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
|
//! Character references are a construct that occurs in the [string][] and
//! [text][] content types.
//!
//! They’re formed with the following BNF:
//!
//! ```bnf
//! character_reference ::= '&' (numeric | named) ';'
//!
//! numeric ::= '#' (hexadecimal | decimal)
//! ; Note: Limit of `6` imposed as all bigger numbers are invalid:
//! hexadecimal ::= ('x' | 'X') 1*6(ascii_hexdigit)
//! ; Note: Limit of `7` imposed as all bigger numbers are invalid:
//! decimal ::= 1*7(ascii_digit)
//! ; Note: Limit of `31` imposed by `CounterClockwiseContourIntegral`:
//! ; Note: Limited to any known named character reference (see `constants.rs`)
//! named ::= 1*31(ascii_alphanumeric)
//! ```
//!
//! Like much of markdown, there are no “invalid” character references.
//! However, for security reasons, several numeric character references parse
//! fine but are not rendered as their corresponding character and they are
//! instead replaced by a U+FFFD REPLACEMENT CHARACTER (`�`).
//! See [`decode_numeric`][decode_numeric] for more info.
//!
//! To escape ASCII punctuation characters, use the terser
//! [character escape][character_escape] construct instead (as in, `\&`).
//!
//! Character references in markdown are not the same as character references
//! in HTML.
//! Notably, HTML allows several character references without a closing
//! semicolon.
//! See [*§ 13.2.5.72 Character reference state* in the HTML spec][html] for more info.
//!
//! Character references are parsed insensitive to casing.
//! The casing of hexadecimal numeric character references has no effect.
//! The casing of named character references does not matter when parsing them,
//! but does affect whether they match.
//! Depending on the name, one or more cases are allowed, such as that `AMP`
//! and `amp` are both allowed but other cases are not.
//! See [`CHARACTER_REFERENCES`][character_references] for which
//! names match.
//!
//! ## Tokens
//!
//! * [`CharacterReference`][Token::CharacterReference]
//! * [`CharacterReferenceMarker`][Token::CharacterReferenceMarker]
//! * [`CharacterReferenceMarkerHexadecimal`][Token::CharacterReferenceMarkerHexadecimal]
//! * [`CharacterReferenceMarkerNumeric`][Token::CharacterReferenceMarkerNumeric]
//! * [`CharacterReferenceMarkerSemi`][Token::CharacterReferenceMarkerSemi]
//! * [`CharacterReferenceValue`][Token::CharacterReferenceValue]
//!
//! ## References
//!
//! * [`character-reference.js` in `micromark`](https://github.com/micromark/micromark/blob/main/packages/micromark-core-commonmark/dev/lib/character-reference.js)
//! * [*§ 2.5 Entity and numeric character references* in `CommonMark`](https://spec.commonmark.org/0.30/#entity-and-numeric-character-references)
//!
//! [string]: crate::content::string
//! [text]: crate::content::text
//! [character_escape]: crate::construct::character_reference
//! [decode_numeric]: crate::util::decode_character_reference::decode_numeric
//! [character_references]: crate::constant::CHARACTER_REFERENCES
//! [html]: https://html.spec.whatwg.org/multipage/parsing.html#character-reference-state
use crate::constant::{
CHARACTER_REFERENCES, CHARACTER_REFERENCE_DECIMAL_SIZE_MAX,
CHARACTER_REFERENCE_HEXADECIMAL_SIZE_MAX, CHARACTER_REFERENCE_NAMED_SIZE_MAX,
};
use crate::token::Token;
use crate::tokenizer::{State, StateName, Tokenizer};
use crate::util::slice::Slice;
/// Start of a character reference.
///
/// ```markdown
/// > | a&b
/// ^
/// > | a{b
/// ^
/// > | a	b
/// ^
/// ```
pub fn start(tokenizer: &mut Tokenizer) -> State {
match tokenizer.current {
Some(b'&') if tokenizer.parse_state.constructs.character_reference => {
tokenizer.enter(Token::CharacterReference);
tokenizer.enter(Token::CharacterReferenceMarker);
tokenizer.consume();
tokenizer.exit(Token::CharacterReferenceMarker);
State::Fn(StateName::CharacterReferenceOpen)
}
_ => State::Nok,
}
}
/// Inside a character reference, after `&`, before `#` for numeric references
/// or an alphanumeric for named references.
