op = func(i *encInstr, state *encoderState, array reflect.Value) { + state.update(i) + state.enc.encodeArray(state.b, array, *elemOp, elemIndir, array.Len(), helper) + } + case reflect.Map: + keyOp, keyIndir := encOpFor(t.Key(), inProgress, building) + elemOp, elemIndir := encOpFor(t.Elem(), inProgress, building) + op = func(i *encInstr, state *encoderState, mv reflect.Value) { + // We send zero-length (but non-nil) maps because the + // receiver might want to use the map. (Maps don't use append.) + if !state.sendZero && mv.IsNil() { + return + } + state.update(i) + state.enc.encodeMap(state.b, mv, *keyOp, *elemOp, keyIndir, elemIndir) + } + case reflect.Struct: + // Generate a closure that calls out to the engine for the nested type. + getEncEngine(userType(typ), building) + info := mustGetTypeInfo(typ) + op = func(i *encInstr, state *encoderState, sv reflect.Value) { + state.update(i) + // indirect through info to delay evaluation for recursive structs + enc := info.encoder.Load() + state.enc.encodeStruct(state.b, enc, sv) + } + case reflect.Interface: + op = func(i *encInstr, state *encoderState, iv reflect.Value) { + if !state.sendZero && (!iv.IsValid() || iv.IsNil()) { + return + } + state.update(i) + state.enc.encodeInterface(state.b, iv) + } + } + } + if op == nil { + errorf("can't happen: encode type %s", rt) + } + return &op, indir +} + +// gobEncodeOpFor returns the op for a type that is known to implement GobEncoder. +func gobEncodeOpFor(ut *userTypeInfo) (*encOp, int) { + rt := ut.user + if ut.encIndir == -1 { + rt = reflect.PointerTo(rt) + } else if ut.encIndir > 0 { + for i := int8(0); i < ut.encIndir; i++ { + rt = rt.Elem() + } + } + var op encOp + op = func(i *encInstr, state *encoderState, v reflect.Value) { + if ut.encIndir == -1 { + // Need to climb up one level to turn value into pointer. + if !v.CanAddr() { + errorf("unaddressable value of type %s", rt) + } + v = v.Addr() + } + if !state.sendZero && v.IsZero() { + return + } + state.update(i) + state.enc.encodeGobEncoder(state.b, ut, v) + } + return &op, int(ut.encIndir) // encIndir: op will get called with p == address of receiver. +} + +// compileEnc returns the engine to compile the type. +func compileEnc(ut *userTypeInfo, building map[*typeInfo]bool) *encEngine { + srt := ut.base + engine := new(encEngine) + seen := make(map[reflect.Type]*encOp) + rt := ut.base + if ut.externalEnc != 0 { + rt = ut.user + } + if ut.externalEnc == 0 && srt.Kind() == reflect.Struct { + for fieldNum, wireFieldNum := 0, 0; fieldNum < srt.NumField(); fieldNum++ { + f := srt.Field(fieldNum) + if !isSent(&f) { + continue + } + op, indir := encOpFor(f.Type, seen, building) + engine.instr = append(engine.instr, encInstr{*op, wireFieldNum, f.Index, indir}) + wireFieldNum++ + } + if srt.NumField() > 0 && len(engine.instr) == 0 { + errorf("type %s has no exported fields", rt) + } + engine.instr = append(engine.instr, encInstr{encStructTerminator, 0, nil, 0}) + } else { + engine.instr = make([]encInstr, 1) + op, indir := encOpFor(rt, seen, building) + engine.instr[0] = encInstr{*op, singletonField, nil, indir} + } + return engine +} + +// getEncEngine returns the engine to compile the type. +func getEncEngine(ut *userTypeInfo, building map[*typeInfo]bool) *encEngine { + info, err := getTypeInfo(ut) + if err != nil { + error_(err) + } + enc := info.encoder.Load() + if enc == nil { + enc = buildEncEngine(info, ut, building) + } + return enc +} + +func buildEncEngine(info *typeInfo, ut *userTypeInfo, building map[*typeInfo]bool) *encEngine { + // Check for recursive types. + if building != nil && building[info] { + return nil + } + info.encInit.Lock() + defer info.encInit.Unlock() + enc := info.encoder.Load() + if enc == nil { + if building == nil { + building = make(map[*typeInfo]bool) + } + building[info] = true + enc = compileEnc(ut, building) + info.encoder.Store(enc) + } + return enc +} + +func (enc *Encoder) encode(b *encBuffer, value reflect.Value, ut *userTypeInfo) { + defer catchError(&enc.err) + engine := getEncEngine(ut, nil) + indir := ut.indir + if ut.externalEnc != 0 { + indir = int(ut.encIndir) + } + for i := 0; i < indir; i++ { + value = reflect.Indirect(value) + } + if ut.externalEnc == 0 && value.Type().Kind() == reflect.Struct { + enc.encodeStruct(b, engine, value) + } else { + enc.encodeSingle(b, engine, value) + } +} diff --git a/contrib/go/_std_1.22/src/encoding/gob/encoder.go b/contrib/go/_std_1.22/src/encoding/gob/encoder.go new file mode 100644 index 0000000000..7d46152aba --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/gob/encoder.go @@ -0,0 +1,258 @@ +// Copyright 2009 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package gob + +import ( + "errors" + "io" + "reflect" + "sync" +) + +// An Encoder manages the transmission of type and data information to the +// other side of a connection. It is safe for concurrent use by multiple +// goroutines. +type Encoder struct { + mutex sync.Mutex // each item must be sent atomically + w []io.Writer // where to send the data + sent map[reflect.Type]typeId // which types we've already sent + countState *encoderState // stage for writing counts + freeList *encoderState // list of free encoderStates; avoids reallocation + byteBuf encBuffer // buffer for top-level encoderState + err error +} + +// Before we encode a message, we reserve space at the head of the +// buffer in which to encode its length. This means we can use the +// buffer to assemble the message without another allocation. +const maxLength = 9 // Maximum size of an encoded length. +var spaceForLength = make([]byte, maxLength) + +// NewEncoder returns a new encoder that will transmit on the [io.Writer]. +func NewEncoder(w io.Writer) *Encoder { + enc := new(Encoder) + enc.w = []io.Writer{w} + enc.sent = make(map[reflect.Type]typeId) + enc.countState = enc.newEncoderState(new(encBuffer)) + return enc +} + +// writer returns the innermost writer the encoder is using. +func (enc *Encoder) writer() io.Writer { + return enc.w[len(enc.w)-1] +} + +// pushWriter adds a writer to the encoder. +func (enc *Encoder) pushWriter(w io.Writer) { + enc.w = append(enc.w, w) +} + +// popWriter pops the innermost writer. +func (enc *Encoder) popWriter() { + enc.w = enc.w[0 : len(enc.w)-1] +} + +func (enc *Encoder) setError(err error) { + if enc.err == nil { // remember the first. + enc.err = err + } +} + +// writeMessage sends the data item preceded by an unsigned count of its length. +func (enc *Encoder) writeMessage(w io.Writer, b *encBuffer) { + // Space has been reserved for the length at the head of the message. + // This is a little dirty: we grab the slice from the bytes.Buffer and massage + // it by hand. + message := b.Bytes() + messageLen := len(message) - maxLength + // Length cannot be bigger than the decoder can handle. + if messageLen >= tooBig { + enc.setError(errors.New("gob: encoder: message too big")) + return + } + // Encode the length. + enc.countState.b.Reset() + enc.countState.encodeUint(uint64(messageLen)) + // Copy the length to be a prefix of the message. + offset := maxLength - enc.countState.b.Len() + copy(message[offset:], enc.countState.b.Bytes()) + // Write the data. + _, err := w.Write(message[offset:]) + // Drain the buffer and restore the space at the front for the count of the next message. + b.Reset() + b.Write(spaceForLength) + if err != nil { + enc.setError(err) + } +} + +// sendActualType sends the requested type, without further investigation, unless +// it's been sent before. +func (enc *Encoder) sendActualType(w io.Writer, state *encoderState, ut *userTypeInfo, actual reflect.Type) (sent bool) { + if _, alreadySent := enc.sent[actual]; alreadySent { + return false + } + info, err := getTypeInfo(ut) + if err != nil { + enc.setError(err) + return + } + // Send the pair (-id, type) + // Id: + state.encodeInt(-int64(info.id)) + // Type: + enc.encode(state.b, reflect.ValueOf(info.wire), wireTypeUserInfo) + enc.writeMessage(w, state.b) + if enc.err != nil { + return + } + + // Remember we've sent this type, both what the user gave us and the base type. + enc.sent[ut.base] = info.id + if ut.user != ut.base { + enc.sent[ut.user] = info.id + } + // Now send the inner types + switch st := actual; st.Kind() { + case reflect.Struct: + for i := 0; i < st.NumField(); i++ { + if isExported(st.Field(i).Name) { + enc.sendType(w, state, st.Field(i).Type) + } + } + case reflect.Array, reflect.Slice: + enc.sendType(w, state, st.Elem()) + case reflect.Map: + enc.sendType(w, state, st.Key()) + enc.sendType(w, state, st.Elem()) + } + return true +} + +// sendType sends the type info to the other side, if necessary. +func (enc *Encoder) sendType(w io.Writer, state *encoderState, origt reflect.Type) (sent bool) { + ut := userType(origt) + if ut.externalEnc != 0 { + // The rules are different: regardless of the underlying type's representation, + // we need to tell the other side that the base type is a GobEncoder. + return enc.sendActualType(w, state, ut, ut.base) + } + + // It's a concrete value, so drill down to the base type. + switch rt := ut.base; rt.Kind() { + default: + // Basic types and interfaces do not need to be described. + return + case reflect.Slice: + // If it's []uint8, don't send; it's considered basic. + if rt.Elem().Kind() == reflect.Uint8 { + return + } + // Otherwise we do send. + break + case reflect.Array: + // arrays must be sent so we know their lengths and element types. + break + case reflect.Map: + // maps must be sent so we know their lengths and key/value types. + break + case reflect.Struct: + // structs must be sent so we know their fields. + break + case reflect.Chan, reflect.Func: + // If we get here, it's a field of a struct; ignore it. + return + } + + return enc.sendActualType(w, state, ut, ut.base) +} + +// Encode transmits the data item represented by the empty interface value, +// guaranteeing that all necessary type information has been transmitted first. +// Passing a nil pointer to Encoder will panic, as they cannot be transmitted by gob. +func (enc *Encoder) Encode(e any) error { + return enc.EncodeValue(reflect.ValueOf(e)) +} + +// sendTypeDescriptor makes sure the remote side knows about this type. +// It will send a descriptor if this is the first time the type has been +// sent. +func (enc *Encoder) sendTypeDescriptor(w io.Writer, state *encoderState, ut *userTypeInfo) { + // Make sure the type is known to the other side. + // First, have we already sent this type? + rt := ut.base + if ut.externalEnc != 0 { + rt = ut.user + } + if _, alreadySent := enc.sent[rt]; !alreadySent { + // No, so send it. + sent := enc.sendType(w, state, rt) + if enc.err != nil { + return + } + // If the type info has still not been transmitted, it means we have + // a singleton basic type (int, []byte etc.) at top level. We don't + // need to send the type info but we do need to update enc.sent. + if !sent { + info, err := getTypeInfo(ut) + if err != nil { + enc.setError(err) + return + } + enc.sent[rt] = info.id + } + } +} + +// sendTypeId sends the id, which must have already been defined. +func (enc *Encoder) sendTypeId(state *encoderState, ut *userTypeInfo) { + // Identify the type of this top-level value. + state.encodeInt(int64(enc.sent[ut.base])) +} + +// EncodeValue transmits the data item represented by the reflection value, +// guaranteeing that all necessary type information has been transmitted first. +// Passing a nil pointer to EncodeValue will panic, as they cannot be transmitted by gob. +func (enc *Encoder) EncodeValue(value reflect.Value) error { + if value.Kind() == reflect.Invalid { + return errors.New("gob: cannot encode nil value") + } + if value.Kind() == reflect.Pointer && value.IsNil() { + panic("gob: cannot encode nil pointer of type " + value.Type().String()) + } + + // Make sure we're single-threaded through here, so multiple + // goroutines can share an encoder. + enc.mutex.Lock() + defer enc.mutex.Unlock() + + // Remove any nested writers remaining due to previous errors. + enc.w = enc.w[0:1] + + ut, err := validUserType(value.Type()) + if err != nil { + return err + } + + enc.err = nil + enc.byteBuf.Reset() + enc.byteBuf.Write(spaceForLength) + state := enc.newEncoderState(&enc.byteBuf) + + enc.sendTypeDescriptor(enc.writer(), state, ut) + enc.sendTypeId(state, ut) + if enc.err != nil { + return enc.err + } + + // Encode the object. + enc.encode(state.b, value, ut) + if enc.err == nil { + enc.writeMessage(enc.writer(), state.b) + } + + enc.freeEncoderState(state) + return enc.err +} diff --git a/contrib/go/_std_1.22/src/encoding/gob/error.go b/contrib/go/_std_1.22/src/encoding/gob/error.go new file mode 100644 index 0000000000..9c614e3e3f --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/gob/error.go @@ -0,0 +1,42 @@ +// Copyright 2009 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package gob + +import "fmt" + +// Errors in decoding and encoding are handled using panic and recover. +// Panics caused by user error (that is, everything except run-time panics +// such as "index out of bounds" errors) do not leave the file that caused +// them, but are instead turned into plain error returns. Encoding and +// decoding functions and methods that do not return an error either use +// panic to report an error or are guaranteed error-free. + +// A gobError is used to distinguish errors (panics) generated in this package. +type gobError struct { + err error +} + +// errorf is like error_ but takes Printf-style arguments to construct an error. +// It always prefixes the message with "gob: ". +func errorf(format string, args ...any) { + error_(fmt.Errorf("gob: "+format, args...)) +} + +// error_ wraps the argument error and uses it as the argument to panic. +func error_(err error) { + panic(gobError{err}) +} + +// catchError is meant to be used as a deferred function to turn a panic(gobError) into a +// plain error. It overwrites the error return of the function that deferred its call. +func catchError(err *error) { + if e := recover(); e != nil { + ge, ok := e.(gobError) + if !ok { + panic(e) + } + *err = ge.err + } +} diff --git a/contrib/go/_std_1.22/src/encoding/gob/type.go b/contrib/go/_std_1.22/src/encoding/gob/type.go new file mode 100644 index 0000000000..30d8ca61c4 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/gob/type.go @@ -0,0 +1,944 @@ +// Copyright 2009 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package gob + +import ( + "encoding" + "errors" + "fmt" + "os" + "reflect" + "sync" + "sync/atomic" + "unicode" + "unicode/utf8" +) + +// userTypeInfo stores the information associated with a type the user has handed +// to the package. It's computed once and stored in a map keyed by reflection +// type. +type userTypeInfo struct { + user reflect.Type // the type the user handed us + base reflect.Type // the base type after all indirections + indir int // number of indirections to reach the base type + externalEnc int // xGob, xBinary, or xText + externalDec int // xGob, xBinary, or xText + encIndir int8 // number of indirections to reach the receiver type; may be negative + decIndir int8 // number of indirections to reach the receiver type; may be negative +} + +// externalEncoding bits +const ( + xGob = 1 + iota // GobEncoder or GobDecoder + xBinary // encoding.BinaryMarshaler or encoding.BinaryUnmarshaler + xText // encoding.TextMarshaler or encoding.TextUnmarshaler +) + +var userTypeCache sync.Map // map[reflect.Type]*userTypeInfo + +// validUserType returns, and saves, the information associated with user-provided type rt. +// If the user type is not valid, err will be non-nil. To be used when the error handler +// is not set up. +func validUserType(rt reflect.Type) (*userTypeInfo, error) { + if ui, ok := userTypeCache.Load(rt); ok { + return ui.(*userTypeInfo), nil + } + + // Construct a new userTypeInfo and atomically add it to the userTypeCache. + // If we lose the race, we'll waste a little CPU and create a little garbage + // but return the existing value anyway. + + ut := new(userTypeInfo) + ut.base = rt + ut.user = rt + // A type that is just a cycle of pointers (such as type T *T) cannot + // be represented in gobs, which need some concrete data. We use a + // cycle detection algorithm from Knuth, Vol 2, Section 3.1, Ex 6, + // pp 539-540. As we step through indirections, run another type at + // half speed. If they meet up, there's a cycle. + slowpoke := ut.base // walks half as fast as ut.base + for { + pt := ut.base + if pt.Kind() != reflect.Pointer { + break + } + ut.base = pt.Elem() + if ut.base == slowpoke { // ut.base lapped slowpoke + // recursive pointer type. + return nil, errors.New("can't represent recursive pointer type " + ut.base.String()) + } + if ut.indir%2 == 0 { + slowpoke = slowpoke.Elem() + } + ut.indir++ + } + + if ok, indir := implementsInterface(ut.user, gobEncoderInterfaceType); ok { + ut.externalEnc, ut.encIndir = xGob, indir + } else if ok, indir := implementsInterface(ut.user, binaryMarshalerInterfaceType); ok { + ut.externalEnc, ut.encIndir = xBinary, indir + } + + // NOTE(rsc): Would like to allow MarshalText here, but results in incompatibility + // with older encodings for net.IP. See golang.org/issue/6760. + // } else if ok, indir := implementsInterface(ut.user, textMarshalerInterfaceType); ok { + // ut.externalEnc, ut.encIndir = xText, indir + // } + + if ok, indir := implementsInterface(ut.user, gobDecoderInterfaceType); ok { + ut.externalDec, ut.decIndir = xGob, indir + } else if ok, indir := implementsInterface(ut.user, binaryUnmarshalerInterfaceType); ok { + ut.externalDec, ut.decIndir = xBinary, indir + } + + // See note above. + // } else if ok, indir := implementsInterface(ut.user, textUnmarshalerInterfaceType); ok { + // ut.externalDec, ut.decIndir = xText, indir + // } + + ui, _ := userTypeCache.LoadOrStore(rt, ut) + return ui.(*userTypeInfo), nil +} + +var ( + gobEncoderInterfaceType = reflect.TypeFor[GobEncoder]() + gobDecoderInterfaceType = reflect.TypeFor[GobDecoder]() + binaryMarshalerInterfaceType = reflect.TypeFor[encoding.BinaryMarshaler]() + binaryUnmarshalerInterfaceType = reflect.TypeFor[encoding.BinaryUnmarshaler]() + textMarshalerInterfaceType = reflect.TypeFor[encoding.TextMarshaler]() + textUnmarshalerInterfaceType = reflect.TypeFor[encoding.TextUnmarshaler]() + + wireTypeType = reflect.TypeFor[wireType]() +) + +// implementsInterface reports whether the type implements the +// gobEncoder/gobDecoder interface. +// It also returns the number of indirections required to get to the +// implementation. +func implementsInterface(typ, gobEncDecType reflect.Type) (success bool, indir int8) { + if typ == nil { + return + } + rt := typ + // The type might be a pointer and we need to keep + // dereferencing to the base type until we find an implementation. + for { + if rt.Implements(gobEncDecType) { + return true, indir + } + if p := rt; p.Kind() == reflect.Pointer { + indir++ + if indir > 100 { // insane number of indirections + return false, 0 + } + rt = p.Elem() + continue + } + break + } + // No luck yet, but if this is a base type (non-pointer), the pointer might satisfy. + if typ.Kind() != reflect.Pointer { + // Not a pointer, but does the pointer work? + if reflect.PointerTo(typ).Implements(gobEncDecType) { + return true, -1 + } + } + return false, 0 +} + +// userType returns, and saves, the information associated with user-provided type rt. +// If the user type is not valid, it calls error. +func userType(rt reflect.Type) *userTypeInfo { + ut, err := validUserType(rt) + if err != nil { + error_(err) + } + return ut +} + +// A typeId represents a gob Type as an integer that can be passed on the wire. +// Internally, typeIds are used as keys to a map to recover the underlying type info. +type typeId int32 + +var typeLock sync.Mutex // set while building a type +const firstUserId = 64 // lowest id number granted to user + +type gobType interface { + id() typeId + setId(id typeId) + name() string + string() string // not public; only for debugging + safeString(seen map[typeId]bool) string +} + +var ( + types = make(map[reflect.Type]gobType, 32) + idToTypeSlice = make([]gobType, 1, firstUserId) + builtinIdToTypeSlice [firstUserId]gobType // set in init() after builtins are established +) + +func idToType(id typeId) gobType { + if id < 0 || int(id) >= len(idToTypeSlice) { + return nil + } + return idToTypeSlice[id] +} + +func builtinIdToType(id typeId) gobType { + if id < 0 || int(id) >= len(builtinIdToTypeSlice) { + return nil + } + return builtinIdToTypeSlice[id] +} + +func setTypeId(typ gobType) { + // When building recursive types, someone may get there before us. + if typ.id() != 0 { + return + } + nextId := typeId(len(idToTypeSlice)) + typ.setId(nextId) + idToTypeSlice = append(idToTypeSlice, typ) +} + +func (t typeId) gobType() gobType { + if t == 0 { + return nil + } + return idToType(t) +} + +// string returns the string representation of the type associated with the typeId. +func (t typeId) string() string { + if t.gobType() == nil { + return "<nil>" + } + return t.gobType().string() +} + +// Name returns the name of the type associated with the typeId. +func (t typeId) name() string { + if t.gobType() == nil { + return "<nil>" + } + return t.gobType().name() +} + +// CommonType holds elements of all types. +// It is a historical artifact, kept for binary compatibility and exported +// only for the benefit of the package's encoding of type descriptors. It is +// not intended for direct use by clients. +type CommonType struct { + Name string + Id typeId +} + +func (t *CommonType) id() typeId { return t.Id } + +func (t *CommonType) setId(id typeId) { t.Id = id } + +func (t *CommonType) string() string { return t.Name } + +func (t *CommonType) safeString(seen map[typeId]bool) string { + return t.Name +} + +func (t *CommonType) name() string { return t.Name } + +// Create and check predefined types +// The string for tBytes is "bytes" not "[]byte" to signify its specialness. + +var ( + // Primordial types, needed during initialization. + // Always passed as pointers so the interface{} type + // goes through without losing its interfaceness. + tBool = bootstrapType("bool", (*bool)(nil)) + tInt = bootstrapType("int", (*int)(nil)) + tUint = bootstrapType("uint", (*uint)(nil)) + tFloat = bootstrapType("float", (*float64)(nil)) + tBytes = bootstrapType("bytes", (*[]byte)(nil)) + tString = bootstrapType("string", (*string)(nil)) + tComplex = bootstrapType("complex", (*complex128)(nil)) + tInterface = bootstrapType("interface", (*any)(nil)) + // Reserve some Ids for compatible expansion + tReserved7 = bootstrapType("_reserved1", (*struct{ r7 int })(nil)) + tReserved6 = bootstrapType("_reserved1", (*struct{ r6 int })(nil)) + tReserved5 = bootstrapType("_reserved1", (*struct{ r5 int })(nil)) + tReserved4 = bootstrapType("_reserved1", (*struct{ r4 int })(nil)) + tReserved3 = bootstrapType("_reserved1", (*struct{ r3 int })(nil)) + tReserved2 = bootstrapType("_reserved1", (*struct{ r2 int })(nil)) + tReserved1 = bootstrapType("_reserved1", (*struct{ r1 int })(nil)) +) + +// Predefined because it's needed by the Decoder +var tWireType = mustGetTypeInfo(wireTypeType).id +var wireTypeUserInfo *userTypeInfo // userTypeInfo of wireType + +func init() { + // Some magic numbers to make sure there are no surprises. + checkId(16, tWireType) + checkId(17, mustGetTypeInfo(reflect.TypeFor[arrayType]()).id) + checkId(18, mustGetTypeInfo(reflect.TypeFor[CommonType]()).id) + checkId(19, mustGetTypeInfo(reflect.TypeFor[sliceType]()).id) + checkId(20, mustGetTypeInfo(reflect.TypeFor[structType]()).id) + checkId(21, mustGetTypeInfo(reflect.TypeFor[fieldType]()).id) + checkId(23, mustGetTypeInfo(reflect.TypeFor[mapType]()).id) + + copy(builtinIdToTypeSlice[:], idToTypeSlice) + + // Move the id space upwards to allow for growth in the predefined world + // without breaking existing files. + if nextId := len(idToTypeSlice); nextId > firstUserId { + panic(fmt.Sprintln("nextId too large:", nextId)) + } + idToTypeSlice = idToTypeSlice[:firstUserId] + registerBasics() + wireTypeUserInfo = userType(wireTypeType) +} + +// Array type +type arrayType struct { + CommonType + Elem typeId + Len int +} + +func newArrayType(name string) *arrayType { + a := &arrayType{CommonType{Name: name}, 0, 0} + return a +} + +func (a *arrayType) init(elem gobType, len int) { + // Set our type id before evaluating the element's, in case it's our own. + setTypeId(a) + a.Elem = elem.id() + a.Len = len +} + +func (a *arrayType) safeString(seen map[typeId]bool) string { + if seen[a.Id] { + return a.Name + } + seen[a.Id] = true + return fmt.Sprintf("[%d]%s", a.Len, a.Elem.gobType().safeString(seen)) +} + +func (a *arrayType) string() string { return a.safeString(make(map[typeId]bool)) } + +// GobEncoder type (something that implements the GobEncoder interface) +type gobEncoderType struct { + CommonType +} + +func newGobEncoderType(name string) *gobEncoderType { + g := &gobEncoderType{CommonType{Name: name}} + setTypeId(g) + return g +} + +func (g *gobEncoderType) safeString(seen map[typeId]bool) string { + return g.Name +} + +func (g *gobEncoderType) string() string { return g.Name } + +// Map type +type mapType struct { + CommonType + Key typeId + Elem typeId +} + +func newMapType(name string) *mapType { + m := &mapType{CommonType{Name: name}, 0, 0} + return m +} + +func (m *mapType) init(key, elem gobType) { + // Set our type id before evaluating the element's, in case it's our own. + setTypeId(m) + m.Key = key.id() + m.Elem = elem.id() +} + +func (m *mapType) safeString(seen map[typeId]bool) string { + if seen[m.Id] { + return m.Name + } + seen[m.Id] = true + key := m.Key.gobType().safeString(seen) + elem := m.Elem.gobType().safeString(seen) + return fmt.Sprintf("map[%s]%s", key, elem) +} + +func (m *mapType) string() string { return m.safeString(make(map[typeId]bool)) } + +// Slice type +type sliceType struct { + CommonType + Elem typeId +} + +func newSliceType(name string) *sliceType { + s := &sliceType{CommonType{Name: name}, 0} + return s +} + +func (s *sliceType) init(elem gobType) { + // Set our type id before evaluating the element's, in case it's our own. + setTypeId(s) + // See the comments about ids in newTypeObject. Only slices and + // structs have mutual recursion. + if elem.id() == 0 { + setTypeId(elem) + } + s.Elem = elem.id() +} + +func (s *sliceType) safeString(seen map[typeId]bool) string { + if seen[s.Id] { + return s.Name + } + seen[s.Id] = true + return fmt.Sprintf("[]%s", s.Elem.gobType().safeString(seen)) +} + +func (s *sliceType) string() string { return s.safeString(make(map[typeId]bool)) } + +// Struct type +type fieldType struct { + Name string + Id typeId +} + +type structType struct { + CommonType + Field []fieldType +} + +func (s *structType) safeString(seen map[typeId]bool) string { + if s == nil { + return "<nil>" + } + if _, ok := seen[s.Id]; ok { + return s.Name + } + seen[s.Id] = true + str := s.Name + " = struct { " + for _, f := range s.Field { + str += fmt.Sprintf("%s %s; ", f.Name, f.Id.gobType().safeString(seen)) + } + str += "}" + return str +} + +func (s *structType) string() string { return s.safeString(make(map[typeId]bool)) } + +func newStructType(name string) *structType { + s := &structType{CommonType{Name: name}, nil} + // For historical reasons we set the id here rather than init. + // See the comment in newTypeObject for details. + setTypeId(s) + return s +} + +// newTypeObject allocates a gobType for the reflection type rt. +// Unless ut represents a GobEncoder, rt should be the base type +// of ut. +// This is only called from the encoding side. The decoding side +// works through typeIds and userTypeInfos alone. +func newTypeObject(name string, ut *userTypeInfo, rt reflect.Type) (gobType, error) { + // Does this type implement GobEncoder? + if ut.externalEnc != 0 { + return newGobEncoderType(name), nil + } + var err error + var type0, type1 gobType + defer func() { + if err != nil { + delete(types, rt) + } + }() + // Install the top-level type before the subtypes (e.g. struct before + // fields) so recursive types can be constructed safely. + switch t := rt; t.Kind() { + // All basic types are easy: they are predefined. + case reflect.Bool: + return tBool.gobType(), nil + + case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64: + return tInt.gobType(), nil + + case reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr: + return tUint.gobType(), nil + + case reflect.Float32, reflect.Float64: + return tFloat.gobType(), nil + + case reflect.Complex64, reflect.Complex128: + return tComplex.gobType(), nil + + case reflect.String: + return tString.gobType(), nil + + case reflect.Interface: + return tInterface.gobType(), nil + + case reflect.Array: + at := newArrayType(name) + types[rt] = at + type0, err = getBaseType("", t.Elem()) + if err != nil { + return nil, err + } + // Historical aside: + // For arrays, maps, and slices, we set the type id after the elements + // are constructed. This is to retain the order of type id allocation after + // a fix made to handle recursive types, which changed the order in + // which types are built. Delaying the setting in this way preserves + // type ids while allowing recursive types to be described. Structs, + // done below, were already handling recursion correctly so they + // assign the top-level id before those of the field. + at.init(type0, t.Len()) + return at, nil + + case reflect.Map: + mt := newMapType(name) + types[rt] = mt + type0, err = getBaseType("", t.Key()) + if err != nil { + return nil, err + } + type1, err = getBaseType("", t.Elem()) + if err != nil { + return nil, err + } + mt.init(type0, type1) + return mt, nil + + case reflect.Slice: + // []byte == []uint8 is a special case + if t.Elem().Kind() == reflect.Uint8 { + return tBytes.gobType(), nil + } + st := newSliceType(name) + types[rt] = st + type0, err = getBaseType(t.Elem().Name(), t.Elem()) + if err != nil { + return nil, err + } + st.init(type0) + return st, nil + + case reflect.Struct: + st := newStructType(name) + types[rt] = st + idToTypeSlice[st.id()] = st + for i := 0; i < t.NumField(); i++ { + f := t.Field(i) + if !isSent(&f) { + continue + } + typ := userType(f.Type).base + tname := typ.Name() + if tname == "" { + t := userType(f.Type).base + tname = t.String() + } + gt, err := getBaseType(tname, f.Type) + if err != nil { + return nil, err + } + // Some mutually recursive types can cause us to be here while + // still defining the element. Fix the element type id here. + // We could do this more neatly by setting the id at the start of + // building every type, but that would break binary compatibility. + if gt.id() == 0 { + setTypeId(gt) + } + st.Field = append(st.Field, fieldType{f.Name, gt.id()}) + } + return st, nil + + default: + return nil, errors.New("gob NewTypeObject can't handle type: " + rt.String()) + } +} + +// isExported reports whether this is an exported - upper case - name. +func isExported(name string) bool { + rune, _ := utf8.DecodeRuneInString(name) + return unicode.IsUpper(rune) +} + +// isSent reports whether this struct field is to be transmitted. +// It will be transmitted only if it is exported and not a chan or func field +// or pointer to chan or func. +func isSent(field *reflect.StructField) bool { + if !isExported(field.Name) { + return false + } + // If the field is a chan or func or pointer thereto, don't send it. + // That is, treat it like an unexported field. + typ := field.Type + for typ.Kind() == reflect.Pointer { + typ = typ.Elem() + } + if typ.Kind() == reflect.Chan || typ.Kind() == reflect.Func { + return false + } + return true +} + +// getBaseType returns the Gob type describing the given reflect.Type's base type. +// typeLock must be held. +func getBaseType(name string, rt reflect.Type) (gobType, error) { + ut := userType(rt) + return getType(name, ut, ut.base) +} + +// getType returns the Gob type describing the given reflect.Type. +// Should be called only when handling GobEncoders/Decoders, +// which may be pointers. All other types are handled through the +// base type, never a pointer. +// typeLock must be held. +func getType(name string, ut *userTypeInfo, rt reflect.Type) (gobType, error) { + typ, present := types[rt] + if present { + return typ, nil + } + typ, err := newTypeObject(name, ut, rt) + if err == nil { + types[rt] = typ + } + return typ, err +} + +func checkId(want, got typeId) { + if want != got { + fmt.Fprintf(os.Stderr, "checkId: %d should be %d\n", int(got), int(want)) + panic("bootstrap type wrong id: " + got.name() + " " + got.string() + " not " + want.string()) + } +} + +// used for building the basic types; called only from init(). the incoming +// interface always refers to a pointer. +func bootstrapType(name string, e any) typeId { + rt := reflect.TypeOf(e).Elem() + _, present := types[rt] + if present { + panic("bootstrap type already present: " + name + ", " + rt.String()) + } + typ := &CommonType{Name: name} + types[rt] = typ + setTypeId(typ) + return typ.id() +} + +// Representation of the information we send and receive about this type. +// Each value we send is preceded by its type definition: an encoded int. +// However, the very first time we send the value, we first send the pair +// (-id, wireType). +// For bootstrapping purposes, we assume that the recipient knows how +// to decode a wireType; it is exactly the wireType struct here, interpreted +// using the gob rules for sending a structure, except that we assume the +// ids for wireType and structType etc. are known. The relevant pieces +// are built in encode.go's init() function. +// To maintain binary compatibility, if you extend this type, always put +// the new fields last. +type wireType struct { + ArrayT *arrayType + SliceT *sliceType + StructT *structType + MapT *mapType + GobEncoderT *gobEncoderType + BinaryMarshalerT *gobEncoderType + TextMarshalerT *gobEncoderType +} + +func (w *wireType) string() string { + const unknown = "unknown type" + if w == nil { + return unknown + } + switch { + case w.ArrayT != nil: + return w.ArrayT.Name + case w.SliceT != nil: + return w.SliceT.Name + case w.StructT != nil: + return w.StructT.Name + case w.MapT != nil: + return w.MapT.Name + case w.GobEncoderT != nil: + return w.GobEncoderT.Name + case w.BinaryMarshalerT != nil: + return w.BinaryMarshalerT.Name + case w.TextMarshalerT != nil: + return w.TextMarshalerT.Name + } + return unknown +} + +type typeInfo struct { + id typeId + encInit sync.Mutex // protects creation of encoder + encoder atomic.Pointer[encEngine] + wire wireType +} + +// typeInfoMap is an atomic pointer to map[reflect.Type]*typeInfo. +// It's updated copy-on-write. Readers just do an atomic load +// to get the current version of the map. Writers make a full copy of +// the map and atomically update the pointer to point to the new map. +// Under heavy read contention, this is significantly faster than a map +// protected by a mutex. +var typeInfoMap atomic.Value + +// typeInfoMapInit is used instead of typeInfoMap during init time, +// as types are registered sequentially during init and we can save +// the overhead of making map copies. +// It is saved to typeInfoMap and set to nil before init finishes. +var typeInfoMapInit = make(map[reflect.Type]*typeInfo, 16) + +func lookupTypeInfo(rt reflect.Type) *typeInfo { + if m := typeInfoMapInit; m != nil { + return m[rt] + } + m, _ := typeInfoMap.Load().(map[reflect.Type]*typeInfo) + return m[rt] +} + +func getTypeInfo(ut *userTypeInfo) (*typeInfo, error) { + rt := ut.base + if ut.externalEnc != 0 { + // We want the user type, not the base type. + rt = ut.user + } + if info := lookupTypeInfo(rt); info != nil { + return info, nil + } + return buildTypeInfo(ut, rt) +} + +// buildTypeInfo constructs the type information for the type +// and stores it in the type info map. +func buildTypeInfo(ut *userTypeInfo, rt reflect.Type) (*typeInfo, error) { + typeLock.Lock() + defer typeLock.Unlock() + + if info := lookupTypeInfo(rt); info != nil { + return info, nil + } + + gt, err := getBaseType(rt.Name(), rt) + if err != nil { + return nil, err + } + info := &typeInfo{id: gt.id()} + + if ut.externalEnc != 0 { + userType, err := getType(rt.Name(), ut, rt) + if err != nil { + return nil, err + } + gt := userType.id().gobType().