1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
|
#include <Columns/ColumnString.h>
#include <Functions/FunctionFactory.h>
#include <Functions/FunctionStringToString.h>
#include <Common/StringUtils/StringUtils.h>
#include <base/hex.h>
#include <base/find_symbols.h>
namespace DB
{
namespace ErrorCodes
{
extern const int ILLEGAL_TYPE_OF_ARGUMENT;
}
namespace
{
struct DecodeXMLComponentName
{
static constexpr auto name = "decodeXMLComponent";
};
class FunctionDecodeXMLComponentImpl
{
public:
static void vector(
const ColumnString::Chars & data,
const ColumnString::Offsets & offsets,
ColumnString::Chars & res_data,
ColumnString::Offsets & res_offsets)
{
/// The size of result is always not more than the size of source.
/// Because entities decodes to the shorter byte sequence.
/// Example: &#xx... &#xx... will decode to UTF-8 byte sequence not longer than 4 bytes.
res_data.resize(data.size());
size_t size = offsets.size();
res_offsets.resize(size);
size_t prev_offset = 0;
size_t res_offset = 0;
for (size_t i = 0; i < size; ++i)
{
const char * src_data = reinterpret_cast<const char *>(&data[prev_offset]);
size_t src_size = offsets[i] - prev_offset;
size_t dst_size = execute(src_data, src_size, reinterpret_cast<char *>(res_data.data() + res_offset));
res_offset += dst_size;
res_offsets[i] = res_offset;
prev_offset = offsets[i];
}
res_data.resize(res_offset);
}
[[noreturn]] static void vectorFixed(const ColumnString::Chars &, size_t, ColumnString::Chars &)
{
throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Function decodeXMLComponent cannot work with FixedString argument");
}
private:
static const int max_legal_unicode_value = 0x10FFFF;
static const int max_decimal_length_of_unicode_point = 7; /// 1114111
static size_t execute(const char * src, size_t src_size, char * dst)
{
const char * src_pos = src;
const char * src_end = src + src_size;
char * dst_pos = dst;
while (true)
{
const char * entity_pos = find_first_symbols<'&'>(src_pos, src_end);
if (entity_pos + strlen("lt;") >= src_end)
break;
/// Copy text between entities.
size_t bytes_to_copy = entity_pos - src_pos;
memcpySmallAllowReadWriteOverflow15(dst_pos, src_pos, bytes_to_copy);
dst_pos += bytes_to_copy;
src_pos = entity_pos;
++entity_pos;
const char * entity_end = find_first_symbols<';'>(entity_pos, src_end);
if (entity_end == src_end)
break;
bool parsed = false;
/// &#NNNN; or &#xNNNN;
uint32_t code_point = 0;
if (isValidNumericEntity(entity_pos, entity_end, code_point))
{
codePointToUTF8(code_point, dst_pos);
parsed = true;
}
else if (entity_end - entity_pos == 2)
{
if (memcmp(entity_pos, "lt", 2) == 0)
{
*dst_pos = '<';
++dst_pos;
parsed = true;
}
else if (memcmp(entity_pos, "gt", 2) == 0)
{
*dst_pos = '>';
++dst_pos;
parsed = true;
}
}
else if (entity_end - entity_pos == 3)
{
if (memcmp(entity_pos, "amp", 3) == 0)
{
*dst_pos = '&';
++dst_pos;
parsed = true;
}
}
else if (entity_end - entity_pos == 4)
{
if (memcmp(entity_pos, "quot", 4) == 0)
{
*dst_pos = '"';
++dst_pos;
parsed = true;
}
else if (memcmp(entity_pos, "apos", 4) == 0)
{
*dst_pos = '\'';
++dst_pos;
parsed = true;
}
}
if (parsed)
{
/// Skip the parsed entity.
src_pos = entity_end + 1;
}
else
{
/// Copy one byte as is and skip it.
*dst_pos = *src_pos;
++dst_pos;
++src_pos;
}
}
/// Copy the rest of the string.
if (src_pos < src_end)
{
size_t bytes_to_copy = src_end - src_pos;
memcpySmallAllowReadWriteOverflow15(dst_pos, src_pos, bytes_to_copy);
dst_pos += bytes_to_copy;
}
return dst_pos - dst;
}
static void codePointToUTF8(uint32_t code_point, char *& dst_pos)
{
if (code_point < (1 << 7))
{
dst_pos[0] = (code_point & 0x7F);
++dst_pos;
}
else if (code_point < (1 << 11))
{
dst_pos[0] = ((code_point >> 6) & 0x1F) + 0xC0;
dst_pos[1] = (code_point & 0x3F) + 0x80;
dst_pos += 2;
}
else if (code_point < (1 << 16))
{
dst_pos[0] = ((code_point >> 12) & 0x0F) + 0xE0;
dst_pos[1] = ((code_point >> 6) & 0x3F) + 0x80;
dst_pos[2] = (code_point & 0x3F) + 0x80;
dst_pos += 3;
}
else
{
dst_pos[0] = ((code_point >> 18) & 0x07) + 0xF0;
dst_pos[1] = ((code_point >> 12) & 0x3F) + 0x80;
dst_pos[2] = ((code_point >> 6) & 0x3F) + 0x80;
dst_pos[3] = (code_point & 0x3F) + 0x80;
dst_pos += 4;
}
}
static bool isValidNumericEntity(const char * src, const char * end, uint32_t & code_point)
{
if (src + strlen("#") >= end)
return false;
if (src[0] != '#' || (end - src > 1 + max_decimal_length_of_unicode_point))
return false;
if (src + 2 < end && (src[1] == 'x' || src[1] == 'X'))
{
src += 2;
for (; src < end; ++src)
{
if (!isHexDigit(*src))
return false;
code_point *= 16;
code_point += unhex(*src);
}
}
else
{
src += 1;
for (; src < end; ++src)
{
if (!isNumericASCII(*src))
return false;
code_point *= 10;
code_point += *src - '0';
}
}
return code_point <= max_legal_unicode_value;
}
};
using FunctionDecodeXMLComponent = FunctionStringToString<FunctionDecodeXMLComponentImpl, DecodeXMLComponentName>;
}
REGISTER_FUNCTION(DecodeXMLComponent)
{
factory.registerFunction<FunctionDecodeXMLComponent>();
}
}
|