///
/// ```markdown
/// > | a&b
/// ^
/// > | a{b
/// ^
/// > | a	b
/// ^
/// ```
// StateName::CharacterReferenceOpen
pub fn open(tokenizer: &mut Tokenizer) -> State {
if let Some(b'#') = tokenizer.current {
tokenizer.enter(Token::CharacterReferenceMarkerNumeric);
tokenizer.consume();
tokenizer.exit(Token::CharacterReferenceMarkerNumeric);
State::Fn(StateName::CharacterReferenceNumeric)
} else {
tokenizer.tokenize_state.marker = b'&';
tokenizer.enter(Token::CharacterReferenceValue);
value(tokenizer)
}
}
/// Inside a numeric character reference, right before `x` for hexadecimals,
/// or a digit for decimals.
///
/// ```markdown
/// > | a{b
/// ^
/// > | a	b
/// ^
/// ```
// StateName::CharacterReferenceNumeric
pub fn numeric(tokenizer: &mut Tokenizer) -> State {
if let Some(b'x' | b'X') = tokenizer.current {
tokenizer.enter(Token::CharacterReferenceMarkerHexadecimal);
tokenizer.consume();
tokenizer.exit(Token::CharacterReferenceMarkerHexadecimal);
tokenizer.enter(Token::CharacterReferenceValue);
tokenizer.tokenize_state.marker = b'x';
State::Fn(StateName::CharacterReferenceValue)
} else {
tokenizer.enter(Token::CharacterReferenceValue);
tokenizer.tokenize_state.marker = b'#';
value(tokenizer)
}
}
/// Inside a character reference value, after the markers (`&#x`, `&#`, or
/// `&`) that define its kind, but before the `;`.
///
/// The character reference kind defines what and how many characters are
/// allowed.
///
/// ```markdown
/// > | a&b
/// ^^^
/// > | a{b
/// ^^^
/// > | a	b
/// ^
/// ```
pub fn value(tokenizer: &mut Tokenizer) -> State {
if matches!(tokenizer.current, Some(b';')) && tokenizer.tokenize_state.size > 0 {
// Named.
if tokenizer.tokenize_state.marker == b'&' {
// Guaranteed to be valid ASCII bytes.
let slice = Slice::from_indices(
tokenizer.parse_state.bytes,
tokenizer.point.index - tokenizer.tokenize_state.size,
tokenizer.point.index,
);
let name = slice.as_str();
if !CHARACTER_REFERENCES.iter().any(|d| d.0 == name) {
tokenizer.tokenize_state.marker = 0;
tokenizer.tokenize_state.size = 0;
return State::Nok;
}
}
tokenizer.exit(Token::CharacterReferenceValue);
tokenizer.enter(Token::CharacterReferenceMarkerSemi);
tokenizer.consume();
tokenizer.exit(Token::CharacterReferenceMarkerSemi);
tokenizer.exit(Token::CharacterReference);
tokenizer.tokenize_state.marker = 0;
tokenizer.tokenize_state.size = 0;
return State::Ok;
}
let max = match tokenizer.tokenize_state.marker {
b'&' => CHARACTER_REFERENCE_NAMED_SIZE_MAX,
b'x' => CHARACTER_REFERENCE_HEXADECIMAL_SIZE_MAX,
b'#' => CHARACTER_REFERENCE_DECIMAL_SIZE_MAX,
_ => unreachable!("Unexpected marker `{}`", tokenizer.tokenize_state.marker),
};
let test = match tokenizer.tokenize_state.marker {
b'&' => u8::is_ascii_alphanumeric,
b'x' => u8::is_ascii_hexdigit,
b'#' => u8::is_ascii_digit,
_ => unreachable!("Unexpected marker `{}`", tokenizer.tokenize_state.marker),
};
if let Some(byte) = tokenizer.current {
if tokenizer.tokenize_state.size < max && test(&byte) {
tokenizer.tokenize_state.size += 1;
tokenizer.consume();
return State::Fn(StateName::CharacterReferenceValue);
}
}
tokenizer.tokenize_state.marker = 0;
tokenizer.tokenize_state.size = 0;
State::Nok
}
|