(*gobEncoderType) + switch ut.externalEnc { + case xGob: + info.wire.GobEncoderT = gt + case xBinary: + info.wire.BinaryMarshalerT = gt + case xText: + info.wire.TextMarshalerT = gt + } + rt = ut.user + } else { + t := info.id.gobType() + switch typ := rt; typ.Kind() { + case reflect.Array: + info.wire.ArrayT = t.(*arrayType) + case reflect.Map: + info.wire.MapT = t.(*mapType) + case reflect.Slice: + // []byte == []uint8 is a special case handled separately + if typ.Elem().Kind() != reflect.Uint8 { + info.wire.SliceT = t.(*sliceType) + } + case reflect.Struct: + info.wire.StructT = t.(*structType) + } + } + + if m := typeInfoMapInit; m != nil { + m[rt] = info + return info, nil + } + + // Create new map with old contents plus new entry. + m, _ := typeInfoMap.Load().(map[reflect.Type]*typeInfo) + newm := make(map[reflect.Type]*typeInfo, len(m)) + for k, v := range m { + newm[k] = v + } + newm[rt] = info + typeInfoMap.Store(newm) + return info, nil +} + +// Called only when a panic is acceptable and unexpected. +func mustGetTypeInfo(rt reflect.Type) *typeInfo { + t, err := getTypeInfo(userType(rt)) + if err != nil { + panic("getTypeInfo: " + err.Error()) + } + return t +} + +// GobEncoder is the interface describing data that provides its own +// representation for encoding values for transmission to a GobDecoder. +// A type that implements GobEncoder and GobDecoder has complete +// control over the representation of its data and may therefore +// contain things such as private fields, channels, and functions, +// which are not usually transmissible in gob streams. +// +// Note: Since gobs can be stored permanently, it is good design +// to guarantee the encoding used by a GobEncoder is stable as the +// software evolves. For instance, it might make sense for GobEncode +// to include a version number in the encoding. +type GobEncoder interface { + // GobEncode returns a byte slice representing the encoding of the + // receiver for transmission to a GobDecoder, usually of the same + // concrete type. + GobEncode() ([]byte, error) +} + +// GobDecoder is the interface describing data that provides its own +// routine for decoding transmitted values sent by a GobEncoder. +type GobDecoder interface { + // GobDecode overwrites the receiver, which must be a pointer, + // with the value represented by the byte slice, which was written + // by GobEncode, usually for the same concrete type. + GobDecode([]byte) error +} + +var ( + nameToConcreteType sync.Map // map[string]reflect.Type + concreteTypeToName sync.Map // map[reflect.Type]string +) + +// RegisterName is like [Register] but uses the provided name rather than the +// type's default. +func RegisterName(name string, value any) { + if name == "" { + // reserved for nil + panic("attempt to register empty name") + } + + ut := userType(reflect.TypeOf(value)) + + // Check for incompatible duplicates. The name must refer to the + // same user type, and vice versa. + + // Store the name and type provided by the user.... + if t, dup := nameToConcreteType.LoadOrStore(name, reflect.TypeOf(value)); dup && t != ut.user { + panic(fmt.Sprintf("gob: registering duplicate types for %q: %s != %s", name, t, ut.user)) + } + + // but the flattened type in the type table, since that's what decode needs. + if n, dup := concreteTypeToName.LoadOrStore(ut.base, name); dup && n != name { + nameToConcreteType.Delete(name) + panic(fmt.Sprintf("gob: registering duplicate names for %s: %q != %q", ut.user, n, name)) + } +} + +// Register records a type, identified by a value for that type, under its +// internal type name. That name will identify the concrete type of a value +// sent or received as an interface variable. Only types that will be +// transferred as implementations of interface values need to be registered. +// Expecting to be used only during initialization, it panics if the mapping +// between types and names is not a bijection. +func Register(value any) { + // Default to printed representation for unnamed types + rt := reflect.TypeOf(value) + name := rt.String() + + // But for named types (or pointers to them), qualify with import path (but see inner comment). + // Dereference one pointer looking for a named type. + star := "" + if rt.Name() == "" { + if pt := rt; pt.Kind() == reflect.Pointer { + star = "*" + // NOTE: The following line should be rt = pt.Elem() to implement + // what the comment above claims, but fixing it would break compatibility + // with existing gobs. + // + // Given package p imported as "full/p" with these definitions: + // package p + // type T1 struct { ... } + // this table shows the intended and actual strings used by gob to + // name the types: + // + // Type Correct string Actual string + // + // T1 full/p.T1 full/p.T1 + // *T1 *full/p.T1 *p.T1 + // + // The missing full path cannot be fixed without breaking existing gob decoders. + rt = pt + } + } + if rt.Name() != "" { + if rt.PkgPath() == "" { + name = star + rt.Name() + } else { + name = star + rt.PkgPath() + "." + rt.Name() + } + } + + RegisterName(name, value) +} + +func registerBasics() { + Register(int(0)) + Register(int8(0)) + Register(int16(0)) + Register(int32(0)) + Register(int64(0)) + Register(uint(0)) + Register(uint8(0)) + Register(uint16(0)) + Register(uint32(0)) + Register(uint64(0)) + Register(float32(0)) + Register(float64(0)) + Register(complex64(0i)) + Register(complex128(0i)) + Register(uintptr(0)) + Register(false) + Register("") + Register([]byte(nil)) + Register([]int(nil)) + Register([]int8(nil)) + Register([]int16(nil)) + Register([]int32(nil)) + Register([]int64(nil)) + Register([]uint(nil)) + Register([]uint8(nil)) + Register([]uint16(nil)) + Register([]uint32(nil)) + Register([]uint64(nil)) + Register([]float32(nil)) + Register([]float64(nil)) + Register([]complex64(nil)) + Register([]complex128(nil)) + Register([]uintptr(nil)) + Register([]bool(nil)) + Register([]string(nil)) +} + +func init() { + typeInfoMap.Store(typeInfoMapInit) + typeInfoMapInit = nil +} diff --git a/contrib/go/_std_1.22/src/encoding/gob/ya.make b/contrib/go/_std_1.22/src/encoding/gob/ya.make new file mode 100644 index 0000000000..384c2ef076 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/gob/ya.make @@ -0,0 +1,15 @@ +GO_LIBRARY() +IF (TRUE) + SRCS( + dec_helpers.go + decode.go + decoder.go + doc.go + enc_helpers.go + encode.go + encoder.go + error.go + type.go + ) +ENDIF() +END() diff --git a/contrib/go/_std_1.22/src/encoding/hex/hex.go b/contrib/go/_std_1.22/src/encoding/hex/hex.go new file mode 100644 index 0000000000..791d2bd4ad --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/hex/hex.go @@ -0,0 +1,355 @@ +// Copyright 2009 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +// Package hex implements hexadecimal encoding and decoding. +package hex + +import ( + "errors" + "fmt" + "io" + "slices" + "strings" +) + +const ( + hextable = "0123456789abcdef" + reverseHexTable = "" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\xff\xff\xff\xff\xff\xff" + + "\xff\x0a\x0b\x0c\x0d\x0e\x0f\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\x0a\x0b\x0c\x0d\x0e\x0f\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" + + "\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff" +) + +// EncodedLen returns the length of an encoding of n source bytes. +// Specifically, it returns n * 2. +func EncodedLen(n int) int { return n * 2 } + +// Encode encodes src into [EncodedLen](len(src)) +// bytes of dst. As a convenience, it returns the number +// of bytes written to dst, but this value is always [EncodedLen](len(src)). +// Encode implements hexadecimal encoding. +func Encode(dst, src []byte) int { + j := 0 + for _, v := range src { + dst[j] = hextable[v>>4] + dst[j+1] = hextable[v&0x0f] + j += 2 + } + return len(src) * 2 +} + +// AppendEncode appends the hexadecimally encoded src to dst +// and returns the extended buffer. +func AppendEncode(dst, src []byte) []byte { + n := EncodedLen(len(src)) + dst = slices.Grow(dst, n) + Encode(dst[len(dst):][:n], src) + return dst[:len(dst)+n] +} + +// ErrLength reports an attempt to decode an odd-length input +// using [Decode] or [DecodeString]. +// The stream-based Decoder returns [io.ErrUnexpectedEOF] instead of ErrLength. +var ErrLength = errors.New("encoding/hex: odd length hex string") + +// InvalidByteError values describe errors resulting from an invalid byte in a hex string. +type InvalidByteError byte + +func (e InvalidByteError) Error() string { + return fmt.Sprintf("encoding/hex: invalid byte: %#U", rune(e)) +} + +// DecodedLen returns the length of a decoding of x source bytes. +// Specifically, it returns x / 2. +func DecodedLen(x int) int { return x / 2 } + +// Decode decodes src into [DecodedLen](len(src)) bytes, +// returning the actual number of bytes written to dst. +// +// Decode expects that src contains only hexadecimal +// characters and that src has even length. +// If the input is malformed, Decode returns the number +// of bytes decoded before the error. +func Decode(dst, src []byte) (int, error) { + i, j := 0, 1 + for ; j < len(src); j += 2 { + p := src[j-1] + q := src[j] + + a := reverseHexTable[p] + b := reverseHexTable[q] + if a > 0x0f { + return i, InvalidByteError(p) + } + if b > 0x0f { + return i, InvalidByteError(q) + } + dst[i] = (a << 4) | b + i++ + } + if len(src)%2 == 1 { + // Check for invalid char before reporting bad length, + // since the invalid char (if present) is an earlier problem. + if reverseHexTable[src[j-1]] > 0x0f { + return i, InvalidByteError(src[j-1]) + } + return i, ErrLength + } + return i, nil +} + +// AppendDecode appends the hexadecimally decoded src to dst +// and returns the extended buffer. +// If the input is malformed, it returns the partially decoded src and an error. +func AppendDecode(dst, src []byte) ([]byte, error) { + n := DecodedLen(len(src)) + dst = slices.Grow(dst, n) + n, err := Decode(dst[len(dst):][:n], src) + return dst[:len(dst)+n], err +} + +// EncodeToString returns the hexadecimal encoding of src. +func EncodeToString(src []byte) string { + dst := make([]byte, EncodedLen(len(src))) + Encode(dst, src) + return string(dst) +} + +// DecodeString returns the bytes represented by the hexadecimal string s. +// +// DecodeString expects that src contains only hexadecimal +// characters and that src has even length. +// If the input is malformed, DecodeString returns +// the bytes decoded before the error. +func DecodeString(s string) ([]byte, error) { + src := []byte(s) + // We can use the source slice itself as the destination + // because the decode loop increments by one and then the 'seen' byte is not used anymore. + n, err := Decode(src, src) + return src[:n], err +} + +// Dump returns a string that contains a hex dump of the given data. The format +// of the hex dump matches the output of `hexdump -C` on the command line. +func Dump(data []byte) string { + if len(data) == 0 { + return "" + } + + var buf strings.Builder + // Dumper will write 79 bytes per complete 16 byte chunk, and at least + // 64 bytes for whatever remains. Round the allocation up, since only a + // maximum of 15 bytes will be wasted. + buf.Grow((1 + ((len(data) - 1) / 16)) * 79) + + dumper := Dumper(&buf) + dumper.Write(data) + dumper.Close() + return buf.String() +} + +// bufferSize is the number of hexadecimal characters to buffer in encoder and decoder. +const bufferSize = 1024 + +type encoder struct { + w io.Writer + err error + out [bufferSize]byte // output buffer +} + +// NewEncoder returns an [io.Writer] that writes lowercase hexadecimal characters to w. +func NewEncoder(w io.Writer) io.Writer { + return &encoder{w: w} +} + +func (e *encoder) Write(p []byte) (n int, err error) { + for len(p) > 0 && e.err == nil { + chunkSize := bufferSize / 2 + if len(p) < chunkSize { + chunkSize = len(p) + } + + var written int + encoded := Encode(e.out[:], p[:chunkSize]) + written, e.err = e.w.Write(e.out[:encoded]) + n += written / 2 + p = p[chunkSize:] + } + return n, e.err +} + +type decoder struct { + r io.Reader + err error + in []byte // input buffer (encoded form) + arr [bufferSize]byte // backing array for in +} + +// NewDecoder returns an [io.Reader] that decodes hexadecimal characters from r. +// NewDecoder expects that r contain only an even number of hexadecimal characters. +func NewDecoder(r io.Reader) io.Reader { + return &decoder{r: r} +} + +func (d *decoder) Read(p []byte) (n int, err error) { + // Fill internal buffer with sufficient bytes to decode + if len(d.in) < 2 && d.err == nil { + var numCopy, numRead int + numCopy = copy(d.arr[:], d.in) // Copies either 0 or 1 bytes + numRead, d.err = d.r.Read(d.arr[numCopy:]) + d.in = d.arr[:numCopy+numRead] + if d.err == io.EOF && len(d.in)%2 != 0 { + + if a := reverseHexTable[d.in[len(d.in)-1]]; a > 0x0f { + d.err = InvalidByteError(d.in[len(d.in)-1]) + } else { + d.err = io.ErrUnexpectedEOF + } + } + } + + // Decode internal buffer into output buffer + if numAvail := len(d.in) / 2; len(p) > numAvail { + p = p[:numAvail] + } + numDec, err := Decode(p, d.in[:len(p)*2]) + d.in = d.in[2*numDec:] + if err != nil { + d.in, d.err = nil, err // Decode error; discard input remainder + } + + if len(d.in) < 2 { + return numDec, d.err // Only expose errors when buffer fully consumed + } + return numDec, nil +} + +// Dumper returns a [io.WriteCloser] that writes a hex dump of all written data to +// w. The format of the dump matches the output of `hexdump -C` on the command +// line. +func Dumper(w io.Writer) io.WriteCloser { + return &dumper{w: w} +} + +type dumper struct { + w io.Writer + rightChars [18]byte + buf [14]byte + used int // number of bytes in the current line + n uint // number of bytes, total + closed bool +} + +func toChar(b byte) byte { + if b < 32 || b > 126 { + return '.' + } + return b +} + +func (h *dumper) Write(data []byte) (n int, err error) { + if h.closed { + return 0, errors.New("encoding/hex: dumper closed") + } + + // Output lines look like: + // 00000010 2e 2f 30 31 32 33 34 35 36 37 38 39 3a 3b 3c 3d |./0123456789:;<=| + // ^ offset ^ extra space ^ ASCII of line. + for i := range data { + if h.used == 0 { + // At the beginning of a line we print the current + // offset in hex. + h.buf[0] = byte(h.n >> 24) + h.buf[1] = byte(h.n >> 16) + h.buf[2] = byte(h.n >> 8) + h.buf[3] = byte(h.n) + Encode(h.buf[4:], h.buf[:4]) + h.buf[12] = ' ' + h.buf[13] = ' ' + _, err = h.w.Write(h.buf[4:]) + if err != nil { + return + } + } + Encode(h.buf[:], data[i:i+1]) + h.buf[2] = ' ' + l := 3 + if h.used == 7 { + // There's an additional space after the 8th byte. + h.buf[3] = ' ' + l = 4 + } else if h.used == 15 { + // At the end of the line there's an extra space and + // the bar for the right column. + h.buf[3] = ' ' + h.buf[4] = '|' + l = 5 + } + _, err = h.w.Write(h.buf[:l]) + if err != nil { + return + } + n++ + h.rightChars[h.used] = toChar(data[i]) + h.used++ + h.n++ + if h.used == 16 { + h.rightChars[16] = '|' + h.rightChars[17] = '\n' + _, err = h.w.Write(h.rightChars[:]) + if err != nil { + return + } + h.used = 0 + } + } + return +} + +func (h *dumper) Close() (err error) { + // See the comments in Write() for the details of this format. + if h.closed { + return + } + h.closed = true + if h.used == 0 { + return + } + h.buf[0] = ' ' + h.buf[1] = ' ' + h.buf[2] = ' ' + h.buf[3] = ' ' + h.buf[4] = '|' + nBytes := h.used + for h.used < 16 { + l := 3 + if h.used == 7 { + l = 4 + } else if h.used == 15 { + l = 5 + } + _, err = h.w.Write(h.buf[:l]) + if err != nil { + return + } + h.used++ + } + h.rightChars[nBytes] = '|' + h.rightChars[nBytes+1] = '\n' + _, err = h.w.Write(h.rightChars[:nBytes+2]) + return +} diff --git a/contrib/go/_std_1.22/src/encoding/hex/ya.make b/contrib/go/_std_1.22/src/encoding/hex/ya.make new file mode 100644 index 0000000000..a6c3609b1c --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/hex/ya.make @@ -0,0 +1,7 @@ +GO_LIBRARY() +IF (TRUE) + SRCS( + hex.go + ) +ENDIF() +END() diff --git a/contrib/go/_std_1.22/src/encoding/json/decode.go b/contrib/go/_std_1.22/src/encoding/json/decode.go new file mode 100644 index 0000000000..bc1891f8ac --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/json/decode.go @@ -0,0 +1,1292 @@ +// Copyright 2010 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +// Represents JSON data structure using native Go types: booleans, floats, +// strings, arrays, and maps. + +package json + +import ( + "encoding" + "encoding/base64" + "fmt" + "reflect" + "strconv" + "strings" + "unicode" + "unicode/utf16" + "unicode/utf8" +) + +// Unmarshal parses the JSON-encoded data and stores the result +// in the value pointed to by v. If v is nil or not a pointer, +// Unmarshal returns an [InvalidUnmarshalError]. +// +// Unmarshal uses the inverse of the encodings that +// [Marshal] uses, allocating maps, slices, and pointers as necessary, +// with the following additional rules: +// +// To unmarshal JSON into a pointer, Unmarshal first handles the case of +// the JSON being the JSON literal null. In that case, Unmarshal sets +// the pointer to nil. Otherwise, Unmarshal unmarshals the JSON into +// the value pointed at by the pointer. If the pointer is nil, Unmarshal +// allocates a new value for it to point to. +// +// To unmarshal JSON into a value implementing [Unmarshaler], +// Unmarshal calls that value's [Unmarshaler.UnmarshalJSON] method, including +// when the input is a JSON null. +// Otherwise, if the value implements [encoding.TextUnmarshaler] +// and the input is a JSON quoted string, Unmarshal calls +// [encoding.TextUnmarshaler.UnmarshalText] with the unquoted form of the string. +// +// To unmarshal JSON into a struct, Unmarshal matches incoming object +// keys to the keys used by [Marshal] (either the struct field name or its tag), +// preferring an exact match but also accepting a case-insensitive match. By +// default, object keys which don't have a corresponding struct field are +// ignored (see [Decoder.DisallowUnknownFields] for an alternative). +// +// To unmarshal JSON into an interface value, +// Unmarshal stores one of these in the interface value: +// +// - bool, for JSON booleans +// - float64, for JSON numbers +// - string, for JSON strings +// - []interface{}, for JSON arrays +// - map[string]interface{}, for JSON objects +// - nil for JSON null +// +// To unmarshal a JSON array into a slice, Unmarshal resets the slice length +// to zero and then appends each element to the slice. +// As a special case, to unmarshal an empty JSON array into a slice, +// Unmarshal replaces the slice with a new empty slice. +// +// To unmarshal a JSON array into a Go array, Unmarshal decodes +// JSON array elements into corresponding Go array elements. +// If the Go array is smaller than the JSON array, +// the additional JSON array elements are discarded. +// If the JSON array is smaller than the Go array, +// the additional Go array elements are set to zero values. +// +// To unmarshal a JSON object into a map, Unmarshal first establishes a map to +// use. If the map is nil, Unmarshal allocates a new map. Otherwise Unmarshal +// reuses the existing map, keeping existing entries. Unmarshal then stores +// key-value pairs from the JSON object into the map. The map's key type must +// either be any string type, an integer, implement [json.Unmarshaler], or +// implement [encoding.TextUnmarshaler]. +// +// If the JSON-encoded data contain a syntax error, Unmarshal returns a [SyntaxError]. +// +// If a JSON value is not appropriate for a given target type, +// or if a JSON number overflows the target type, Unmarshal +// skips that field and completes the unmarshaling as best it can. +// If no more serious errors are encountered, Unmarshal returns +// an [UnmarshalTypeError] describing the earliest such error. In any +// case, it's not guaranteed that all the remaining fields following +// the problematic one will be unmarshaled into the target object. +// +// The JSON null value unmarshals into an interface, map, pointer, or slice +// by setting that Go value to nil. Because null is often used in JSON to mean +// “not present,” unmarshaling a JSON null into any other Go type has no effect +// on the value and produces no error. +// +// When unmarshaling quoted strings, invalid UTF-8 or +// invalid UTF-16 surrogate pairs are not treated as an error. +// Instead, they are replaced by the Unicode replacement +// character U+FFFD. +func Unmarshal(data []byte, v any) error { + // Check for well-formedness. + // Avoids filling out half a data structure + // before discovering a JSON syntax error. + var d decodeState + err := checkValid(data, &d.scan) + if err != nil { + return err + } + + d.init(data) + return d.unmarshal(v) +} + +// Unmarshaler is the interface implemented by types +// that can unmarshal a JSON description of themselves. +// The input can be assumed to be a valid encoding of +// a JSON value. UnmarshalJSON must copy the JSON data +// if it wishes to retain the data after returning. +// +// By convention, to approximate the behavior of [Unmarshal] itself, +// Unmarshalers implement UnmarshalJSON([]byte("null")) as a no-op. +type Unmarshaler interface { + UnmarshalJSON([]byte) error +} + +// An UnmarshalTypeError describes a JSON value that was +// not appropriate for a value of a specific Go type. +type UnmarshalTypeError struct { + Value string // description of JSON value - "bool", "array", "number -5" + Type reflect.Type // type of Go value it could not be assigned to + Offset int64 // error occurred after reading Offset bytes + Struct string // name of the struct type containing the field + Field string // the full path from root node to the field +} + +func (e *UnmarshalTypeError) Error() string { + if e.Struct != "" || e.Field != "" { + return "json: cannot unmarshal " + e.Value + " into Go struct field " + e.Struct + "." + e.Field + " of type " + e.Type.String() + } + return "json: cannot unmarshal " + e.Value + " into Go value of type " + e.Type.String() +} + +// An UnmarshalFieldError describes a JSON object key that +// led to an unexported (and therefore unwritable) struct field. +// +// Deprecated: No longer used; kept for compatibility. +type UnmarshalFieldError struct { + Key string + Type reflect.Type + Field reflect.StructField +} + +func (e *UnmarshalFieldError) Error() string { + return "json: cannot unmarshal object key " + strconv.Quote(e.Key) + " into unexported field " + e.Field.Name + " of type " + e.Type.String() +} + +// An InvalidUnmarshalError describes an invalid argument passed to [Unmarshal]. +// (The argument to [Unmarshal] must be a non-nil pointer.) +type InvalidUnmarshalError struct { + Type reflect.Type +} + +func (e *InvalidUnmarshalError) Error() string { + if e.Type == nil { + return "json: Unmarshal(nil)" + } + + if e.Type.Kind() != reflect.Pointer { + return "json: Unmarshal(non-pointer " + e.Type.String() + ")" + } + return "json: Unmarshal(nil " + e.Type.String() + ")" +} + +func (d *decodeState) unmarshal(v any) error { + rv := reflect.ValueOf(v) + if rv.Kind() != reflect.Pointer || rv.IsNil() { + return &InvalidUnmarshalError{reflect.TypeOf(v)} + } + + d.scan.reset() + d.scanWhile(scanSkipSpace) + // We decode rv not rv.Elem because the Unmarshaler interface + // test must be applied at the top level of the value. + err := d.value(rv) + if err != nil { + return d.addErrorContext(err) + } + return d.savedError +} + +// A Number represents a JSON number literal. +type Number string + +// String returns the literal text of the number. +func (n Number) String() string { return string(n) } + +// Float64 returns the number as a float64. +func (n Number) Float64() (float64, error) { + return strconv.ParseFloat(string(n), 64) +} + +// Int64 returns the number as an int64. +func (n Number) Int64() (int64, error) { + return strconv.ParseInt(string(n), 10, 64) +} + +// An errorContext provides context for type errors during decoding. +type errorContext struct { + Struct reflect.Type + FieldStack []string +} + +// decodeState represents the state while decoding a JSON value. +type decodeState struct { + data []byte + off int // next read offset in data + opcode int // last read result + scan scanner + errorContext *errorContext + savedError error + useNumber bool + disallowUnknownFields bool +} + +// readIndex returns the position of the last byte read. +func (d *decodeState) readIndex() int { + return d.off - 1 +} + +// phasePanicMsg is used as a panic message when we end up with something that +// shouldn't happen. It can indicate a bug in the JSON decoder, or that +// something is editing the data slice while the decoder executes. +const phasePanicMsg = "JSON decoder out of sync - data changing underfoot?" + +func (d *decodeState) init(data []byte) *decodeState { + d.data = data + d.off = 0 + d.savedError = nil + if d.errorContext != nil { + d.errorContext.Struct = nil + // Reuse the allocated space for the FieldStack slice. + d.errorContext.FieldStack = d.errorContext.FieldStack[:0] + } + return d +} + +// saveError saves the first err it is called with, +// for reporting at the end of the unmarshal. +func (d *decodeState) saveError(err error) { + if d.savedError == nil { + d.savedError = d.addErrorContext(err) + } +} + +// addErrorContext returns a new error enhanced with information from d.errorContext +func (d *decodeState) addErrorContext(err error) error { + if d.errorContext != nil && (d.errorContext.Struct != nil || len(d.errorContext.FieldStack) > 0) { + switch err := err.(type) { + case *UnmarshalTypeError: + err.Struct = d.errorContext.Struct.Name() + err.Field = strings.Join(d.errorContext.FieldStack, ".") + } + } + return err +} + +// skip scans to the end of what was started. +func (d *decodeState) skip() { + s, data, i := &d.scan, d.data, d.off + depth := len(s.parseState) + for { + op := s.step(s, data[i]) + i++ + if len(s.parseState) < depth { + d.off = i + d.opcode = op + return + } + } +} + +// scanNext processes the byte at d.data[d.off]. +func (d *decodeState) scanNext() { + if d.off < len(d.data) { + d.opcode = d.scan.step(&d.scan, d.data[d.off]) + d.off++ + } else { + d.opcode = d.scan.eof() + d.off = len(d.data) + 1 // mark processed EOF with len+1 + } +} + +// scanWhile processes bytes in d.data[d.off:] until it +// receives a scan code not equal to op. +func (d *decodeState) scanWhile(op int) { + s, data, i := &d.scan, d.data, d.off + for i < len(data) { + newOp := s.step(s, data[i]) + i++ + if newOp != op { + d.opcode = newOp + d.off = i + return + } + } + + d.off = len(data) + 1 // mark processed EOF with len+1 + d.opcode = d.scan.eof() +} + +// rescanLiteral is similar to scanWhile(scanContinue), but it specialises the +// common case where we're decoding a literal. The decoder scans the input +// twice, once for syntax errors and to check the length of the value, and the +// second to perform the decoding. +// +// Only in the second step do we use decodeState to tokenize literals, so we +// know there aren't any syntax errors. We can take advantage of that knowledge, +// and scan a literal's bytes much more quickly. +func (d *decodeState) rescanLiteral() { + data, i := d.data, d.off +Switch: + switch data[i-1] { + case '"': // string + for ; i < len(data); i++ { + switch data[i] { + case '\\': + i++ // escaped char + case '"': + i++ // tokenize the closing quote too + break Switch + } + } + case '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '-': // number + for ; i < len(data); i++ { + switch data[i] { + case '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', + '.', 'e', 'E', '+', '-': + default: + break Switch + } + } + case 't': // true + i += len("rue") + case 'f': // false + i += len("alse") + case 'n': // null + i += len("ull") + } + if i < len(data) { + d.opcode = stateEndValue(&d.scan, data[i]) + } else { + d.opcode = scanEnd + } + d.off = i + 1 +} + +// value consumes a JSON value from d.data[d.off-1:], decoding into v, and +// reads the following byte ahead. If v is invalid, the value is discarded. +// The first byte of the value has been read already. +func (d *decodeState) value(v reflect.Value) error { + switch d.opcode { + default: + panic(phasePanicMsg) + + case scanBeginArray: + if v.IsValid() { + if err := d.array(v); err != nil { + return err + } + } else { + d.skip() + } + d.scanNext() + + case scanBeginObject: + if v.IsValid() { + if err := d.object(v); err != nil { + return err + } + } else { + d.skip() + } + d.scanNext() + + case scanBeginLiteral: + // All bytes inside literal return scanContinue op code. + start := d.readIndex() + d.rescanLiteral() + + if v.IsValid() { + if err := d.literalStore(d.data[start:d.readIndex()], v, false); err != nil { + return err + } + } + } + return nil +} + +type unquotedValue struct{} + +// valueQuoted is like value but decodes a +// quoted string literal or literal null into an interface value. +// If it finds anything other than a quoted string literal or null, +// valueQuoted returns unquotedValue{}. +func (d *decodeState) valueQuoted() any { + switch d.opcode { + default: + panic(phasePanicMsg) + + case scanBeginArray, scanBeginObject: + d.skip() + d.scanNext() + + case scanBeginLiteral: + v := d.literalInterface() + switch v.(type) { + case nil, string: + return v + } + } + return unquotedValue{} +} + +// indirect walks down v allocating pointers as needed, +// until it gets to a non-pointer. +// If it encounters an Unmarshaler, indirect stops and returns that. +// If decodingNull is true, indirect stops at the first settable pointer so it +// can be set to nil. +func indirect(v reflect.Value, decodingNull bool) (Unmarshaler, encoding.TextUnmarshaler, reflect.Value) { + // Issue #24153 indicates that it is generally not a guaranteed property + // that you may round-trip a reflect.Value by calling Value.Addr().Elem() + // and expect the value to still be settable for values derived from + // unexported embedded struct fields. + // + // The logic below effectively does this when it first addresses the value + // (to satisfy possible pointer methods) and continues to dereference + // subsequent pointers as necessary. + // + // After the first round-trip, we set v back to the original value to + // preserve the original RW flags contained in reflect.Value. + v0 := v + haveAddr := false + + // If v is a named type and is addressable, + // start with its address, so that if the type has pointer methods, + // we find them. + if v.Kind() != reflect.Pointer && v.Type().Name() != "" && v.CanAddr() { + haveAddr = true + v = v.Addr() + } + for { + // Load value from interface, but only if the result will be + // usefully addressable. + if v.Kind() == reflect.Interface && !v.IsNil() { + e := v.Elem() + if e.Kind() == reflect.Pointer && !e.IsNil() && (!decodingNull || e.Elem().Kind() == reflect.Pointer) { + haveAddr = false + v = e + continue + } + } + + if v.Kind() != reflect.Pointer { + break + } + + if decodingNull && v.CanSet() { + break + } + + // Prevent infinite loop if v is an interface pointing to its own address: + // var v interface{} + // v = &v + if v.Elem().Kind() == reflect.Interface && v.Elem().Elem() == v { + v = v.Elem() + break + } + if v.IsNil() { + v.Set(reflect.New(v.Type().Elem())) + } + if v.Type().NumMethod() > 0 && v.CanInterface() { + if u, ok := v.Interface().(Unmarshaler); ok { + return u, nil, reflect.Value{} + } + if !decodingNull { + if u, ok := v.Interface().(encoding.TextUnmarshaler); ok { + return nil, u, reflect.Value{} + } + } + } + + if haveAddr { + v = v0 // restore original value after round-trip Value.Addr().Elem() + haveAddr = false + } else { + v = v.Elem() + } + } + return nil, nil, v +} + +// array consumes an array from d.data[d.off-1:], decoding into v. +// The first byte of the array ('[') has been read already. +func (d *decodeState) array(v reflect.Value) error { + // Check for unmarshaler. + u, ut, pv := indirect(v, false) + if u != nil { + start := d.readIndex() + d.skip() + return u.UnmarshalJSON(d.data[start:d.off]) + } + if ut != nil { + d.saveError(&UnmarshalTypeError{Value: "array", Type: v.Type(), Offset: int64(d.off)}) + d.skip() + return nil + } + v = pv + + // Check type of target. + switch v.Kind() { + case reflect.Interface: + if v.NumMethod() == 0 { + // Decoding into nil interface? Switch to non-reflect code. + ai := d.arrayInterface() + v.Set(reflect.ValueOf(ai)) + return nil + } + // Otherwise it's invalid. + fallthrough + default: + d.saveError(&UnmarshalTypeError{Value: "array", Type: v.Type(), Offset: int64(d.off)}) + d.skip() + return nil + case reflect.Array, reflect.Slice: + break + } + + i := 0 + for { + // Look ahead for ] - can only happen on first iteration. + d.scanWhile(scanSkipSpace) + if d.opcode == scanEndArray { + break + } + + // Expand slice length, growing the slice if necessary. + if v.Kind() == reflect.Slice { + if i >= v.Cap() { + v.Grow(1) + } + if i >= v.Len() { + v.SetLen(i + 1) + } + } + + if i < v.Len() { + // Decode into element. + if err := d.value(v.Index(i)); err != nil { + return err + } + } else { + // Ran out of fixed array: skip. + if err := d.value(reflect.Value{}); err != nil { + return err + } + } + i++ + + // Next token must be , or ]. + if d.opcode == scanSkipSpace { + d.scanWhile(scanSkipSpace) + } + if d.opcode == scanEndArray { + break + } + if d.opcode != scanArrayValue { + panic(phasePanicMsg) + } + } + + if i < v.Len() { + if v.Kind() == reflect.Array { + for ; i < v.Len(); i++ { + v.Index(i).SetZero() // zero remainder of array + } + } else { + v.SetLen(i) // truncate the slice + } + } + if i == 0 && v.Kind() == reflect.Slice { + v.Set(reflect.MakeSlice(v.Type(), 0, 0)) + } + return nil +} + +var nullLiteral = []byte("null") +var textUnmarshalerType = reflect.TypeFor[encoding.TextUnmarshaler]() + +// object consumes an object from d.data[d.off-1:], decoding into v. +// The first byte ('{') of the object has been read already. +func (d *decodeState) object(v reflect.Value) error { + // Check for unmarshaler. + u, ut, pv := indirect(v, false) + if u != nil { + start := d.readIndex() + d.skip() + return u.UnmarshalJSON(d.data[start:d.off]) + } + if ut != nil { + d.saveError(&UnmarshalTypeError{Value: "object", Type: v.Type(), Offset: int64(d.off)}) + d.skip() + return nil + } + v = pv + t := v.Type() + + // Decoding into nil interface? Switch to non-reflect code. + if v.Kind() == reflect.Interface && v.NumMethod() == 0 { + oi := d.objectInterface() + v.Set(reflect.ValueOf(oi)) + return nil + } + + var fields structFields + + // Check type of target: + // struct or + // map[T1]T2 where T1 is string, an integer type, + // or an encoding.TextUnmarshaler + switch v.Kind() { + case reflect.Map: + // Map key must either have string kind, have an integer kind, + // or be an encoding.TextUnmarshaler. + switch t.Key().Kind() { + case reflect.String, + reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64, + reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr: + default: + if !reflect.PointerTo(t.Key()).Implements(textUnmarshalerType) { + d.saveError(&UnmarshalTypeError{Value: "object", Type: t, Offset: int64(d.off)}) + d.skip() + return nil + } + } + if v.IsNil() { + v.Set(reflect.MakeMap(t)) + } + case reflect.Struct: + fields = cachedTypeFields(t) + // ok + default: + d.saveError(&UnmarshalTypeError{Value: "object", Type: t, Offset: int64(d.off)}) + d.skip() + return nil + } + + var mapElem reflect.Value + var origErrorContext errorContext + if d.errorContext != nil { + origErrorContext = *d.errorContext + } + + for { + // Read opening " of string key or closing }. + d.scanWhile(scanSkipSpace) + if d.opcode == scanEndObject { + // closing } - can only happen on first iteration. + break + } + if d.opcode != scanBeginLiteral { + panic(phasePanicMsg) + } + + // Read key. + start := d.readIndex() + d.rescanLiteral() + item := d.data[start:d.readIndex()] + key, ok := unquoteBytes(item) + if !ok { + panic(phasePanicMsg) + } + + // Figure out field corresponding to key. + var subv reflect.Value + destring := false // whether the value is wrapped in a string to be decoded first + + if v.Kind() == reflect.Map { + elemType := t.Elem() + if !mapElem.IsValid() { + mapElem = reflect.New(elemType).Elem() + } else { + mapElem.SetZero() + } + subv = mapElem + } else { + f := fields.byExactName[string(key)] + if f == nil { + f = fields.byFoldedName[string(foldName(key))] + } + if f != nil { + subv = v + destring = f.quoted + for _, i := range f.index { + if subv.Kind() == reflect.Pointer { + if subv.IsNil() { + // If a struct embeds a pointer to an unexported type, + // it is not possible to set a newly allocated value + // since the field is unexported. + // + // See https://golang.org/issue/21357 + if !subv.CanSet() { + d.saveError(fmt.Errorf("json: cannot set embedded pointer to unexported struct: %v", subv.Type().Elem())) + // Invalidate subv to ensure d.value(subv) skips over + // the JSON value without assigning it to subv. + subv = reflect.Value{} + destring = false + break + } + subv.Set(reflect.New(subv.Type().Elem())) + } + subv = subv.Elem() + } + subv = subv.Field(i) + } + if d.errorContext == nil { + d.errorContext = new(errorContext) + } + d.errorContext.FieldStack = append(d.errorContext.FieldStack, f.name) + d.errorContext.Struct = t + } else if d.disallowUnknownFields { + d.saveError(fmt.Errorf("json: unknown field %q", key)) + } + } + + // Read : before value. + if d.opcode == scanSkipSpace { + d.scanWhile(scanSkipSpace) + } + if d.opcode != scanObjectKey { + panic(phasePanicMsg) + } + d.scanWhile(scanSkipSpace) + + if destring { + switch qv := d.valueQuoted().