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
//! Link reference definition
//!
//! `[label]: /url "title"`
//!
//! <https://spec.commonmark.org/0.30/#link-reference-definition>
//!
//! This plugin parses markdown link references. Check documentation on [ReferenceMap]
//! to see how you can use and/or extend it if you have external source for references.
//!
use std::collections::HashMap;
use derivative::Derivative;

use crate::{MarkdownIt, Node};
use crate::common::utils::normalize_reference;
use crate::generics::inline::full_link;
use crate::parser::block::{BlockRule, BlockState};
use crate::parser::extset::RootExt;

/// Storage for parsed references
///
/// if you have some external source for your link references, you can add them like this:
///
/// ```rust
/// use markdown_it::parser::block::builtin::BlockParserRule;
/// use markdown_it::parser::core::{CoreRule, Root};
/// use markdown_it::plugins::cmark::block::reference::{
///     ReferenceMap, ReferenceMapEntry, ReferenceMapKey
/// };
/// use markdown_it::{MarkdownIt, Node};
///
/// let md = &mut MarkdownIt::new();
/// markdown_it::plugins::cmark::add(md);
///
/// struct ReferencePatcher;
/// impl CoreRule for ReferencePatcher {
///     fn run(root: &mut Node, _: &MarkdownIt) {
///         let data = root.cast_mut::<Root>().unwrap();
///         let references = data.ext.get_or_insert_default::<ReferenceMap>();
///         references.insert(
///             ReferenceMapKey::new("rust".into()),
///             ReferenceMapEntry::new(
///                 "https://www.rust-lang.org/".into(),
///                 Some("The Rust Language".into())
///             )
///         );
///     }
/// }
///
/// md.add_rule::<ReferencePatcher>()
///     .before::<BlockParserRule>();
///
/// let html = md.parse("[rust]").render();
/// assert_eq!(
///     html.trim(),
///     r#"<p><a href="https://www.rust-lang.org/" title="The Rust Language">rust</a></p>"#
/// );
/// ```
///
/// It is possible to support callback for external link references in the future,
/// please tell us whether that's useful for you.
///
pub type ReferenceMap = HashMap<ReferenceMapKey, ReferenceMapEntry>;
impl RootExt for ReferenceMap {}

#[derive(Derivative)]
#[derivative(Debug, Default, Hash, PartialEq, Eq)]
/// Reference label
pub struct ReferenceMapKey {
    #[derivative(PartialEq = "ignore")]
    #[derivative(Hash = "ignore")]
    pub label: String,
    normalized: String,
}

impl ReferenceMapKey {
    pub fn new(label: String) -> Self {
        let normalized = normalize_reference(&label);
        Self { label, normalized }
    }
}

#[derive(Debug, Default)]
/// Reference value
pub struct ReferenceMapEntry {
    pub destination: String,
    pub title: Option<String>,
}

impl ReferenceMapEntry {
    pub fn new(destination: String, title: Option<String>) -> Self {
        Self { destination, title }
    }
}

/// Add plugin that parses markdown link references
pub fn add(md: &mut MarkdownIt) {
    md.block.add_rule::<ReferenceScanner>();
}

#[doc(hidden)]
pub struct ReferenceScanner;
impl BlockRule for ReferenceScanner {
    fn check(_: &mut BlockState) -> Option<()> {
        None // can't interrupt anything
    }

    fn run(state: &mut BlockState) -> Option<(Node, usize)> {
        // if it's indented more than 3 spaces, it should be a code block
        if state.line_indent(state.line) >= 4 { return None; }

        let mut chars = state.get_line(state.line).chars();

        if let Some('[') = chars.next() {} else { return None; }

        // Simple check to quickly interrupt scan on [link](url) at the start of line.
        // Can be useful on practice: https://github.com/markdown-it/markdown-it/issues/54
        loop {
            match chars.next() {
                Some('\\') => { chars.next(); },
                Some(']') => {
                    if let Some(':') = chars.next() {
                        break;
                    } else {
                        return None;
                    }
                }
                Some(_) => {},
                None => break,
            }
        }

        let start_line = state.line;
        let mut next_line = start_line;

