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
|
// Copyright 2020 The Chromium Authors. All rights reserved.
// Copyright 2014 Blake Embrey (hello@blakeembrey.com)
// Use of this source code is governed by an MIT-style license that can be
// found in the LICENSE file or at https://opensource.org/licenses/MIT.
#ifndef THIRD_PARTY_LIBURLPATTERN_LEXER_H_
#define THIRD_PARTY_LIBURLPATTERN_LEXER_H_
#include <vector>
#include "base/component_export.h"
#include "third_party/abseil-cpp/absl/status/statusor.h"
#include "third_party/abseil-cpp/absl/strings/string_view.h"
namespace liburlpattern {
enum class TokenType {
// Open a scope with a '{'.
kOpen,
// Close a scope with a '}'.
kClose,
// A regular expression group like '(...)'.
kRegex,
// A named group like ':foo'.
kName,
// A single character.
kChar,
// The '\' escape character.
kEscapedChar,
// A '+' or '?' modifier.
kOtherModifier,
// A '*' character which can be a wildcard or modifier.
kAsterisk,
// The end of the token stream.
kEnd,
// A character that is not valid in a properly formed pattern; e.g. the colon
// in `https://`. This is only generated when TokenizerPolicy::kLenient is
// used.
kInvalidChar,
};
const char* TokenTypeToString(TokenType type);
// Simple structure representing a single lexical token.
struct COMPONENT_EXPORT(LIBURLPATTERN) Token {
// Indicate the token type.
TokenType type = TokenType::kEnd;
// Index of the start of this token in the original pattern string.
size_t index = 0;
// The value of the token. May be one or many characters depending on type.
// May be null zero characters for the kEnd type.
absl::string_view value;
Token(TokenType t, size_t i, absl::string_view v)
: type(t), index(i), value(v) {}
Token() = default;
};
enum class TokenizePolicy {
// The strict policy causes any problems found during tokenization to be
// thrown as errors.
kStrict,
// The lenient policy converts problems detected during tokenization into
// kInvalidChar tokens in the returned token list. For something like a
// `\` at the end of the string, this simply returns the immediate `\`
// character. For validation errors that cause a group to be invalid, the
// first character of the group is instead returned. For example, `https://`
// returns the `:` as a kInvalidChar. For `(foo(bar))` where capture groups
// are illegal it causes the first `(` to be returned as a kInvalidChar.
// Tokenization then continues with the next character after the kInvalidChar.
kLenient,
};
COMPONENT_EXPORT(LIBURLPATTERN)
inline bool operator==(const Token& lh, const Token& rh) {
return lh.type == rh.type && lh.index == rh.index && lh.value == rh.value;
}
inline bool operator!=(const Token& lh, const Token& rh) {
return !(lh == rh);
}
COMPONENT_EXPORT(LIBURLPATTERN)
std::ostream& operator<<(std::ostream& o, Token token);
// Split the given input pattern string into a list of lexical tokens. Note,
// the generated Token objects simply reference positions within the input
// |pattern|. The |pattern| must be kept alive as long as the Token objects.
COMPONENT_EXPORT(LIBURLPATTERN)
absl::StatusOr<std::vector<Token>> Tokenize(
absl::string_view pattern,
TokenizePolicy policy = TokenizePolicy::kStrict);
} // namespace liburlpattern
#endif // THIRD_PARTY_LIBURLPATTERN_LEXER_H_
|