(type) { + case nil: + if err := d.literalStore(nullLiteral, subv, false); err != nil { + return err + } + case string: + if err := d.literalStore([]byte(qv), subv, true); err != nil { + return err + } + default: + d.saveError(fmt.Errorf("json: invalid use of ,string struct tag, trying to unmarshal unquoted value into %v", subv.Type())) + } + } else { + if err := d.value(subv); err != nil { + return err + } + } + + // Write value back to map; + // if using struct, subv points into struct already. + if v.Kind() == reflect.Map { + kt := t.Key() + var kv reflect.Value + if reflect.PointerTo(kt).Implements(textUnmarshalerType) { + kv = reflect.New(kt) + if err := d.literalStore(item, kv, true); err != nil { + return err + } + kv = kv.Elem() + } else { + switch kt.Kind() { + case reflect.String: + kv = reflect.New(kt).Elem() + kv.SetString(string(key)) + case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64: + s := string(key) + n, err := strconv.ParseInt(s, 10, 64) + if err != nil || reflect.Zero(kt).OverflowInt(n) { + d.saveError(&UnmarshalTypeError{Value: "number " + s, Type: kt, Offset: int64(start + 1)}) + break + } + kv = reflect.New(kt).Elem() + kv.SetInt(n) + case reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr: + s := string(key) + n, err := strconv.ParseUint(s, 10, 64) + if err != nil || reflect.Zero(kt).OverflowUint(n) { + d.saveError(&UnmarshalTypeError{Value: "number " + s, Type: kt, Offset: int64(start + 1)}) + break + } + kv = reflect.New(kt).Elem() + kv.SetUint(n) + default: + panic("json: Unexpected key type") // should never occur + } + } + if kv.IsValid() { + v.SetMapIndex(kv, subv) + } + } + + // Next token must be , or }. + if d.opcode == scanSkipSpace { + d.scanWhile(scanSkipSpace) + } + if d.errorContext != nil { + // Reset errorContext to its original state. + // Keep the same underlying array for FieldStack, to reuse the + // space and avoid unnecessary allocs. + d.errorContext.FieldStack = d.errorContext.FieldStack[:len(origErrorContext.FieldStack)] + d.errorContext.Struct = origErrorContext.Struct + } + if d.opcode == scanEndObject { + break + } + if d.opcode != scanObjectValue { + panic(phasePanicMsg) + } + } + return nil +} + +// convertNumber converts the number literal s to a float64 or a Number +// depending on the setting of d.useNumber. +func (d *decodeState) convertNumber(s string) (any, error) { + if d.useNumber { + return Number(s), nil + } + f, err := strconv.ParseFloat(s, 64) + if err != nil { + return nil, &UnmarshalTypeError{Value: "number " + s, Type: reflect.TypeFor[float64](), Offset: int64(d.off)} + } + return f, nil +} + +var numberType = reflect.TypeFor[Number]() + +// literalStore decodes a literal stored in item into v. +// +// fromQuoted indicates whether this literal came from unwrapping a +// string from the ",string" struct tag option. this is used only to +// produce more helpful error messages. +func (d *decodeState) literalStore(item []byte, v reflect.Value, fromQuoted bool) error { + // Check for unmarshaler. + if len(item) == 0 { + // Empty string given. + d.saveError(fmt.Errorf("json: invalid use of ,string struct tag, trying to unmarshal %q into %v", item, v.Type())) + return nil + } + isNull := item[0] == 'n' // null + u, ut, pv := indirect(v, isNull) + if u != nil { + return u.UnmarshalJSON(item) + } + if ut != nil { + if item[0] != '"' { + if fromQuoted { + d.saveError(fmt.Errorf("json: invalid use of ,string struct tag, trying to unmarshal %q into %v", item, v.Type())) + return nil + } + val := "number" + switch item[0] { + case 'n': + val = "null" + case 't', 'f': + val = "bool" + } + d.saveError(&UnmarshalTypeError{Value: val, Type: v.Type(), Offset: int64(d.readIndex())}) + return nil + } + s, ok := unquoteBytes(item) + if !ok { + if fromQuoted { + return fmt.Errorf("json: invalid use of ,string struct tag, trying to unmarshal %q into %v", item, v.Type()) + } + panic(phasePanicMsg) + } + return ut.UnmarshalText(s) + } + + v = pv + + switch c := item[0]; c { + case 'n': // null + // The main parser checks that only true and false can reach here, + // but if this was a quoted string input, it could be anything. + if fromQuoted && string(item) != "null" { + d.saveError(fmt.Errorf("json: invalid use of ,string struct tag, trying to unmarshal %q into %v", item, v.Type())) + break + } + switch v.Kind() { + case reflect.Interface, reflect.Pointer, reflect.Map, reflect.Slice: + v.SetZero() + // otherwise, ignore null for primitives/string + } + case 't', 'f': // true, false + value := item[0] == 't' + // The main parser checks that only true and false can reach here, + // but if this was a quoted string input, it could be anything. + if fromQuoted && string(item) != "true" && string(item) != "false" { + d.saveError(fmt.Errorf("json: invalid use of ,string struct tag, trying to unmarshal %q into %v", item, v.Type())) + break + } + switch v.Kind() { + default: + if fromQuoted { + d.saveError(fmt.Errorf("json: invalid use of ,string struct tag, trying to unmarshal %q into %v", item, v.Type())) + } else { + d.saveError(&UnmarshalTypeError{Value: "bool", Type: v.Type(), Offset: int64(d.readIndex())}) + } + case reflect.Bool: + v.SetBool(value) + case reflect.Interface: + if v.NumMethod() == 0 { + v.Set(reflect.ValueOf(value)) + } else { + d.saveError(&UnmarshalTypeError{Value: "bool", Type: v.Type(), Offset: int64(d.readIndex())}) + } + } + + case '"': // string + s, ok := unquoteBytes(item) + if !ok { + if fromQuoted { + return fmt.Errorf("json: invalid use of ,string struct tag, trying to unmarshal %q into %v", item, v.Type()) + } + panic(phasePanicMsg) + } + switch v.Kind() { + default: + d.saveError(&UnmarshalTypeError{Value: "string", Type: v.Type(), Offset: int64(d.readIndex())}) + case reflect.Slice: + if v.Type().Elem().Kind() != reflect.Uint8 { + d.saveError(&UnmarshalTypeError{Value: "string", Type: v.Type(), Offset: int64(d.readIndex())}) + break + } + b := make([]byte, base64.StdEncoding.DecodedLen(len(s))) + n, err := base64.StdEncoding.Decode(b, s) + if err != nil { + d.saveError(err) + break + } + v.SetBytes(b[:n]) + case reflect.String: + if v.Type() == numberType && !isValidNumber(string(s)) { + return fmt.Errorf("json: invalid number literal, trying to unmarshal %q into Number", item) + } + v.SetString(string(s)) + case reflect.Interface: + if v.NumMethod() == 0 { + v.Set(reflect.ValueOf(string(s))) + } else { + d.saveError(&UnmarshalTypeError{Value: "string", Type: v.Type(), Offset: int64(d.readIndex())}) + } + } + + default: // number + if c != '-' && (c < '0' || c > '9') { + if fromQuoted { + return fmt.Errorf("json: invalid use of ,string struct tag, trying to unmarshal %q into %v", item, v.Type()) + } + panic(phasePanicMsg) + } + switch v.Kind() { + default: + if v.Kind() == reflect.String && v.Type() == numberType { + // s must be a valid number, because it's + // already been tokenized. + v.SetString(string(item)) + break + } + if fromQuoted { + return fmt.Errorf("json: invalid use of ,string struct tag, trying to unmarshal %q into %v", item, v.Type()) + } + d.saveError(&UnmarshalTypeError{Value: "number", Type: v.Type(), Offset: int64(d.readIndex())}) + case reflect.Interface: + n, err := d.convertNumber(string(item)) + if err != nil { + d.saveError(err) + break + } + if v.NumMethod() != 0 { + d.saveError(&UnmarshalTypeError{Value: "number", Type: v.Type(), Offset: int64(d.readIndex())}) + break + } + v.Set(reflect.ValueOf(n)) + + case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64: + n, err := strconv.ParseInt(string(item), 10, 64) + if err != nil || v.OverflowInt(n) { + d.saveError(&UnmarshalTypeError{Value: "number " + string(item), Type: v.Type(), Offset: int64(d.readIndex())}) + break + } + v.SetInt(n) + + case reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr: + n, err := strconv.ParseUint(string(item), 10, 64) + if err != nil || v.OverflowUint(n) { + d.saveError(&UnmarshalTypeError{Value: "number " + string(item), Type: v.Type(), Offset: int64(d.readIndex())}) + break + } + v.SetUint(n) + + case reflect.Float32, reflect.Float64: + n, err := strconv.ParseFloat(string(item), v.Type().Bits()) + if err != nil || v.OverflowFloat(n) { + d.saveError(&UnmarshalTypeError{Value: "number " + string(item), Type: v.Type(), Offset: int64(d.readIndex())}) + break + } + v.SetFloat(n) + } + } + return nil +} + +// The xxxInterface routines build up a value to be stored +// in an empty interface. They are not strictly necessary, +// but they avoid the weight of reflection in this common case. + +// valueInterface is like value but returns interface{} +func (d *decodeState) valueInterface() (val any) { + switch d.opcode { + default: + panic(phasePanicMsg) + case scanBeginArray: + val = d.arrayInterface() + d.scanNext() + case scanBeginObject: + val = d.objectInterface() + d.scanNext() + case scanBeginLiteral: + val = d.literalInterface() + } + return +} + +// arrayInterface is like array but returns []interface{}. +func (d *decodeState) arrayInterface() []any { + var v = make([]any, 0) + for { + // Look ahead for ] - can only happen on first iteration. + d.scanWhile(scanSkipSpace) + if d.opcode == scanEndArray { + break + } + + v = append(v, d.valueInterface()) + + // Next token must be , or ]. + if d.opcode == scanSkipSpace { + d.scanWhile(scanSkipSpace) + } + if d.opcode == scanEndArray { + break + } + if d.opcode != scanArrayValue { + panic(phasePanicMsg) + } + } + return v +} + +// objectInterface is like object but returns map[string]interface{}. +func (d *decodeState) objectInterface() map[string]any { + m := make(map[string]any) + for { + // Read opening " of string key or closing }. + d.scanWhile(scanSkipSpace) + if d.opcode == scanEndObject { + // closing } - can only happen on first iteration. + break + } + if d.opcode != scanBeginLiteral { + panic(phasePanicMsg) + } + + // Read string key. + start := d.readIndex() + d.rescanLiteral() + item := d.data[start:d.readIndex()] + key, ok := unquote(item) + if !ok { + panic(phasePanicMsg) + } + + // Read : before value. + if d.opcode == scanSkipSpace { + d.scanWhile(scanSkipSpace) + } + if d.opcode != scanObjectKey { + panic(phasePanicMsg) + } + d.scanWhile(scanSkipSpace) + + // Read value. + m[key] = d.valueInterface() + + // Next token must be , or }. + if d.opcode == scanSkipSpace { + d.scanWhile(scanSkipSpace) + } + if d.opcode == scanEndObject { + break + } + if d.opcode != scanObjectValue { + panic(phasePanicMsg) + } + } + return m +} + +// literalInterface consumes and returns a literal from d.data[d.off-1:] and +// it reads the following byte ahead. The first byte of the literal has been +// read already (that's how the caller knows it's a literal). +func (d *decodeState) literalInterface() any { + // All bytes inside literal return scanContinue op code. + start := d.readIndex() + d.rescanLiteral() + + item := d.data[start:d.readIndex()] + + switch c := item[0]; c { + case 'n': // null + return nil + + case 't', 'f': // true, false + return c == 't' + + case '"': // string + s, ok := unquote(item) + if !ok { + panic(phasePanicMsg) + } + return s + + default: // number + if c != '-' && (c < '0' || c > '9') { + panic(phasePanicMsg) + } + n, err := d.convertNumber(string(item)) + if err != nil { + d.saveError(err) + } + return n + } +} + +// getu4 decodes \uXXXX from the beginning of s, returning the hex value, +// or it returns -1. +func getu4(s []byte) rune { + if len(s) < 6 || s[0] != '\\' || s[1] != 'u' { + return -1 + } + var r rune + for _, c := range s[2:6] { + switch { + case '0' <= c && c <= '9': + c = c - '0' + case 'a' <= c && c <= 'f': + c = c - 'a' + 10 + case 'A' <= c && c <= 'F': + c = c - 'A' + 10 + default: + return -1 + } + r = r*16 + rune(c) + } + return r +} + +// unquote converts a quoted JSON string literal s into an actual string t. +// The rules are different than for Go, so cannot use strconv.Unquote. +func unquote(s []byte) (t string, ok bool) { + s, ok = unquoteBytes(s) + t = string(s) + return +} + +func unquoteBytes(s []byte) (t []byte, ok bool) { + if len(s) < 2 || s[0] != '"' || s[len(s)-1] != '"' { + return + } + s = s[1 : len(s)-1] + + // Check for unusual characters. If there are none, + // then no unquoting is needed, so return a slice of the + // original bytes. + r := 0 + for r < len(s) { + c := s[r] + if c == '\\' || c == '"' || c < ' ' { + break + } + if c < utf8.RuneSelf { + r++ + continue + } + rr, size := utf8.DecodeRune(s[r:]) + if rr == utf8.RuneError && size == 1 { + break + } + r += size + } + if r == len(s) { + return s, true + } + + b := make([]byte, len(s)+2*utf8.UTFMax) + w := copy(b, s[0:r]) + for r < len(s) { + // Out of room? Can only happen if s is full of + // malformed UTF-8 and we're replacing each + // byte with RuneError. + if w >= len(b)-2*utf8.UTFMax { + nb := make([]byte, (len(b)+utf8.UTFMax)*2) + copy(nb, b[0:w]) + b = nb + } + switch c := s[r]; { + case c == '\\': + r++ + if r >= len(s) { + return + } + switch s[r] { + default: + return + case '"', '\\', '/', '\'': + b[w] = s[r] + r++ + w++ + case 'b': + b[w] = '\b' + r++ + w++ + case 'f': + b[w] = '\f' + r++ + w++ + case 'n': + b[w] = '\n' + r++ + w++ + case 'r': + b[w] = '\r' + r++ + w++ + case 't': + b[w] = '\t' + r++ + w++ + case 'u': + r-- + rr := getu4(s[r:]) + if rr < 0 { + return + } + r += 6 + if utf16.IsSurrogate(rr) { + rr1 := getu4(s[r:]) + if dec := utf16.DecodeRune(rr, rr1); dec != unicode.ReplacementChar { + // A valid pair; consume. + r += 6 + w += utf8.EncodeRune(b[w:], dec) + break + } + // Invalid surrogate; fall back to replacement rune. + rr = unicode.ReplacementChar + } + w += utf8.EncodeRune(b[w:], rr) + } + + // Quote, control characters are invalid. + case c == '"', c < ' ': + return + + // ASCII + case c < utf8.RuneSelf: + b[w] = c + r++ + w++ + + // Coerce to well-formed UTF-8. + default: + rr, size := utf8.DecodeRune(s[r:]) + r += size + w += utf8.EncodeRune(b[w:], rr) + } + } + return b[0:w], true +} diff --git a/contrib/go/_std_1.22/src/encoding/json/encode.go b/contrib/go/_std_1.22/src/encoding/json/encode.go new file mode 100644 index 0000000000..d6f6900dc9 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/json/encode.go @@ -0,0 +1,1280 @@ +// Copyright 2010 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +// Package json implements encoding and decoding of JSON as defined in +// RFC 7159. The mapping between JSON and Go values is described +// in the documentation for the Marshal and Unmarshal functions. +// +// See "JSON and Go" for an introduction to this package: +// https://golang.org/doc/articles/json_and_go.html +package json + +import ( + "bytes" + "encoding" + "encoding/base64" + "fmt" + "math" + "reflect" + "slices" + "sort" + "strconv" + "strings" + "sync" + "unicode" + "unicode/utf8" +) + +// Marshal returns the JSON encoding of v. +// +// Marshal traverses the value v recursively. +// If an encountered value implements [Marshaler] +// and is not a nil pointer, Marshal calls [Marshaler.MarshalJSON] +// to produce JSON. If no [Marshaler.MarshalJSON] method is present but the +// value implements [encoding.TextMarshaler] instead, Marshal calls +// [encoding.TextMarshaler.MarshalText] and encodes the result as a JSON string. +// The nil pointer exception is not strictly necessary +// but mimics a similar, necessary exception in the behavior of +// [Unmarshaler.UnmarshalJSON]. +// +// Otherwise, Marshal uses the following type-dependent default encodings: +// +// Boolean values encode as JSON booleans. +// +// Floating point, integer, and [Number] values encode as JSON numbers. +// NaN and +/-Inf values will return an [UnsupportedValueError]. +// +// String values encode as JSON strings coerced to valid UTF-8, +// replacing invalid bytes with the Unicode replacement rune. +// So that the JSON will be safe to embed inside HTML <script> tags, +// the string is encoded using [HTMLEscape], +// which replaces "<", ">", "&", U+2028, and U+2029 are escaped +// to "\u003c","\u003e", "\u0026", "\u2028", and "\u2029". +// This replacement can be disabled when using an [Encoder], +// by calling [Encoder.SetEscapeHTML](false). +// +// Array and slice values encode as JSON arrays, except that +// []byte encodes as a base64-encoded string, and a nil slice +// encodes as the null JSON value. +// +// Struct values encode as JSON objects. +// Each exported struct field becomes a member of the object, using the +// field name as the object key, unless the field is omitted for one of the +// reasons given below. +// +// The encoding of each struct field can be customized by the format string +// stored under the "json" key in the struct field's tag. +// The format string gives the name of the field, possibly followed by a +// comma-separated list of options. The name may be empty in order to +// specify options without overriding the default field name. +// +// The "omitempty" option specifies that the field should be omitted +// from the encoding if the field has an empty value, defined as +// false, 0, a nil pointer, a nil interface value, and any empty array, +// slice, map, or string. +// +// As a special case, if the field tag is "-", the field is always omitted. +// Note that a field with name "-" can still be generated using the tag "-,". +// +// Examples of struct field tags and their meanings: +// +// // Field appears in JSON as key "myName". +// Field int `json:"myName"` +// +// // Field appears in JSON as key "myName" and +// // the field is omitted from the object if its value is empty, +// // as defined above. +// Field int `json:"myName,omitempty"` +// +// // Field appears in JSON as key "Field" (the default), but +// // the field is skipped if empty. +// // Note the leading comma. +// Field int `json:",omitempty"` +// +// // Field is ignored by this package. +// Field int `json:"-"` +// +// // Field appears in JSON as key "-". +// Field int `json:"-,"` +// +// The "string" option signals that a field is stored as JSON inside a +// JSON-encoded string. It applies only to fields of string, floating point, +// integer, or boolean types. This extra level of encoding is sometimes used +// when communicating with JavaScript programs: +// +// Int64String int64 `json:",string"` +// +// The key name will be used if it's a non-empty string consisting of +// only Unicode letters, digits, and ASCII punctuation except quotation +// marks, backslash, and comma. +// +// Embedded struct fields are usually marshaled as if their inner exported fields +// were fields in the outer struct, subject to the usual Go visibility rules amended +// as described in the next paragraph. +// An anonymous struct field with a name given in its JSON tag is treated as +// having that name, rather than being anonymous. +// An anonymous struct field of interface type is treated the same as having +// that type as its name, rather than being anonymous. +// +// The Go visibility rules for struct fields are amended for JSON when +// deciding which field to marshal or unmarshal. If there are +// multiple fields at the same level, and that level is the least +// nested (and would therefore be the nesting level selected by the +// usual Go rules), the following extra rules apply: +// +// 1) Of those fields, if any are JSON-tagged, only tagged fields are considered, +// even if there are multiple untagged fields that would otherwise conflict. +// +// 2) If there is exactly one field (tagged or not according to the first rule), that is selected. +// +// 3) Otherwise there are multiple fields, and all are ignored; no error occurs. +// +// Handling of anonymous struct fields is new in Go 1.1. +// Prior to Go 1.1, anonymous struct fields were ignored. To force ignoring of +// an anonymous struct field in both current and earlier versions, give the field +// a JSON tag of "-". +// +// Map values encode as JSON objects. The map's key type must either be a +// string, an integer type, or implement [encoding.TextMarshaler]. The map keys +// are sorted and used as JSON object keys by applying the following rules, +// subject to the UTF-8 coercion described for string values above: +// - keys of any string type are used directly +// - [encoding.TextMarshalers] are marshaled +// - integer keys are converted to strings +// +// Pointer values encode as the value pointed to. +// A nil pointer encodes as the null JSON value. +// +// Interface values encode as the value contained in the interface. +// A nil interface value encodes as the null JSON value. +// +// Channel, complex, and function values cannot be encoded in JSON. +// Attempting to encode such a value causes Marshal to return +// an [UnsupportedTypeError]. +// +// JSON cannot represent cyclic data structures and Marshal does not +// handle them. Passing cyclic structures to Marshal will result in +// an error. +func Marshal(v any) ([]byte, error) { + e := newEncodeState() + defer encodeStatePool.Put(e) + + err := e.marshal(v, encOpts{escapeHTML: true}) + if err != nil { + return nil, err + } + buf := append([]byte(nil), e.Bytes()...) + + return buf, nil +} + +// MarshalIndent is like [Marshal] but applies [Indent] to format the output. +// Each JSON element in the output will begin on a new line beginning with prefix +// followed by one or more copies of indent according to the indentation nesting. +func MarshalIndent(v any, prefix, indent string) ([]byte, error) { + b, err := Marshal(v) + if err != nil { + return nil, err + } + b2 := make([]byte, 0, indentGrowthFactor*len(b)) + b2, err = appendIndent(b2, b, prefix, indent) + if err != nil { + return nil, err + } + return b2, nil +} + +// Marshaler is the interface implemented by types that +// can marshal themselves into valid JSON. +type Marshaler interface { + MarshalJSON() ([]byte, error) +} + +// An UnsupportedTypeError is returned by [Marshal] when attempting +// to encode an unsupported value type. +type UnsupportedTypeError struct { + Type reflect.Type +} + +func (e *UnsupportedTypeError) Error() string { + return "json: unsupported type: " + e.Type.String() +} + +// An UnsupportedValueError is returned by [Marshal] when attempting +// to encode an unsupported value. +type UnsupportedValueError struct { + Value reflect.Value + Str string +} + +func (e *UnsupportedValueError) Error() string { + return "json: unsupported value: " + e.Str +} + +// Before Go 1.2, an InvalidUTF8Error was returned by [Marshal] when +// attempting to encode a string value with invalid UTF-8 sequences. +// As of Go 1.2, [Marshal] instead coerces the string to valid UTF-8 by +// replacing invalid bytes with the Unicode replacement rune U+FFFD. +// +// Deprecated: No longer used; kept for compatibility. +type InvalidUTF8Error struct { + S string // the whole string value that caused the error +} + +func (e *InvalidUTF8Error) Error() string { + return "json: invalid UTF-8 in string: " + strconv.Quote(e.S) +} + +// A MarshalerError represents an error from calling a +// [Marshaler.MarshalJSON] or [encoding.TextMarshaler.MarshalText] method. +type MarshalerError struct { + Type reflect.Type + Err error + sourceFunc string +} + +func (e *MarshalerError) Error() string { + srcFunc := e.sourceFunc + if srcFunc == "" { + srcFunc = "MarshalJSON" + } + return "json: error calling " + srcFunc + + " for type " + e.Type.String() + + ": " + e.Err.Error() +} + +// Unwrap returns the underlying error. +func (e *MarshalerError) Unwrap() error { return e.Err } + +const hex = "0123456789abcdef" + +// An encodeState encodes JSON into a bytes.Buffer. +type encodeState struct { + bytes.Buffer // accumulated output + + // Keep track of what pointers we've seen in the current recursive call + // path, to avoid cycles that could lead to a stack overflow. Only do + // the relatively expensive map operations if ptrLevel is larger than + // startDetectingCyclesAfter, so that we skip the work if we're within a + // reasonable amount of nested pointers deep. + ptrLevel uint + ptrSeen map[any]struct{} +} + +const startDetectingCyclesAfter = 1000 + +var encodeStatePool sync.Pool + +func newEncodeState() *encodeState { + if v := encodeStatePool.Get(); v != nil { + e := v.(*encodeState) + e.Reset() + if len(e.ptrSeen) > 0 { + panic("ptrEncoder.encode should have emptied ptrSeen via defers") + } + e.ptrLevel = 0 + return e + } + return &encodeState{ptrSeen: make(map[any]struct{})} +} + +// jsonError is an error wrapper type for internal use only. +// Panics with errors are wrapped in jsonError so that the top-level recover +// can distinguish intentional panics from this package. +type jsonError struct{ error } + +func (e *encodeState) marshal(v any, opts encOpts) (err error) { + defer func() { + if r := recover(); r != nil { + if je, ok := r.(jsonError); ok { + err = je.error + } else { + panic(r) + } + } + }() + e.reflectValue(reflect.ValueOf(v), opts) + return nil +} + +// error aborts the encoding by panicking with err wrapped in jsonError. +func (e *encodeState) error(err error) { + panic(jsonError{err}) +} + +func isEmptyValue(v reflect.Value) bool { + switch v.Kind() { + case reflect.Array, reflect.Map, reflect.Slice, reflect.String: + return v.Len() == 0 + case reflect.Bool, + reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64, + reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr, + reflect.Float32, reflect.Float64, + reflect.Interface, reflect.Pointer: + return v.IsZero() + } + return false +} + +func (e *encodeState) reflectValue(v reflect.Value, opts encOpts) { + valueEncoder(v)(e, v, opts) +} + +type encOpts struct { + // quoted causes primitive fields to be encoded inside JSON strings. + quoted bool + // escapeHTML causes '<', '>', and '&' to be escaped in JSON strings. + escapeHTML bool +} + +type encoderFunc func(e *encodeState, v reflect.Value, opts encOpts) + +var encoderCache sync.Map // map[reflect.Type]encoderFunc + +func valueEncoder(v reflect.Value) encoderFunc { + if !v.IsValid() { + return invalidValueEncoder + } + return typeEncoder(v.Type()) +} + +func typeEncoder(t reflect.Type) encoderFunc { + if fi, ok := encoderCache.Load(t); ok { + return fi.(encoderFunc) + } + + // To deal with recursive types, populate the map with an + // indirect func before we build it. This type waits on the + // real func (f) to be ready and then calls it. This indirect + // func is only used for recursive types. + var ( + wg sync.WaitGroup + f encoderFunc + ) + wg.Add(1) + fi, loaded := encoderCache.LoadOrStore(t, encoderFunc(func(e *encodeState, v reflect.Value, opts encOpts) { + wg.Wait() + f(e, v, opts) + })) + if loaded { + return fi.(encoderFunc) + } + + // Compute the real encoder and replace the indirect func with it. + f = newTypeEncoder(t, true) + wg.Done() + encoderCache.Store(t, f) + return f +} + +var ( + marshalerType = reflect.TypeFor[Marshaler]() + textMarshalerType = reflect.TypeFor[encoding.TextMarshaler]() +) + +// newTypeEncoder constructs an encoderFunc for a type. +// The returned encoder only checks CanAddr when allowAddr is true. +func newTypeEncoder(t reflect.Type, allowAddr bool) encoderFunc { + // If we have a non-pointer value whose type implements + // Marshaler with a value receiver, then we're better off taking + // the address of the value - otherwise we end up with an + // allocation as we cast the value to an interface. + if t.Kind() != reflect.Pointer && allowAddr && reflect.PointerTo(t).Implements(marshalerType) { + return newCondAddrEncoder(addrMarshalerEncoder, newTypeEncoder(t, false)) + } + if t.Implements(marshalerType) { + return marshalerEncoder + } + if t.Kind() != reflect.Pointer && allowAddr && reflect.PointerTo(t).Implements(textMarshalerType) { + return newCondAddrEncoder(addrTextMarshalerEncoder, newTypeEncoder(t, false)) + } + if t.Implements(textMarshalerType) { + return textMarshalerEncoder + } + + switch t.Kind() { + case reflect.Bool: + return boolEncoder + case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64: + return intEncoder + case reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr: + return uintEncoder + case reflect.Float32: + return float32Encoder + case reflect.Float64: + return float64Encoder + case reflect.String: + return stringEncoder + case reflect.Interface: + return interfaceEncoder + case reflect.Struct: + return newStructEncoder(t) + case reflect.Map: + return newMapEncoder(t) + case reflect.Slice: + return newSliceEncoder(t) + case reflect.Array: + return newArrayEncoder(t) + case reflect.Pointer: + return newPtrEncoder(t) + default: + return unsupportedTypeEncoder + } +} + +func invalidValueEncoder(e *encodeState, v reflect.Value, _ encOpts) { + e.WriteString("null") +} + +func marshalerEncoder(e *encodeState, v reflect.Value, opts encOpts) { + if v.Kind() == reflect.Pointer && v.IsNil() { + e.WriteString("null") + return + } + m, ok := v.Interface().(Marshaler) + if !ok { + e.WriteString("null") + return + } + b, err := m.MarshalJSON() + if err == nil { + e.Grow(len(b)) + out := e.AvailableBuffer() + out, err = appendCompact(out, b, opts.escapeHTML) + e.Buffer.Write(out) + } + if err != nil { + e.error(&MarshalerError{v.Type(), err, "MarshalJSON"}) + } +} + +func addrMarshalerEncoder(e *encodeState, v reflect.Value, opts encOpts) { + va := v.Addr() + if va.IsNil() { + e.WriteString("null") + return + } + m := va.Interface().(Marshaler) + b, err := m.MarshalJSON() + if err == nil { + e.Grow(len(b)) + out := e.AvailableBuffer() + out, err = appendCompact(out, b, opts.escapeHTML) + e.Buffer.Write(out) + } + if err != nil { + e.error(&MarshalerError{v.Type(), err, "MarshalJSON"}) + } +} + +func textMarshalerEncoder(e *encodeState, v reflect.Value, opts encOpts) { + if v.Kind() == reflect.Pointer && v.IsNil() { + e.WriteString("null") + return + } + m, ok := v.Interface().(encoding.TextMarshaler) + if !ok { + e.WriteString("null") + return + } + b, err := m.MarshalText() + if err != nil { + e.error(&MarshalerError{v.Type(), err, "MarshalText"}) + } + e.Write(appendString(e.AvailableBuffer(), b, opts.escapeHTML)) +} + +func addrTextMarshalerEncoder(e *encodeState, v reflect.Value, opts encOpts) { + va := v.Addr() + if va.IsNil() { + e.WriteString("null") + return + } + m := va.Interface().