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
|
/*
* Copyright (c) 2020, Emanuel Sprung <emanuel.sprung@gmail.com>
*
* SPDX-License-Identifier: BSD-2-Clause
*/
#pragma once
#include "RegexByteCode.h"
#include "RegexError.h"
#include "RegexLexer.h"
#include "RegexOptions.h"
#include <AK/Forward.h>
#include <AK/StringBuilder.h>
#include <AK/Types.h>
#include <AK/Vector.h>
namespace regex {
class PosixExtendedParser;
class ECMA262Parser;
template<typename T>
struct GenericParserTraits {
using OptionsType = T;
};
template<typename T>
struct ParserTraits : public GenericParserTraits<T> {
};
template<>
struct ParserTraits<PosixExtendedParser> : public GenericParserTraits<PosixOptions> {
};
template<>
struct ParserTraits<ECMA262Parser> : public GenericParserTraits<ECMAScriptOptions> {
};
class Parser {
public:
struct Result {
ByteCode bytecode;
size_t capture_groups_count;
size_t named_capture_groups_count;
size_t match_length_minimum;
Error error;
Token error_token;
};
explicit Parser(Lexer& lexer)
: m_parser_state(lexer)
{
}
Parser(Lexer& lexer, AllOptions regex_options)
: m_parser_state(lexer, regex_options)
{
}
virtual ~Parser() = default;
Result parse(Optional<AllOptions> regex_options = {});
bool has_error() const { return m_parser_state.error != Error::NoError; }
Error error() const { return m_parser_state.error; }
protected:
virtual bool parse_internal(ByteCode&, size_t& match_length_minimum) = 0;
ALWAYS_INLINE bool match(TokenType type) const;
ALWAYS_INLINE bool match(char ch) const;
ALWAYS_INLINE bool match_ordinary_characters();
ALWAYS_INLINE Token consume();
ALWAYS_INLINE Token consume(TokenType type, Error error);
ALWAYS_INLINE bool consume(const String&);
ALWAYS_INLINE bool try_skip(StringView);
ALWAYS_INLINE bool lookahead_any(StringView);
ALWAYS_INLINE char skip();
ALWAYS_INLINE void back(size_t = 1);
ALWAYS_INLINE void reset();
ALWAYS_INLINE bool done() const;
ALWAYS_INLINE bool set_error(Error error);
struct ParserState {
Lexer& lexer;
Token current_token;
Error error = Error::NoError;
Token error_token { TokenType::Eof, 0, StringView(nullptr) };
ByteCode bytecode;
size_t capture_groups_count { 0 };
size_t named_capture_groups_count { 0 };
size_t match_length_minimum { 0 };
AllOptions regex_options;
HashMap<int, size_t> capture_group_minimum_lengths;
HashMap<FlyString, size_t> named_capture_group_minimum_lengths;
HashMap<size_t, FlyString> named_capture_groups;
explicit ParserState(Lexer& lexer)
: lexer(lexer)
, current_token(lexer.next())
{
}
explicit ParserState(Lexer& lexer, AllOptions regex_options)
: lexer(lexer)
, current_token(lexer.next())
, regex_options(regex_options)
{
}
};
ParserState m_parser_state;
};
class PosixExtendedParser final : public Parser {
public:
explicit PosixExtendedParser(Lexer& lexer)
: Parser(lexer)
{
}
PosixExtendedParser(Lexer& lexer, Optional<typename ParserTraits<PosixExtendedParser>::OptionsType> regex_options)
: Parser(lexer, regex_options.value_or({}))
{
}
~PosixExtendedParser() = default;
private:
ALWAYS_INLINE bool match_repetition_symbol();
bool parse_internal(ByteCode&, size_t&) override;
bool parse_root(ByteCode&, size_t&);
ALWAYS_INLINE bool parse_sub_expression(ByteCode&, size_t&);
ALWAYS_INLINE bool parse_bracket_expression(ByteCode&, size_t&);
ALWAYS_INLINE bool parse_repetition_symbol(ByteCode&, size_t&);
};
class ECMA262Parser final : public Parser {
public:
explicit ECMA262Parser(Lexer& lexer)
: Parser(lexer)
{
}
ECMA262Parser(Lexer& lexer, Optional<typename ParserTraits<ECMA262Parser>::OptionsType> regex_options)
: Parser(lexer, regex_options.value_or({}))
{
m_should_use_browser_extended_grammar = regex_options.has_value() && regex_options->has_flag_set(ECMAScriptFlags::BrowserExtended);
}
~ECMA262Parser() = default;
private:
bool parse_internal(ByteCode&, size_t&) override;
enum class ReadDigitsInitialZeroState {
Allow,
Disallow,
};
StringView read_digits_as_string(ReadDigitsInitialZeroState initial_zero = ReadDigitsInitialZeroState::Allow, bool hex = false, int max_count = -1);
Optional<unsigned> read_digits(ReadDigitsInitialZeroState initial_zero = ReadDigitsInitialZeroState::Allow, bool hex = false, int max_count = -1);
StringView read_capture_group_specifier(bool take_starting_angle_bracket = false);
bool parse_pattern(ByteCode&, size_t&, bool unicode, bool named);
bool parse_disjunction(ByteCode&, size_t&, bool unicode, bool named);
bool parse_alternative(ByteCode&, size_t&, bool unicode, bool named);
bool parse_term(ByteCode&, size_t&, bool unicode, bool named);
bool parse_assertion(ByteCode&, size_t&, bool unicode, bool named);
bool parse_atom(ByteCode&, size_t&, bool unicode, bool named);
bool parse_quantifier(ByteCode&, size_t&, bool unicode, bool named);
bool parse_atom_escape(ByteCode&, size_t&, bool unicode, bool named);
bool parse_character_class(ByteCode&, size_t&, bool unicode, bool named);
bool parse_capture_group(ByteCode&, size_t&, bool unicode, bool named);
Optional<CharClass> parse_character_class_escape(bool& out_inverse, bool expect_backslash = false);
bool parse_nonempty_class_ranges(Vector<CompareTypeAndValuePair>&, bool unicode);
// Used only by B.1.4, Regular Expression Patterns (Extended for use in browsers)
bool parse_quantifiable_assertion(ByteCode&, size_t&, bool named);
bool parse_extended_atom(ByteCode&, size_t&, bool named);
bool parse_inner_disjunction(ByteCode& bytecode_stack, size_t& length, bool unicode, bool named);
bool parse_invalid_braced_quantifier(); // Note: This function either parses and *fails*, or doesn't parse anything and returns false.
bool parse_legacy_octal_escape_sequence(ByteCode& bytecode_stack, size_t& length);
Optional<u8> parse_legacy_octal_escape();
size_t ensure_total_number_of_capturing_parenthesis();
// ECMA-262's flavour of regex is a bit weird in that it allows backrefs to reference "future" captures, and such backrefs
// always match the empty string. So we have to know how many capturing parenthesis there are, but we don't want to always
// parse it twice, so we'll just do so when it's actually needed.
// Most patterns should have no need to ever populate this field.
Optional<size_t> m_total_number_of_capturing_parenthesis;
// Keep the Annex B. behaviour behind a flag, the users can enable it by passing the `ECMAScriptFlags::BrowserExtended` flag.
bool m_should_use_browser_extended_grammar { false };
};
using PosixExtended = PosixExtendedParser;
using ECMA262 = ECMA262Parser;
}
using regex::ECMA262;
using regex::PosixExtended;
|
