crashkeys.dev
Markdown parser fork with extended syntax for personal use.
1//! GFM: Task list item check occurs in the [text][] content type.
2//!
3//! ## Grammar
4//!
5//! Checks form with the following BNF
6//! (<small>see [construct][crate::construct] for character groups</small>):
7//!
8//! ```bnf
9//! gfm_task_list_item_check ::= '[' (0x09 | ' ' | 'X' | 'x') ']'
10//! ```
11//!
12//! The check is only allowed at the start of the first paragraph, optionally
13//! following zero or more definitions or a blank line, in a list item.
14//! The check must be followed by whitespace, which is in turn followed by
15//! non-whitespace.
16//!
17//! ## HTML
18//!
19//! Checks relate to the `<input>` element, in the checkbox state
20//! (`type=checkbox`), in HTML.
21//! See [*§ 4.10.5.1.15 Checkbox state (`type=checkbox`)*][html-input-checkbox]
22//! in the HTML spec for more info.
23//!
24//! ## Recommendation
25//!
26//! It is recommended to use lowercase `x` (instead of uppercase `X`), because
27//! in markdown, it is more common to use lowercase in places where casing does
28//! not matter.
29//! It is also recommended to use a space (instead of a tab), as there is no
30//! benefit of using tabs in this case.
31//!
32//! ## Tokens
33//!
34//! * [`GfmTaskListItemCheck`][Name::GfmTaskListItemCheck]
35//! * [`GfmTaskListItemMarker`][Name::GfmTaskListItemMarker]
36//! * [`GfmTaskListItemValueChecked`][Name::GfmTaskListItemValueChecked]
37//! * [`GfmTaskListItemValueUnchecked`][Name::GfmTaskListItemValueUnchecked]
38//!
39//! ## References
40//!
41//! * [`micromark-extension-gfm-task-list-item`](https://github.com/micromark/micromark-extension-gfm-task-list-item)
42//! * [*§ 5.3 Task list items (extension)* in `GFM`](https://github.github.com/gfm/#task-list-items-extension-)
43//!
44//! [text]: crate::construct::text
45//! [html-input-checkbox]: https://html.spec.whatwg.org/multipage/input.html#checkbox-state-(type=checkbox)
46
47use crate::construct::partial_space_or_tab::space_or_tab;
48use crate::event::Name;
49use crate::state::{Name as StateName, State};
50use crate::tokenizer::Tokenizer;
51
52/// At start of task list item check.
53///
54/// ```markdown
55/// > | * [x] y.
56/// ^
57/// ```
58pub fn start(tokenizer: &mut Tokenizer) -> State {
59 if tokenizer.parse_state.options.constructs.gfm_task_list_item
60 && tokenizer
61 .tokenize_state
62 .document_at_first_paragraph_of_list_item
63 && tokenizer.current == Some(b'[')
64 && tokenizer.previous.is_none()
65 {
66 tokenizer.enter(Name::GfmTaskListItemCheck);
67 tokenizer.enter(Name::GfmTaskListItemMarker);
68 tokenizer.consume();
69 tokenizer.exit(Name::GfmTaskListItemMarker);
70 State::Next(StateName::GfmTaskListItemCheckInside)
71 } else {
72 State::Nok
73 }
74}
75
76/// In task list item check.
77///
78/// ```markdown
79/// > | * [x] y.
80/// ^
81/// ```
82pub fn inside(tokenizer: &mut Tokenizer) -> State {
83 match tokenizer.current {
84 Some(b'\t' | b'\n' | b' ') => {
85 tokenizer.enter(Name::GfmTaskListItemValueUnchecked);
86 tokenizer.consume();
87 tokenizer.exit(Name::GfmTaskListItemValueUnchecked);
88 State::Next(StateName::GfmTaskListItemCheckClose)
89 }
90 Some(b'X' | b'x') => {
91 tokenizer.enter(Name::GfmTaskListItemValueChecked);
92 tokenizer.consume();
93 tokenizer.exit(Name::GfmTaskListItemValueChecked);
94 State::Next(StateName::GfmTaskListItemCheckClose)
95 }
96 _ => State::Nok,
97 }
98}
99
100/// At close of task list item check.
101///
102/// ```markdown
103/// > | * [x] y.
104/// ^
105/// ```
106pub fn close(tokenizer: &mut Tokenizer) -> State {
107 match tokenizer.current {
108 Some(b']') => {
109 tokenizer.enter(Name::GfmTaskListItemMarker);
110 tokenizer.consume();
111 tokenizer.exit(Name::GfmTaskListItemMarker);
112 tokenizer.exit(Name::GfmTaskListItemCheck);
113 State::Next(StateName::GfmTaskListItemCheckAfter)
114 }
115 _ => State::Nok,
116 }
117}
118
119/// After task list item check.
120///
121/// ```markdown
122/// > | * [x] y.
123/// ^
124/// ```
125pub fn after(tokenizer: &mut Tokenizer) -> State {
126 match tokenizer.current {
127 // EOL in paragraph means there must be something else after it.
128 Some(b'\n') => State::Ok,
129 // Space or tab?
130 // Check what comes after.
131 Some(b'\t' | b' ') => {
132 tokenizer.check(State::Ok, State::Nok);
133 tokenizer.attempt(
134 State::Next(StateName::GfmTaskListItemCheckAfterSpaceOrTab),
135 State::Nok,
136 );
137 State::Retry(space_or_tab(tokenizer))
138 }
139 // EOF, or non-whitespace, both wrong.
140 _ => State::Nok,
141 }
142}
143
144/// After whitespace, after task list item check.
145///
146/// ```markdown
147/// > | * [x] y.
148/// ^
149/// ```
150pub fn after_space_or_tab(tokenizer: &mut Tokenizer) -> State {
151 // End of paragraph, after whitespace, after check, is not okay.
152 if tokenizer.current.is_none() {
153 State::Nok
154 } else {
155 State::Ok
156 }
157}