(encoding.TextMarshaler) + b, err := m.MarshalText() + if err != nil { + e.error(&MarshalerError{v.Type(), err, "MarshalText"}) + } + e.Write(appendString(e.AvailableBuffer(), b, opts.escapeHTML)) +} + +func boolEncoder(e *encodeState, v reflect.Value, opts encOpts) { + b := e.AvailableBuffer() + b = mayAppendQuote(b, opts.quoted) + b = strconv.AppendBool(b, v.Bool()) + b = mayAppendQuote(b, opts.quoted) + e.Write(b) +} + +func intEncoder(e *encodeState, v reflect.Value, opts encOpts) { + b := e.AvailableBuffer() + b = mayAppendQuote(b, opts.quoted) + b = strconv.AppendInt(b, v.Int(), 10) + b = mayAppendQuote(b, opts.quoted) + e.Write(b) +} + +func uintEncoder(e *encodeState, v reflect.Value, opts encOpts) { + b := e.AvailableBuffer() + b = mayAppendQuote(b, opts.quoted) + b = strconv.AppendUint(b, v.Uint(), 10) + b = mayAppendQuote(b, opts.quoted) + e.Write(b) +} + +type floatEncoder int // number of bits + +func (bits floatEncoder) encode(e *encodeState, v reflect.Value, opts encOpts) { + f := v.Float() + if math.IsInf(f, 0) || math.IsNaN(f) { + e.error(&UnsupportedValueError{v, strconv.FormatFloat(f, 'g', -1, int(bits))}) + } + + // Convert as if by ES6 number to string conversion. + // This matches most other JSON generators. + // See golang.org/issue/6384 and golang.org/issue/14135. + // Like fmt %g, but the exponent cutoffs are different + // and exponents themselves are not padded to two digits. + b := e.AvailableBuffer() + b = mayAppendQuote(b, opts.quoted) + abs := math.Abs(f) + fmt := byte('f') + // Note: Must use float32 comparisons for underlying float32 value to get precise cutoffs right. + if abs != 0 { + if bits == 64 && (abs < 1e-6 || abs >= 1e21) || bits == 32 && (float32(abs) < 1e-6 || float32(abs) >= 1e21) { + fmt = 'e' + } + } + b = strconv.AppendFloat(b, f, fmt, -1, int(bits)) + if fmt == 'e' { + // clean up e-09 to e-9 + n := len(b) + if n >= 4 && b[n-4] == 'e' && b[n-3] == '-' && b[n-2] == '0' { + b[n-2] = b[n-1] + b = b[:n-1] + } + } + b = mayAppendQuote(b, opts.quoted) + e.Write(b) +} + +var ( + float32Encoder = (floatEncoder(32)).encode + float64Encoder = (floatEncoder(64)).encode +) + +func stringEncoder(e *encodeState, v reflect.Value, opts encOpts) { + if v.Type() == numberType { + numStr := v.String() + // In Go1.5 the empty string encodes to "0", while this is not a valid number literal + // we keep compatibility so check validity after this. + if numStr == "" { + numStr = "0" // Number's zero-val + } + if !isValidNumber(numStr) { + e.error(fmt.Errorf("json: invalid number literal %q", numStr)) + } + b := e.AvailableBuffer() + b = mayAppendQuote(b, opts.quoted) + b = append(b, numStr...) + b = mayAppendQuote(b, opts.quoted) + e.Write(b) + return + } + if opts.quoted { + b := appendString(nil, v.String(), opts.escapeHTML) + e.Write(appendString(e.AvailableBuffer(), b, false)) // no need to escape again since it is already escaped + } else { + e.Write(appendString(e.AvailableBuffer(), v.String(), opts.escapeHTML)) + } +} + +// isValidNumber reports whether s is a valid JSON number literal. +func isValidNumber(s string) bool { + // This function implements the JSON numbers grammar. + // See https://tools.ietf.org/html/rfc7159#section-6 + // and https://www.json.org/img/number.png + + if s == "" { + return false + } + + // Optional - + if s[0] == '-' { + s = s[1:] + if s == "" { + return false + } + } + + // Digits + switch { + default: + return false + + case s[0] == '0': + s = s[1:] + + case '1' <= s[0] && s[0] <= '9': + s = s[1:] + for len(s) > 0 && '0' <= s[0] && s[0] <= '9' { + s = s[1:] + } + } + + // . followed by 1 or more digits. + if len(s) >= 2 && s[0] == '.' && '0' <= s[1] && s[1] <= '9' { + s = s[2:] + for len(s) > 0 && '0' <= s[0] && s[0] <= '9' { + s = s[1:] + } + } + + // e or E followed by an optional - or + and + // 1 or more digits. + if len(s) >= 2 && (s[0] == 'e' || s[0] == 'E') { + s = s[1:] + if s[0] == '+' || s[0] == '-' { + s = s[1:] + if s == "" { + return false + } + } + for len(s) > 0 && '0' <= s[0] && s[0] <= '9' { + s = s[1:] + } + } + + // Make sure we are at the end. + return s == "" +} + +func interfaceEncoder(e *encodeState, v reflect.Value, opts encOpts) { + if v.IsNil() { + e.WriteString("null") + return + } + e.reflectValue(v.Elem(), opts) +} + +func unsupportedTypeEncoder(e *encodeState, v reflect.Value, _ encOpts) { + e.error(&UnsupportedTypeError{v.Type()}) +} + +type structEncoder struct { + fields structFields +} + +type structFields struct { + list []field + byExactName map[string]*field + byFoldedName map[string]*field +} + +func (se structEncoder) encode(e *encodeState, v reflect.Value, opts encOpts) { + next := byte('{') +FieldLoop: + for i := range se.fields.list { + f := &se.fields.list[i] + + // Find the nested struct field by following f.index. + fv := v + for _, i := range f.index { + if fv.Kind() == reflect.Pointer { + if fv.IsNil() { + continue FieldLoop + } + fv = fv.Elem() + } + fv = fv.Field(i) + } + + if f.omitEmpty && isEmptyValue(fv) { + continue + } + e.WriteByte(next) + next = ',' + if opts.escapeHTML { + e.WriteString(f.nameEscHTML) + } else { + e.WriteString(f.nameNonEsc) + } + opts.quoted = f.quoted + f.encoder(e, fv, opts) + } + if next == '{' { + e.WriteString("{}") + } else { + e.WriteByte('}') + } +} + +func newStructEncoder(t reflect.Type) encoderFunc { + se := structEncoder{fields: cachedTypeFields(t)} + return se.encode +} + +type mapEncoder struct { + elemEnc encoderFunc +} + +func (me mapEncoder) encode(e *encodeState, v reflect.Value, opts encOpts) { + if v.IsNil() { + e.WriteString("null") + return + } + if e.ptrLevel++; e.ptrLevel > startDetectingCyclesAfter { + // We're a large number of nested ptrEncoder.encode calls deep; + // start checking if we've run into a pointer cycle. + ptr := v.UnsafePointer() + if _, ok := e.ptrSeen[ptr]; ok { + e.error(&UnsupportedValueError{v, fmt.Sprintf("encountered a cycle via %s", v.Type())}) + } + e.ptrSeen[ptr] = struct{}{} + defer delete(e.ptrSeen, ptr) + } + e.WriteByte('{') + + // Extract and sort the keys. + var ( + sv = make([]reflectWithString, v.Len()) + mi = v.MapRange() + err error + ) + for i := 0; mi.Next(); i++ { + if sv[i].ks, err = resolveKeyName(mi.Key()); err != nil { + e.error(fmt.Errorf("json: encoding error for type %q: %q", v.Type().String(), err.Error())) + } + sv[i].v = mi.Value() + } + slices.SortFunc(sv, func(i, j reflectWithString) int { + return strings.Compare(i.ks, j.ks) + }) + + for i, kv := range sv { + if i > 0 { + e.WriteByte(',') + } + e.Write(appendString(e.AvailableBuffer(), kv.ks, opts.escapeHTML)) + e.WriteByte(':') + me.elemEnc(e, kv.v, opts) + } + e.WriteByte('}') + e.ptrLevel-- +} + +func newMapEncoder(t reflect.Type) encoderFunc { + switch t.Key().Kind() { + case reflect.String, + reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64, + reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr: + default: + if !t.Key().Implements(textMarshalerType) { + return unsupportedTypeEncoder + } + } + me := mapEncoder{typeEncoder(t.Elem())} + return me.encode +} + +func encodeByteSlice(e *encodeState, v reflect.Value, _ encOpts) { + if v.IsNil() { + e.WriteString("null") + return + } + + s := v.Bytes() + b := e.AvailableBuffer() + b = append(b, '"') + b = base64.StdEncoding.AppendEncode(b, s) + b = append(b, '"') + e.Write(b) +} + +// sliceEncoder just wraps an arrayEncoder, checking to make sure the value isn't nil. +type sliceEncoder struct { + arrayEnc encoderFunc +} + +func (se sliceEncoder) encode(e *encodeState, v reflect.Value, opts encOpts) { + if v.IsNil() { + e.WriteString("null") + return + } + if e.ptrLevel++; e.ptrLevel > startDetectingCyclesAfter { + // We're a large number of nested ptrEncoder.encode calls deep; + // start checking if we've run into a pointer cycle. + // Here we use a struct to memorize the pointer to the first element of the slice + // and its length. + ptr := struct { + ptr interface{} // always an unsafe.Pointer, but avoids a dependency on package unsafe + len int + }{v.UnsafePointer(), v.Len()} + if _, ok := e.ptrSeen[ptr]; ok { + e.error(&UnsupportedValueError{v, fmt.Sprintf("encountered a cycle via %s", v.Type())}) + } + e.ptrSeen[ptr] = struct{}{} + defer delete(e.ptrSeen, ptr) + } + se.arrayEnc(e, v, opts) + e.ptrLevel-- +} + +func newSliceEncoder(t reflect.Type) encoderFunc { + // Byte slices get special treatment; arrays don't. + if t.Elem().Kind() == reflect.Uint8 { + p := reflect.PointerTo(t.Elem()) + if !p.Implements(marshalerType) && !p.Implements(textMarshalerType) { + return encodeByteSlice + } + } + enc := sliceEncoder{newArrayEncoder(t)} + return enc.encode +} + +type arrayEncoder struct { + elemEnc encoderFunc +} + +func (ae arrayEncoder) encode(e *encodeState, v reflect.Value, opts encOpts) { + e.WriteByte('[') + n := v.Len() + for i := 0; i < n; i++ { + if i > 0 { + e.WriteByte(',') + } + ae.elemEnc(e, v.Index(i), opts) + } + e.WriteByte(']') +} + +func newArrayEncoder(t reflect.Type) encoderFunc { + enc := arrayEncoder{typeEncoder(t.Elem())} + return enc.encode +} + +type ptrEncoder struct { + elemEnc encoderFunc +} + +func (pe ptrEncoder) encode(e *encodeState, v reflect.Value, opts encOpts) { + if v.IsNil() { + e.WriteString("null") + return + } + if e.ptrLevel++; e.ptrLevel > startDetectingCyclesAfter { + // We're a large number of nested ptrEncoder.encode calls deep; + // start checking if we've run into a pointer cycle. + ptr := v.Interface() + if _, ok := e.ptrSeen[ptr]; ok { + e.error(&UnsupportedValueError{v, fmt.Sprintf("encountered a cycle via %s", v.Type())}) + } + e.ptrSeen[ptr] = struct{}{} + defer delete(e.ptrSeen, ptr) + } + pe.elemEnc(e, v.Elem(), opts) + e.ptrLevel-- +} + +func newPtrEncoder(t reflect.Type) encoderFunc { + enc := ptrEncoder{typeEncoder(t.Elem())} + return enc.encode +} + +type condAddrEncoder struct { + canAddrEnc, elseEnc encoderFunc +} + +func (ce condAddrEncoder) encode(e *encodeState, v reflect.Value, opts encOpts) { + if v.CanAddr() { + ce.canAddrEnc(e, v, opts) + } else { + ce.elseEnc(e, v, opts) + } +} + +// newCondAddrEncoder returns an encoder that checks whether its value +// CanAddr and delegates to canAddrEnc if so, else to elseEnc. +func newCondAddrEncoder(canAddrEnc, elseEnc encoderFunc) encoderFunc { + enc := condAddrEncoder{canAddrEnc: canAddrEnc, elseEnc: elseEnc} + return enc.encode +} + +func isValidTag(s string) bool { + if s == "" { + return false + } + for _, c := range s { + switch { + case strings.ContainsRune("!#$%&()*+-./:;<=>?@[]^_{|}~ ", c): + // Backslash and quote chars are reserved, but + // otherwise any punctuation chars are allowed + // in a tag name. + case !unicode.IsLetter(c) && !unicode.IsDigit(c): + return false + } + } + return true +} + +func typeByIndex(t reflect.Type, index []int) reflect.Type { + for _, i := range index { + if t.Kind() == reflect.Pointer { + t = t.Elem() + } + t = t.Field(i).Type + } + return t +} + +type reflectWithString struct { + v reflect.Value + ks string +} + +func resolveKeyName(k reflect.Value) (string, error) { + if k.Kind() == reflect.String { + return k.String(), nil + } + if tm, ok := k.Interface().(encoding.TextMarshaler); ok { + if k.Kind() == reflect.Pointer && k.IsNil() { + return "", nil + } + buf, err := tm.MarshalText() + return string(buf), err + } + switch k.Kind() { + case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64: + return strconv.FormatInt(k.Int(), 10), nil + case reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr: + return strconv.FormatUint(k.Uint(), 10), nil + } + panic("unexpected map key type") +} + +func appendString[Bytes []byte | string](dst []byte, src Bytes, escapeHTML bool) []byte { + dst = append(dst, '"') + start := 0 + for i := 0; i < len(src); { + if b := src[i]; b < utf8.RuneSelf { + if htmlSafeSet[b] || (!escapeHTML && safeSet[b]) { + i++ + continue + } + dst = append(dst, src[start:i]...) + switch b { + case '\\', '"': + dst = append(dst, '\\', b) + case '\b': + dst = append(dst, '\\', 'b') + case '\f': + dst = append(dst, '\\', 'f') + case '\n': + dst = append(dst, '\\', 'n') + case '\r': + dst = append(dst, '\\', 'r') + case '\t': + dst = append(dst, '\\', 't') + default: + // This encodes bytes < 0x20 except for \b, \f, \n, \r and \t. + // If escapeHTML is set, it also escapes <, >, and & + // because they can lead to security holes when + // user-controlled strings are rendered into JSON + // and served to some browsers. + dst = append(dst, '\\', 'u', '0', '0', hex[b>>4], hex[b&0xF]) + } + i++ + start = i + continue + } + // TODO(https://go.dev/issue/56948): Use generic utf8 functionality. + // For now, cast only a small portion of byte slices to a string + // so that it can be stack allocated. This slows down []byte slightly + // due to the extra copy, but keeps string performance roughly the same. + n := len(src) - i + if n > utf8.UTFMax { + n = utf8.UTFMax + } + c, size := utf8.DecodeRuneInString(string(src[i : i+n])) + if c == utf8.RuneError && size == 1 { + dst = append(dst, src[start:i]...) + dst = append(dst, `\ufffd`...) + i += size + start = i + continue + } + // U+2028 is LINE SEPARATOR. + // U+2029 is PARAGRAPH SEPARATOR. + // They are both technically valid characters in JSON strings, + // but don't work in JSONP, which has to be evaluated as JavaScript, + // and can lead to security holes there. It is valid JSON to + // escape them, so we do so unconditionally. + // See https://en.wikipedia.org/wiki/JSON#Safety. + if c == '\u2028' || c == '\u2029' { + dst = append(dst, src[start:i]...) + dst = append(dst, '\\', 'u', '2', '0', '2', hex[c&0xF]) + i += size + start = i + continue + } + i += size + } + dst = append(dst, src[start:]...) + dst = append(dst, '"') + return dst +} + +// A field represents a single field found in a struct. +type field struct { + name string + nameBytes []byte // []byte(name) + + nameNonEsc string // `"` + name + `":` + nameEscHTML string // `"` + HTMLEscape(name) + `":` + + tag bool + index []int + typ reflect.Type + omitEmpty bool + quoted bool + + encoder encoderFunc +} + +// byIndex sorts field by index sequence. +type byIndex []field + +func (x byIndex) Len() int { return len(x) } + +func (x byIndex) Swap(i, j int) { x[i], x[j] = x[j], x[i] } + +func (x byIndex) Less(i, j int) bool { + for k, xik := range x[i].index { + if k >= len(x[j].index) { + return false + } + if xik != x[j].index[k] { + return xik < x[j].index[k] + } + } + return len(x[i].index) < len(x[j].index) +} + +// typeFields returns a list of fields that JSON should recognize for the given type. +// The algorithm is breadth-first search over the set of structs to include - the top struct +// and then any reachable anonymous structs. +func typeFields(t reflect.Type) structFields { + // Anonymous fields to explore at the current level and the next. + current := []field{} + next := []field{{typ: t}} + + // Count of queued names for current level and the next. + var count, nextCount map[reflect.Type]int + + // Types already visited at an earlier level. + visited := map[reflect.Type]bool{} + + // Fields found. + var fields []field + + // Buffer to run appendHTMLEscape on field names. + var nameEscBuf []byte + + for len(next) > 0 { + current, next = next, current[:0] + count, nextCount = nextCount, map[reflect.Type]int{} + + for _, f := range current { + if visited[f.typ] { + continue + } + visited[f.typ] = true + + // Scan f.typ for fields to include. + for i := 0; i < f.typ.NumField(); i++ { + sf := f.typ.Field(i) + if sf.Anonymous { + t := sf.Type + if t.Kind() == reflect.Pointer { + t = t.Elem() + } + if !sf.IsExported() && t.Kind() != reflect.Struct { + // Ignore embedded fields of unexported non-struct types. + continue + } + // Do not ignore embedded fields of unexported struct types + // since they may have exported fields. + } else if !sf.IsExported() { + // Ignore unexported non-embedded fields. + continue + } + tag := sf.Tag.Get("json") + if tag == "-" { + continue + } + name, opts := parseTag(tag) + if !isValidTag(name) { + name = "" + } + index := make([]int, len(f.index)+1) + copy(index, f.index) + index[len(f.index)] = i + + ft := sf.Type + if ft.Name() == "" && ft.Kind() == reflect.Pointer { + // Follow pointer. + ft = ft.Elem() + } + + // Only strings, floats, integers, and booleans can be quoted. + quoted := false + if opts.Contains("string") { + switch ft.Kind() { + case reflect.Bool, + reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64, + reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr, + reflect.Float32, reflect.Float64, + reflect.String: + quoted = true + } + } + + // Record found field and index sequence. + if name != "" || !sf.Anonymous || ft.Kind() != reflect.Struct { + tagged := name != "" + if name == "" { + name = sf.Name + } + field := field{ + name: name, + tag: tagged, + index: index, + typ: ft, + omitEmpty: opts.Contains("omitempty"), + quoted: quoted, + } + field.nameBytes = []byte(field.name) + + // Build nameEscHTML and nameNonEsc ahead of time. + nameEscBuf = appendHTMLEscape(nameEscBuf[:0], field.nameBytes) + field.nameEscHTML = `"` + string(nameEscBuf) + `":` + field.nameNonEsc = `"` + field.name + `":` + + fields = append(fields, field) + if count[f.typ] > 1 { + // If there were multiple instances, add a second, + // so that the annihilation code will see a duplicate. + // It only cares about the distinction between 1 and 2, + // so don't bother generating any more copies. + fields = append(fields, fields[len(fields)-1]) + } + continue + } + + // Record new anonymous struct to explore in next round. + nextCount[ft]++ + if nextCount[ft] == 1 { + next = append(next, field{name: ft.Name(), index: index, typ: ft}) + } + } + } + } + + sort.Slice(fields, func(i, j int) bool { + x := fields + // sort field by name, breaking ties with depth, then + // breaking ties with "name came from json tag", then + // breaking ties with index sequence. + if x[i].name != x[j].name { + return x[i].name < x[j].name + } + if len(x[i].index) != len(x[j].index) { + return len(x[i].index) < len(x[j].index) + } + if x[i].tag != x[j].tag { + return x[i].tag + } + return byIndex(x).Less(i, j) + }) + + // Delete all fields that are hidden by the Go rules for embedded fields, + // except that fields with JSON tags are promoted. + + // The fields are sorted in primary order of name, secondary order + // of field index length. Loop over names; for each name, delete + // hidden fields by choosing the one dominant field that survives. + out := fields[:0] + for advance, i := 0, 0; i < len(fields); i += advance { + // One iteration per name. + // Find the sequence of fields with the name of this first field. + fi := fields[i] + name := fi.name + for advance = 1; i+advance < len(fields); advance++ { + fj := fields[i+advance] + if fj.name != name { + break + } + } + if advance == 1 { // Only one field with this name + out = append(out, fi) + continue + } + dominant, ok := dominantField(fields[i : i+advance]) + if ok { + out = append(out, dominant) + } + } + + fields = out + sort.Sort(byIndex(fields)) + + for i := range fields { + f := &fields[i] + f.encoder = typeEncoder(typeByIndex(t, f.index)) + } + exactNameIndex := make(map[string]*field, len(fields)) + foldedNameIndex := make(map[string]*field, len(fields)) + for i, field := range fields { + exactNameIndex[field.name] = &fields[i] + // For historical reasons, first folded match takes precedence. + if _, ok := foldedNameIndex[string(foldName(field.nameBytes))]; !ok { + foldedNameIndex[string(foldName(field.nameBytes))] = &fields[i] + } + } + return structFields{fields, exactNameIndex, foldedNameIndex} +} + +// dominantField looks through the fields, all of which are known to +// have the same name, to find the single field that dominates the +// others using Go's embedding rules, modified by the presence of +// JSON tags. If there are multiple top-level fields, the boolean +// will be false: This condition is an error in Go and we skip all +// the fields. +func dominantField(fields []field) (field, bool) { + // The fields are sorted in increasing index-length order, then by presence of tag. + // That means that the first field is the dominant one. We need only check + // for error cases: two fields at top level, either both tagged or neither tagged. + if len(fields) > 1 && len(fields[0].index) == len(fields[1].index) && fields[0].tag == fields[1].tag { + return field{}, false + } + return fields[0], true +} + +var fieldCache sync.Map // map[reflect.Type]structFields + +// cachedTypeFields is like typeFields but uses a cache to avoid repeated work. +func cachedTypeFields(t reflect.Type) structFields { + if f, ok := fieldCache.Load(t); ok { + return f.(structFields) + } + f, _ := fieldCache.LoadOrStore(t, typeFields(t)) + return f.(structFields) +} + +func mayAppendQuote(b []byte, quoted bool) []byte { + if quoted { + b = append(b, '"') + } + return b +} diff --git a/contrib/go/_std_1.22/src/encoding/json/fold.go b/contrib/go/_std_1.22/src/encoding/json/fold.go new file mode 100644 index 0000000000..c4c671b527 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/json/fold.go @@ -0,0 +1,48 @@ +// Copyright 2013 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package json + +import ( + "unicode" + "unicode/utf8" +) + +// foldName returns a folded string such that foldName(x) == foldName(y) +// is identical to bytes.EqualFold(x, y). +func foldName(in []byte) []byte { + // This is inlinable to take advantage of "function outlining". + var arr [32]byte // large enough for most JSON names + return appendFoldedName(arr[:0], in) +} + +func appendFoldedName(out, in []byte) []byte { + for i := 0; i < len(in); { + // Handle single-byte ASCII. + if c := in[i]; c < utf8.RuneSelf { + if 'a' <= c && c <= 'z' { + c -= 'a' - 'A' + } + out = append(out, c) + i++ + continue + } + // Handle multi-byte Unicode. + r, n := utf8.DecodeRune(in[i:]) + out = utf8.AppendRune(out, foldRune(r)) + i += n + } + return out +} + +// foldRune is returns the smallest rune for all runes in the same fold set. +func foldRune(r rune) rune { + for { + r2 := unicode.SimpleFold(r) + if r2 <= r { + return r2 + } + r = r2 + } +} diff --git a/contrib/go/_std_1.22/src/encoding/json/indent.go b/contrib/go/_std_1.22/src/encoding/json/indent.go new file mode 100644 index 0000000000..01bfdf65e7 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/json/indent.go @@ -0,0 +1,182 @@ +// Copyright 2010 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package json + +import "bytes" + +// HTMLEscape appends to dst the JSON-encoded src with <, >, &, U+2028 and U+2029 +// characters inside string literals changed to \u003c, \u003e, \u0026, \u2028, \u2029 +// so that the JSON will be safe to embed inside HTML <script> tags. +// For historical reasons, web browsers don't honor standard HTML +// escaping within <script> tags, so an alternative JSON encoding must be used. +func HTMLEscape(dst *bytes.Buffer, src []byte) { + dst.Grow(len(src)) + dst.Write(appendHTMLEscape(dst.AvailableBuffer(), src)) +} + +func appendHTMLEscape(dst, src []byte) []byte { + // The characters can only appear in string literals, + // so just scan the string one byte at a time. + start := 0 + for i, c := range src { + if c == '<' || c == '>' || c == '&' { + dst = append(dst, src[start:i]...) + dst = append(dst, '\\', 'u', '0', '0', hex[c>>4], hex[c&0xF]) + start = i + 1 + } + // Convert U+2028 and U+2029 (E2 80 A8 and E2 80 A9). + if c == 0xE2 && i+2 < len(src) && src[i+1] == 0x80 && src[i+2]&^1 == 0xA8 { + dst = append(dst, src[start:i]...) + dst = append(dst, '\\', 'u', '2', '0', '2', hex[src[i+2]&0xF]) + start = i + len("\u2029") + } + } + return append(dst, src[start:]...) +} + +// Compact appends to dst the JSON-encoded src with +// insignificant space characters elided. +func Compact(dst *bytes.Buffer, src []byte) error { + dst.Grow(len(src)) + b := dst.AvailableBuffer() + b, err := appendCompact(b, src, false) + dst.Write(b) + return err +} + +func appendCompact(dst, src []byte, escape bool) ([]byte, error) { + origLen := len(dst) + scan := newScanner() + defer freeScanner(scan) + start := 0 + for i, c := range src { + if escape && (c == '<' || c == '>' || c == '&') { + if start < i { + dst = append(dst, src[start:i]...) + } + dst = append(dst, '\\', 'u', '0', '0', hex[c>>4], hex[c&0xF]) + start = i + 1 + } + // Convert U+2028 and U+2029 (E2 80 A8 and E2 80 A9). + if escape && c == 0xE2 && i+2 < len(src) && src[i+1] == 0x80 && src[i+2]&^1 == 0xA8 { + if start < i { + dst = append(dst, src[start:i]...) + } + dst = append(dst, '\\', 'u', '2', '0', '2', hex[src[i+2]&0xF]) + start = i + 3 + } + v := scan.step(scan, c) + if v >= scanSkipSpace { + if v == scanError { + break + } + if start < i { + dst = append(dst, src[start:i]...) + } + start = i + 1 + } + } + if scan.eof() == scanError { + return dst[:origLen], scan.err + } + if start < len(src) { + dst = append(dst, src[start:]...) + } + return dst, nil +} + +func appendNewline(dst []byte, prefix, indent string, depth int) []byte { + dst = append(dst, '\n') + dst = append(dst, prefix...) + for i := 0; i < depth; i++ { + dst = append(dst, indent...) + } + return dst +} + +// indentGrowthFactor specifies the growth factor of indenting JSON input. +// Empirically, the growth factor was measured to be between 1.4x to 1.8x +// for some set of compacted JSON with the indent being a single tab. +// Specify a growth factor slightly larger than what is observed +// to reduce probability of allocation in appendIndent. +// A factor no higher than 2 ensures that wasted space never exceeds 50%. +const indentGrowthFactor = 2 + +// Indent appends to dst an indented form of the JSON-encoded src. +// Each element in a JSON object or array begins on a new, +// indented line beginning with prefix followed by one or more +// copies of indent according to the indentation nesting. +// The data appended to dst does not begin with the prefix nor +// any indentation, to make it easier to embed inside other formatted JSON data. +// Although leading space characters (space, tab, carriage return, newline) +// at the beginning of src are dropped, trailing space characters +// at the end of src are preserved and copied to dst. +// For example, if src has no trailing spaces, neither will dst; +// if src ends in a trailing newline, so will dst. +func Indent(dst *bytes.Buffer, src []byte, prefix, indent string) error { + dst.Grow(indentGrowthFactor * len(src)) + b := dst.AvailableBuffer() + b, err := appendIndent(b, src, prefix, indent) + dst.Write(b) + return err +} + +func appendIndent(dst, src []byte, prefix, indent string) ([]byte, error) { + origLen := len(dst) + scan := newScanner() + defer freeScanner(scan) + needIndent := false + depth := 0 + for _, c := range src { + scan.bytes++ + v := scan.step(scan, c) + if v == scanSkipSpace { + continue + } + if v == scanError { + break + } + if needIndent && v != scanEndObject && v != scanEndArray { + needIndent = false + depth++ + dst = appendNewline(dst, prefix, indent, depth) + } + + // Emit semantically uninteresting bytes + // (in particular, punctuation in strings) unmodified. + if v == scanContinue { + dst = append(dst, c) + continue + } + + // Add spacing around real punctuation. + switch c { + case '{', '[': + // delay indent so that empty object and array are formatted as {} and []. + needIndent = true + dst = append(dst, c) + case ',': + dst = append(dst, c) + dst = appendNewline(dst, prefix, indent, depth) + case ':': + dst = append(dst, c, ' ') + case '}', ']': + if needIndent { + // suppress indent in empty object/array + needIndent = false + } else { + depth-- + dst = appendNewline(dst, prefix, indent, depth) + } + dst = append(dst, c) + default: + dst = append(dst, c) + } + } + if scan.eof() == scanError { + return dst[:origLen], scan.err + } + return dst, nil +} diff --git a/contrib/go/_std_1.22/src/encoding/json/scanner.go b/contrib/go/_std_1.22/src/encoding/json/scanner.go new file mode 100644 index 0000000000..da6ea2ac8f --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/json/scanner.go @@ -0,0 +1,610 @@ +// Copyright 2010 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package json + +// JSON value parser state machine. +// Just about at the limit of what is reasonable to write by hand. +// Some parts are a bit tedious, but overall it nicely factors out the +// otherwise common code from the multiple scanning functions +// in this package (Compact, Indent, checkValid, etc). +// +// This file starts with two simple examples using the scanner +// before diving into the scanner itself. + +import ( + "strconv" + "sync" +) + +// Valid reports whether data is a valid JSON encoding. +func Valid(data []byte) bool { + scan := newScanner() + defer freeScanner(scan) + return checkValid(data, scan) == nil +} + +// checkValid verifies that data is valid JSON-encoded data. +// scan is passed in for use by checkValid to avoid an allocation. +// checkValid returns nil or a SyntaxError. +func checkValid(data []byte, scan *scanner) error { + scan.reset() + for _, c := range data { + scan.bytes++ + if scan.step(scan, c) == scanError { + return scan.err + } + } + if scan.eof() == scanError { + return scan.err + } + return nil +} + +// A SyntaxError is a description of a JSON syntax error. +// [Unmarshal] will return a SyntaxError if the JSON can't be parsed. +type SyntaxError struct { + msg string // description of error + Offset int64 // error occurred after reading Offset bytes +} + +func (e *SyntaxError) Error() string { return e.msg } + +// A scanner is a JSON scanning state machine. +// Callers call scan.reset and then pass bytes in one at a time +// by calling scan.step(&scan, c) for each byte. +// The return value, referred to as an opcode, tells the +// caller about significant parsing events like beginning +// and ending literals, objects, and arrays, so that the +// caller can follow along if it wishes. +// The return value scanEnd indicates that a single top-level +// JSON value has been completed, *before* the byte that +// just got passed in. (The indication must be delayed in order +// to recognize the end of numbers: is 123 a whole value or +// the beginning of 12345e+6?). +type scanner struct { + // The step is a func to be called to execute the next transition. + // Also tried using an integer constant and a single func + // with a switch, but using the func directly was 10% faster + // on a 64-bit Mac Mini, and it's nicer to read. + step func(*scanner, byte) int + + // Reached end of top-level value. + endTop bool + + // Stack of what we're in the middle of - array values, object keys, object values. + parseState []int + + // Error that happened, if any. + err error + + // total bytes consumed, updated by decoder.Decode (and deliberately + // not set to zero by scan.reset) + bytes int64 +} + +var scannerPool = sync.Pool{ + New: func() any { + return &scanner{} + }, +} + +func newScanner() *scanner { + scan := scannerPool.Get().(*scanner) + // scan.reset by design doesn't set bytes to zero + scan.bytes = 0 + scan.reset() + return scan +} + +func freeScanner(scan *scanner) { + // Avoid hanging on to too much memory in extreme cases. + if len(scan.parseState) > 1024 { + scan.parseState = nil + } + scannerPool.Put(scan) +} + +// These values are returned by the state transition functions +// assigned to scanner.state and the method scanner.eof. +// They give details about the current state of the scan that +// callers might be interested to know about. +// It is okay to ignore the return value of any particular +// call to scanner.state: if one call returns scanError, +// every subsequent call will return scanError too. +const ( + // Continue. + scanContinue = iota // uninteresting byte + scanBeginLiteral // end implied by next result != scanContinue + scanBeginObject // begin object + scanObjectKey // just finished object key (string) + scanObjectValue // just finished non-last object value + scanEndObject // end object (implies scanObjectValue if possible) + scanBeginArray // begin array + scanArrayValue // just finished array value + scanEndArray // end array (implies scanArrayValue if possible) + scanSkipSpace // space byte; can skip; known to be last "continue" result + + // Stop. + scanEnd // top-level value ended *before* this byte; known to be first "stop" result + scanError // hit an error, scanner.err. +) + +// These values are stored in the parseState stack. +// They give the current state of a composite value +// being scanned. If the parser is inside a nested value +// the parseState describes the nested state, outermost at entry 0. +const ( + parseObjectKey = iota // parsing object key (before colon) + parseObjectValue // parsing object value (after colon) + parseArrayValue // parsing array value +) + +// This limits the max nesting depth to prevent stack overflow. +// This is permitted by https://tools.ietf.org/html/rfc7159#section-9 +const maxNestingDepth = 10000 + +// reset prepares the scanner for use. +// It must be called before calling s.step. +func (s *scanner) reset() { + s.step = stateBeginValue + s.parseState = s.parseState[0:0] + s.err = nil + s.endTop = false +} + +// eof tells the scanner that the end of input has been reached. +// It returns a scan status just as s.step does. +func (s *scanner) eof() int { + if s.err != nil { + return scanError + } + if s.endTop { + return scanEnd + } + s.step(s, ' ') + if s.endTop { + return scanEnd + } + if s.err == nil { + s.err = &SyntaxError{"unexpected end of JSON input", s.bytes} + } + return scanError +} + +// pushParseState pushes a new parse state p onto the parse stack. +// an error state is returned if maxNestingDepth was exceeded, otherwise successState is returned. +func (s *scanner) pushParseState(c byte, newParseState int, successState int) int { + s.parseState = append(s.parseState, newParseState) + if len(s.parseState) <= maxNestingDepth { + return successState + } + return s.error(c, "exceeded max depth") +} + +// popParseState pops a parse state (already obtained) off the stack +// and updates s.step accordingly. +func (s *scanner) popParseState() { + n := len(s.parseState) - 1 + s.parseState = s.parseState[0:n] + if n == 0 { + s.step = stateEndTop + s.endTop = true + } else { + s.step = stateEndValue + } +} + +func isSpace(c byte) bool { + return c <= ' ' && (c == ' ' || c == '\t' || c == '\r' || c == '\n') +} + +// stateBeginValueOrEmpty is the state after reading `[`. +func stateBeginValueOrEmpty(s *scanner, c byte) int { + if isSpace(c) { + return scanSkipSpace + } + if c == ']' { + return stateEndValue(s, c) + } + return stateBeginValue(s, c) +} + +// stateBeginValue is the state at the beginning of the input. +func stateBeginValue(s *scanner, c byte) int { + if isSpace(c) { + return scanSkipSpace + } + switch c { + case '{': + s.step = stateBeginStringOrEmpty + return s.pushParseState(c, parseObjectKey, scanBeginObject) + case '[': + s.step = stateBeginValueOrEmpty + return s.pushParseState(c, parseArrayValue, scanBeginArray) + case '"': + s.step = stateInString + return scanBeginLiteral + case '-': + s.step = stateNeg + return scanBeginLiteral + case '0': // beginning of 0.123 + s.step = state0 + return scanBeginLiteral + case 't': // beginning of true + s.step = stateT + return scanBeginLiteral + case 'f': // beginning of false + s.step = stateF + return scanBeginLiteral + case 'n': // beginning of null + s.step = stateN + return scanBeginLiteral + } + if '1' <= c && c <= '9' { // beginning of 1234.5 + s.step = state1 + return scanBeginLiteral + } + return s.error(c, "looking for beginning of value") +} + +// stateBeginStringOrEmpty is the state after reading `{`. +func stateBeginStringOrEmpty(s *scanner, c byte) int { + if isSpace(c) { + return scanSkipSpace + } + if c == '}' { + n := len(s.parseState) + s.parseState[n-1] = parseObjectValue + return stateEndValue(s, c) + } + return stateBeginString(s, c) +} + +// stateBeginString is the state after reading `{"key": value,`. +func stateBeginString(s *scanner, c byte) int { + if isSpace(c) { + return scanSkipSpace + } + if c == '"' { + s.step = stateInString + return scanBeginLiteral + } + return s.error(c, "looking for beginning of object key string") +} + +// stateEndValue is the state after completing a value, +// such as after reading `{}` or `true` or `["x"`. +func stateEndValue(s *scanner, c byte) int { + n := len(s.parseState) + if n == 0 { + // Completed top-level before the current byte. + s.step = stateEndTop + s.endTop = true + return stateEndTop(s, c) + } + if isSpace(c) { + s.step = stateEndValue + return scanSkipSpace + } + ps := s.parseState[n-1] + switch ps { + case parseObjectKey: + if c == ':' { + s.parseState[n-1] = parseObjectValue + s.step = stateBeginValue + return scanObjectKey + } + return s.error(c, "after object key") + case parseObjectValue: + if c == ',' { + s.parseState[n-1] = parseObjectKey + s.step = stateBeginString + return scanObjectValue + } + if c == '}' { + s.popParseState() + return scanEndObject + } + return s.error(c, "after object key:value pair") + case parseArrayValue: + if c == ',' { + s.step = stateBeginValue + return scanArrayValue + } + if c == ']' { + s.popParseState() + return scanEndArray + } + return s.error(c, "after array element") + } + return s.error(c, "") +} + +// stateEndTop is the state after finishing the top-level value, +// such as after reading `{}` or `[1,2,3]`. +// Only space characters should be seen now. +func stateEndTop(s *scanner, c byte) int { + if !isSpace(c) { + // Complain about non-space byte on next call. + s.error(c, "after top-level value") + } + return scanEnd +} + +// stateInString is the state after reading `"`. +func stateInString(s *scanner, c byte) int { + if c == '"' { + s.step = stateEndValue + return scanContinue + } + if c == '\\' { + s.step = stateInStringEsc + return scanContinue + } + if c < 0x20 { + return s.error(c, "in string literal") + } + return scanContinue +} + +// stateInStringEsc is the state after reading `"\` during a quoted string. +func stateInStringEsc(s *scanner, c byte) int { + switch c { + case 'b', 'f', 'n', 'r', 't', '\\', '/', '"': + s.step = stateInString + return scanContinue + case 'u': + s.step = stateInStringEscU + return scanContinue + } + return s.error(c, "in string escape code") +} + +// stateInStringEscU is the state after reading `"\u` during a quoted string. +func stateInStringEscU(s *scanner, c byte) int { + if '0' <= c && c <= '9' || 'a' <= c && c <= 'f' || 'A' <= c && c <= 'F' { + s.step = stateInStringEscU1 + return scanContinue + } + // numbers + return s.error(c, "in \\u hexadecimal character escape") +} + +// stateInStringEscU1 is the state after reading `"\u1` during a quoted string. +func stateInStringEscU1(s *scanner, c byte) int { + if '0' <= c && c <= '9' || 'a' <= c && c <= 'f' || 'A' <= c && c <= 'F' { + s.step = stateInStringEscU12 + return scanContinue + } + // numbers + return s.error(c, "in \\u hexadecimal character escape") +} + +// stateInStringEscU12 is the state after reading `"\u12` during a quoted string. +func stateInStringEscU12(s *scanner, c byte) int { + if '0' <= c && c <= '9' || 'a' <= c && c <= 'f' || 'A' <= c && c <= 'F' { + s.step = stateInStringEscU123 + return scanContinue + } + // numbers + return s.error(c, "in \\u hexadecimal character escape") +} + +// stateInStringEscU123 is the state after reading `"\u123` during a quoted string. +func stateInStringEscU123(s *scanner, c byte) int { + if '0' <= c && c <= '9' || 'a' <= c && c <= 'f' || 'A' <= c && c <= 'F' { + s.step = stateInString + return scanContinue + } + // numbers + return s.error(c, "in \\u hexadecimal character escape") +} + +// stateNeg is the state after reading `-` during a number. +func stateNeg(s *scanner, c byte) int { + if c == '0' { + s.step = state0 + return scanContinue + } + if '1' <= c && c <= '9' { + s.step = state1 + return scanContinue + } + return s.error(c, "in numeric literal") +} + +// state1 is the state after reading a non-zero integer during a number, +// such as after reading `1` or `100` but not `0`. +func state1(s *scanner, c byte) int { + if '0' <= c && c <= '9' { + s.step = state1 + return scanContinue + } + return state0(s, c) +} + +// state0 is the state after reading `0` during a number. +func state0(s *scanner, c byte) int { + if c == '.' { + s.step = stateDot + return scanContinue + } + if c == 'e' || c == 'E' { + s.step = stateE + return scanContinue + } + return stateEndValue(s, c) +} + +// stateDot is the state after reading the integer and decimal point in a number, +// such as after reading `1.