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
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
|
//! Heading (setext) is a construct that occurs in the [flow] content type.
//!
//! They’re formed with the following BNF:
//!
//! ```bnf
//! heading_setext ::= line *(eol line) eol whitespace_optional (1*'-' | 1*'=') whitespace_optional
//!
//! whitespace ::= 1*space_or_tab
//! whitespace_optional ::= [ whitespace ]
//! line ::= code - eol
//! eol ::= '\r' | '\r\n' | '\n'
//! ```
//!
//! Heading (setext) in markdown relates to the `<h1>` and `<h2>` elements in
//! HTML.
//! See [*§ 4.3.6 The `h1`, `h2`, `h3`, `h4`, `h5`, and `h6` elements* in the
//! HTML spec][html] for more info.
//!
//! In markdown, it is also possible to create headings with a
//! [heading (atx)][heading_atx] construct.
//! The benefit of setext headings is that their text can include line endings,
//! and by extensions also hard breaks (e.g., with
//! [hard break (escape)][hard_break_escape]).
//! However, their limit is that they cannot form `<h3>` through `<h6>`
//! headings.
//! Due to this limitation, it is recommended to use atx headings.
//!
//! [Thematic breaks][thematic_break] formed with dashes (without whitespace)
//! can also form heading (setext).
//!
//! > 🏛 **Background**: the word *setext* originates from a small markup
//! > language by Ian Feldman from 1991.
//! > See [*§ Setext* on Wikipedia][wiki-setext] for more info.
//! > The word *atx* originates from a tiny markup language by Aaron Swartz
//! > from 2002.
//! > See [*§ atx, the true structured text format* on `aaronsw.com`][atx] for
//! > more info.
//!
//! ## References
//!
//! * [`setext-underline.js` in `micromark`](https://github.com/micromark/micromark/blob/main/packages/micromark-core-commonmark/dev/lib/setext-underline.js)
//! * [*§ 4.3 Setext headings* in `CommonMark`](https://spec.commonmark.org/0.30/#setext-headings)
//!
//! [flow]: crate::content::flow
//! [heading_atx]: crate::construct::heading_atx
//! [thematic_break]: crate::construct::thematic_break
//! [hard_break_escape]: crate::construct::hard_break_escape
//! [html]: https://html.spec.whatwg.org/multipage/sections.html#the-h1,-h2,-h3,-h4,-h5,-and-h6-elements
//! [wiki-setext]: https://en.wikipedia.org/wiki/Setext
//! [atx]: http://www.aaronsw.com/2002/atx/
use crate::constant::TAB_SIZE;
use crate::construct::partial_space_or_tab::space_or_tab_opt;
use crate::tokenizer::{Code, State, StateFnResult, TokenType, Tokenizer};
use crate::util::span::from_exit_event;
/// Kind of underline.
#[derive(Debug, Clone, PartialEq)]
pub enum Kind {
/// Grave accent (tick) code.
Dash,
/// Tilde code.
EqualsTo,
}
/// Start of a heading (setext).
///
/// ```markdown
/// |alpha
/// ==
/// ```
pub fn start(tokenizer: &mut Tokenizer, code: Code) -> StateFnResult {
tokenizer.enter(TokenType::HeadingSetext);
tokenizer.go(space_or_tab_opt(), before)(tokenizer, code)
}
/// Start of a heading (setext), after whitespace.
///
/// ```markdown
/// |alpha
/// ==
/// ```
pub fn before(tokenizer: &mut Tokenizer, code: Code) -> StateFnResult {
match code {
Code::None | Code::CarriageReturnLineFeed | Code::Char('\n' | '\r') => {
unreachable!("expected non-eol/eof");
}
_ => {
tokenizer.enter(TokenType::HeadingSetextText);
tokenizer.enter(TokenType::ChunkText);
text_inside(tokenizer, code)
}
}
}
/// Inside text.
///
/// ```markdown
/// al|pha
/// bra|vo
/// ==
/// ```
fn text_inside(tokenizer: &mut Tokenizer, code: Code) -> StateFnResult {
match code {
Code::None => (State::Nok, None),
Code::CarriageReturnLineFeed | Code::Char('\n' | '\r') => {
tokenizer.exit(TokenType::ChunkText);
tokenizer.exit(TokenType::HeadingSetextText);
tokenizer.attempt(underline_before, |ok| {
Box::new(if ok { after } else { text_continue })
})(tokenizer, code)
}
_ => {
tokenizer.consume(code);
(State::Fn(Box::new(text_inside)), None)
}
}
}
/// At a line ending, not at an underline.
///
/// ```markdown
/// alpha
/// |bravo
/// ==
/// ```
fn text_continue(tokenizer: &mut Tokenizer, code: Code) -> StateFnResult {
// Needed to connect the text.
