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
|
#include "generic.h"
#include <contrib/libs/re2/re2/re2.h>
#include <util/string/builder.h>
namespace NSQLTranslationV1 {
namespace {
TMaybe<TStringBuf> Match(TStringBuf prefix, const RE2& regex) {
re2::StringPiece input(prefix.data(), prefix.size());
if (RE2::Consume(&input, regex)) {
return TStringBuf(prefix.data(), input.data());
}
return Nothing();
}
} // namespace
class TGenericLexer: public IGenericLexer {
private:
static constexpr TStringBuf Utf8BOM = "\xEF\xBB\xBF";
public:
explicit TGenericLexer(TGenericLexerGrammar grammar)
: Grammar_(std::move(grammar))
{
}
virtual bool Tokenize(
TStringBuf text,
const TTokenCallback& onNext,
size_t maxErrors) const override {
Y_ENSURE(0 < maxErrors);
size_t errors = 0;
size_t pos = 0;
if (text.StartsWith(Utf8BOM)) {
pos += Utf8BOM.size();
}
while (pos < text.size() && errors < maxErrors) {
TGenericToken matched = Match(TStringBuf(text, pos));
matched.Begin = pos;
pos += matched.Content.size();
if (matched.Name == TGenericToken::Error) {
errors += 1;
}
onNext(std::move(matched));
}
if (errors == maxErrors) {
return false;
}
onNext(TGenericToken{
.Name = "EOF",
.Content = "<EOF>",
.Begin = pos,
});
return errors == 0;
}
private:
TGenericToken Match(TStringBuf prefix) const {
TMaybe<TGenericToken> max;
Match(prefix, [&](TGenericToken&& token) {
if (max.Empty() || max->Content.size() < token.Content.size()) {
max = std::move(token);
}
});
if (max) {
return *max;
}
return {
.Name = TGenericToken::Error,
.Content = prefix.substr(0, 1),
};
}
void Match(TStringBuf prefix, auto onMatch) const {
for (const auto& matcher : Grammar_) {
if (auto token = matcher(prefix)) {
onMatch(std::move(*token));
}
}
}
TGenericLexerGrammar Grammar_;
};
TTokenMatcher Compile(TString name, const TRegexPattern& regex) {
RE2::Options options;
options.set_case_sensitive(!regex.IsCaseInsensitive);
return [bodyRe = MakeAtomicShared<RE2>(regex.Body, options),
afterRe = MakeAtomicShared<RE2>(regex.After, options),
name = std::move(name)](TStringBuf prefix) -> TMaybe<TGenericToken> {
TMaybe<TStringBuf> body, after;
if ((body = Match(prefix, *bodyRe)) &&
(after = Match(prefix.Tail(body->size()), *afterRe))) {
return TGenericToken{
.Name = name,
.Content = *body,
};
}
return Nothing();
};
}
TRegexPattern Merged(TVector<TRegexPattern> patterns) {
Y_ENSURE(!patterns.empty());
const TRegexPattern& sample = patterns.back();
Y_ENSURE(AllOf(patterns, [&](const TRegexPattern& pattern) {
return std::tie(pattern.After, pattern.IsCaseInsensitive) ==
std::tie(sample.After, sample.IsCaseInsensitive);
}));
Sort(patterns, [](const TRegexPattern& lhs, const TRegexPattern& rhs) {
return lhs.Body.length() > rhs.Body.length();
});
TStringBuilder body;
for (const auto& pattern : patterns) {
body << "(" << pattern.Body << ")|";
}
Y_ENSURE(body.back() == '|');
body.pop_back();
return TRegexPattern{
.Body = std::move(body),
.After = sample.After,
.IsCaseInsensitive = sample.IsCaseInsensitive,
};
}
IGenericLexer::TPtr MakeGenericLexer(TGenericLexerGrammar grammar) {
return IGenericLexer::TPtr(new TGenericLexer(std::move(grammar)));
}
TVector<TGenericToken> Tokenize(IGenericLexer::TPtr& lexer, TStringBuf text) {
TVector<TGenericToken> tokens;
lexer->Tokenize(text, [&](TGenericToken&& token) {
tokens.emplace_back(std::move(token));
});
return tokens;
}
} // namespace NSQLTranslationV1
|