`. +func stateDot(s *scanner, c byte) int { + if '0' <= c && c <= '9' { + s.step = stateDot0 + return scanContinue + } + return s.error(c, "after decimal point in numeric literal") +} + +// stateDot0 is the state after reading the integer, decimal point, and subsequent +// digits of a number, such as after reading `3.14`. +func stateDot0(s *scanner, c byte) int { + if '0' <= c && c <= '9' { + return scanContinue + } + if c == 'e' || c == 'E' { + s.step = stateE + return scanContinue + } + return stateEndValue(s, c) +} + +// stateE is the state after reading the mantissa and e in a number, +// such as after reading `314e` or `0.314e`. +func stateE(s *scanner, c byte) int { + if c == '+' || c == '-' { + s.step = stateESign + return scanContinue + } + return stateESign(s, c) +} + +// stateESign is the state after reading the mantissa, e, and sign in a number, +// such as after reading `314e-` or `0.314e+`. +func stateESign(s *scanner, c byte) int { + if '0' <= c && c <= '9' { + s.step = stateE0 + return scanContinue + } + return s.error(c, "in exponent of numeric literal") +} + +// stateE0 is the state after reading the mantissa, e, optional sign, +// and at least one digit of the exponent in a number, +// such as after reading `314e-2` or `0.314e+1` or `3.14e0`. +func stateE0(s *scanner, c byte) int { + if '0' <= c && c <= '9' { + return scanContinue + } + return stateEndValue(s, c) +} + +// stateT is the state after reading `t`. +func stateT(s *scanner, c byte) int { + if c == 'r' { + s.step = stateTr + return scanContinue + } + return s.error(c, "in literal true (expecting 'r')") +} + +// stateTr is the state after reading `tr`. +func stateTr(s *scanner, c byte) int { + if c == 'u' { + s.step = stateTru + return scanContinue + } + return s.error(c, "in literal true (expecting 'u')") +} + +// stateTru is the state after reading `tru`. +func stateTru(s *scanner, c byte) int { + if c == 'e' { + s.step = stateEndValue + return scanContinue + } + return s.error(c, "in literal true (expecting 'e')") +} + +// stateF is the state after reading `f`. +func stateF(s *scanner, c byte) int { + if c == 'a' { + s.step = stateFa + return scanContinue + } + return s.error(c, "in literal false (expecting 'a')") +} + +// stateFa is the state after reading `fa`. +func stateFa(s *scanner, c byte) int { + if c == 'l' { + s.step = stateFal + return scanContinue + } + return s.error(c, "in literal false (expecting 'l')") +} + +// stateFal is the state after reading `fal`. +func stateFal(s *scanner, c byte) int { + if c == 's' { + s.step = stateFals + return scanContinue + } + return s.error(c, "in literal false (expecting 's')") +} + +// stateFals is the state after reading `fals`. +func stateFals(s *scanner, c byte) int { + if c == 'e' { + s.step = stateEndValue + return scanContinue + } + return s.error(c, "in literal false (expecting 'e')") +} + +// stateN is the state after reading `n`. +func stateN(s *scanner, c byte) int { + if c == 'u' { + s.step = stateNu + return scanContinue + } + return s.error(c, "in literal null (expecting 'u')") +} + +// stateNu is the state after reading `nu`. +func stateNu(s *scanner, c byte) int { + if c == 'l' { + s.step = stateNul + return scanContinue + } + return s.error(c, "in literal null (expecting 'l')") +} + +// stateNul is the state after reading `nul`. +func stateNul(s *scanner, c byte) int { + if c == 'l' { + s.step = stateEndValue + return scanContinue + } + return s.error(c, "in literal null (expecting 'l')") +} + +// stateError is the state after reaching a syntax error, +// such as after reading `[1}` or `5.1.2`. +func stateError(s *scanner, c byte) int { + return scanError +} + +// error records an error and switches to the error state. +func (s *scanner) error(c byte, context string) int { + s.step = stateError + s.err = &SyntaxError{"invalid character " + quoteChar(c) + " " + context, s.bytes} + return scanError +} + +// quoteChar formats c as a quoted character literal. +func quoteChar(c byte) string { + // special cases - different from quoted strings + if c == '\'' { + return `'\''` + } + if c == '"' { + return `'"'` + } + + // use quoted string with different quotation marks + s := strconv.Quote(string(c)) + return "'" + s[1:len(s)-1] + "'" +} diff --git a/contrib/go/_std_1.22/src/encoding/json/stream.go b/contrib/go/_std_1.22/src/encoding/json/stream.go new file mode 100644 index 0000000000..5c98d1de04 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/json/stream.go @@ -0,0 +1,511 @@ +// Copyright 2010 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package json + +import ( + "bytes" + "errors" + "io" +) + +// A Decoder reads and decodes JSON values from an input stream. +type Decoder struct { + r io.Reader + buf []byte + d decodeState + scanp int // start of unread data in buf + scanned int64 // amount of data already scanned + scan scanner + err error + + tokenState int + tokenStack []int +} + +// NewDecoder returns a new decoder that reads from r. +// +// The decoder introduces its own buffering and may +// read data from r beyond the JSON values requested. +func NewDecoder(r io.Reader) *Decoder { + return &Decoder{r: r} +} + +// UseNumber causes the Decoder to unmarshal a number into an interface{} as a +// [Number] instead of as a float64. +func (dec *Decoder) UseNumber() { dec.d.useNumber = true } + +// DisallowUnknownFields causes the Decoder to return an error when the destination +// is a struct and the input contains object keys which do not match any +// non-ignored, exported fields in the destination. +func (dec *Decoder) DisallowUnknownFields() { dec.d.disallowUnknownFields = true } + +// Decode reads the next JSON-encoded value from its +// input and stores it in the value pointed to by v. +// +// See the documentation for [Unmarshal] for details about +// the conversion of JSON into a Go value. +func (dec *Decoder) Decode(v any) error { + if dec.err != nil { + return dec.err + } + + if err := dec.tokenPrepareForDecode(); err != nil { + return err + } + + if !dec.tokenValueAllowed() { + return &SyntaxError{msg: "not at beginning of value", Offset: dec.InputOffset()} + } + + // Read whole value into buffer. + n, err := dec.readValue() + if err != nil { + return err + } + dec.d.init(dec.buf[dec.scanp : dec.scanp+n]) + dec.scanp += n + + // Don't save err from unmarshal into dec.err: + // the connection is still usable since we read a complete JSON + // object from it before the error happened. + err = dec.d.unmarshal(v) + + // fixup token streaming state + dec.tokenValueEnd() + + return err +} + +// Buffered returns a reader of the data remaining in the Decoder's +// buffer. The reader is valid until the next call to [Decoder.Decode]. +func (dec *Decoder) Buffered() io.Reader { + return bytes.NewReader(dec.buf[dec.scanp:]) +} + +// readValue reads a JSON value into dec.buf. +// It returns the length of the encoding. +func (dec *Decoder) readValue() (int, error) { + dec.scan.reset() + + scanp := dec.scanp + var err error +Input: + // help the compiler see that scanp is never negative, so it can remove + // some bounds checks below. + for scanp >= 0 { + + // Look in the buffer for a new value. + for ; scanp < len(dec.buf); scanp++ { + c := dec.buf[scanp] + dec.scan.bytes++ + switch dec.scan.step(&dec.scan, c) { + case scanEnd: + // scanEnd is delayed one byte so we decrement + // the scanner bytes count by 1 to ensure that + // this value is correct in the next call of Decode. + dec.scan.bytes-- + break Input + case scanEndObject, scanEndArray: + // scanEnd is delayed one byte. + // We might block trying to get that byte from src, + // so instead invent a space byte. + if stateEndValue(&dec.scan, ' ') == scanEnd { + scanp++ + break Input + } + case scanError: + dec.err = dec.scan.err + return 0, dec.scan.err + } + } + + // Did the last read have an error? + // Delayed until now to allow buffer scan. + if err != nil { + if err == io.EOF { + if dec.scan.step(&dec.scan, ' ') == scanEnd { + break Input + } + if nonSpace(dec.buf) { + err = io.ErrUnexpectedEOF + } + } + dec.err = err + return 0, err + } + + n := scanp - dec.scanp + err = dec.refill() + scanp = dec.scanp + n + } + return scanp - dec.scanp, nil +} + +func (dec *Decoder) refill() error { + // Make room to read more into the buffer. + // First slide down data already consumed. + if dec.scanp > 0 { + dec.scanned += int64(dec.scanp) + n := copy(dec.buf, dec.buf[dec.scanp:]) + dec.buf = dec.buf[:n] + dec.scanp = 0 + } + + // Grow buffer if not large enough. + const minRead = 512 + if cap(dec.buf)-len(dec.buf) < minRead { + newBuf := make([]byte, len(dec.buf), 2*cap(dec.buf)+minRead) + copy(newBuf, dec.buf) + dec.buf = newBuf + } + + // Read. Delay error for next iteration (after scan). + n, err := dec.r.Read(dec.buf[len(dec.buf):cap(dec.buf)]) + dec.buf = dec.buf[0 : len(dec.buf)+n] + + return err +} + +func nonSpace(b []byte) bool { + for _, c := range b { + if !isSpace(c) { + return true + } + } + return false +} + +// An Encoder writes JSON values to an output stream. +type Encoder struct { + w io.Writer + err error + escapeHTML bool + + indentBuf []byte + indentPrefix string + indentValue string +} + +// NewEncoder returns a new encoder that writes to w. +func NewEncoder(w io.Writer) *Encoder { + return &Encoder{w: w, escapeHTML: true} +} + +// Encode writes the JSON encoding of v to the stream, +// followed by a newline character. +// +// See the documentation for [Marshal] for details about the +// conversion of Go values to JSON. +func (enc *Encoder) Encode(v any) error { + if enc.err != nil { + return enc.err + } + + e := newEncodeState() + defer encodeStatePool.Put(e) + + err := e.marshal(v, encOpts{escapeHTML: enc.escapeHTML}) + if err != nil { + return err + } + + // Terminate each value with a newline. + // This makes the output look a little nicer + // when debugging, and some kind of space + // is required if the encoded value was a number, + // so that the reader knows there aren't more + // digits coming. + e.WriteByte('\n') + + b := e.Bytes() + if enc.indentPrefix != "" || enc.indentValue != "" { + enc.indentBuf, err = appendIndent(enc.indentBuf[:0], b, enc.indentPrefix, enc.indentValue) + if err != nil { + return err + } + b = enc.indentBuf + } + if _, err = enc.w.Write(b); err != nil { + enc.err = err + } + return err +} + +// SetIndent instructs the encoder to format each subsequent encoded +// value as if indented by the package-level function Indent(dst, src, prefix, indent). +// Calling SetIndent("", "") disables indentation. +func (enc *Encoder) SetIndent(prefix, indent string) { + enc.indentPrefix = prefix + enc.indentValue = indent +} + +// SetEscapeHTML specifies whether problematic HTML characters +// should be escaped inside JSON quoted strings. +// The default behavior is to escape &, <, and > to \u0026, \u003c, and \u003e +// to avoid certain safety problems that can arise when embedding JSON in HTML. +// +// In non-HTML settings where the escaping interferes with the readability +// of the output, SetEscapeHTML(false) disables this behavior. +func (enc *Encoder) SetEscapeHTML(on bool) { + enc.escapeHTML = on +} + +// RawMessage is a raw encoded JSON value. +// It implements [Marshaler] and [Unmarshaler] and can +// be used to delay JSON decoding or precompute a JSON encoding. +type RawMessage []byte + +// MarshalJSON returns m as the JSON encoding of m. +func (m RawMessage) MarshalJSON() ([]byte, error) { + if m == nil { + return []byte("null"), nil + } + return m, nil +} + +// UnmarshalJSON sets *m to a copy of data. +func (m *RawMessage) UnmarshalJSON(data []byte) error { + if m == nil { + return errors.New("json.RawMessage: UnmarshalJSON on nil pointer") + } + *m = append((*m)[0:0], data...) + return nil +} + +var _ Marshaler = (*RawMessage)(nil) +var _ Unmarshaler = (*RawMessage)(nil) + +// A Token holds a value of one of these types: +// +// - [Delim], for the four JSON delimiters [ ] { } +// - bool, for JSON booleans +// - float64, for JSON numbers +// - [Number], for JSON numbers +// - string, for JSON string literals +// - nil, for JSON null +type Token any + +const ( + tokenTopValue = iota + tokenArrayStart + tokenArrayValue + tokenArrayComma + tokenObjectStart + tokenObjectKey + tokenObjectColon + tokenObjectValue + tokenObjectComma +) + +// advance tokenstate from a separator state to a value state +func (dec *Decoder) tokenPrepareForDecode() error { + // Note: Not calling peek before switch, to avoid + // putting peek into the standard Decode path. + // peek is only called when using the Token API. + switch dec.tokenState { + case tokenArrayComma: + c, err := dec.peek() + if err != nil { + return err + } + if c != ',' { + return &SyntaxError{"expected comma after array element", dec.InputOffset()} + } + dec.scanp++ + dec.tokenState = tokenArrayValue + case tokenObjectColon: + c, err := dec.peek() + if err != nil { + return err + } + if c != ':' { + return &SyntaxError{"expected colon after object key", dec.InputOffset()} + } + dec.scanp++ + dec.tokenState = tokenObjectValue + } + return nil +} + +func (dec *Decoder) tokenValueAllowed() bool { + switch dec.tokenState { + case tokenTopValue, tokenArrayStart, tokenArrayValue, tokenObjectValue: + return true + } + return false +} + +func (dec *Decoder) tokenValueEnd() { + switch dec.tokenState { + case tokenArrayStart, tokenArrayValue: + dec.tokenState = tokenArrayComma + case tokenObjectValue: + dec.tokenState = tokenObjectComma + } +} + +// A Delim is a JSON array or object delimiter, one of [ ] { or }. +type Delim rune + +func (d Delim) String() string { + return string(d) +} + +// Token returns the next JSON token in the input stream. +// At the end of the input stream, Token returns nil, [io.EOF]. +// +// Token guarantees that the delimiters [ ] { } it returns are +// properly nested and matched: if Token encounters an unexpected +// delimiter in the input, it will return an error. +// +// The input stream consists of basic JSON values—bool, string, +// number, and null—along with delimiters [ ] { } of type [Delim] +// to mark the start and end of arrays and objects. +// Commas and colons are elided. +func (dec *Decoder) Token() (Token, error) { + for { + c, err := dec.peek() + if err != nil { + return nil, err + } + switch c { + case '[': + if !dec.tokenValueAllowed() { + return dec.tokenError(c) + } + dec.scanp++ + dec.tokenStack = append(dec.tokenStack, dec.tokenState) + dec.tokenState = tokenArrayStart + return Delim('['), nil + + case ']': + if dec.tokenState != tokenArrayStart && dec.tokenState != tokenArrayComma { + return dec.tokenError(c) + } + dec.scanp++ + dec.tokenState = dec.tokenStack[len(dec.tokenStack)-1] + dec.tokenStack = dec.tokenStack[:len(dec.tokenStack)-1] + dec.tokenValueEnd() + return Delim(']'), nil + + case '{': + if !dec.tokenValueAllowed() { + return dec.tokenError(c) + } + dec.scanp++ + dec.tokenStack = append(dec.tokenStack, dec.tokenState) + dec.tokenState = tokenObjectStart + return Delim('{'), nil + + case '}': + if dec.tokenState != tokenObjectStart && dec.tokenState != tokenObjectComma { + return dec.tokenError(c) + } + dec.scanp++ + dec.tokenState = dec.tokenStack[len(dec.tokenStack)-1] + dec.tokenStack = dec.tokenStack[:len(dec.tokenStack)-1] + dec.tokenValueEnd() + return Delim('}'), nil + + case ':': + if dec.tokenState != tokenObjectColon { + return dec.tokenError(c) + } + dec.scanp++ + dec.tokenState = tokenObjectValue + continue + + case ',': + if dec.tokenState == tokenArrayComma { + dec.scanp++ + dec.tokenState = tokenArrayValue + continue + } + if dec.tokenState == tokenObjectComma { + dec.scanp++ + dec.tokenState = tokenObjectKey + continue + } + return dec.tokenError(c) + + case '"': + if dec.tokenState == tokenObjectStart || dec.tokenState == tokenObjectKey { + var x string + old := dec.tokenState + dec.tokenState = tokenTopValue + err := dec.Decode(&x) + dec.tokenState = old + if err != nil { + return nil, err + } + dec.tokenState = tokenObjectColon + return x, nil + } + fallthrough + + default: + if !dec.tokenValueAllowed() { + return dec.tokenError(c) + } + var x any + if err := dec.Decode(&x); err != nil { + return nil, err + } + return x, nil + } + } +} + +func (dec *Decoder) tokenError(c byte) (Token, error) { + var context string + switch dec.tokenState { + case tokenTopValue: + context = " looking for beginning of value" + case tokenArrayStart, tokenArrayValue, tokenObjectValue: + context = " looking for beginning of value" + case tokenArrayComma: + context = " after array element" + case tokenObjectKey: + context = " looking for beginning of object key string" + case tokenObjectColon: + context = " after object key" + case tokenObjectComma: + context = " after object key:value pair" + } + return nil, &SyntaxError{"invalid character " + quoteChar(c) + context, dec.InputOffset()} +} + +// More reports whether there is another element in the +// current array or object being parsed. +func (dec *Decoder) More() bool { + c, err := dec.peek() + return err == nil && c != ']' && c != '}' +} + +func (dec *Decoder) peek() (byte, error) { + var err error + for { + for i := dec.scanp; i < len(dec.buf); i++ { + c := dec.buf[i] + if isSpace(c) { + continue + } + dec.scanp = i + return c, nil + } + // buffer has been scanned, now report any error + if err != nil { + return 0, err + } + err = dec.refill() + } +} + +// InputOffset returns the input stream byte offset of the current decoder position. +// The offset gives the location of the end of the most recently returned token +// and the beginning of the next token. +func (dec *Decoder) InputOffset() int64 { + return dec.scanned + int64(dec.scanp) +} diff --git a/contrib/go/_std_1.22/src/encoding/json/tables.go b/contrib/go/_std_1.22/src/encoding/json/tables.go new file mode 100644 index 0000000000..10acdc18c6 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/json/tables.go @@ -0,0 +1,218 @@ +// Copyright 2016 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package json + +import "unicode/utf8" + +// safeSet holds the value true if the ASCII character with the given array +// position can be represented inside a JSON string without any further +// escaping. +// +// All values are true except for the ASCII control characters (0-31), the +// double quote ("), and the backslash character ("\"). +var safeSet = [utf8.RuneSelf]bool{ + ' ': true, + '!': true, + '"': false, + '#': true, + '$': true, + '%': true, + '&': true, + '\'': true, + '(': true, + ')': true, + '*': true, + '+': true, + ',': true, + '-': true, + '.': true, + '/': true, + '0': true, + '1': true, + '2': true, + '3': true, + '4': true, + '5': true, + '6': true, + '7': true, + '8': true, + '9': true, + ':': true, + ';': true, + '<': true, + '=': true, + '>': true, + '?': true, + '@': true, + 'A': true, + 'B': true, + 'C': true, + 'D': true, + 'E': true, + 'F': true, + 'G': true, + 'H': true, + 'I': true, + 'J': true, + 'K': true, + 'L': true, + 'M': true, + 'N': true, + 'O': true, + 'P': true, + 'Q': true, + 'R': true, + 'S': true, + 'T': true, + 'U': true, + 'V': true, + 'W': true, + 'X': true, + 'Y': true, + 'Z': true, + '[': true, + '\\': false, + ']': true, + '^': true, + '_': true, + '`': true, + 'a': true, + 'b': true, + 'c': true, + 'd': true, + 'e': true, + 'f': true, + 'g': true, + 'h': true, + 'i': true, + 'j': true, + 'k': true, + 'l': true, + 'm': true, + 'n': true, + 'o': true, + 'p': true, + 'q': true, + 'r': true, + 's': true, + 't': true, + 'u': true, + 'v': true, + 'w': true, + 'x': true, + 'y': true, + 'z': true, + '{': true, + '|': true, + '}': true, + '~': true, + '\u007f': true, +} + +// htmlSafeSet holds the value true if the ASCII character with the given +// array position can be safely represented inside a JSON string, embedded +// inside of HTML <script> tags, without any additional escaping. +// +// All values are true except for the ASCII control characters (0-31), the +// double quote ("), the backslash character ("\"), HTML opening and closing +// tags ("<" and ">"), and the ampersand ("&"). +var htmlSafeSet = [utf8.RuneSelf]bool{ + ' ': true, + '!': true, + '"': false, + '#': true, + '$': true, + '%': true, + '&': false, + '\'': true, + '(': true, + ')': true, + '*': true, + '+': true, + ',': true, + '-': true, + '.': true, + '/': true, + '0': true, + '1': true, + '2': true, + '3': true, + '4': true, + '5': true, + '6': true, + '7': true, + '8': true, + '9': true, + ':': true, + ';': true, + '<': false, + '=': true, + '>': false, + '?': true, + '@': true, + 'A': true, + 'B': true, + 'C': true, + 'D': true, + 'E': true, + 'F': true, + 'G': true, + 'H': true, + 'I': true, + 'J': true, + 'K': true, + 'L': true, + 'M': true, + 'N': true, + 'O': true, + 'P': true, + 'Q': true, + 'R': true, + 'S': true, + 'T': true, + 'U': true, + 'V': true, + 'W': true, + 'X': true, + 'Y': true, + 'Z': true, + '[': true, + '\\': false, + ']': true, + '^': true, + '_': true, + '`': true, + 'a': true, + 'b': true, + 'c': true, + 'd': true, + 'e': true, + 'f': true, + 'g': true, + 'h': true, + 'i': true, + 'j': true, + 'k': true, + 'l': true, + 'm': true, + 'n': true, + 'o': true, + 'p': true, + 'q': true, + 'r': true, + 's': true, + 't': true, + 'u': true, + 'v': true, + 'w': true, + 'x': true, + 'y': true, + 'z': true, + '{': true, + '|': true, + '}': true, + '~': true, + '\u007f': true, +} diff --git a/contrib/go/_std_1.22/src/encoding/json/tags.go b/contrib/go/_std_1.22/src/encoding/json/tags.go new file mode 100644 index 0000000000..b490328f4c --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/json/tags.go @@ -0,0 +1,38 @@ +// Copyright 2011 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package json + +import ( + "strings" +) + +// tagOptions is the string following a comma in a struct field's "json" +// tag, or the empty string. It does not include the leading comma. +type tagOptions string + +// parseTag splits a struct field's json tag into its name and +// comma-separated options. +func parseTag(tag string) (string, tagOptions) { + tag, opt, _ := strings.Cut(tag, ",") + return tag, tagOptions(opt) +} + +// Contains reports whether a comma-separated list of options +// contains a particular substr flag. substr must be surrounded by a +// string boundary or commas. +func (o tagOptions) Contains(optionName string) bool { + if len(o) == 0 { + return false + } + s := string(o) + for s != "" { + var name string + name, s, _ = strings.Cut(s, ",") + if name == optionName { + return true + } + } + return false +} diff --git a/contrib/go/_std_1.22/src/encoding/json/ya.make b/contrib/go/_std_1.22/src/encoding/json/ya.make new file mode 100644 index 0000000000..9294755bdc --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/json/ya.make @@ -0,0 +1,14 @@ +GO_LIBRARY() +IF (TRUE) + SRCS( + decode.go + encode.go + fold.go + indent.go + scanner.go + stream.go + tables.go + tags.go + ) +ENDIF() +END() diff --git a/contrib/go/_std_1.22/src/encoding/pem/pem.go b/contrib/go/_std_1.22/src/encoding/pem/pem.go new file mode 100644 index 0000000000..4b4f749021 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/pem/pem.go @@ -0,0 +1,316 @@ +// Copyright 2009 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +// Package pem implements the PEM data encoding, which originated in Privacy +// Enhanced Mail. The most common use of PEM encoding today is in TLS keys and +// certificates. See RFC 1421. +package pem + +import ( + "bytes" + "encoding/base64" + "errors" + "io" + "sort" + "strings" +) + +// A Block represents a PEM encoded structure. +// +// The encoded form is: +// +// -----BEGIN Type----- +// Headers +// base64-encoded Bytes +// -----END Type----- +// +// where [Block.Headers] is a possibly empty sequence of Key: Value lines. +type Block struct { + Type string // The type, taken from the preamble (i.e. "RSA PRIVATE KEY"). + Headers map[string]string // Optional headers. + Bytes []byte // The decoded bytes of the contents. Typically a DER encoded ASN.1 structure. +} + +// getLine results the first \r\n or \n delineated line from the given byte +// array. The line does not include trailing whitespace or the trailing new +// line bytes. The remainder of the byte array (also not including the new line +// bytes) is also returned and this will always be smaller than the original +// argument. +func getLine(data []byte) (line, rest []byte) { + i := bytes.IndexByte(data, '\n') + var j int + if i < 0 { + i = len(data) + j = i + } else { + j = i + 1 + if i > 0 && data[i-1] == '\r' { + i-- + } + } + return bytes.TrimRight(data[0:i], " \t"), data[j:] +} + +// removeSpacesAndTabs returns a copy of its input with all spaces and tabs +// removed, if there were any. Otherwise, the input is returned unchanged. +// +// The base64 decoder already skips newline characters, so we don't need to +// filter them out here. +func removeSpacesAndTabs(data []byte) []byte { + if !bytes.ContainsAny(data, " \t") { + // Fast path; most base64 data within PEM contains newlines, but + // no spaces nor tabs. Skip the extra alloc and work. + return data + } + result := make([]byte, len(data)) + n := 0 + + for _, b := range data { + if b == ' ' || b == '\t' { + continue + } + result[n] = b + n++ + } + + return result[0:n] +} + +var pemStart = []byte("\n-----BEGIN ") +var pemEnd = []byte("\n-----END ") +var pemEndOfLine = []byte("-----") +var colon = []byte(":") + +// Decode will find the next PEM formatted block (certificate, private key +// etc) in the input. It returns that block and the remainder of the input. If +// no PEM data is found, p is nil and the whole of the input is returned in +// rest. +func Decode(data []byte) (p *Block, rest []byte) { + // pemStart begins with a newline. However, at the very beginning of + // the byte array, we'll accept the start string without it. + rest = data + for { + if bytes.HasPrefix(rest, pemStart[1:]) { + rest = rest[len(pemStart)-1:] + } else if _, after, ok := bytes.Cut(rest, pemStart); ok { + rest = after + } else { + return nil, data + } + + var typeLine []byte + typeLine, rest = getLine(rest) + if !bytes.HasSuffix(typeLine, pemEndOfLine) { + continue + } + typeLine = typeLine[0 : len(typeLine)-len(pemEndOfLine)] + + p = &Block{ + Headers: make(map[string]string), + Type: string(typeLine), + } + + for { + // This loop terminates because getLine's second result is + // always smaller than its argument. + if len(rest) == 0 { + return nil, data + } + line, next := getLine(rest) + + key, val, ok := bytes.Cut(line, colon) + if !ok { + break + } + + // TODO(agl): need to cope with values that spread across lines. + key = bytes.TrimSpace(key) + val = bytes.TrimSpace(val) + p.Headers[string(key)] = string(val) + rest = next + } + + var endIndex, endTrailerIndex int + + // If there were no headers, the END line might occur + // immediately, without a leading newline. + if len(p.Headers) == 0 && bytes.HasPrefix(rest, pemEnd[1:]) { + endIndex = 0 + endTrailerIndex = len(pemEnd) - 1 + } else { + endIndex = bytes.Index(rest, pemEnd) + endTrailerIndex = endIndex + len(pemEnd) + } + + if endIndex < 0 { + continue + } + + // After the "-----" of the ending line, there should be the same type + // and then a final five dashes. + endTrailer := rest[endTrailerIndex:] + endTrailerLen := len(typeLine) + len(pemEndOfLine) + if len(endTrailer) < endTrailerLen { + continue + } + + restOfEndLine := endTrailer[endTrailerLen:] + endTrailer = endTrailer[:endTrailerLen] + if !bytes.HasPrefix(endTrailer, typeLine) || + !bytes.HasSuffix(endTrailer, pemEndOfLine) { + continue + } + + // The line must end with only whitespace. + if s, _ := getLine(restOfEndLine); len(s) != 0 { + continue + } + + base64Data := removeSpacesAndTabs(rest[:endIndex]) + p.Bytes = make([]byte, base64.StdEncoding.DecodedLen(len(base64Data))) + n, err := base64.StdEncoding.Decode(p.Bytes, base64Data) + if err != nil { + continue + } + p.Bytes = p.Bytes[:n] + + // the -1 is because we might have only matched pemEnd without the + // leading newline if the PEM block was empty. + _, rest = getLine(rest[endIndex+len(pemEnd)-1:]) + return p, rest + } +} + +const pemLineLength = 64 + +type lineBreaker struct { + line [pemLineLength]byte + used int + out io.Writer +} + +var nl = []byte{'\n'} + +func (l *lineBreaker) Write(b []byte) (n int, err error) { + if l.used+len(b) < pemLineLength { + copy(l.line[l.used:], b) + l.used += len(b) + return len(b), nil + } + + n, err = l.out.Write(l.line[0:l.used]) + if err != nil { + return + } + excess := pemLineLength - l.used + l.used = 0 + + n, err = l.out.Write(b[0:excess]) + if err != nil { + return + } + + n, err = l.out.Write(nl) + if err != nil { + return + } + + return l.Write(b[excess:]) +} + +func (l *lineBreaker) Close() (err error) { + if l.used > 0 { + _, err = l.out.Write(l.line[0:l.used]) + if err != nil { + return + } + _, err = l.out.Write(nl) + } + + return +} + +func writeHeader(out io.Writer, k, v string) error { + _, err := out.Write([]byte(k + ": " + v + "\n")) + return err +} + +// Encode writes the PEM encoding of b to out. +func Encode(out io.Writer, b *Block) error { + // Check for invalid block before writing any output. + for k := range b.Headers { + if strings.Contains(k, ":") { + return errors.New("pem: cannot encode a header key that contains a colon") + } + } + + // All errors below are relayed from underlying io.Writer, + // so it is now safe to write data. + + if _, err := out.Write(pemStart[1:]); err != nil { + return err + } + if _, err := out.Write([]byte(b.Type + "-----\n")); err != nil { + return err + } + + if len(b.Headers) > 0 { + const procType = "Proc-Type" + h := make([]string, 0, len(b.Headers)) + hasProcType := false + for k := range b.Headers { + if k == procType { + hasProcType = true + continue + } + h = append(h, k) + } + // The Proc-Type header must be written first. + // See RFC 1421, section 4.6.1.1 + if hasProcType { + if err := writeHeader(out, procType, b.Headers[procType]); err != nil { + return err + } + } + // For consistency of output, write other headers sorted by key. + sort.Strings(h) + for _, k := range h { + if err := writeHeader(out, k, b.Headers[k]); err != nil { + return err + } + } + if _, err := out.Write(nl); err != nil { + return err + } + } + + var breaker lineBreaker + breaker.out = out + + b64 := base64.NewEncoder(base64.StdEncoding, &breaker) + if _, err := b64.Write(b.Bytes); err != nil { + return err + } + b64.Close() + breaker.Close() + + if _, err := out.Write(pemEnd[1:]); err != nil { + return err + } + _, err := out.Write([]byte(b.Type + "-----\n")) + return err +} + +// EncodeToMemory returns the PEM encoding of b. +// +// If b has invalid headers and cannot be encoded, +// EncodeToMemory returns nil. If it is important to +// report details about this error case, use [Encode] instead. +func EncodeToMemory(b *Block) []byte { + var buf bytes.Buffer + if err := Encode(&buf, b); err != nil { + return nil + } + return buf.Bytes() +} diff --git a/contrib/go/_std_1.22/src/encoding/pem/ya.make b/contrib/go/_std_1.22/src/encoding/pem/ya.make new file mode 100644 index 0000000000..534626b3a5 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/pem/ya.make @@ -0,0 +1,7 @@ +GO_LIBRARY() +IF (TRUE) + SRCS( + pem.go + ) +ENDIF() +END() diff --git a/contrib/go/_std_1.22/src/encoding/xml/marshal.go b/contrib/go/_std_1.22/src/encoding/xml/marshal.go new file mode 100644 index 0000000000..05b5542dfb --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/xml/marshal.go @@ -0,0 +1,1131 @@ +// Copyright 2011 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package xml + +import ( + "bufio" + "bytes" + "encoding" + "errors" + "fmt" + "io" + "reflect" + "strconv" + "strings" +) + +const ( + // Header is a generic XML header suitable for use with the output of [Marshal]. + // This is not automatically added to any output of this package, + // it is provided as a convenience. + Header = `<?xml version="1.0" encoding="UTF-8"?>` + "\n" +) + +// Marshal returns the XML encoding of v. +// +// Marshal handles an array or slice by marshaling each of the elements. +// Marshal handles a pointer by marshaling the value it points at or, if the +// pointer is nil, by writing nothing. Marshal handles an interface value by +// marshaling the value it contains or, if the interface value is nil, by +// writing nothing. Marshal handles all other data by writing one or more XML +// elements containing the data. +// +// The name for the XML elements is taken from, in order of preference: +// - the tag on the XMLName field, if the data is a struct +// - the value of the XMLName field of type [Name] +// - the tag of the struct field used to obtain the data +// - the name of the struct field used to obtain the data +// - the name of the marshaled type +// +// The XML element for a struct contains marshaled elements for each of the +// exported fields of the struct, with these exceptions: +// - the XMLName field, described above, is omitted. +// - a field with tag "-" is omitted. +// - a field with tag "name,attr" becomes an attribute with +// the given name in the XML element. +// - a field with tag ",attr" becomes an attribute with the +// field name in the XML element. +// - a field with tag ",chardata" is written as character data, +// not as an XML element. +// - a field with tag ",cdata" is written as character data +// wrapped in one or more <![CDATA[ ... ]]> tags, not as an XML element. +// - a field with tag ",innerxml" is written verbatim, not subject +// to the usual marshaling procedure. +// - a field with tag ",comment" is written as an XML comment, not +// subject to the usual marshaling procedure. It must not contain +// the "--" string within it. +// - a field with a tag including the "omitempty" option is omitted +// if the field value is empty. The empty values are false, 0, any +// nil pointer or interface value, and any array, slice, map, or +// string of length zero. +// - an anonymous struct field is handled as if the fields of its +// value were part of the outer struct. +// - a field implementing [Marshaler] is written by calling its MarshalXML +// method. +// - a field implementing [encoding.TextMarshaler] is written by encoding the +// result of its MarshalText method as text. +// +// If a field uses a tag "a>b>c", then the element c will be nested inside +// parent elements a and b. Fields that appear next to each other that name +// the same parent will be enclosed in one XML element. +// +// If the XML name for a struct field is defined by both the field tag and the +// struct's XMLName field, the names must match. +// +// See [MarshalIndent] for an example. +// +// Marshal will return an error if asked to marshal a channel, function, or map. +func Marshal(v any) ([]byte, error) { + var b bytes.Buffer + enc := NewEncoder(&b) + if err := enc.Encode(v); err != nil { + return nil, err + } + if err := enc.Close(); err != nil { + return nil, err + } + return b.Bytes(), nil +} + +// Marshaler is the interface implemented by objects that can marshal +// themselves into valid XML elements. +// +// MarshalXML encodes the receiver as zero or more XML elements. +// By convention, arrays or slices are typically encoded as a sequence +// of elements, one per entry. +// Using start as the element tag is not required, but doing so +// will enable [Unmarshal] to match the XML elements to the correct +// struct field. +// One common implementation strategy is to construct a separate +// value with a layout corresponding to the desired XML and then +// to encode it using e.EncodeElement. +// Another common strategy is to use repeated calls to e.EncodeToken +// to generate the XML output one token at a time. +// The sequence of encoded tokens must make up zero or more valid +// XML elements. +type Marshaler interface { + MarshalXML(e *Encoder, start StartElement) error +} + +// MarshalerAttr is the interface implemented by objects that can marshal +// themselves into valid XML attributes. +// +// MarshalXMLAttr returns an XML attribute with the encoded value of the receiver. +// Using name as the attribute name is not required, but doing so +// will enable [Unmarshal] to match the attribute to the correct +// struct field. +// If MarshalXMLAttr returns the zero attribute [Attr]{}, no attribute +// will be generated in the output. +// MarshalXMLAttr is used only for struct fields with the +// "attr" option in the field tag. +type MarshalerAttr interface { + MarshalXMLAttr(name Name) (Attr, error) +} + +// MarshalIndent works like [Marshal], but each XML element begins on a new +// indented line that starts with prefix and is followed by one or more +// copies of indent according to the nesting depth. +func MarshalIndent(v any, prefix, indent string) ([]byte, error) { + var b bytes.Buffer + enc := NewEncoder(&b) + enc.Indent(prefix, indent) + if err := enc.Encode(v); err != nil { + return nil, err + } + if err := enc.Close(); err != nil { + return nil, err + } + return b.Bytes(), nil +} + +// An Encoder writes XML data to an output stream. +type Encoder struct { + p printer +} + +// NewEncoder returns a new encoder that writes to w. +func NewEncoder(w io.Writer) *Encoder { + e := &Encoder{printer{w: bufio.NewWriter(w)}} + e.p.encoder = e + return e +} + +// Indent sets the encoder to generate XML in which each element +// begins on a new indented line that starts with prefix and is followed by +// one or more copies of indent according to the nesting depth. +func (enc *Encoder) Indent(prefix, indent string) { + enc.p.prefix = prefix + enc.p.indent = indent +} + +// Encode writes the XML encoding of v to the stream. +// +// See the documentation for [Marshal] for details about the conversion +// of Go values to XML. +// +// Encode calls [Encoder.Flush] before returning. +func (enc *Encoder) Encode(v any) error { + err := enc.p.marshalValue(reflect.ValueOf(v), nil, nil) + if err != nil { + return err + } + return enc.p.w.Flush() +} + +// EncodeElement writes the XML encoding of v to the stream, +// using start as the outermost tag in the encoding. +// +// See the documentation for [Marshal] for details about the conversion +// of Go values to XML. +// +// EncodeElement calls [Encoder.Flush] before returning. +func (enc *Encoder) EncodeElement(v any, start StartElement) error { + err := enc.p.marshalValue(reflect.ValueOf(v), nil, &start) + if err != nil { + return err + } + return enc.p.w.Flush() +} + +var ( + begComment = []byte("<!