// To do: does it work?
tokenizer.enter(TokenType::HeadingSetextText);
tokenizer.events.pop();
tokenizer.events.pop();
match code {
Code::CarriageReturnLineFeed | Code::Char('\n' | '\r') => {
let next = tokenizer.events.len();
let previous = next - 2;
tokenizer.enter(TokenType::LineEnding);
tokenizer.consume(code);
tokenizer.exit(TokenType::LineEnding);
tokenizer.events[previous].next = Some(next);
tokenizer.events[next].previous = Some(previous);
(
State::Fn(Box::new(tokenizer.go(space_or_tab_opt(), text_line_start))),
None,
)
}
_ => unreachable!("expected eol"),
}
}
/// At a line ending after whitespace, not at an underline.
///
/// ```markdown
/// alpha
/// |bravo
/// ==
/// ```
fn text_line_start(tokenizer: &mut Tokenizer, code: Code) -> StateFnResult {
let next = tokenizer.events.len() - 2;
let previous = next - 2;
// Link the whitespace, if it exists.
if tokenizer.events[next].token_type == TokenType::Whitespace {
tokenizer.events[previous].next = Some(next);
tokenizer.events[next].previous = Some(previous);
}
match code {
// Blank lines not allowed.
Code::None | Code::CarriageReturnLineFeed | Code::Char('\n' | '\r') => (State::Nok, None),
_ => {
let next = tokenizer.events.len();
let previous = next - 2;
tokenizer.enter(TokenType::ChunkText);
tokenizer.events[previous].next = Some(next);
tokenizer.events[next].previous = Some(previous);
text_inside(tokenizer, code)
}
}
}
/// After a heading (setext).
///
/// ```markdown
/// alpha
/// ==|
/// ```
fn after(tokenizer: &mut Tokenizer, code: Code) -> StateFnResult {
tokenizer.exit(TokenType::HeadingSetext);
(State::Ok, Some(vec![code]))
}
/// At a line ending, presumably an underline.
///
/// ```markdown
/// alpha|
/// ==
/// ```
fn underline_before(tokenizer: &mut Tokenizer, code: Code) -> StateFnResult {
match code {
Code::CarriageReturnLineFeed | Code::Char('\n' | '\r') => {
tokenizer.enter(TokenType::LineEnding);
tokenizer.consume(code);
tokenizer.exit(TokenType::LineEnding);
(
State::Fn(Box::new(
tokenizer.go(space_or_tab_opt(), underline_sequence_start),
)),
None,
)
}
_ => unreachable!("expected eol"),
}
}
/// After optional whitespace, presumably an underline.
///
/// ```markdown
/// alpha
/// |==
/// ```
fn underline_sequence_start(tokenizer: &mut Tokenizer, code: Code) -> StateFnResult {
let tail = tokenizer.events.last();
let mut prefix = 0;
if let Some(event) = tail {
if event.token_type == TokenType::Whitespace {
let span = from_exit_event(&tokenizer.events, tokenizer.events.len() - 1);
prefix = span.end_index - span.start_index;
}
}
// To do: 4+ should be okay if code (indented) is turned off!
if prefix >= TAB_SIZE {
return (State::Nok, None);
}
match code {
Code::Char(char) if char == '-' || char == '=' => {
let marker = if char == '-' {
Kind::Dash
} else {
Kind::EqualsTo
};
tokenizer.enter(TokenType::HeadingSetextUnderline);
underline_sequence_inside(tokenizer, code, marker)
}
_ => (State::Nok, None),
}
}
/// In an underline sequence.
///
/// ```markdown
/// alpha
/// =|=
/// ```
fn underline_sequence_inside(tokenizer: &mut Tokenizer, code: Code, kind: Kind) -> StateFnResult {
let marker = if kind == Kind::Dash { '-' } else { '=' };
match code {
Code::Char(char) if char == marker => {
tokenizer.consume(code);
(
State::Fn(Box::new(move |tokenizer, code| {
underline_sequence_inside(tokenizer, code, kind)
})),
None,
)
}
_ => tokenizer.go(space_or_tab_opt(), underline_after)(tokenizer, code),
}
}
/// After an underline sequence, after optional whitespace.
///
/// ```markdown
/// alpha
/// ==|
/// ```
fn underline_after(tokenizer: &mut Tokenizer, code: Code) -> StateFnResult {
match code {
Code::None | Code::CarriageReturnLineFeed | Code::Char('\n' | '\r') => {
tokenizer.exit(TokenType::HeadingSetextUnderline);
(State::Ok, Some(vec![code]))
}
_ => (State::Nok, None),
}
}
|