        // jump line-by-line until empty one or EOF
        'outer: loop {
            next_line += 1;

            if next_line >= state.line_max || state.is_empty(next_line) { break; }

            // this would be a code block normally, but after paragraph
            // it's considered a lazy continuation regardless of what's there
            if state.line_indent(next_line) >= 4 { continue; }

            // quirk for blockquotes, this line should already be checked by that rule
            if state.line_offsets[next_line].indent_nonspace < 0 { continue; }

            // Some tags can terminate paragraph without empty line.
            let old_state_line = state.line;
            state.line = next_line;
            if state.test_rules_at_line() {
                state.line = old_state_line;
                break 'outer;
            }
            state.line = old_state_line;
        }

        let (str_before_trim, _) = state.get_lines(start_line, next_line, state.blk_indent, false);
        let str = str_before_trim.trim();
        let mut chars = str.char_indices();
        chars.next(); // skip '['
        let label_end;
        let mut lines = 0;

        loop {
            match chars.next() {
                Some((_, '[')) => return None,
                Some((p, ']')) => {
                    label_end = p;
                    break;
                }
                Some((_, '\n')) => lines += 1,
                Some((_, '\\')) => {
                    if let Some((_, '\n')) = chars.next() {
                        lines += 1;
                    }
                }
                Some(_) => {},
                None => return None,
            }
        }

        if let Some((_, ':')) = chars.next() {} else { return None; }

        // [label]:   destination   'title'
        //         ^^^ skip optional whitespace here
        let mut pos = label_end + 2;
        while let Some((_, ch @ (' ' | '\t' | '\n'))) = chars.next() {
            if ch == '\n' { lines += 1; }
            pos += 1;
        }

        // [label]:   destination   'title'
        //            ^^^^^^^^^^^ parse this
        let href;
        if let Some(res) = full_link::parse_link_destination(str, pos, str.len()) {
            if pos == res.pos { return None; }
            href = (state.md.normalize_link)(&res.str);
            if !(state.md.validate_link)(&href) { return None; }
            pos = res.pos;
            lines += res.lines;
        } else {
            return None;
        }

        // save cursor state, we could require to rollback later
        let dest_end_pos = pos;
        let dest_end_lines = lines;

        // [label]:   destination   'title'
        //                       ^^^ skipping those spaces
        let start = pos;
        let mut chars = str[pos..].chars();
        while let Some(ch @ (' ' | '\t' | '\n')) = chars.next() {
            if ch == '\n' { lines += 1; }
            pos += 1;
        }

        // [label]:   destination   'title'
        //                          ^^^^^^^ parse this
        let mut title = None;
        if pos != start {
            if let Some(res) = full_link::parse_link_title(str, pos, str.len()) {
                title = Some(res.str);
                pos = res.pos;
                lines += res.lines;
            } else {
                pos = dest_end_pos;
                lines = dest_end_lines;
            }
        }

        // skip trailing spaces until the rest of the line
        let mut chars = str[pos..].chars();
        loop {
            match chars.next() {
                Some(' ' | '\t') => pos += 1,
                Some('\n') | None => break,
                Some(_) if title.is_some() => {
                    // garbage at the end of the line after title,
                    // but it could still be a valid reference if we roll back
                    title = None;
                    pos = dest_end_pos;
                    lines = dest_end_lines;
                    chars = str[pos..].chars();
                }
                Some(_) => {
                    // garbage at the end of the line
                    return None;
                }
            }
        }

        let label = normalize_reference(&str[1..label_end]);
        if label.is_empty() {
            // CommonMark 0.20 disallows empty labels
            return None;
        }

        let references = &mut state.root_ext.get_or_insert_default::<ReferenceMap>();

        references.entry(ReferenceMapKey::new(label)).or_insert_with(|| ReferenceMapEntry::new(href, title));

        Some((Node::default(), lines + 1))
    }
}