--") + endComment = []byte("-->") + endProcInst = []byte("?>") +) + +// EncodeToken writes the given XML token to the stream. +// It returns an error if [StartElement] and [EndElement] tokens are not properly matched. +// +// EncodeToken does not call [Encoder.Flush], because usually it is part of a larger operation +// such as [Encoder.Encode] or [Encoder.EncodeElement] (or a custom [Marshaler]'s MarshalXML invoked +// during those), and those will call Flush when finished. +// Callers that create an Encoder and then invoke EncodeToken directly, without +// using Encode or EncodeElement, need to call Flush when finished to ensure +// that the XML is written to the underlying writer. +// +// EncodeToken allows writing a [ProcInst] with Target set to "xml" only as the first token +// in the stream. +func (enc *Encoder) EncodeToken(t Token) error { + + p := &enc.p + switch t := t.(type) { + case StartElement: + if err := p.writeStart(&t); err != nil { + return err + } + case EndElement: + if err := p.writeEnd(t.Name); err != nil { + return err + } + case CharData: + escapeText(p, t, false) + case Comment: + if bytes.Contains(t, endComment) { + return fmt.Errorf("xml: EncodeToken of Comment containing --> marker") + } + p.WriteString("<!--") + p.Write(t) + p.WriteString("-->") + return p.cachedWriteError() + case ProcInst: + // First token to be encoded which is also a ProcInst with target of xml + // is the xml declaration. The only ProcInst where target of xml is allowed. + if t.Target == "xml" && p.w.Buffered() != 0 { + return fmt.Errorf("xml: EncodeToken of ProcInst xml target only valid for xml declaration, first token encoded") + } + if !isNameString(t.Target) { + return fmt.Errorf("xml: EncodeToken of ProcInst with invalid Target") + } + if bytes.Contains(t.Inst, endProcInst) { + return fmt.Errorf("xml: EncodeToken of ProcInst containing ?> marker") + } + p.WriteString("<?") + p.WriteString(t.Target) + if len(t.Inst) > 0 { + p.WriteByte(' ') + p.Write(t.Inst) + } + p.WriteString("?>") + case Directive: + if !isValidDirective(t) { + return fmt.Errorf("xml: EncodeToken of Directive containing wrong < or > markers") + } + p.WriteString("<!") + p.Write(t) + p.WriteString(">") + default: + return fmt.Errorf("xml: EncodeToken of invalid token type") + + } + return p.cachedWriteError() +} + +// isValidDirective reports whether dir is a valid directive text, +// meaning angle brackets are matched, ignoring comments and strings. +func isValidDirective(dir Directive) bool { + var ( + depth int + inquote uint8 + incomment bool + ) + for i, c := range dir { + switch { + case incomment: + if c == '>' { + if n := 1 + i - len(endComment); n >= 0 && bytes.Equal(dir[n:i+1], endComment) { + incomment = false + } + } + // Just ignore anything in comment + case inquote != 0: + if c == inquote { + inquote = 0 + } + // Just ignore anything within quotes + case c == '\'' || c == '"': + inquote = c + case c == '<': + if i+len(begComment) < len(dir) && bytes.Equal(dir[i:i+len(begComment)], begComment) { + incomment = true + } else { + depth++ + } + case c == '>': + if depth == 0 { + return false + } + depth-- + } + } + return depth == 0 && inquote == 0 && !incomment +} + +// Flush flushes any buffered XML to the underlying writer. +// See the [Encoder.EncodeToken] documentation for details about when it is necessary. +func (enc *Encoder) Flush() error { + return enc.p.w.Flush() +} + +// Close the Encoder, indicating that no more data will be written. It flushes +// any buffered XML to the underlying writer and returns an error if the +// written XML is invalid (e.g. by containing unclosed elements). +func (enc *Encoder) Close() error { + return enc.p.Close() +} + +type printer struct { + w *bufio.Writer + encoder *Encoder + seq int + indent string + prefix string + depth int + indentedIn bool + putNewline bool + attrNS map[string]string // map prefix -> name space + attrPrefix map[string]string // map name space -> prefix + prefixes []string + tags []Name + closed bool + err error +} + +// createAttrPrefix finds the name space prefix attribute to use for the given name space, +// defining a new prefix if necessary. It returns the prefix. +func (p *printer) createAttrPrefix(url string) string { + if prefix := p.attrPrefix[url]; prefix != "" { + return prefix + } + + // The "http://www.w3.org/XML/1998/namespace" name space is predefined as "xml" + // and must be referred to that way. + // (The "http://www.w3.org/2000/xmlns/" name space is also predefined as "xmlns", + // but users should not be trying to use that one directly - that's our job.) + if url == xmlURL { + return xmlPrefix + } + + // Need to define a new name space. + if p.attrPrefix == nil { + p.attrPrefix = make(map[string]string) + p.attrNS = make(map[string]string) + } + + // Pick a name. We try to use the final element of the path + // but fall back to _. + prefix := strings.TrimRight(url, "/") + if i := strings.LastIndex(prefix, "/"); i >= 0 { + prefix = prefix[i+1:] + } + if prefix == "" || !isName([]byte(prefix)) || strings.Contains(prefix, ":") { + prefix = "_" + } + // xmlanything is reserved and any variant of it regardless of + // case should be matched, so: + // (('X'|'x') ('M'|'m') ('L'|'l')) + // See Section 2.3 of https://www.w3.org/TR/REC-xml/ + if len(prefix) >= 3 && strings.EqualFold(prefix[:3], "xml") { + prefix = "_" + prefix + } + if p.attrNS[prefix] != "" { + // Name is taken. Find a better one. + for p.seq++; ; p.seq++ { + if id := prefix + "_" + strconv.Itoa(p.seq); p.attrNS[id] == "" { + prefix = id + break + } + } + } + + p.attrPrefix[url] = prefix + p.attrNS[prefix] = url + + p.WriteString(`xmlns:`) + p.WriteString(prefix) + p.WriteString(`="`) + EscapeText(p, []byte(url)) + p.WriteString(`" `) + + p.prefixes = append(p.prefixes, prefix) + + return prefix +} + +// deleteAttrPrefix removes an attribute name space prefix. +func (p *printer) deleteAttrPrefix(prefix string) { + delete(p.attrPrefix, p.attrNS[prefix]) + delete(p.attrNS, prefix) +} + +func (p *printer) markPrefix() { + p.prefixes = append(p.prefixes, "") +} + +func (p *printer) popPrefix() { + for len(p.prefixes) > 0 { + prefix := p.prefixes[len(p.prefixes)-1] + p.prefixes = p.prefixes[:len(p.prefixes)-1] + if prefix == "" { + break + } + p.deleteAttrPrefix(prefix) + } +} + +var ( + marshalerType = reflect.TypeFor[Marshaler]() + marshalerAttrType = reflect.TypeFor[MarshalerAttr]() + textMarshalerType = reflect.TypeFor[encoding.TextMarshaler]() +) + +// marshalValue writes one or more XML elements representing val. +// If val was obtained from a struct field, finfo must have its details. +func (p *printer) marshalValue(val reflect.Value, finfo *fieldInfo, startTemplate *StartElement) error { + if startTemplate != nil && startTemplate.Name.Local == "" { + return fmt.Errorf("xml: EncodeElement of StartElement with missing name") + } + + if !val.IsValid() { + return nil + } + if finfo != nil && finfo.flags&fOmitEmpty != 0 && isEmptyValue(val) { + return nil + } + + // Drill into interfaces and pointers. + // This can turn into an infinite loop given a cyclic chain, + // but it matches the Go 1 behavior. + for val.Kind() == reflect.Interface || val.Kind() == reflect.Pointer { + if val.IsNil() { + return nil + } + val = val.Elem() + } + + kind := val.Kind() + typ := val.Type() + + // Check for marshaler. + if val.CanInterface() && typ.Implements(marshalerType) { + return p.marshalInterface(val.Interface().(Marshaler), defaultStart(typ, finfo, startTemplate)) + } + if val.CanAddr() { + pv := val.Addr() + if pv.CanInterface() && pv.Type().Implements(marshalerType) { + return p.marshalInterface(pv.Interface().(Marshaler), defaultStart(pv.Type(), finfo, startTemplate)) + } + } + + // Check for text marshaler. + if val.CanInterface() && typ.Implements(textMarshalerType) { + return p.marshalTextInterface(val.Interface().(encoding.TextMarshaler), defaultStart(typ, finfo, startTemplate)) + } + if val.CanAddr() { + pv := val.Addr() + if pv.CanInterface() && pv.Type().Implements(textMarshalerType) { + return p.marshalTextInterface(pv.Interface().(encoding.TextMarshaler), defaultStart(pv.Type(), finfo, startTemplate)) + } + } + + // Slices and arrays iterate over the elements. They do not have an enclosing tag. + if (kind == reflect.Slice || kind == reflect.Array) && typ.Elem().Kind() != reflect.Uint8 { + for i, n := 0, val.Len(); i < n; i++ { + if err := p.marshalValue(val.Index(i), finfo, startTemplate); err != nil { + return err + } + } + return nil + } + + tinfo, err := getTypeInfo(typ) + if err != nil { + return err + } + + // Create start element. + // Precedence for the XML element name is: + // 0. startTemplate + // 1. XMLName field in underlying struct; + // 2. field name/tag in the struct field; and + // 3. type name + var start StartElement + + if startTemplate != nil { + start.Name = startTemplate.Name + start.Attr = append(start.Attr, startTemplate.Attr...) + } else if tinfo.xmlname != nil { + xmlname := tinfo.xmlname + if xmlname.name != "" { + start.Name.Space, start.Name.Local = xmlname.xmlns, xmlname.name + } else { + fv := xmlname.value(val, dontInitNilPointers) + if v, ok := fv.Interface().(Name); ok && v.Local != "" { + start.Name = v + } + } + } + if start.Name.Local == "" && finfo != nil { + start.Name.Space, start.Name.Local = finfo.xmlns, finfo.name + } + if start.Name.Local == "" { + name := typ.Name() + if i := strings.IndexByte(name, '['); i >= 0 { + // Truncate generic instantiation name. See issue 48318. + name = name[:i] + } + if name == "" { + return &UnsupportedTypeError{typ} + } + start.Name.Local = name + } + + // Attributes + for i := range tinfo.fields { + finfo := &tinfo.fields[i] + if finfo.flags&fAttr == 0 { + continue + } + fv := finfo.value(val, dontInitNilPointers) + + if finfo.flags&fOmitEmpty != 0 && (!fv.IsValid() || isEmptyValue(fv)) { + continue + } + + if fv.Kind() == reflect.Interface && fv.IsNil() { + continue + } + + name := Name{Space: finfo.xmlns, Local: finfo.name} + if err := p.marshalAttr(&start, name, fv); err != nil { + return err + } + } + + // If an empty name was found, namespace is overridden with an empty space + if tinfo.xmlname != nil && start.Name.Space == "" && + tinfo.xmlname.xmlns == "" && tinfo.xmlname.name == "" && + len(p.tags) != 0 && p.tags[len(p.tags)-1].Space != "" { + start.Attr = append(start.Attr, Attr{Name{"", xmlnsPrefix}, ""}) + } + if err := p.writeStart(&start); err != nil { + return err + } + + if val.Kind() == reflect.Struct { + err = p.marshalStruct(tinfo, val) + } else { + s, b, err1 := p.marshalSimple(typ, val) + if err1 != nil { + err = err1 + } else if b != nil { + EscapeText(p, b) + } else { + p.EscapeString(s) + } + } + if err != nil { + return err + } + + if err := p.writeEnd(start.Name); err != nil { + return err + } + + return p.cachedWriteError() +} + +// marshalAttr marshals an attribute with the given name and value, adding to start.Attr. +func (p *printer) marshalAttr(start *StartElement, name Name, val reflect.Value) error { + if val.CanInterface() && val.Type().Implements(marshalerAttrType) { + attr, err := val.Interface().(MarshalerAttr).MarshalXMLAttr(name) + if err != nil { + return err + } + if attr.Name.Local != "" { + start.Attr = append(start.Attr, attr) + } + return nil + } + + if val.CanAddr() { + pv := val.Addr() + if pv.CanInterface() && pv.Type().Implements(marshalerAttrType) { + attr, err := pv.Interface().(MarshalerAttr).MarshalXMLAttr(name) + if err != nil { + return err + } + if attr.Name.Local != "" { + start.Attr = append(start.Attr, attr) + } + return nil + } + } + + if val.CanInterface() && val.Type().Implements(textMarshalerType) { + text, err := val.Interface().(encoding.TextMarshaler).MarshalText() + if err != nil { + return err + } + start.Attr = append(start.Attr, Attr{name, string(text)}) + return nil + } + + if val.CanAddr() { + pv := val.Addr() + if pv.CanInterface() && pv.Type().Implements(textMarshalerType) { + text, err := pv.Interface().(encoding.TextMarshaler).MarshalText() + if err != nil { + return err + } + start.Attr = append(start.Attr, Attr{name, string(text)}) + return nil + } + } + + // Dereference or skip nil pointer, interface values. + switch val.Kind() { + case reflect.Pointer, reflect.Interface: + if val.IsNil() { + return nil + } + val = val.Elem() + } + + // Walk slices. + if val.Kind() == reflect.Slice && val.Type().Elem().Kind() != reflect.Uint8 { + n := val.Len() + for i := 0; i < n; i++ { + if err := p.marshalAttr(start, name, val.Index(i)); err != nil { + return err + } + } + return nil + } + + if val.Type() == attrType { + start.Attr = append(start.Attr, val.Interface().(Attr)) + return nil + } + + s, b, err := p.marshalSimple(val.Type(), val) + if err != nil { + return err + } + if b != nil { + s = string(b) + } + start.Attr = append(start.Attr, Attr{name, s}) + return nil +} + +// defaultStart returns the default start element to use, +// given the reflect type, field info, and start template. +func defaultStart(typ reflect.Type, finfo *fieldInfo, startTemplate *StartElement) StartElement { + var start StartElement + // Precedence for the XML element name is as above, + // except that we do not look inside structs for the first field. + if startTemplate != nil { + start.Name = startTemplate.Name + start.Attr = append(start.Attr, startTemplate.Attr...) + } else if finfo != nil && finfo.name != "" { + start.Name.Local = finfo.name + start.Name.Space = finfo.xmlns + } else if typ.Name() != "" { + start.Name.Local = typ.Name() + } else { + // Must be a pointer to a named type, + // since it has the Marshaler methods. + start.Name.Local = typ.Elem().Name() + } + return start +} + +// marshalInterface marshals a Marshaler interface value. +func (p *printer) marshalInterface(val Marshaler, start StartElement) error { + // Push a marker onto the tag stack so that MarshalXML + // cannot close the XML tags that it did not open. + p.tags = append(p.tags, Name{}) + n := len(p.tags) + + err := val.MarshalXML(p.encoder, start) + if err != nil { + return err + } + + // Make sure MarshalXML closed all its tags. p.tags[n-1] is the mark. + if len(p.tags) > n { + return fmt.Errorf("xml: %s.MarshalXML wrote invalid XML: <%s> not closed", receiverType(val), p.tags[len(p.tags)-1].Local) + } + p.tags = p.tags[:n-1] + return nil +} + +// marshalTextInterface marshals a TextMarshaler interface value. +func (p *printer) marshalTextInterface(val encoding.TextMarshaler, start StartElement) error { + if err := p.writeStart(&start); err != nil { + return err + } + text, err := val.MarshalText() + if err != nil { + return err + } + EscapeText(p, text) + return p.writeEnd(start.Name) +} + +// writeStart writes the given start element. +func (p *printer) writeStart(start *StartElement) error { + if start.Name.Local == "" { + return fmt.Errorf("xml: start tag with no name") + } + + p.tags = append(p.tags, start.Name) + p.markPrefix() + + p.writeIndent(1) + p.WriteByte('<') + p.WriteString(start.Name.Local) + + if start.Name.Space != "" { + p.WriteString(` xmlns="`) + p.EscapeString(start.Name.Space) + p.WriteByte('"') + } + + // Attributes + for _, attr := range start.Attr { + name := attr.Name + if name.Local == "" { + continue + } + p.WriteByte(' ') + if name.Space != "" { + p.WriteString(p.createAttrPrefix(name.Space)) + p.WriteByte(':') + } + p.WriteString(name.Local) + p.WriteString(`="`) + p.EscapeString(attr.Value) + p.WriteByte('"') + } + p.WriteByte('>') + return nil +} + +func (p *printer) writeEnd(name Name) error { + if name.Local == "" { + return fmt.Errorf("xml: end tag with no name") + } + if len(p.tags) == 0 || p.tags[len(p.tags)-1].Local == "" { + return fmt.Errorf("xml: end tag </%s> without start tag", name.Local) + } + if top := p.tags[len(p.tags)-1]; top != name { + if top.Local != name.Local { + return fmt.Errorf("xml: end tag </%s> does not match start tag <%s>", name.Local, top.Local) + } + return fmt.Errorf("xml: end tag </%s> in namespace %s does not match start tag <%s> in namespace %s", name.Local, name.Space, top.Local, top.Space) + } + p.tags = p.tags[:len(p.tags)-1] + + p.writeIndent(-1) + p.WriteByte('<') + p.WriteByte('/') + p.WriteString(name.Local) + p.WriteByte('>') + p.popPrefix() + return nil +} + +func (p *printer) marshalSimple(typ reflect.Type, val reflect.Value) (string, []byte, error) { + switch val.Kind() { + case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64: + return strconv.FormatInt(val.Int(), 10), nil, nil + case reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr: + return strconv.FormatUint(val.Uint(), 10), nil, nil + case reflect.Float32, reflect.Float64: + return strconv.FormatFloat(val.Float(), 'g', -1, val.Type().Bits()), nil, nil + case reflect.String: + return val.String(), nil, nil + case reflect.Bool: + return strconv.FormatBool(val.Bool()), nil, nil + case reflect.Array: + if typ.Elem().Kind() != reflect.Uint8 { + break + } + // [...]byte + var bytes []byte + if val.CanAddr() { + bytes = val.Bytes() + } else { + bytes = make([]byte, val.Len()) + reflect.Copy(reflect.ValueOf(bytes), val) + } + return "", bytes, nil + case reflect.Slice: + if typ.Elem().Kind() != reflect.Uint8 { + break + } + // []byte + return "", val.Bytes(), nil + } + return "", nil, &UnsupportedTypeError{typ} +} + +var ddBytes = []byte("--") + +// indirect drills into interfaces and pointers, returning the pointed-at value. +// If it encounters a nil interface or pointer, indirect returns that nil value. +// This can turn into an infinite loop given a cyclic chain, +// but it matches the Go 1 behavior. +func indirect(vf reflect.Value) reflect.Value { + for vf.Kind() == reflect.Interface || vf.Kind() == reflect.Pointer { + if vf.IsNil() { + return vf + } + vf = vf.Elem() + } + return vf +} + +func (p *printer) marshalStruct(tinfo *typeInfo, val reflect.Value) error { + s := parentStack{p: p} + for i := range tinfo.fields { + finfo := &tinfo.fields[i] + if finfo.flags&fAttr != 0 { + continue + } + vf := finfo.value(val, dontInitNilPointers) + if !vf.IsValid() { + // The field is behind an anonymous struct field that's + // nil. Skip it. + continue + } + + switch finfo.flags & fMode { + case fCDATA, fCharData: + emit := EscapeText + if finfo.flags&fMode == fCDATA { + emit = emitCDATA + } + if err := s.trim(finfo.parents); err != nil { + return err + } + if vf.CanInterface() && vf.Type().Implements(textMarshalerType) { + data, err := vf.Interface().(encoding.TextMarshaler).MarshalText() + if err != nil { + return err + } + if err := emit(p, data); err != nil { + return err + } + continue + } + if vf.CanAddr() { + pv := vf.Addr() + if pv.CanInterface() && pv.Type().Implements(textMarshalerType) { + data, err := pv.Interface().(encoding.TextMarshaler).MarshalText() + if err != nil { + return err + } + if err := emit(p, data); err != nil { + return err + } + continue + } + } + + var scratch [64]byte + vf = indirect(vf) + switch vf.Kind() { + case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64: + if err := emit(p, strconv.AppendInt(scratch[:0], vf.Int(), 10)); err != nil { + return err + } + case reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr: + if err := emit(p, strconv.AppendUint(scratch[:0], vf.Uint(), 10)); err != nil { + return err + } + case reflect.Float32, reflect.Float64: + if err := emit(p, strconv.AppendFloat(scratch[:0], vf.Float(), 'g', -1, vf.Type().Bits())); err != nil { + return err + } + case reflect.Bool: + if err := emit(p, strconv.AppendBool(scratch[:0], vf.Bool())); err != nil { + return err + } + case reflect.String: + if err := emit(p, []byte(vf.String())); err != nil { + return err + } + case reflect.Slice: + if elem, ok := vf.Interface().([]byte); ok { + if err := emit(p, elem); err != nil { + return err + } + } + } + continue + + case fComment: + if err := s.trim(finfo.parents); err != nil { + return err + } + vf = indirect(vf) + k := vf.Kind() + if !(k == reflect.String || k == reflect.Slice && vf.Type().Elem().Kind() == reflect.Uint8) { + return fmt.Errorf("xml: bad type for comment field of %s", val.Type()) + } + if vf.Len() == 0 { + continue + } + p.writeIndent(0) + p.WriteString("<!--") + dashDash := false + dashLast := false + switch k { + case reflect.String: + s := vf.String() + dashDash = strings.Contains(s, "--") + dashLast = s[len(s)-1] == '-' + if !dashDash { + p.WriteString(s) + } + case reflect.Slice: + b := vf.Bytes() + dashDash = bytes.Contains(b, ddBytes) + dashLast = b[len(b)-1] == '-' + if !dashDash { + p.Write(b) + } + default: + panic("can't happen") + } + if dashDash { + return fmt.Errorf(`xml: comments must not contain "--"`) + } + if dashLast { + // "--->" is invalid grammar. Make it "- -->" + p.WriteByte(' ') + } + p.WriteString("-->") + continue + + case fInnerXML: + vf = indirect(vf) + iface := vf.Interface() + switch raw := iface.(type) { + case []byte: + p.Write(raw) + continue + case string: + p.WriteString(raw) + continue + } + + case fElement, fElement | fAny: + if err := s.trim(finfo.parents); err != nil { + return err + } + if len(finfo.parents) > len(s.stack) { + if vf.Kind() != reflect.Pointer && vf.Kind() != reflect.Interface || !vf.IsNil() { + if err := s.push(finfo.parents[len(s.stack):]); err != nil { + return err + } + } + } + } + if err := p.marshalValue(vf, finfo, nil); err != nil { + return err + } + } + s.trim(nil) + return p.cachedWriteError() +} + +// Write implements io.Writer +func (p *printer) Write(b []byte) (n int, err error) { + if p.closed && p.err == nil { + p.err = errors.New("use of closed Encoder") + } + if p.err == nil { + n, p.err = p.w.Write(b) + } + return n, p.err +} + +// WriteString implements io.StringWriter +func (p *printer) WriteString(s string) (n int, err error) { + if p.closed && p.err == nil { + p.err = errors.New("use of closed Encoder") + } + if p.err == nil { + n, p.err = p.w.WriteString(s) + } + return n, p.err +} + +// WriteByte implements io.ByteWriter +func (p *printer) WriteByte(c byte) error { + if p.closed && p.err == nil { + p.err = errors.New("use of closed Encoder") + } + if p.err == nil { + p.err = p.w.WriteByte(c) + } + return p.err +} + +// Close the Encoder, indicating that no more data will be written. It flushes +// any buffered XML to the underlying writer and returns an error if the +// written XML is invalid (e.g. by containing unclosed elements). +func (p *printer) Close() error { + if p.closed { + return nil + } + p.closed = true + if err := p.w.Flush(); err != nil { + return err + } + if len(p.tags) > 0 { + return fmt.Errorf("unclosed tag <%s>", p.tags[len(p.tags)-1].Local) + } + return nil +} + +// return the bufio Writer's cached write error +func (p *printer) cachedWriteError() error { + _, err := p.Write(nil) + return err +} + +func (p *printer) writeIndent(depthDelta int) { + if len(p.prefix) == 0 && len(p.indent) == 0 { + return + } + if depthDelta < 0 { + p.depth-- + if p.indentedIn { + p.indentedIn = false + return + } + p.indentedIn = false + } + if p.putNewline { + p.WriteByte('\n') + } else { + p.putNewline = true + } + if len(p.prefix) > 0 { + p.WriteString(p.prefix) + } + if len(p.indent) > 0 { + for i := 0; i < p.depth; i++ { + p.WriteString(p.indent) + } + } + if depthDelta > 0 { + p.depth++ + p.indentedIn = true + } +} + +type parentStack struct { + p *printer + stack []string +} + +// trim updates the XML context to match the longest common prefix of the stack +// and the given parents. A closing tag will be written for every parent +// popped. Passing a zero slice or nil will close all the elements. +func (s *parentStack) trim(parents []string) error { + split := 0 + for ; split < len(parents) && split < len(s.stack); split++ { + if parents[split] != s.stack[split] { + break + } + } + for i := len(s.stack) - 1; i >= split; i-- { + if err := s.p.writeEnd(Name{Local: s.stack[i]}); err != nil { + return err + } + } + s.stack = s.stack[:split] + return nil +} + +// push adds parent elements to the stack and writes open tags. +func (s *parentStack) push(parents []string) error { + for i := 0; i < len(parents); i++ { + if err := s.p.writeStart(&StartElement{Name: Name{Local: parents[i]}}); err != nil { + return err + } + } + s.stack = append(s.stack, parents...) + return nil +} + +// UnsupportedTypeError is returned when [Marshal] encounters a type +// that cannot be converted into XML. +type UnsupportedTypeError struct { + Type reflect.Type +} + +func (e *UnsupportedTypeError) Error() string { + return "xml: unsupported type: " + e.Type.String() +} + +func isEmptyValue(v reflect.Value) bool { + switch v.Kind() { + case reflect.Array, reflect.Map, reflect.Slice, reflect.String: + return v.Len() == 0 + case reflect.Bool, + reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64, + reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr, + reflect.Float32, reflect.Float64, + reflect.Interface, reflect.Pointer: + return v.IsZero() + } + return false +} diff --git a/contrib/go/_std_1.22/src/encoding/xml/read.go b/contrib/go/_std_1.22/src/encoding/xml/read.go new file mode 100644 index 0000000000..3cc4968c76 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/xml/read.go @@ -0,0 +1,777 @@ +// Copyright 2009 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package xml + +import ( + "bytes" + "encoding" + "errors" + "fmt" + "reflect" + "runtime" + "strconv" + "strings" +) + +// BUG(rsc): Mapping between XML elements and data structures is inherently flawed: +// an XML element is an order-dependent collection of anonymous +// values, while a data structure is an order-independent collection +// of named values. +// See [encoding/json] for a textual representation more suitable +// to data structures. + +// Unmarshal parses the XML-encoded data and stores the result in +// the value pointed to by v, which must be an arbitrary struct, +// slice, or string. Well-formed data that does not fit into v is +// discarded. +// +// Because Unmarshal uses the reflect package, it can only assign +// to exported (upper case) fields. Unmarshal uses a case-sensitive +// comparison to match XML element names to tag values and struct +// field names. +// +// Unmarshal maps an XML element to a struct using the following rules. +// In the rules, the tag of a field refers to the value associated with the +// key 'xml' in the struct field's tag (see the example above). +// +// - If the struct has a field of type []byte or string with tag +// ",innerxml", Unmarshal accumulates the raw XML nested inside the +// element in that field. The rest of the rules still apply. +// +// - If the struct has a field named XMLName of type Name, +// Unmarshal records the element name in that field. +// +// - If the XMLName field has an associated tag of the form +// "name" or "namespace-URL name", the XML element must have +// the given name (and, optionally, name space) or else Unmarshal +// returns an error. +// +// - If the XML element has an attribute whose name matches a +// struct field name with an associated tag containing ",attr" or +// the explicit name in a struct field tag of the form "name,attr", +// Unmarshal records the attribute value in that field. +// +// - If the XML element has an attribute not handled by the previous +// rule and the struct has a field with an associated tag containing +// ",any,attr", Unmarshal records the attribute value in the first +// such field. +// +// - If the XML element contains character data, that data is +// accumulated in the first struct field that has tag ",chardata". +// The struct field may have type []byte or string. +// If there is no such field, the character data is discarded. +// +// - If the XML element contains comments, they are accumulated in +// the first struct field that has tag ",comment". The struct +// field may have type []byte or string. If there is no such +// field, the comments are discarded. +// +// - If the XML element contains a sub-element whose name matches +// the prefix of a tag formatted as "a" or "a>b>c", unmarshal +// will descend into the XML structure looking for elements with the +// given names, and will map the innermost elements to that struct +// field. A tag starting with ">" is equivalent to one starting +// with the field name followed by ">". +// +// - If the XML element contains a sub-element whose name matches +// a struct field's XMLName tag and the struct field has no +// explicit name tag as per the previous rule, unmarshal maps +// the sub-element to that struct field. +// +// - If the XML element contains a sub-element whose name matches a +// field without any mode flags (",attr", ",chardata", etc), Unmarshal +// maps the sub-element to that struct field. +// +// - If the XML element contains a sub-element that hasn't matched any +// of the above rules and the struct has a field with tag ",any", +// unmarshal maps the sub-element to that struct field. +// +// - An anonymous struct field is handled as if the fields of its +// value were part of the outer struct. +// +// - A struct field with tag "-" is never unmarshaled into. +// +// If Unmarshal encounters a field type that implements the Unmarshaler +// interface, Unmarshal calls its UnmarshalXML method to produce the value from +// the XML element. Otherwise, if the value implements +// [encoding.TextUnmarshaler], Unmarshal calls that value's UnmarshalText method. +// +// Unmarshal maps an XML element to a string or []byte by saving the +// concatenation of that element's character data in the string or +// []byte. The saved []byte is never nil. +// +// Unmarshal maps an attribute value to a string or []byte by saving +// the value in the string or slice. +// +// Unmarshal maps an attribute value to an [Attr] by saving the attribute, +// including its name, in the Attr. +// +// Unmarshal maps an XML element or attribute value to a slice by +// extending the length of the slice and mapping the element or attribute +// to the newly created value. +// +// Unmarshal maps an XML element or attribute value to a bool by +// setting it to the boolean value represented by the string. Whitespace +// is trimmed and ignored. +// +// Unmarshal maps an XML element or attribute value to an integer or +// floating-point field by setting the field to the result of +// interpreting the string value in decimal. There is no check for +// overflow. Whitespace is trimmed and ignored. +// +// Unmarshal maps an XML element to a Name by recording the element +// name. +// +// Unmarshal maps an XML element to a pointer by setting the pointer +// to a freshly allocated value and then mapping the element to that value. +// +// A missing element or empty attribute value will be unmarshaled as a zero value. +// If the field is a slice, a zero value will be appended to the field. Otherwise, the +// field will be set to its zero value. +func Unmarshal(data []byte, v any) error { + return NewDecoder(bytes.NewReader(data)).Decode(v) +} + +// Decode works like [Unmarshal], except it reads the decoder +// stream to find the start element. +func (d *Decoder) Decode(v any) error { + return d.DecodeElement(v, nil) +} + +// DecodeElement works like [Unmarshal] except that it takes +// a pointer to the start XML element to decode into v. +// It is useful when a client reads some raw XML tokens itself +// but also wants to defer to [Unmarshal] for some elements. +func (d *Decoder) DecodeElement(v any, start *StartElement) error { + val := reflect.ValueOf(v) + if val.Kind() != reflect.Pointer { + return errors.New("non-pointer passed to Unmarshal") + } + + if val.IsNil() { + return errors.New("nil pointer passed to Unmarshal") + } + return d.unmarshal(val.Elem(), start, 0) +} + +// An UnmarshalError represents an error in the unmarshaling process. +type UnmarshalError string + +func (e UnmarshalError) Error() string { return string(e) } + +// Unmarshaler is the interface implemented by objects that can unmarshal +// an XML element description of themselves. +// +// UnmarshalXML decodes a single XML element +// beginning with the given start element. +// If it returns an error, the outer call to Unmarshal stops and +// returns that error. +// UnmarshalXML must consume exactly one XML element. +// One common implementation strategy is to unmarshal into +// a separate value with a layout matching the expected XML +// using d.DecodeElement, and then to copy the data from +// that value into the receiver. +// Another common strategy is to use d.Token to process the +// XML object one token at a time. +// UnmarshalXML may not use d.RawToken. +type Unmarshaler interface { + UnmarshalXML(d *Decoder, start StartElement) error +} + +// UnmarshalerAttr is the interface implemented by objects that can unmarshal +// an XML attribute description of themselves. +// +// UnmarshalXMLAttr decodes a single XML attribute. +// If it returns an error, the outer call to [Unmarshal] stops and +// returns that error. +// UnmarshalXMLAttr is used only for struct fields with the +// "attr" option in the field tag. +type UnmarshalerAttr interface { + UnmarshalXMLAttr(attr Attr) error +} + +// receiverType returns the receiver type to use in an expression like "%s.MethodName". +func receiverType(val any) string { + t := reflect.TypeOf(val) + if t.Name() != "" { + return t.String() + } + return "(" + t.String() + ")" +} + +// unmarshalInterface unmarshals a single XML element into val. +// start is the opening tag of the element. +func (d *Decoder) unmarshalInterface(val Unmarshaler, start *StartElement) error { + // Record that decoder must stop at end tag corresponding to start. + d.pushEOF() + + d.unmarshalDepth++ + err := val.UnmarshalXML(d, *start) + d.unmarshalDepth-- + if err != nil { + d.popEOF() + return err + } + + if !d.popEOF() { + return fmt.Errorf("xml: %s.UnmarshalXML did not consume entire <%s> element", receiverType(val), start.Name.Local) + } + + return nil +} + +// unmarshalTextInterface unmarshals a single XML element into val. +// The chardata contained in the element (but not its children) +// is passed to the text unmarshaler. +func (d *Decoder) unmarshalTextInterface(val encoding.TextUnmarshaler) error { + var buf []byte + depth := 1 + for depth > 0 { + t, err := d.Token() + if err != nil { + return err + } + switch t := t.