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

// SPDX-License-Identifier: Apache-2.0

use crate::pt::Comment;

#[derive(Debug, PartialEq, Eq, Clone)]
pub enum DocComment {
    Line { comment: DocCommentTag },
    Block { comments: Vec<DocCommentTag> },
}

impl DocComment {
    pub fn comments(&self) -> Vec<&DocCommentTag> {
        match self {
            DocComment::Line { comment } => vec![comment],
            DocComment::Block { comments } => comments.iter().collect(),
        }
    }
}

#[derive(Debug, PartialEq, Eq, Clone)]
pub struct DocCommentTag {
    pub tag: String,
    pub tag_offset: usize,
    pub value: String,
    pub value_offset: usize,
}

enum CommentType {
    Line,
    Block,
}

/// From the start to end offset, filter all the doc comments out of the comments and parse
/// them into tags with values.
pub fn parse_doccomments(comments: &[Comment], start: usize, end: usize) -> Vec<DocComment> {
    // first extract the tags
    let mut tags = Vec::new();

    let lines = filter_comments(comments, start, end);

    for (ty, comment_lines) in lines {
        let mut single_tags = Vec::new();

        for (start_offset, line) in comment_lines {
            let mut chars = line.char_indices().peekable();

            if let Some((_, '@')) = chars.peek() {
                // step over @
                let (tag_start, _) = chars.next().unwrap();
                let mut tag_end = tag_start;

                while let Some((offset, c)) = chars.peek() {
                    if c.is_whitespace() {
                        break;
                    }

                    tag_end = *offset;

                    chars.next();
                }

                let leading = line[tag_end + 1..]
                    .chars()
                    .take_while(|ch| ch.is_whitespace())
                    .count();

                // tag value
                single_tags.push(DocCommentTag {
                    tag_offset: start_offset + tag_start + 1,
                    tag: line[tag_start + 1..tag_end + 1].to_owned(),
                    value_offset: start_offset + tag_end + leading + 1,
                    value: line[tag_end + 1..].trim().to_owned(),
                });
            } else if !single_tags.is_empty() || !tags.is_empty() {
                let line = line.trim();
                if !line.is_empty() {
                    let single_doc_comment = if let Some(single_tag) = single_tags.last_mut() {
                        Some(single_tag)
                    } else if let Some(tag) = tags.last_mut() {
                        match tag {
                            DocComment::Line { comment } => Some(comment),
                            DocComment::Block { comments } => comments.last_mut(),
                        }
                    } else {
                        None
                    };

                    if let Some(comment) = single_doc_comment {
                        comment.value.push('\n');
                        comment.value.push_str(line);
                    }
                }
            } else {
                let leading = line.chars().take_while(|ch| ch.is_whitespace()).count();

                single_tags.push(DocCommentTag {
                    tag_offset: start_offset + start_offset + leading,
                    tag: String::from("notice"),
                    value_offset: start_offset + start_offset + leading,
                    value: line.trim().to_owned(),
                });
            }
        }

        match ty {
            CommentType::Line if !single_tags.is_empty() => tags.push(DocComment::Line {
                comment: single_tags[0].to_owned(),
            }),
            CommentType::Block => tags.push(DocComment::Block {
                comments: single_tags,
            }),
            _ => {}
        }
    }

    tags
}

/// Convert the comment to lines, stripping whitespace, comment characters and leading * in block comments
fn filter_comments(
    comments: &[Comment],
    start: usize,
    end: usize,
) -> Vec<(CommentType, Vec<(usize, &str)>)> {
    let mut res = Vec::new();

    for comment in comments.iter() {
        let mut grouped_comments = Vec::new();

        match comment {
            Comment::DocLine(loc, comment) => {
                if loc.start() >= end || loc.end() < start {
                    continue;
                }

                // remove the leading ///
                let leading = comment[3..]
                    .chars()
                    .take_while(|ch| ch.is_whitespace())
                    .count();

                grouped_comments.push((loc.start() + leading + 3, comment[3..].trim()));

                res.push((CommentType::Line, grouped_comments));
            }
            Comment::DocBlock(loc, comment) => {
                if loc.start() >= end || loc.end() < start {
                    continue;
                }

                let mut start = loc.start() + 3;

                let len = comment.len();

                // remove the leading /** and tailing */
                for s in comment[3..len - 2].lines() {
                    if let Some((i, _)) = s
                        .char_indices()
                        .find(|(_, ch)| !ch.is_whitespace() && *ch != '*')
                    {
                        grouped_comments.push((start + i, s[i..].trim_end()));
                    }

                    start += s.len() + 1;
                }

                res.push((CommentType::Block, grouped_comments));
            }
            _ => (),
        }
    }

    res
}