(type) { + case CharData: + if depth == 1 { + buf = append(buf, t...) + } + case StartElement: + depth++ + case EndElement: + depth-- + } + } + return val.UnmarshalText(buf) +} + +// unmarshalAttr unmarshals a single XML attribute into val. +func (d *Decoder) unmarshalAttr(val reflect.Value, attr Attr) error { + if val.Kind() == reflect.Pointer { + if val.IsNil() { + val.Set(reflect.New(val.Type().Elem())) + } + val = val.Elem() + } + if val.CanInterface() && val.Type().Implements(unmarshalerAttrType) { + // This is an unmarshaler with a non-pointer receiver, + // so it's likely to be incorrect, but we do what we're told. + return val.Interface().(UnmarshalerAttr).UnmarshalXMLAttr(attr) + } + if val.CanAddr() { + pv := val.Addr() + if pv.CanInterface() && pv.Type().Implements(unmarshalerAttrType) { + return pv.Interface().(UnmarshalerAttr).UnmarshalXMLAttr(attr) + } + } + + // Not an UnmarshalerAttr; try encoding.TextUnmarshaler. + if val.CanInterface() && val.Type().Implements(textUnmarshalerType) { + // This is an unmarshaler with a non-pointer receiver, + // so it's likely to be incorrect, but we do what we're told. + return val.Interface().(encoding.TextUnmarshaler).UnmarshalText([]byte(attr.Value)) + } + if val.CanAddr() { + pv := val.Addr() + if pv.CanInterface() && pv.Type().Implements(textUnmarshalerType) { + return pv.Interface().(encoding.TextUnmarshaler).UnmarshalText([]byte(attr.Value)) + } + } + + if val.Type().Kind() == reflect.Slice && val.Type().Elem().Kind() != reflect.Uint8 { + // Slice of element values. + // Grow slice. + n := val.Len() + val.Grow(1) + val.SetLen(n + 1) + + // Recur to read element into slice. + if err := d.unmarshalAttr(val.Index(n), attr); err != nil { + val.SetLen(n) + return err + } + return nil + } + + if val.Type() == attrType { + val.Set(reflect.ValueOf(attr)) + return nil + } + + return copyValue(val, []byte(attr.Value)) +} + +var ( + attrType = reflect.TypeFor[Attr]() + unmarshalerType = reflect.TypeFor[Unmarshaler]() + unmarshalerAttrType = reflect.TypeFor[UnmarshalerAttr]() + textUnmarshalerType = reflect.TypeFor[encoding.TextUnmarshaler]() +) + +const ( + maxUnmarshalDepth = 10000 + maxUnmarshalDepthWasm = 5000 // go.dev/issue/56498 +) + +var errUnmarshalDepth = errors.New("exceeded max depth") + +// Unmarshal a single XML element into val. +func (d *Decoder) unmarshal(val reflect.Value, start *StartElement, depth int) error { + if depth >= maxUnmarshalDepth || runtime.GOARCH == "wasm" && depth >= maxUnmarshalDepthWasm { + return errUnmarshalDepth + } + // Find start element if we need it. + if start == nil { + for { + tok, err := d.Token() + if err != nil { + return err + } + if t, ok := tok.(StartElement); ok { + start = &t + break + } + } + } + + // Load value from interface, but only if the result will be + // usefully addressable. + if val.Kind() == reflect.Interface && !val.IsNil() { + e := val.Elem() + if e.Kind() == reflect.Pointer && !e.IsNil() { + val = e + } + } + + if val.Kind() == reflect.Pointer { + if val.IsNil() { + val.Set(reflect.New(val.Type().Elem())) + } + val = val.Elem() + } + + if val.CanInterface() && val.Type().Implements(unmarshalerType) { + // This is an unmarshaler with a non-pointer receiver, + // so it's likely to be incorrect, but we do what we're told. + return d.unmarshalInterface(val.Interface().(Unmarshaler), start) + } + + if val.CanAddr() { + pv := val.Addr() + if pv.CanInterface() && pv.Type().Implements(unmarshalerType) { + return d.unmarshalInterface(pv.Interface().(Unmarshaler), start) + } + } + + if val.CanInterface() && val.Type().Implements(textUnmarshalerType) { + return d.unmarshalTextInterface(val.Interface().(encoding.TextUnmarshaler)) + } + + if val.CanAddr() { + pv := val.Addr() + if pv.CanInterface() && pv.Type().Implements(textUnmarshalerType) { + return d.unmarshalTextInterface(pv.Interface().(encoding.TextUnmarshaler)) + } + } + + var ( + data []byte + saveData reflect.Value + comment []byte + saveComment reflect.Value + saveXML reflect.Value + saveXMLIndex int + saveXMLData []byte + saveAny reflect.Value + sv reflect.Value + tinfo *typeInfo + err error + ) + + switch v := val; v.Kind() { + default: + return errors.New("unknown type " + v.Type().String()) + + case reflect.Interface: + // TODO: For now, simply ignore the field. In the near + // future we may choose to unmarshal the start + // element on it, if not nil. + return d.Skip() + + case reflect.Slice: + typ := v.Type() + if typ.Elem().Kind() == reflect.Uint8 { + // []byte + saveData = v + break + } + + // Slice of element values. + // Grow slice. + n := v.Len() + v.Grow(1) + v.SetLen(n + 1) + + // Recur to read element into slice. + if err := d.unmarshal(v.Index(n), start, depth+1); err != nil { + v.SetLen(n) + return err + } + return nil + + case reflect.Bool, reflect.Float32, reflect.Float64, reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64, reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr, reflect.String: + saveData = v + + case reflect.Struct: + typ := v.Type() + if typ == nameType { + v.Set(reflect.ValueOf(start.Name)) + break + } + + sv = v + tinfo, err = getTypeInfo(typ) + if err != nil { + return err + } + + // Validate and assign element name. + if tinfo.xmlname != nil { + finfo := tinfo.xmlname + if finfo.name != "" && finfo.name != start.Name.Local { + return UnmarshalError("expected element type <" + finfo.name + "> but have <" + start.Name.Local + ">") + } + if finfo.xmlns != "" && finfo.xmlns != start.Name.Space { + e := "expected element <" + finfo.name + "> in name space " + finfo.xmlns + " but have " + if start.Name.Space == "" { + e += "no name space" + } else { + e += start.Name.Space + } + return UnmarshalError(e) + } + fv := finfo.value(sv, initNilPointers) + if _, ok := fv.Interface().(Name); ok { + fv.Set(reflect.ValueOf(start.Name)) + } + } + + // Assign attributes. + for _, a := range start.Attr { + handled := false + any := -1 + for i := range tinfo.fields { + finfo := &tinfo.fields[i] + switch finfo.flags & fMode { + case fAttr: + strv := finfo.value(sv, initNilPointers) + if a.Name.Local == finfo.name && (finfo.xmlns == "" || finfo.xmlns == a.Name.Space) { + if err := d.unmarshalAttr(strv, a); err != nil { + return err + } + handled = true + } + + case fAny | fAttr: + if any == -1 { + any = i + } + } + } + if !handled && any >= 0 { + finfo := &tinfo.fields[any] + strv := finfo.value(sv, initNilPointers) + if err := d.unmarshalAttr(strv, a); err != nil { + return err + } + } + } + + // Determine whether we need to save character data or comments. + for i := range tinfo.fields { + finfo := &tinfo.fields[i] + switch finfo.flags & fMode { + case fCDATA, fCharData: + if !saveData.IsValid() { + saveData = finfo.value(sv, initNilPointers) + } + + case fComment: + if !saveComment.IsValid() { + saveComment = finfo.value(sv, initNilPointers) + } + + case fAny, fAny | fElement: + if !saveAny.IsValid() { + saveAny = finfo.value(sv, initNilPointers) + } + + case fInnerXML: + if !saveXML.IsValid() { + saveXML = finfo.value(sv, initNilPointers) + if d.saved == nil { + saveXMLIndex = 0 + d.saved = new(bytes.Buffer) + } else { + saveXMLIndex = d.savedOffset() + } + } + } + } + } + + // Find end element. + // Process sub-elements along the way. +Loop: + for { + var savedOffset int + if saveXML.IsValid() { + savedOffset = d.savedOffset() + } + tok, err := d.Token() + if err != nil { + return err + } + switch t := tok.(type) { + case StartElement: + consumed := false + if sv.IsValid() { + // unmarshalPath can call unmarshal, so we need to pass the depth through so that + // we can continue to enforce the maximum recursion limit. + consumed, err = d.unmarshalPath(tinfo, sv, nil, &t, depth) + if err != nil { + return err + } + if !consumed && saveAny.IsValid() { + consumed = true + if err := d.unmarshal(saveAny, &t, depth+1); err != nil { + return err + } + } + } + if !consumed { + if err := d.Skip(); err != nil { + return err + } + } + + case EndElement: + if saveXML.IsValid() { + saveXMLData = d.saved.Bytes()[saveXMLIndex:savedOffset] + if saveXMLIndex == 0 { + d.saved = nil + } + } + break Loop + + case CharData: + if saveData.IsValid() { + data = append(data, t...) + } + + case Comment: + if saveComment.IsValid() { + comment = append(comment, t...) + } + } + } + + if saveData.IsValid() && saveData.CanInterface() && saveData.Type().Implements(textUnmarshalerType) { + if err := saveData.Interface().(encoding.TextUnmarshaler).UnmarshalText(data); err != nil { + return err + } + saveData = reflect.Value{} + } + + if saveData.IsValid() && saveData.CanAddr() { + pv := saveData.Addr() + if pv.CanInterface() && pv.Type().Implements(textUnmarshalerType) { + if err := pv.Interface().(encoding.TextUnmarshaler).UnmarshalText(data); err != nil { + return err + } + saveData = reflect.Value{} + } + } + + if err := copyValue(saveData, data); err != nil { + return err + } + + switch t := saveComment; t.Kind() { + case reflect.String: + t.SetString(string(comment)) + case reflect.Slice: + t.Set(reflect.ValueOf(comment)) + } + + switch t := saveXML; t.Kind() { + case reflect.String: + t.SetString(string(saveXMLData)) + case reflect.Slice: + if t.Type().Elem().Kind() == reflect.Uint8 { + t.Set(reflect.ValueOf(saveXMLData)) + } + } + + return nil +} + +func copyValue(dst reflect.Value, src []byte) (err error) { + dst0 := dst + + if dst.Kind() == reflect.Pointer { + if dst.IsNil() { + dst.Set(reflect.New(dst.Type().Elem())) + } + dst = dst.Elem() + } + + // Save accumulated data. + switch dst.Kind() { + case reflect.Invalid: + // Probably a comment. + default: + return errors.New("cannot unmarshal into " + dst0.Type().String()) + case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64: + if len(src) == 0 { + dst.SetInt(0) + return nil + } + itmp, err := strconv.ParseInt(strings.TrimSpace(string(src)), 10, dst.Type().Bits()) + if err != nil { + return err + } + dst.SetInt(itmp) + case reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64, reflect.Uintptr: + if len(src) == 0 { + dst.SetUint(0) + return nil + } + utmp, err := strconv.ParseUint(strings.TrimSpace(string(src)), 10, dst.Type().Bits()) + if err != nil { + return err + } + dst.SetUint(utmp) + case reflect.Float32, reflect.Float64: + if len(src) == 0 { + dst.SetFloat(0) + return nil + } + ftmp, err := strconv.ParseFloat(strings.TrimSpace(string(src)), dst.Type().Bits()) + if err != nil { + return err + } + dst.SetFloat(ftmp) + case reflect.Bool: + if len(src) == 0 { + dst.SetBool(false) + return nil + } + value, err := strconv.ParseBool(strings.TrimSpace(string(src))) + if err != nil { + return err + } + dst.SetBool(value) + case reflect.String: + dst.SetString(string(src)) + case reflect.Slice: + if len(src) == 0 { + // non-nil to flag presence + src = []byte{} + } + dst.SetBytes(src) + } + return nil +} + +// unmarshalPath walks down an XML structure looking for wanted +// paths, and calls unmarshal on them. +// The consumed result tells whether XML elements have been consumed +// from the Decoder until start's matching end element, or if it's +// still untouched because start is uninteresting for sv's fields. +func (d *Decoder) unmarshalPath(tinfo *typeInfo, sv reflect.Value, parents []string, start *StartElement, depth int) (consumed bool, err error) { + recurse := false +Loop: + for i := range tinfo.fields { + finfo := &tinfo.fields[i] + if finfo.flags&fElement == 0 || len(finfo.parents) < len(parents) || finfo.xmlns != "" && finfo.xmlns != start.Name.Space { + continue + } + for j := range parents { + if parents[j] != finfo.parents[j] { + continue Loop + } + } + if len(finfo.parents) == len(parents) && finfo.name == start.Name.Local { + // It's a perfect match, unmarshal the field. + return true, d.unmarshal(finfo.value(sv, initNilPointers), start, depth+1) + } + if len(finfo.parents) > len(parents) && finfo.parents[len(parents)] == start.Name.Local { + // It's a prefix for the field. Break and recurse + // since it's not ok for one field path to be itself + // the prefix for another field path. + recurse = true + + // We can reuse the same slice as long as we + // don't try to append to it. + parents = finfo.parents[:len(parents)+1] + break + } + } + if !recurse { + // We have no business with this element. + return false, nil + } + // The element is not a perfect match for any field, but one + // or more fields have the path to this element as a parent + // prefix. Recurse and attempt to match these. + for { + var tok Token + tok, err = d.Token() + if err != nil { + return true, err + } + switch t := tok.(type) { + case StartElement: + // the recursion depth of unmarshalPath is limited to the path length specified + // by the struct field tag, so we don't increment the depth here. + consumed2, err := d.unmarshalPath(tinfo, sv, parents, &t, depth) + if err != nil { + return true, err + } + if !consumed2 { + if err := d.Skip(); err != nil { + return true, err + } + } + case EndElement: + return true, nil + } + } +} + +// Skip reads tokens until it has consumed the end element +// matching the most recent start element already consumed, +// skipping nested structures. +// It returns nil if it finds an end element matching the start +// element; otherwise it returns an error describing the problem. +func (d *Decoder) Skip() error { + var depth int64 + for { + tok, err := d.Token() + if err != nil { + return err + } + switch tok.(type) { + case StartElement: + depth++ + case EndElement: + if depth == 0 { + return nil + } + depth-- + } + } +} diff --git a/contrib/go/_std_1.22/src/encoding/xml/typeinfo.go b/contrib/go/_std_1.22/src/encoding/xml/typeinfo.go new file mode 100644 index 0000000000..b18ed284a6 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/xml/typeinfo.go @@ -0,0 +1,367 @@ +// Copyright 2011 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +package xml + +import ( + "fmt" + "reflect" + "strings" + "sync" +) + +// typeInfo holds details for the xml representation of a type. +type typeInfo struct { + xmlname *fieldInfo + fields []fieldInfo +} + +// fieldInfo holds details for the xml representation of a single field. +type fieldInfo struct { + idx []int + name string + xmlns string + flags fieldFlags + parents []string +} + +type fieldFlags int + +const ( + fElement fieldFlags = 1 << iota + fAttr + fCDATA + fCharData + fInnerXML + fComment + fAny + + fOmitEmpty + + fMode = fElement | fAttr | fCDATA | fCharData | fInnerXML | fComment | fAny + + xmlName = "XMLName" +) + +var tinfoMap sync.Map // map[reflect.Type]*typeInfo + +var nameType = reflect.TypeFor[Name]() + +// getTypeInfo returns the typeInfo structure with details necessary +// for marshaling and unmarshaling typ. +func getTypeInfo(typ reflect.Type) (*typeInfo, error) { + if ti, ok := tinfoMap.Load(typ); ok { + return ti.(*typeInfo), nil + } + + tinfo := &typeInfo{} + if typ.Kind() == reflect.Struct && typ != nameType { + n := typ.NumField() + for i := 0; i < n; i++ { + f := typ.Field(i) + if (!f.IsExported() && !f.Anonymous) || f.Tag.Get("xml") == "-" { + continue // Private field + } + + // For embedded structs, embed its fields. + if f.Anonymous { + t := f.Type + if t.Kind() == reflect.Pointer { + t = t.Elem() + } + if t.Kind() == reflect.Struct { + inner, err := getTypeInfo(t) + if err != nil { + return nil, err + } + if tinfo.xmlname == nil { + tinfo.xmlname = inner.xmlname + } + for _, finfo := range inner.fields { + finfo.idx = append([]int{i}, finfo.idx...) + if err := addFieldInfo(typ, tinfo, &finfo); err != nil { + return nil, err + } + } + continue + } + } + + finfo, err := structFieldInfo(typ, &f) + if err != nil { + return nil, err + } + + if f.Name == xmlName { + tinfo.xmlname = finfo + continue + } + + // Add the field if it doesn't conflict with other fields. + if err := addFieldInfo(typ, tinfo, finfo); err != nil { + return nil, err + } + } + } + + ti, _ := tinfoMap.LoadOrStore(typ, tinfo) + return ti.(*typeInfo), nil +} + +// structFieldInfo builds and returns a fieldInfo for f. +func structFieldInfo(typ reflect.Type, f *reflect.StructField) (*fieldInfo, error) { + finfo := &fieldInfo{idx: f.Index} + + // Split the tag from the xml namespace if necessary. + tag := f.Tag.Get("xml") + if ns, t, ok := strings.Cut(tag, " "); ok { + finfo.xmlns, tag = ns, t + } + + // Parse flags. + tokens := strings.Split(tag, ",") + if len(tokens) == 1 { + finfo.flags = fElement + } else { + tag = tokens[0] + for _, flag := range tokens[1:] { + switch flag { + case "attr": + finfo.flags |= fAttr + case "cdata": + finfo.flags |= fCDATA + case "chardata": + finfo.flags |= fCharData + case "innerxml": + finfo.flags |= fInnerXML + case "comment": + finfo.flags |= fComment + case "any": + finfo.flags |= fAny + case "omitempty": + finfo.flags |= fOmitEmpty + } + } + + // Validate the flags used. + valid := true + switch mode := finfo.flags & fMode; mode { + case 0: + finfo.flags |= fElement + case fAttr, fCDATA, fCharData, fInnerXML, fComment, fAny, fAny | fAttr: + if f.Name == xmlName || tag != "" && mode != fAttr { + valid = false + } + default: + // This will also catch multiple modes in a single field. + valid = false + } + if finfo.flags&fMode == fAny { + finfo.flags |= fElement + } + if finfo.flags&fOmitEmpty != 0 && finfo.flags&(fElement|fAttr) == 0 { + valid = false + } + if !valid { + return nil, fmt.Errorf("xml: invalid tag in field %s of type %s: %q", + f.Name, typ, f.Tag.Get("xml")) + } + } + + // Use of xmlns without a name is not allowed. + if finfo.xmlns != "" && tag == "" { + return nil, fmt.Errorf("xml: namespace without name in field %s of type %s: %q", + f.Name, typ, f.Tag.Get("xml")) + } + + if f.Name == xmlName { + // The XMLName field records the XML element name. Don't + // process it as usual because its name should default to + // empty rather than to the field name. + finfo.name = tag + return finfo, nil + } + + if tag == "" { + // If the name part of the tag is completely empty, get + // default from XMLName of underlying struct if feasible, + // or field name otherwise. + if xmlname := lookupXMLName(f.Type); xmlname != nil { + finfo.xmlns, finfo.name = xmlname.xmlns, xmlname.name + } else { + finfo.name = f.Name + } + return finfo, nil + } + + // Prepare field name and parents. + parents := strings.Split(tag, ">") + if parents[0] == "" { + parents[0] = f.Name + } + if parents[len(parents)-1] == "" { + return nil, fmt.Errorf("xml: trailing '>' in field %s of type %s", f.Name, typ) + } + finfo.name = parents[len(parents)-1] + if len(parents) > 1 { + if (finfo.flags & fElement) == 0 { + return nil, fmt.Errorf("xml: %s chain not valid with %s flag", tag, strings.Join(tokens[1:], ",")) + } + finfo.parents = parents[:len(parents)-1] + } + + // If the field type has an XMLName field, the names must match + // so that the behavior of both marshaling and unmarshaling + // is straightforward and unambiguous. + if finfo.flags&fElement != 0 { + ftyp := f.Type + xmlname := lookupXMLName(ftyp) + if xmlname != nil && xmlname.name != finfo.name { + return nil, fmt.Errorf("xml: name %q in tag of %s.%s conflicts with name %q in %s.XMLName", + finfo.name, typ, f.Name, xmlname.name, ftyp) + } + } + return finfo, nil +} + +// lookupXMLName returns the fieldInfo for typ's XMLName field +// in case it exists and has a valid xml field tag, otherwise +// it returns nil. +func lookupXMLName(typ reflect.Type) (xmlname *fieldInfo) { + for typ.Kind() == reflect.Pointer { + typ = typ.Elem() + } + if typ.Kind() != reflect.Struct { + return nil + } + for i, n := 0, typ.NumField(); i < n; i++ { + f := typ.Field(i) + if f.Name != xmlName { + continue + } + finfo, err := structFieldInfo(typ, &f) + if err == nil && finfo.name != "" { + return finfo + } + // Also consider errors as a non-existent field tag + // and let getTypeInfo itself report the error. + break + } + return nil +} + +// addFieldInfo adds finfo to tinfo.fields if there are no +// conflicts, or if conflicts arise from previous fields that were +// obtained from deeper embedded structures than finfo. In the latter +// case, the conflicting entries are dropped. +// A conflict occurs when the path (parent + name) to a field is +// itself a prefix of another path, or when two paths match exactly. +// It is okay for field paths to share a common, shorter prefix. +func addFieldInfo(typ reflect.Type, tinfo *typeInfo, newf *fieldInfo) error { + var conflicts []int +Loop: + // First, figure all conflicts. Most working code will have none. + for i := range tinfo.fields { + oldf := &tinfo.fields[i] + if oldf.flags&fMode != newf.flags&fMode { + continue + } + if oldf.xmlns != "" && newf.xmlns != "" && oldf.xmlns != newf.xmlns { + continue + } + minl := min(len(newf.parents), len(oldf.parents)) + for p := 0; p < minl; p++ { + if oldf.parents[p] != newf.parents[p] { + continue Loop + } + } + if len(oldf.parents) > len(newf.parents) { + if oldf.parents[len(newf.parents)] == newf.name { + conflicts = append(conflicts, i) + } + } else if len(oldf.parents) < len(newf.parents) { + if newf.parents[len(oldf.parents)] == oldf.name { + conflicts = append(conflicts, i) + } + } else { + if newf.name == oldf.name && newf.xmlns == oldf.xmlns { + conflicts = append(conflicts, i) + } + } + } + // Without conflicts, add the new field and return. + if conflicts == nil { + tinfo.fields = append(tinfo.fields, *newf) + return nil + } + + // If any conflict is shallower, ignore the new field. + // This matches the Go field resolution on embedding. + for _, i := range conflicts { + if len(tinfo.fields[i].idx) < len(newf.idx) { + return nil + } + } + + // Otherwise, if any of them is at the same depth level, it's an error. + for _, i := range conflicts { + oldf := &tinfo.fields[i] + if len(oldf.idx) == len(newf.idx) { + f1 := typ.FieldByIndex(oldf.idx) + f2 := typ.FieldByIndex(newf.idx) + return &TagPathError{typ, f1.Name, f1.Tag.Get("xml"), f2.Name, f2.Tag.Get("xml")} + } + } + + // Otherwise, the new field is shallower, and thus takes precedence, + // so drop the conflicting fields from tinfo and append the new one. + for c := len(conflicts) - 1; c >= 0; c-- { + i := conflicts[c] + copy(tinfo.fields[i:], tinfo.fields[i+1:]) + tinfo.fields = tinfo.fields[:len(tinfo.fields)-1] + } + tinfo.fields = append(tinfo.fields, *newf) + return nil +} + +// A TagPathError represents an error in the unmarshaling process +// caused by the use of field tags with conflicting paths. +type TagPathError struct { + Struct reflect.Type + Field1, Tag1 string + Field2, Tag2 string +} + +func (e *TagPathError) Error() string { + return fmt.Sprintf("%s field %q with tag %q conflicts with field %q with tag %q", e.Struct, e.Field1, e.Tag1, e.Field2, e.Tag2) +} + +const ( + initNilPointers = true + dontInitNilPointers = false +) + +// value returns v's field value corresponding to finfo. +// It's equivalent to v.FieldByIndex(finfo.idx), but when passed +// initNilPointers, it initializes and dereferences pointers as necessary. +// When passed dontInitNilPointers and a nil pointer is reached, the function +// returns a zero reflect.Value. +func (finfo *fieldInfo) value(v reflect.Value, shouldInitNilPointers bool) reflect.Value { + for i, x := range finfo.idx { + if i > 0 { + t := v.Type() + if t.Kind() == reflect.Pointer && t.Elem().Kind() == reflect.Struct { + if v.IsNil() { + if !shouldInitNilPointers { + return reflect.Value{} + } + v.Set(reflect.New(v.Type().Elem())) + } + v = v.Elem() + } + } + v = v.Field(x) + } + return v +} diff --git a/contrib/go/_std_1.22/src/encoding/xml/xml.go b/contrib/go/_std_1.22/src/encoding/xml/xml.go new file mode 100644 index 0000000000..73eedad290 --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/xml/xml.go @@ -0,0 +1,2060 @@ +// Copyright 2009 The Go Authors. All rights reserved. +// Use of this source code is governed by a BSD-style +// license that can be found in the LICENSE file. + +// Package xml implements a simple XML 1.0 parser that +// understands XML name spaces. +package xml + +// References: +// Annotated XML spec: https://www.xml.com/axml/testaxml.htm +// XML name spaces: https://www.w3.org/TR/REC-xml-names/ + +import ( + "bufio" + "bytes" + "errors" + "fmt" + "io" + "strconv" + "strings" + "unicode" + "unicode/utf8" +) + +// A SyntaxError represents a syntax error in the XML input stream. +type SyntaxError struct { + Msg string + Line int +} + +func (e *SyntaxError) Error() string { + return "XML syntax error on line " + strconv.Itoa(e.Line) + ": " + e.Msg +} + +// A Name represents an XML name (Local) annotated +// with a name space identifier (Space). +// In tokens returned by [Decoder.Token], the Space identifier +// is given as a canonical URL, not the short prefix used +// in the document being parsed. +type Name struct { + Space, Local string +} + +// An Attr represents an attribute in an XML element (Name=Value). +type Attr struct { + Name Name + Value string +} + +// A Token is an interface holding one of the token types: +// [StartElement], [EndElement], [CharData], [Comment], [ProcInst], or [Directive]. +type Token any + +// A StartElement represents an XML start element. +type StartElement struct { + Name Name + Attr []Attr +} + +// Copy creates a new copy of StartElement. +func (e StartElement) Copy() StartElement { + attrs := make([]Attr, len(e.Attr)) + copy(attrs, e.Attr) + e.Attr = attrs + return e +} + +// End returns the corresponding XML end element. +func (e StartElement) End() EndElement { + return EndElement{e.Name} +} + +// An EndElement represents an XML end element. +type EndElement struct { + Name Name +} + +// A CharData represents XML character data (raw text), +// in which XML escape sequences have been replaced by +// the characters they represent. +type CharData []byte + +// Copy creates a new copy of CharData. +func (c CharData) Copy() CharData { return CharData(bytes.Clone(c)) } + +// A Comment represents an XML comment of the form <!--comment-->. +// The bytes do not include the <!-- and --> comment markers. +type Comment []byte + +// Copy creates a new copy of Comment. +func (c Comment) Copy() Comment { return Comment(bytes.Clone(c)) } + +// A ProcInst represents an XML processing instruction of the form <?target inst?> +type ProcInst struct { + Target string + Inst []byte +} + +// Copy creates a new copy of ProcInst. +func (p ProcInst) Copy() ProcInst { + p.Inst = bytes.Clone(p.Inst) + return p +} + +// A Directive represents an XML directive of the form <!text>. +// The bytes do not include the <! and > markers. +type Directive []byte + +// Copy creates a new copy of Directive. +func (d Directive) Copy() Directive { return Directive(bytes.Clone(d)) } + +// CopyToken returns a copy of a Token. +func CopyToken(t Token) Token { + switch v := t.(type) { + case CharData: + return v.Copy() + case Comment: + return v.Copy() + case Directive: + return v.Copy() + case ProcInst: + return v.Copy() + case StartElement: + return v.Copy() + } + return t +} + +// A TokenReader is anything that can decode a stream of XML tokens, including a +// [Decoder]. +// +// When Token encounters an error or end-of-file condition after successfully +// reading a token, it returns the token. It may return the (non-nil) error from +// the same call or return the error (and a nil token) from a subsequent call. +// An instance of this general case is that a TokenReader returning a non-nil +// token at the end of the token stream may return either io.EOF or a nil error. +// The next Read should return nil, [io.EOF]. +// +// Implementations of Token are discouraged from returning a nil token with a +// nil error. Callers should treat a return of nil, nil as indicating that +// nothing happened; in particular it does not indicate EOF. +type TokenReader interface { + Token() (Token, error) +} + +// A Decoder represents an XML parser reading a particular input stream. +// The parser assumes that its input is encoded in UTF-8. +type Decoder struct { + // Strict defaults to true, enforcing the requirements + // of the XML specification. + // If set to false, the parser allows input containing common + // mistakes: + // * If an element is missing an end tag, the parser invents + // end tags as necessary to keep the return values from Token + // properly balanced. + // * In attribute values and character data, unknown or malformed + // character entities (sequences beginning with &) are left alone. + // + // Setting: + // + // d.Strict = false + // d.AutoClose = xml.HTMLAutoClose + // d.Entity = xml.HTMLEntity + // + // creates a parser that can handle typical HTML. + // + // Strict mode does not enforce the requirements of the XML name spaces TR. + // In particular it does not reject name space tags using undefined prefixes. + // Such tags are recorded with the unknown prefix as the name space URL. + Strict bool + + // When Strict == false, AutoClose indicates a set of elements to + // consider closed immediately after they are opened, regardless + // of whether an end element is present. + AutoClose []string + + // Entity can be used to map non-standard entity names to string replacements. + // The parser behaves as if these standard mappings are present in the map, + // regardless of the actual map content: + // + // "lt": "<", + // "gt": ">", + // "amp": "&", + // "apos": "'", + // "quot": `"`, + Entity map[string]string + + // CharsetReader, if non-nil, defines a function to generate + // charset-conversion readers, converting from the provided + // non-UTF-8 charset into UTF-8. If CharsetReader is nil or + // returns an error, parsing stops with an error. One of the + // CharsetReader's result values must be non-nil. + CharsetReader func(charset string, input io.Reader) (io.Reader, error) + + // DefaultSpace sets the default name space used for unadorned tags, + // as if the entire XML stream were wrapped in an element containing + // the attribute xmlns="DefaultSpace". + DefaultSpace string + + r io.ByteReader + t TokenReader + buf bytes.Buffer + saved *bytes.Buffer + stk *stack + free *stack + needClose bool + toClose Name + nextToken Token + nextByte int + ns map[string]string + err error + line int + linestart int64 + offset int64 + unmarshalDepth int +} + +// NewDecoder creates a new XML parser reading from r. +// If r does not implement [io.ByteReader], NewDecoder will +// do its own buffering. +func NewDecoder(r io.Reader) *Decoder { + d := &Decoder{ + ns: make(map[string]string), + nextByte: -1, + line: 1, + Strict: true, + } + d.switchToReader(r) + return d +} + +// NewTokenDecoder creates a new XML parser using an underlying token stream. +func NewTokenDecoder(t TokenReader) *Decoder { + // Is it already a Decoder? + if d, ok := t.(*Decoder); ok { + return d + } + d := &Decoder{ + ns: make(map[string]string), + t: t, + nextByte: -1, + line: 1, + Strict: true, + } + return d +} + +// Token returns the next XML token in the input stream. +// At the end of the input stream, Token returns nil, [io.EOF]. +// +// Slices of bytes in the returned token data refer to the +// parser's internal buffer and remain valid only until the next +// call to Token. To acquire a copy of the bytes, call [CopyToken] +// or the token's Copy method. +// +// Token expands self-closing elements such as <br> +// into separate start and end elements returned by successive calls. +// +// Token guarantees that the [StartElement] and [EndElement] +// tokens it returns are properly nested and matched: +// if Token encounters an unexpected end element +// or EOF before all expected end elements, +// it will return an error. +// +// If [Decoder.CharsetReader] is called and returns an error, +// the error is wrapped and returned. +// +// Token implements XML name spaces as described by +// https://www.w3.org/TR/REC-xml-names/. Each of the +// [Name] structures contained in the Token has the Space +// set to the URL identifying its name space when known. +// If Token encounters an unrecognized name space prefix, +// it uses the prefix as the Space rather than report an error. +func (d *Decoder) Token() (Token, error) { + var t Token + var err error + if d.stk != nil && d.stk.kind == stkEOF { + return nil, io.EOF + } + if d.nextToken != nil { + t = d.nextToken + d.nextToken = nil + } else { + if t, err = d.rawToken(); t == nil && err != nil { + if err == io.EOF && d.stk != nil && d.stk.kind != stkEOF { + err = d.syntaxError("unexpected EOF") + } + return nil, err + } + // We still have a token to process, so clear any + // errors (e.g. EOF) and proceed. + err = nil + } + if !d.Strict { + if t1, ok := d.autoClose(t); ok { + d.nextToken = t + t = t1 + } + } + switch t1 := t.(type) { + case StartElement: + // In XML name spaces, the translations listed in the + // attributes apply to the element name and + // to the other attribute names, so process + // the translations first. + for _, a := range t1.Attr { + if a.Name.Space == xmlnsPrefix { + v, ok := d.ns[a.Name.Local] + d.pushNs(a.Name.Local, v, ok) + d.ns[a.Name.Local] = a.Value + } + if a.Name.Space == "" && a.Name.Local == xmlnsPrefix { + // Default space for untagged names + v, ok := d.ns[""] + d.pushNs("", v, ok) + d.ns[""] = a.Value + } + } + + d.pushElement(t1.Name) + d.translate(&t1.Name, true) + for i := range t1.Attr { + d.translate(&t1.Attr[i].Name, false) + } + t = t1 + + case EndElement: + if !d.popElement(&t1) { + return nil, d.err + } + t = t1 + } + return t, err +} + +const ( + xmlURL = "http://www.w3.org/XML/1998/namespace" + xmlnsPrefix = "xmlns" + xmlPrefix = "xml" +) + +// Apply name space translation to name n. +// The default name space (for Space=="") +// applies only to element names, not to attribute names. +func (d *Decoder) translate(n *Name, isElementName bool) { + switch { + case n.Space == xmlnsPrefix: + return + case n.Space == "" && !isElementName: + return + case n.Space == xmlPrefix: + n.Space = xmlURL + case n.Space == "" && n.Local == xmlnsPrefix: + return + } + if v, ok := d.ns[n.Space]; ok { + n.Space = v + } else if n.Space == "" { + n.Space = d.DefaultSpace + } +} + +func (d *Decoder) switchToReader(r io.Reader) { + // Get efficient byte at a time reader. + // Assume that if reader has its own + // ReadByte, it's efficient enough. + // Otherwise, use bufio. + if rb, ok := r.(io.ByteReader); ok { + d.r = rb + } else { + d.r = bufio.NewReader(r) + } +} + +// Parsing state - stack holds old name space translations +// and the current set of open elements. The translations to pop when +// ending a given tag are *below* it on the stack, which is +// more work but forced on us by XML. +type stack struct { + next *stack + kind int + name Name + ok bool +} + +const ( + stkStart = iota + stkNs + stkEOF +) + +func (d *Decoder) push(kind int) *stack { + s := d.free + if s != nil { + d.free = s.next + } else { + s = new(stack) + } + s.next = d.stk + s.kind = kind + d.stk = s + return s +} + +func (d *Decoder) pop() *stack { + s := d.stk + if s != nil { + d.stk = s.next + s.next = d.free + d.free = s + } + return s +} + +// Record that after the current element is finished +// (that element is already pushed on the stack) +// Token should return EOF until popEOF is called. +func (d *Decoder) pushEOF() { + // Walk down stack to find Start. + // It might not be the top, because there might be stkNs + // entries above it. + start := d.stk + for start.kind != stkStart { + start = start.next + } + // The stkNs entries below a start are associated with that + // element too; skip over them. + for start.next != nil && start.next.kind == stkNs { + start = start.next + } + s := d.free + if s != nil { + d.free = s.next + } else { + s = new(stack) + } + s.kind = stkEOF + s.next = start.next + start.next = s +} + +// Undo a pushEOF. +// The element must have been finished, so the EOF should be at the top of the stack. +func (d *Decoder) popEOF() bool { + if d.stk == nil || d.stk.kind != stkEOF { + return false + } + d.pop() + return true +} + +// Record that we are starting an element with the given name. +func (d *Decoder) pushElement(name Name) { + s := d.push(stkStart) + s.name = name +} + +// Record that we are changing the value of ns[local]. +// The old value is url, ok. +func (d *Decoder) pushNs(local string, url string, ok bool) { + s := d.push(stkNs) + s.name.Local = local + s.name.Space = url + s.ok = ok +} + +// Creates a SyntaxError with the current line number. +func (d *Decoder) syntaxError(msg string) error { + return &SyntaxError{Msg: msg, Line: d.line} +} + +// Record that we are ending an element with the given name. +// The name must match the record at the top of the stack, +// which must be a pushElement record. +// After popping the element, apply any undo records from +// the stack to restore the name translations that existed +// before we saw this element. +func (d *Decoder) popElement(t *EndElement) bool { + s := d.pop() + name := t.Name + switch { + case s == nil || s.kind != stkStart: + d.err = d.syntaxError("unexpected end element </" + name.Local + ">") + return false + case s.name.Local != name.Local: + if !d.Strict { + d.needClose = true + d.toClose = t.Name + t.Name = s.name + return true + } + d.err = d.syntaxError("element <" + s.name.Local + "> closed by </" + name.Local + ">") + return false + case s.name.Space != name.Space: + d.err = d.syntaxError("element <" + s.name.Local + "> in space " + s.name.Space + + " closed by </" + name.Local + "> in space " + name.Space) + return false + } + + d.translate(&t.Name, true) + + // Pop stack until a Start or EOF is on the top, undoing the + // translations that were associated with the element we just closed. + for d.stk != nil && d.stk.kind != stkStart && d.stk.kind != stkEOF { + s := d.pop() + if s.ok { + d.ns[s.name.Local] = s.name.Space + } else { + delete(d.ns, s.name.Local) + } + } + + return true +} + +// If the top element on the stack is autoclosing and +// t is not the end tag, invent the end tag. +func (d *Decoder) autoClose(t Token) (Token, bool) { + if d.stk == nil || d.stk.kind != stkStart { + return nil, false + } + for _, s := range d.AutoClose { + if strings.EqualFold(s, d.stk.name.Local) { + // This one should be auto closed if t doesn't close it. + et, ok := t.(EndElement) + if !ok || !strings.EqualFold(et.Name.Local, d.stk.name.Local) { + return EndElement{d.stk.name}, true + } + break + } + } + return nil, false +} + +var errRawToken = errors.New("xml: cannot use RawToken from UnmarshalXML method") + +// RawToken is like [Decoder.Token] but does not verify that +// start and end elements match and does not translate +// name space prefixes to their corresponding URLs. +func (d *Decoder) RawToken() (Token, error) { + if d.unmarshalDepth > 0 { + return nil, errRawToken + } + return d.rawToken() +} + +func (d *Decoder) rawToken() (Token, error) { + if d.t != nil { + return d.t.Token() + } + if d.err != nil { + return nil, d.err + } + if d.needClose { + // The last element we read was self-closing and + // we returned just the StartElement half. + // Return the EndElement half now. + d.needClose = false + return EndElement{d.toClose}, nil + } + + b, ok := d.getc() + if !ok { + return nil, d.err + } + + if b != '<' { + // Text section. + d.ungetc(b) + data := d.text(-1, false) + if data == nil { + return nil, d.err + } + return CharData(data), nil + } + + if b, ok = d.mustgetc(); !ok { + return nil, d.err + } + switch b { + case '/': + // </: End element + var name Name + if name, ok = d.nsname(); !ok { + if d.err == nil { + d.err = d.syntaxError("expected element name after </") + } + return nil, d.err + } + d.space() + if b, ok = d.mustgetc(); !ok { + return nil, d.err + } + if b != '>' { + d.err = d.syntaxError("invalid characters between </" + name.Local + " and >") + return nil, d.err + } + return EndElement{name}, nil + + case '?': + // <?: Processing instruction. + var target string + if target, ok = d.name(); !ok { + if d.err == nil { + d.err = d.syntaxError("expected target name after <?") + } + return nil, d.err + } + d.space() + d.buf.Reset() + var b0 byte + for { + if b, ok = d.mustgetc(); !ok { + return nil, d.err + } + d.buf.WriteByte(b) + if b0 == '?' && b == '>' { + break + } + b0 = b + } + data := d.buf.Bytes() + data = data[0 : len(data)-2] // chop ?> + + if target == "xml" { + content := string(data) + ver := procInst("version", content) + if ver != "" && ver != "1.0" { + d.err = fmt.Errorf("xml: unsupported version %q; only version 1.0 is supported", ver) + return nil, d.err + } + enc := procInst("encoding", content) + if enc != "" && enc != "utf-8" && enc != "UTF-8" && !strings.EqualFold(enc, "utf-8") { + if d.CharsetReader == nil { + d.err = fmt.Errorf("xml: encoding %q declared but Decoder.CharsetReader is nil", enc) + return nil, d.err + } + newr, err := d.CharsetReader(enc, d.r.(io.Reader)) + if err != nil { + d.err = fmt.Errorf("xml: opening charset %q: %w", enc, err) + return nil, d.err + } + if newr == nil { + panic("CharsetReader returned a nil Reader for charset " + enc) + } + d.switchToReader(newr) + } + } + return ProcInst{target, data}, nil + + case '!': + // <!: Maybe comment, maybe CDATA. + if b, ok = d.mustgetc(); !ok { + return nil, d.err + } + switch b { + case '-': // <!- + // Probably <!-- for a comment. + if b, ok = d.mustgetc(); !ok { + return nil, d.err + } + if b != '-' { + d.err = d.syntaxError("invalid sequence <!- not part of <!--") + return nil, d.err + } + // Look for terminator. + d.buf.Reset() + var b0, b1 byte + for { + if b, ok = d.mustgetc(); !ok { + return nil, d.err + } + d.buf.WriteByte(b) + if b0 == '-' && b1 == '-' { + if b != '>' { + d.err = d.syntaxError( + `invalid sequence "--" not allowed in comments`) + return nil, d.err + } + break + } + b0, b1 = b1, b + } + data := d.buf.Bytes() + data = data[0 : len(data)-3] // chop --> + return Comment(data), nil + + case '[': // <![ + // Probably <![CDATA[. + for i := 0; i < 6; i++ { + if b, ok = d.mustgetc(); !ok { + return nil, d.err + } + if b != "CDATA["[i] { + d.err = d.syntaxError("invalid <![ sequence") + return nil, d.err + } + } + // Have <![CDATA[. Read text until ]]>. + data := d.text(-1, true) + if data == nil { + return nil, d.err + } + return CharData(data), nil + } + + // Probably a directive: <!DOCTYPE ...>, <!ENTITY ...>, etc. + // We don't care, but accumulate for caller. Quoted angle + // brackets do not count for nesting. + d.buf.Reset() + d.buf.WriteByte(b) + inquote := uint8(0) + depth := 0 + for { + if b, ok = d.mustgetc(); !ok { + return nil, d.err + } + if inquote == 0 && b == '>' && depth == 0 { + break + } + HandleB: + d.buf.WriteByte(b) + switch { + case b == inquote: + inquote = 0 + + case inquote != 0: + // in quotes, no special action + + case b == '\'' || b == '"': + inquote = b + + case b == '>' && inquote == 0: + depth-- + + case b == '<' && inquote == 0: + // Look for <!-- to begin comment. + s := "!--" + for i := 0; i < len(s); i++ { + if b, ok = d.mustgetc(); !ok { + return nil, d.err + } + if b != s[i] { + for j := 0; j < i; j++ { + d.buf.WriteByte(s[j]) + } + depth++ + goto HandleB + } + } + + // Remove < that was written above. + d.buf.Truncate(d.buf.Len() - 1) + + // Look for terminator. + var b0, b1 byte + for { + if b, ok = d.mustgetc(); !ok { + return nil, d.err + } + if b0 == '-' && b1 == '-' && b == '>' { + break + } + b0, b1 = b1, b + } + + // Replace the comment with a space in the returned Directive + // body, so that markup parts that were separated by the comment + // (like a "<" and a "!") don't get joined when re-encoding the + // Directive, taking new semantic meaning. + d.buf.WriteByte(' ') + } + } + return Directive(d.buf.Bytes()), nil + } + + // Must be an open element like <a href="foo"> + d.ungetc(b) + + var ( + name Name + empty bool + attr []Attr + ) + if name, ok = d.nsname(); !ok { + if d.err == nil { + d.err = d.syntaxError("expected element name after <") + } + return nil, d.err + } + + attr = []Attr{} + for { + d.space() + if b, ok = d.mustgetc(); !ok { + return nil, d.err + } + if b == '/' { + empty = true + if b, ok = d.mustgetc(); !ok { + return nil, d.err + } + if b != '>' { + d.err = d.syntaxError("expected /> in element") + return nil, d.err + } + break + } + if b == '>' { + break + } + d.ungetc(b) + + a := Attr{} + if a.Name, ok = d.nsname(); !ok { + if d.err == nil { + d.err = d.syntaxError("expected attribute name in element") + } + return nil, d.err + } + d.space() + if b, ok = d.mustgetc(); !ok { + return nil, d.err + } + if b != '=' { + if d.Strict { + d.err = d.syntaxError("attribute name without = in element") + return nil, d.err + } + d.ungetc(b) + a.Value = a.Name.Local + } else { + d.space() + data := d.attrval() + if data == nil { + return nil, d.err + } + a.Value = string(data) + } + attr = append(attr, a) + } + if empty { + d.needClose = true + d.toClose = name + } + return StartElement{name, attr}, nil +} + +func (d *Decoder) attrval() []byte { + b, ok := d.mustgetc() + if !ok { + return nil + } + // Handle quoted attribute values + if b == '"' || b == '\'' { + return d.text(int(b), false) + } + // Handle unquoted attribute values for strict parsers + if d.Strict { + d.err = d.syntaxError("unquoted or missing attribute value in element") + return nil + } + // Handle unquoted attribute values for unstrict parsers + d.ungetc(b) + d.buf.Reset() + for { + b, ok = d.mustgetc() + if !ok { + return nil + } + // https://www.w3.org/TR/REC-html40/intro/sgmltut.html#h-3.2.2 + if 'a' <= b && b <= 'z' || 'A' <= b && b <= 'Z' || + '0' <= b && b <= '9' || b == '_' || b == ':' || b == '-' { + d.buf.WriteByte(b) + } else { + d.ungetc(b) + break + } + } + return d.buf.Bytes() +} + +// Skip spaces if any +func (d *Decoder) space() { + for { + b, ok := d.getc() + if !ok { + return + } + switch b { + case ' ', '\r', '\n', '\t': + default: + d.ungetc(b) + return + } + } +} + +// Read a single byte. +// If there is no byte to read, return ok==false +// and leave the error in d.err. +// Maintain line number. +func (d *Decoder) getc() (b byte, ok bool) { + if d.err != nil { + return 0, false + } + if d.nextByte >= 0 { + b = byte(d.nextByte) + d.nextByte = -1 + } else { + b, d.err = d.r.ReadByte() + if d.err != nil { + return 0, false + } + if d.saved != nil { + d.saved.WriteByte(b) + } + } + if b == '\n' { + d.line++ + d.linestart = d.offset + 1 + } + d.offset++ + return b, true +} + +// InputOffset returns the input stream byte offset of the current decoder position. +// The offset gives the location of the end of the most recently returned token +// and the beginning of the next token. +func (d *Decoder) InputOffset() int64 { + return d.offset +} + +// InputPos returns the line of the current decoder position and the 1 based +// input position of the line. The position gives the location of the end of the +// most recently returned token. +func (d *Decoder) InputPos() (line, column int) { + return d.line, int(d.offset-d.linestart) + 1 +} + +// Return saved offset. +// If we did ungetc (nextByte >= 0), have to back up one. +func (d *Decoder) savedOffset() int { + n := d.saved.Len() + if d.nextByte >= 0 { + n-- + } + return n +} + +// Must read a single byte. +// If there is no byte to read, +// set d.err to SyntaxError("unexpected EOF") +// and return ok==false +func (d *Decoder) mustgetc() (b byte, ok bool) { + if b, ok = d.getc(); !ok { + if d.err == io.EOF { + d.err = d.syntaxError("unexpected EOF") + } + } + return +} + +// Unread a single byte. +func (d *Decoder) ungetc(b byte) { + if b == '\n' { + d.line-- + } + d.nextByte = int(b) + d.offset-- +} + +var entity = map[string]rune{ + "lt": '<', + "gt": '>', + "amp": '&', + "apos": '\'', + "quot": '"', +} + +// Read plain text section (XML calls it character data). +// If quote >= 0, we are in a quoted string and need to find the matching quote. +// If cdata == true, we are in a <![CDATA[ section and need to find ]]>. +// On failure return nil and leave the error in d.err. +func (d *Decoder) text(quote int, cdata bool) []byte { + var b0, b1 byte + var trunc int + d.buf.Reset() +Input: + for { + b, ok := d.getc() + if !ok { + if cdata { + if d.err == io.EOF { + d.err = d.syntaxError("unexpected EOF in CDATA section") + } + return nil + } + break Input + } + + // <![CDATA[ section ends with ]]>. + // It is an error for ]]> to appear in ordinary text. + if b0 == ']' && b1 == ']' && b == '>' { + if cdata { + trunc = 2 + break Input + } + d.err = d.syntaxError("unescaped ]]> not in CDATA section") + return nil + } + + // Stop reading text if we see a <. + if b == '<' && !cdata { + if quote >= 0 { + d.err = d.syntaxError("unescaped < inside quoted string") + return nil + } + d.ungetc('<') + break Input + } + if quote >= 0 && b == byte(quote) { + break Input + } + if b == '&' && !cdata { + // Read escaped character expression up to semicolon. + // XML in all its glory allows a document to define and use + // its own character names with <!ENTITY ...> directives. + // Parsers are required to recognize lt, gt, amp, apos, and quot + // even if they have not been declared. + before := d.buf.Len() + d.buf.WriteByte('&') + var ok bool + var text string + var haveText bool + if b, ok = d.mustgetc(); !ok { + return nil + } + if b == '#' { + d.buf.WriteByte(b) + if b, ok = d.mustgetc(); !ok { + return nil + } + base := 10 + if b == 'x' { + base = 16 + d.buf.WriteByte(b) + if b, ok = d.mustgetc(); !ok { + return nil + } + } + start := d.buf.Len() + for '0' <= b && b <= '9' || + base == 16 && 'a' <= b && b <= 'f' || + base == 16 && 'A' <= b && b <= 'F' { + d.buf.WriteByte(b) + if b, ok = d.mustgetc(); !ok { + return nil + } + } + if b != ';' { + d.ungetc(b) + } else { + s := string(d.buf.Bytes()[start:]) + d.buf.WriteByte(';') + n, err := strconv.ParseUint(s, base, 64) + if err == nil && n <= unicode.MaxRune { + text = string(rune(n)) + haveText = true + } + } + } else { + d.ungetc(b) + if !d.readName() { + if d.err != nil { + return nil + } + } + if b, ok = d.mustgetc(); !ok { + return nil + } + if b != ';' { + d.ungetc(b) + } else { + name := d.buf.Bytes()[before+1:] + d.buf.WriteByte(';') + if isName(name) { + s := string(name) + if r, ok := entity[s]; ok { + text = string(r) + haveText = true + } else if d.Entity != nil { + text, haveText = d.Entity[s] + } + } + } + } + + if haveText { + d.buf.Truncate(before) + d.buf.WriteString(text) + b0, b1 = 0, 0 + continue Input + } + if !d.Strict { + b0, b1 = 0, 0 + continue Input + } + ent := string(d.buf.Bytes()[before:]) + if ent[len(ent)-1] != ';' { + ent += " (no semicolon)" + } + d.err = d.syntaxError("invalid character entity " + ent) + return nil + } + + // We must rewrite unescaped \r and \r\n into \n. + if b == '\r' { + d.buf.WriteByte('\n') + } else if b1 == '\r' && b == '\n' { + // Skip \r\n--we already wrote \n. + } else { + d.buf.WriteByte(b) + } + + b0, b1 = b1, b + } + data := d.buf.Bytes() + data = data[0 : len(data)-trunc] + + // Inspect each rune for being a disallowed character. + buf := data + for len(buf) > 0 { + r, size := utf8.DecodeRune(buf) + if r == utf8.RuneError && size == 1 { + d.err = d.syntaxError("invalid UTF-8") + return nil + } + buf = buf[size:] + if !isInCharacterRange(r) { + d.err = d.syntaxError(fmt.Sprintf("illegal character code %U", r)) + return nil + } + } + + return data +} + +// Decide whether the given rune is in the XML Character Range, per +// the Char production of https://www.xml.com/axml/testaxml.htm, +// Section 2.2 Characters. +func isInCharacterRange(r rune) (inrange bool) { + return r == 0x09 || + r == 0x0A || + r == 0x0D || + r >= 0x20 && r <= 0xD7FF || + r >= 0xE000 && r <= 0xFFFD || + r >= 0x10000 && r <= 0x10FFFF +} + +// Get name space name: name with a : stuck in the middle. +// The part before the : is the name space identifier. +func (d *Decoder) nsname() (name Name, ok bool) { + s, ok := d.name() + if !ok { + return + } + if strings.Count(s, ":") > 1 { + return name, false + } else if space, local, ok := strings.Cut(s, ":"); !ok || space == "" || local == "" { + name.Local = s + } else { + name.Space = space + name.Local = local + } + return name, true +} + +// Get name: /first(first|second)*/ +// Do not set d.err if the name is missing (unless unexpected EOF is received): +// let the caller provide better context. +func (d *Decoder) name() (s string, ok bool) { + d.buf.Reset() + if !d.readName() { + return "", false + } + + // Now we check the characters. + b := d.buf.Bytes() + if !isName(b) { + d.err = d.syntaxError("invalid XML name: " + string(b)) + return "", false + } + return string(b), true +} + +// Read a name and append its bytes to d.buf. +// The name is delimited by any single-byte character not valid in names. +// All multi-byte characters are accepted; the caller must check their validity. +func (d *Decoder) readName() (ok bool) { + var b byte + if b, ok = d.mustgetc(); !ok { + return + } + if b < utf8.RuneSelf && !isNameByte(b) { + d.ungetc(b) + return false + } + d.buf.WriteByte(b) + + for { + if b, ok = d.mustgetc(); !ok { + return + } + if b < utf8.RuneSelf && !isNameByte(b) { + d.ungetc(b) + break + } + d.buf.WriteByte(b) + } + return true +} + +func isNameByte(c byte) bool { + return 'A' <= c && c <= 'Z' || + 'a' <= c && c <= 'z' || + '0' <= c && c <= '9' || + c == '_' || c == ':' || c == '.' || c == '-' +} + +func isName(s []byte) bool { + if len(s) == 0 { + return false + } + c, n := utf8.DecodeRune(s) + if c == utf8.RuneError && n == 1 { + return false + } + if !unicode.Is(first, c) { + return false + } + for n < len(s) { + s = s[n:] + c, n = utf8.DecodeRune(s) + if c == utf8.RuneError && n == 1 { + return false + } + if !unicode.Is(first, c) && !unicode.Is(second, c) { + return false + } + } + return true +} + +func isNameString(s string) bool { + if len(s) == 0 { + return false + } + c, n := utf8.DecodeRuneInString(s) + if c == utf8.RuneError && n == 1 { + return false + } + if !unicode.Is(first, c) { + return false + } + for n < len(s) { + s = s[n:] + c, n = utf8.DecodeRuneInString(s) + if c == utf8.RuneError && n == 1 { + return false + } + if !unicode.Is(first, c) && !unicode.Is(second, c) { + return false + } + } + return true +} + +// These tables were generated by cut and paste from Appendix B of +// the XML spec at https://www.xml.com/axml/testaxml.htm +// and then reformatting. First corresponds to (Letter | '_' | ':') +// and second corresponds to NameChar. + +var first = &unicode.RangeTable{ + R16: []unicode.Range16{ + {0x003A, 0x003A, 1}, + {0x0041, 0x005A, 1}, + {0x005F, 0x005F, 1}, + {0x0061, 0x007A, 1}, + {0x00C0, 0x00D6, 1}, + {0x00D8, 0x00F6, 1}, + {0x00F8, 0x00FF, 1}, + {0x0100, 0x0131, 1}, + {0x0134, 0x013E, 1}, + {0x0141, 0x0148, 1}, + {0x014A, 0x017E, 1}, + {0x0180, 0x01C3, 1}, + {0x01CD, 0x01F0, 1}, + {0x01F4, 0x01F5, 1}, + {0x01FA, 0x0217, 1}, + {0x0250, 0x02A8, 1}, + {0x02BB, 0x02C1, 1}, + {0x0386, 0x0386, 1}, + {0x0388, 0x038A, 1}, + {0x038C, 0x038C, 1}, + {0x038E, 0x03A1, 1}, + {0x03A3, 0x03CE, 1}, + {0x03D0, 0x03D6, 1}, + {0x03DA, 0x03E0, 2}, + {0x03E2, 0x03F3, 1}, + {0x0401, 0x040C, 1}, + {0x040E, 0x044F, 1}, + {0x0451, 0x045C, 1}, + {0x045E, 0x0481, 1}, + {0x0490, 0x04C4, 1}, + {0x04C7, 0x04C8, 1}, + {0x04CB, 0x04CC, 1}, + {0x04D0, 0x04EB, 1}, + {0x04EE, 0x04F5, 1}, + {0x04F8, 0x04F9, 1}, + {0x0531, 0x0556, 1}, + {0x0559, 0x0559, 1}, + {0x0561, 0x0586, 1}, + {0x05D0, 0x05EA, 1}, + {0x05F0, 0x05F2, 1}, + {0x0621, 0x063A, 1}, + {0x0641, 0x064A, 1}, + {0x0671, 0x06B7, 1}, + {0x06BA, 0x06BE, 1}, + {0x06C0, 0x06CE, 1}, + {0x06D0, 0x06D3, 1}, + {0x06D5, 0x06D5, 1}, + {0x06E5, 0x06E6, 1}, + {0x0905, 0x0939, 1}, + {0x093D, 0x093D, 1}, + {0x0958, 0x0961, 1}, + {0x0985, 0x098C, 1}, + {0x098F, 0x0990, 1}, + {0x0993, 0x09A8, 1}, + {0x09AA, 0x09B0, 1}, + {0x09B2, 0x09B2, 1}, + {0x09B6, 0x09B9, 1}, + {0x09DC, 0x09DD, 1}, + {0x09DF, 0x09E1, 1}, + {0x09F0, 0x09F1, 1}, + {0x0A05, 0x0A0A, 1}, + {0x0A0F, 0x0A10, 1}, + {0x0A13, 0x0A28, 1}, + {0x0A2A, 0x0A30, 1}, + {0x0A32, 0x0A33, 1}, + {0x0A35, 0x0A36, 1}, + {0x0A38, 0x0A39, 1}, + {0x0A59, 0x0A5C, 1}, + {0x0A5E, 0x0A5E, 1}, + {0x0A72, 0x0A74, 1}, + {0x0A85, 0x0A8B, 1}, + {0x0A8D, 0x0A8D, 1}, + {0x0A8F, 0x0A91, 1}, + {0x0A93, 0x0AA8, 1}, + {0x0AAA, 0x0AB0, 1}, + {0x0AB2, 0x0AB3, 1}, + {0x0AB5, 0x0AB9, 1}, + {0x0ABD, 0x0AE0, 0x23}, + {0x0B05, 0x0B0C, 1}, + {0x0B0F, 0x0B10, 1}, + {0x0B13, 0x0B28, 1}, + {0x0B2A, 0x0B30, 1}, + {0x0B32, 0x0B33, 1}, + {0x0B36, 0x0B39, 1}, + {0x0B3D, 0x0B3D, 1}, + {0x0B5C, 0x0B5D, 1}, + {0x0B5F, 0x0B61, 1}, + {0x0B85, 0x0B8A, 1}, + {0x0B8E, 0x0B90, 1}, + {0x0B92, 0x0B95, 1}, + {0x0B99, 0x0B9A, 1}, + {0x0B9C, 0x0B9C, 1}, + {0x0B9E, 0x0B9F, 1}, + {0x0BA3, 0x0BA4, 1}, + {0x0BA8, 0x0BAA, 1}, + {0x0BAE, 0x0BB5, 1}, + {0x0BB7, 0x0BB9, 1}, + {0x0C05, 0x0C0C, 1}, + {0x0C0E, 0x0C10, 1}, + {0x0C12, 0x0C28, 1}, + {0x0C2A, 0x0C33, 1}, + {0x0C35, 0x0C39, 1}, + {0x0C60, 0x0C61, 1}, + {0x0C85, 0x0C8C, 1}, + {0x0C8E, 0x0C90, 1}, + {0x0C92, 0x0CA8, 1}, + {0x0CAA, 0x0CB3, 1}, + {0x0CB5, 0x0CB9, 1}, + {0x0CDE, 0x0CDE, 1}, + {0x0CE0, 0x0CE1, 1}, + {0x0D05, 0x0D0C, 1}, + {0x0D0E, 0x0D10, 1}, + {0x0D12, 0x0D28, 1}, + {0x0D2A, 0x0D39, 1}, + {0x0D60, 0x0D61, 1}, + {0x0E01, 0x0E2E, 1}, + {0x0E30, 0x0E30, 1}, + {0x0E32, 0x0E33, 1}, + {0x0E40, 0x0E45, 1}, + {0x0E81, 0x0E82, 1}, + {0x0E84, 0x0E84, 1}, + {0x0E87, 0x0E88, 1}, + {0x0E8A, 0x0E8D, 3}, + {0x0E94, 0x0E97, 1}, + {0x0E99, 0x0E9F, 1}, + {0x0EA1, 0x0EA3, 1}, + {0x0EA5, 0x0EA7, 2}, + {0x0EAA, 0x0EAB, 1}, + {0x0EAD, 0x0EAE, 1}, + {0x0EB0, 0x0EB0, 1}, + {0x0EB2, 0x0EB3, 1}, + {0x0EBD, 0x0EBD, 1}, + {0x0EC0, 0x0EC4, 1}, + {0x0F40, 0x0F47, 1}, + {0x0F49, 0x0F69, 1}, + {0x10A0, 0x10C5, 1}, + {0x10D0, 0x10F6, 1}, + {0x1100, 0x1100, 1}, + {0x1102, 0x1103, 1}, + {0x1105, 0x1107, 1}, + {0x1109, 0x1109, 1}, + {0x110B, 0x110C, 1}, + {0x110E, 0x1112, 1}, + {0x113C, 0x1140, 2}, + {0x114C, 0x1150, 2}, + {0x1154, 0x1155, 1}, + {0x1159, 0x1159, 1}, + {0x115F, 0x1161, 1}, + {0x1163, 0x1169, 2}, + {0x116D, 0x116E, 1}, + {0x1172, 0x1173, 1}, + {0x1175, 0x119E, 0x119E - 0x1175}, + {0x11A8, 0x11AB, 0x11AB - 0x11A8}, + {0x11AE, 0x11AF, 1}, + {0x11B7, 0x11B8, 1}, + {0x11BA, 0x11BA, 1}, + {0x11BC, 0x11C2, 1}, + {0x11EB, 0x11F0, 0x11F0 - 0x11EB}, + {0x11F9, 0x11F9, 1}, + {0x1E00, 0x1E9B, 1}, + {0x1EA0, 0x1EF9, 1}, + {0x1F00, 0x1F15, 1}, + {0x1F18, 0x1F1D, 1}, + {0x1F20, 0x1F45, 1}, + {0x1F48, 0x1F4D, 1}, + {0x1F50, 0x1F57, 1}, + {0x1F59, 0x1F5B, 0x1F5B - 0x1F59}, + {0x1F5D, 0x1F5D, 1}, + {0x1F5F, 0x1F7D, 1}, + {0x1F80, 0x1FB4, 1}, + {0x1FB6, 0x1FBC, 1}, + {0x1FBE, 0x1FBE, 1}, + {0x1FC2, 0x1FC4, 1}, + {0x1FC6, 0x1FCC, 1}, + {0x1FD0, 0x1FD3, 1}, + {0x1FD6, 0x1FDB, 1}, + {0x1FE0, 0x1FEC, 1}, + {0x1FF2, 0x1FF4, 1}, + {0x1FF6, 0x1FFC, 1}, + {0x2126, 0x2126, 1}, + {0x212A, 0x212B, 1}, + {0x212E, 0x212E, 1}, + {0x2180, 0x2182, 1}, + {0x3007, 0x3007, 1}, + {0x3021, 0x3029, 1}, + {0x3041, 0x3094, 1}, + {0x30A1, 0x30FA, 1}, + {0x3105, 0x312C, 1}, + {0x4E00, 0x9FA5, 1}, + {0xAC00, 0xD7A3, 1}, + }, +} + +var second = &unicode.RangeTable{ + R16: []unicode.Range16{ + {0x002D, 0x002E, 1}, + {0x0030, 0x0039, 1}, + {0x00B7, 0x00B7, 1}, + {0x02D0, 0x02D1, 1}, + {0x0300, 0x0345, 1}, + {0x0360, 0x0361, 1}, + {0x0387, 0x0387, 1}, + {0x0483, 0x0486, 1}, + {0x0591, 0x05A1, 1}, + {0x05A3, 0x05B9, 1}, + {0x05BB, 0x05BD, 1}, + {0x05BF, 0x05BF, 1}, + {0x05C1, 0x05C2, 1}, + {0x05C4, 0x0640, 0x0640 - 0x05C4}, + {0x064B, 0x0652, 1}, + {0x0660, 0x0669, 1}, + {0x0670, 0x0670, 1}, + {0x06D6, 0x06DC, 1}, + {0x06DD, 0x06DF, 1}, + {0x06E0, 0x06E4, 1}, + {0x06E7, 0x06E8, 1}, + {0x06EA, 0x06ED, 1}, + {0x06F0, 0x06F9, 1}, + {0x0901, 0x0903, 1}, + {0x093C, 0x093C, 1}, + {0x093E, 0x094C, 1}, + {0x094D, 0x094D, 1}, + {0x0951, 0x0954, 1}, + {0x0962, 0x0963, 1}, + {0x0966, 0x096F, 1}, + {0x0981, 0x0983, 1}, + {0x09BC, 0x09BC, 1}, + {0x09BE, 0x09BF, 1}, + {0x09C0, 0x09C4, 1}, + {0x09C7, 0x09C8, 1}, + {0x09CB, 0x09CD, 1}, + {0x09D7, 0x09D7, 1}, + {0x09E2, 0x09E3, 1}, + {0x09E6, 0x09EF, 1}, + {0x0A02, 0x0A3C, 0x3A}, + {0x0A3E, 0x0A3F, 1}, + {0x0A40, 0x0A42, 1}, + {0x0A47, 0x0A48, 1}, + {0x0A4B, 0x0A4D, 1}, + {0x0A66, 0x0A6F, 1}, + {0x0A70, 0x0A71, 1}, + {0x0A81, 0x0A83, 1}, + {0x0ABC, 0x0ABC, 1}, + {0x0ABE, 0x0AC5, 1}, + {0x0AC7, 0x0AC9, 1}, + {0x0ACB, 0x0ACD, 1}, + {0x0AE6, 0x0AEF, 1}, + {0x0B01, 0x0B03, 1}, + {0x0B3C, 0x0B3C, 1}, + {0x0B3E, 0x0B43, 1}, + {0x0B47, 0x0B48, 1}, + {0x0B4B, 0x0B4D, 1}, + {0x0B56, 0x0B57, 1}, + {0x0B66, 0x0B6F, 1}, + {0x0B82, 0x0B83, 1}, + {0x0BBE, 0x0BC2, 1}, + {0x0BC6, 0x0BC8, 1}, + {0x0BCA, 0x0BCD, 1}, + {0x0BD7, 0x0BD7, 1}, + {0x0BE7, 0x0BEF, 1}, + {0x0C01, 0x0C03, 1}, + {0x0C3E, 0x0C44, 1}, + {0x0C46, 0x0C48, 1}, + {0x0C4A, 0x0C4D, 1}, + {0x0C55, 0x0C56, 1}, + {0x0C66, 0x0C6F, 1}, + {0x0C82, 0x0C83, 1}, + {0x0CBE, 0x0CC4, 1}, + {0x0CC6, 0x0CC8, 1}, + {0x0CCA, 0x0CCD, 1}, + {0x0CD5, 0x0CD6, 1}, + {0x0CE6, 0x0CEF, 1}, + {0x0D02, 0x0D03, 1}, + {0x0D3E, 0x0D43, 1}, + {0x0D46, 0x0D48, 1}, + {0x0D4A, 0x0D4D, 1}, + {0x0D57, 0x0D57, 1}, + {0x0D66, 0x0D6F, 1}, + {0x0E31, 0x0E31, 1}, + {0x0E34, 0x0E3A, 1}, + {0x0E46, 0x0E46, 1}, + {0x0E47, 0x0E4E, 1}, + {0x0E50, 0x0E59, 1}, + {0x0EB1, 0x0EB1, 1}, + {0x0EB4, 0x0EB9, 1}, + {0x0EBB, 0x0EBC, 1}, + {0x0EC6, 0x0EC6, 1}, + {0x0EC8, 0x0ECD, 1}, + {0x0ED0, 0x0ED9, 1}, + {0x0F18, 0x0F19, 1}, + {0x0F20, 0x0F29, 1}, + {0x0F35, 0x0F39, 2}, + {0x0F3E, 0x0F3F, 1}, + {0x0F71, 0x0F84, 1}, + {0x0F86, 0x0F8B, 1}, + {0x0F90, 0x0F95, 1}, + {0x0F97, 0x0F97, 1}, + {0x0F99, 0x0FAD, 1}, + {0x0FB1, 0x0FB7, 1}, + {0x0FB9, 0x0FB9, 1}, + {0x20D0, 0x20DC, 1}, + {0x20E1, 0x3005, 0x3005 - 0x20E1}, + {0x302A, 0x302F, 1}, + {0x3031, 0x3035, 1}, + {0x3099, 0x309A, 1}, + {0x309D, 0x309E, 1}, + {0x30FC, 0x30FE, 1}, + }, +} + +// HTMLEntity is an entity map containing translations for the +// standard HTML entity characters. +// +// See the [Decoder.Strict] and [Decoder.Entity] fields' documentation. +var HTMLEntity map[string]string = htmlEntity + +var htmlEntity = map[string]string{ + /* + hget http://www.w3.org/TR/html4/sgml/entities.html | + ssam ' + ,y /\>/ x/\<(.|\n)+/ s/\n/ /g + ,x v/^\<!ENTITY/d + ,s/\<!ENTITY ([^ ]+) .*U\+([0-9A-F][0-9A-F][0-9A-F][0-9A-F]) .+/ "\1": "\\u\2",/g + ' + */ + "nbsp": "\u00A0", + "iexcl": "\u00A1", + "cent": "\u00A2", + "pound": "\u00A3", + "curren": "\u00A4", + "yen": "\u00A5", + "brvbar": "\u00A6", + "sect": "\u00A7", + "uml": "\u00A8", + "copy": "\u00A9", + "ordf": "\u00AA", + "laquo": "\u00AB", + "not": "\u00AC", + "shy": "\u00AD", + "reg": "\u00AE", + "macr": "\u00AF", + "deg": "\u00B0", + "plusmn": "\u00B1", + "sup2": "\u00B2", + "sup3": "\u00B3", + "acute": "\u00B4", + "micro": "\u00B5", + "para": "\u00B6", + "middot": "\u00B7", + "cedil": "\u00B8", + "sup1": "\u00B9", + "ordm": "\u00BA", + "raquo": "\u00BB", + "frac14": "\u00BC", + "frac12": "\u00BD", + "frac34": "\u00BE", + "iquest": "\u00BF", + "Agrave": "\u00C0", + "Aacute": "\u00C1", + "Acirc": "\u00C2", + "Atilde": "\u00C3", + "Auml": "\u00C4", + "Aring": "\u00C5", + "AElig": "\u00C6", + "Ccedil": "\u00C7", + "Egrave": "\u00C8", + "Eacute": "\u00C9", + "Ecirc": "\u00CA", + "Euml": "\u00CB", + "Igrave": "\u00CC", + "Iacute": "\u00CD", + "Icirc": "\u00CE", + "Iuml": "\u00CF", + "ETH": "\u00D0", + "Ntilde": "\u00D1", + "Ograve": "\u00D2", + "Oacute": "\u00D3", + "Ocirc": "\u00D4", + "Otilde": "\u00D5", + "Ouml": "\u00D6", + "times": "\u00D7", + "Oslash": "\u00D8", + "Ugrave": "\u00D9", + "Uacute": "\u00DA", + "Ucirc": "\u00DB", + "Uuml": "\u00DC", + "Yacute": "\u00DD", + "THORN": "\u00DE", + "szlig": "\u00DF", + "agrave": "\u00E0", + "aacute": "\u00E1", + "acirc": "\u00E2", + "atilde": "\u00E3", + "auml": "\u00E4", + "aring": "\u00E5", + "aelig": "\u00E6", + "ccedil": "\u00E7", + "egrave": "\u00E8", + "eacute": "\u00E9", + "ecirc": "\u00EA", + "euml": "\u00EB", + "igrave": "\u00EC", + "iacute": "\u00ED", + "icirc": "\u00EE", + "iuml": "\u00EF", + "eth": "\u00F0", + "ntilde": "\u00F1", + "ograve": "\u00F2", + "oacute": "\u00F3", + "ocirc": "\u00F4", + "otilde": "\u00F5", + "ouml": "\u00F6", + "divide": "\u00F7", + "oslash": "\u00F8", + "ugrave": "\u00F9", + "uacute": "\u00FA", + "ucirc": "\u00FB", + "uuml": "\u00FC", + "yacute": "\u00FD", + "thorn": "\u00FE", + "yuml": "\u00FF", + "fnof": "\u0192", + "Alpha": "\u0391", + "Beta": "\u0392", + "Gamma": "\u0393", + "Delta": "\u0394", + "Epsilon": "\u0395", + "Zeta": "\u0396", + "Eta": "\u0397", + "Theta": "\u0398", + "Iota": "\u0399", + "Kappa": "\u039A", + "Lambda": "\u039B", + "Mu": "\u039C", + "Nu": "\u039D", + "Xi": "\u039E", + "Omicron": "\u039F", + "Pi": "\u03A0", + "Rho": "\u03A1", + "Sigma": "\u03A3", + "Tau": "\u03A4", + "Upsilon": "\u03A5", + "Phi": "\u03A6", + "Chi": "\u03A7", + "Psi": "\u03A8", + "Omega": "\u03A9", + "alpha": "\u03B1", + "beta": "\u03B2", + "gamma": "\u03B3", + "delta": "\u03B4", + "epsilon": "\u03B5", + "zeta": "\u03B6", + "eta": "\u03B7", + "theta": "\u03B8", + "iota": "\u03B9", + "kappa": "\u03BA", + "lambda": "\u03BB", + "mu": "\u03BC", + "nu": "\u03BD", + "xi": "\u03BE", + "omicron": "\u03BF", + "pi": "\u03C0", + "rho": "\u03C1", + "sigmaf": "\u03C2", + "sigma": "\u03C3", + "tau": "\u03C4", + "upsilon": "\u03C5", + "phi": "\u03C6", + "chi": "\u03C7", + "psi": "\u03C8", + "omega": "\u03C9", + "thetasym": "\u03D1", + "upsih": "\u03D2", + "piv": "\u03D6", + "bull": "\u2022", + "hellip": "\u2026", + "prime": "\u2032", + "Prime": "\u2033", + "oline": "\u203E", + "frasl": "\u2044", + "weierp": "\u2118", + "image": "\u2111", + "real": "\u211C", + "trade": "\u2122", + "alefsym": "\u2135", + "larr": "\u2190", + "uarr": "\u2191", + "rarr": "\u2192", + "darr": "\u2193", + "harr": "\u2194", + "crarr": "\u21B5", + "lArr": "\u21D0", + "uArr": "\u21D1", + "rArr": "\u21D2", + "dArr": "\u21D3", + "hArr": "\u21D4", + "forall": "\u2200", + "part": "\u2202", + "exist": "\u2203", + "empty": "\u2205", + "nabla": "\u2207", + "isin": "\u2208", + "notin": "\u2209", + "ni": "\u220B", + "prod": "\u220F", + "sum": "\u2211", + "minus": "\u2212", + "lowast": "\u2217", + "radic": "\u221A", + "prop": "\u221D", + "infin": "\u221E", + "ang": "\u2220", + "and": "\u2227", + "or": "\u2228", + "cap": "\u2229", + "cup": "\u222A", + "int": "\u222B", + "there4": "\u2234", + "sim": "\u223C", + "cong": "\u2245", + "asymp": "\u2248", + "ne": "\u2260", + "equiv": "\u2261", + "le": "\u2264", + "ge": "\u2265", + "sub": "\u2282", + "sup": "\u2283", + "nsub": "\u2284", + "sube": "\u2286", + "supe": "\u2287", + "oplus": "\u2295", + "otimes": "\u2297", + "perp": "\u22A5", + "sdot": "\u22C5", + "lceil": "\u2308", + "rceil": "\u2309", + "lfloor": "\u230A", + "rfloor": "\u230B", + "lang": "\u2329", + "rang": "\u232A", + "loz": "\u25CA", + "spades": "\u2660", + "clubs": "\u2663", + "hearts": "\u2665", + "diams": "\u2666", + "quot": "\u0022", + "amp": "\u0026", + "lt": "\u003C", + "gt": "\u003E", + "OElig": "\u0152", + "oelig": "\u0153", + "Scaron": "\u0160", + "scaron": "\u0161", + "Yuml": "\u0178", + "circ": "\u02C6", + "tilde": "\u02DC", + "ensp": "\u2002", + "emsp": "\u2003", + "thinsp": "\u2009", + "zwnj": "\u200C", + "zwj": "\u200D", + "lrm": "\u200E", + "rlm": "\u200F", + "ndash": "\u2013", + "mdash": "\u2014", + "lsquo": "\u2018", + "rsquo": "\u2019", + "sbquo": "\u201A", + "ldquo": "\u201C", + "rdquo": "\u201D", + "bdquo": "\u201E", + "dagger": "\u2020", + "Dagger": "\u2021", + "permil": "\u2030", + "lsaquo": "\u2039", + "rsaquo": "\u203A", + "euro": "\u20AC", +} + +// HTMLAutoClose is the set of HTML elements that +// should be considered to close automatically. +// +// See the [Decoder.Strict] and [Decoder.Entity] fields' documentation. +var HTMLAutoClose []string = htmlAutoClose + +var htmlAutoClose = []string{ + /* + hget http://www.w3.org/TR/html4/loose.dtd | + 9 sed -n 's/<!ELEMENT ([^ ]*) +- O EMPTY.+/ "\1",/p' | tr A-Z a-z + */ + "basefont", + "br", + "area", + "link", + "img", + "param", + "hr", + "input", + "col", + "frame", + "isindex", + "base", + "meta", +} + +var ( + escQuot = []byte(""") // shorter than """ + escApos = []byte("'") // shorter than "'" + escAmp = []byte("&") + escLT = []byte("<") + escGT = []byte(">") + escTab = []byte("	") + escNL = []byte("
") + escCR = []byte("
") + escFFFD = []byte("\uFFFD") // Unicode replacement character +) + +// EscapeText writes to w the properly escaped XML equivalent +// of the plain text data s. +func EscapeText(w io.Writer, s []byte) error { + return escapeText(w, s, true) +} + +// escapeText writes to w the properly escaped XML equivalent +// of the plain text data s. If escapeNewline is true, newline +// characters will be escaped. +func escapeText(w io.Writer, s []byte, escapeNewline bool) error { + var esc []byte + last := 0 + for i := 0; i < len(s); { + r, width := utf8.DecodeRune(s[i:]) + i += width + switch r { + case '"': + esc = escQuot + case '\'': + esc = escApos + case '&': + esc = escAmp + case '<': + esc = escLT + case '>': + esc = escGT + case '\t': + esc = escTab + case '\n': + if !escapeNewline { + continue + } + esc = escNL + case '\r': + esc = escCR + default: + if !isInCharacterRange(r) || (r == 0xFFFD && width == 1) { + esc = escFFFD + break + } + continue + } + if _, err := w.Write(s[last : i-width]); err != nil { + return err + } + if _, err := w.Write(esc); err != nil { + return err + } + last = i + } + _, err := w.Write(s[last:]) + return err +} + +// EscapeString writes to p the properly escaped XML equivalent +// of the plain text data s. +func (p *printer) EscapeString(s string) { + var esc []byte + last := 0 + for i := 0; i < len(s); { + r, width := utf8.DecodeRuneInString(s[i:]) + i += width + switch r { + case '"': + esc = escQuot + case '\'': + esc = escApos + case '&': + esc = escAmp + case '<': + esc = escLT + case '>': + esc = escGT + case '\t': + esc = escTab + case '\n': + esc = escNL + case '\r': + esc = escCR + default: + if !isInCharacterRange(r) || (r == 0xFFFD && width == 1) { + esc = escFFFD + break + } + continue + } + p.WriteString(s[last : i-width]) + p.Write(esc) + last = i + } + p.WriteString(s[last:]) +} + +// Escape is like [EscapeText] but omits the error return value. +// It is provided for backwards compatibility with Go 1.0. +// Code targeting Go 1.1 or later should use [EscapeText]. +func Escape(w io.Writer, s []byte) { + EscapeText(w, s) +} + +var ( + cdataStart = []byte("<![CDATA[") + cdataEnd = []byte("]]>") + cdataEscape = []byte("]]]]><![CDATA[>") +) + +// emitCDATA writes to w the CDATA-wrapped plain text data s. +// It escapes CDATA directives nested in s. +func emitCDATA(w io.Writer, s []byte) error { + if len(s) == 0 { + return nil + } + if _, err := w.Write(cdataStart); err != nil { + return err + } + + for { + before, after, ok := bytes.Cut(s, cdataEnd) + if !ok { + break + } + // Found a nested CDATA directive end. + if _, err := w.Write(before); err != nil { + return err + } + if _, err := w.Write(cdataEscape); err != nil { + return err + } + s = after + } + + if _, err := w.Write(s); err != nil { + return err + } + + _, err := w.Write(cdataEnd) + return err +} + +// procInst parses the `param="..."` or `param='...'` +// value out of the provided string, returning "" if not found. +func procInst(param, s string) string { + // TODO: this parsing is somewhat lame and not exact. + // It works for all actual cases, though. + param = param + "=" + _, v, _ := strings.Cut(s, param) + if v == "" { + return "" + } + if v[0] != '\'' && v[0] != '"' { + return "" + } + unquote, _, ok := strings.Cut(v[1:], v[:1]) + if !ok { + return "" + } + return unquote +} diff --git a/contrib/go/_std_1.22/src/encoding/xml/ya.make b/contrib/go/_std_1.22/src/encoding/xml/ya.make new file mode 100644 index 0000000000..03d6e0045f --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/xml/ya.make @@ -0,0 +1,10 @@ +GO_LIBRARY() +IF (TRUE) + SRCS( + marshal.go + read.go + typeinfo.go + xml.go + ) +ENDIF() +END() diff --git a/contrib/go/_std_1.22/src/encoding/ya.make b/contrib/go/_std_1.22/src/encoding/ya.make new file mode 100644 index 0000000000..8e021ee14c --- /dev/null +++ b/contrib/go/_std_1.22/src/encoding/ya.make @@ -0,0 +1,7 @@ +GO_LIBRARY() +IF (TRUE) + SRCS( + encoding.go + ) +ENDIF() +END() |