Unnamed repository; edit this file 'description' to name the repository.
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
//! Renders a bit of code as HTML.

use hir::Semantics;
use ide_db::ra_fixture::RaFixtureConfig;
use oorandom::Rand32;
use stdx::format_to;
use syntax::AstNode;

use crate::{
    FileId, RootDatabase,
    syntax_highlighting::{HighlightConfig, highlight},
};

pub(crate) fn highlight_as_html_with_config(
    db: &RootDatabase,
    config: &HighlightConfig<'_>,
    file_id: FileId,
    rainbow: bool,
) -> String {
    let sema = Semantics::new(db);
    let file_id = sema.attach_first_edition(file_id);
    let file = sema.parse(file_id);
    let file = file.syntax();
    fn rainbowify(seed: u64) -> String {
        let mut rng = Rand32::new(seed);
        format!(
            "hsl({h},{s}%,{l}%)",
            h = rng.rand_range(0..361),
            s = rng.rand_range(42..99),
            l = rng.rand_range(40..91),
        )
    }

    let hl_ranges = highlight(db, config, file_id.file_id(db), None);
    let text = file.to_string();
    let mut buf = String::new();
    buf.push_str(STYLE);
    buf.push_str("<pre><code>");
    for r in &hl_ranges {
        let chunk = html_escape(&text[r.range]);
        if r.highlight.is_empty() {
            format_to!(buf, "{}", chunk);
            continue;
        }

        let class = r.highlight.to_string().replace('.', " ");
        let color = match (rainbow, r.binding_hash) {
            (true, Some(hash)) => {
                format!(" data-binding-hash=\"{hash}\" style=\"color: {};\"", rainbowify(hash))
            }
            _ => "".into(),
        };
        format_to!(buf, "<span class=\"{}\"{}>{}</span>", class, color, chunk);
    }
    buf.push_str("</code></pre>");
    buf
}

pub(crate) fn highlight_as_html(db: &RootDatabase, file_id: FileId, rainbow: bool) -> String {
    highlight_as_html_with_config(
        db,
        &HighlightConfig {
            strings: true,
            comments: true,
            punctuation: true,
            specialize_punctuation: true,
            specialize_operator: true,
            operator: true,
            inject_doc_comment: true,
            macro_bang: true,
            syntactic_name_ref_highlighting: false,
            ra_fixture: RaFixtureConfig::default(),
        },
        file_id,
        rainbow,
    )
}

//FIXME: like, real html escaping
fn html_escape(text: &str) -> String {
    text.replace('<', "&lt;").replace('>', "&gt;")
}

const STYLE: &str = "
<style>
body                { margin: 0; }
pre                 { color: #DCDCCC; background: #3F3F3F; font-size: 22px; padding: 0.4em; }

.lifetime           { color: #DFAF8F; font-style: italic; }
.label              { color: #DFAF8F; font-style: italic; }
.comment            { color: #7F9F7F; }
.documentation      { color: #629755; }
.intra_doc_link     { font-style: italic; }
.injected           { opacity: 0.65 ; }
.struct, .enum      { color: #7CB8BB; }
.enum_variant       { color: #BDE0F3; }
.string_literal     { color: #CC9393; }
.field              { color: #94BFF3; }
.function           { color: #93E0E3; }
.parameter          { color: #94BFF3; }
.text               { color: #DCDCCC; }
.type               { color: #7CB8BB; }
.builtin_type       { color: #8CD0D3; }
.type_param         { color: #DFAF8F; }
.attribute          { color: #94BFF3; }
.numeric_literal    { color: #BFEBBF; }
.bool_literal       { color: #BFE6EB; }
.macro              { color: #94BFF3; }
.proc_macro         { color: #94BFF3; text-decoration: underline; }
.derive             { color: #94BFF3; font-style: italic; }
.module             { color: #AFD8AF; }
.value_param        { color: #DCDCCC; }
.variable           { color: #DCDCCC; }
.format_specifier   { color: #CC696B; }
.mutable            { text-decoration: underline; }
.escape_sequence    { color: #94BFF3; }
.keyword            { color: #F0DFAF; font-weight: bold; }
.control            { font-style: italic; }
.reference          { font-style: italic; font-weight: bold; }
.const              { font-weight: bolder; }
.unsafe             { color: #BC8383; }
.deprecated         { text-decoration: line-through; }

.invalid_escape_sequence { color: #FC5555; text-decoration: wavy underline; }
.unresolved_reference    { color: #FC5555; text-decoration: wavy underline; }
</style>
";
00 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348
// Copied from https://github.com/rust-lang/cargo/blob/367fd9f213750cd40317803dd0a5a3ce3f0c676d/src/cargo/util/frontmatter.rs
#![expect(dead_code)] // avoid editing
#![expect(unreachable_pub)] // avoid editing
#![expect(clippy::useless_format)] // avoid editing

type Span = std::ops::Range<usize>;

#[derive(Debug)]
pub struct ScriptSource<'s> {
    /// The full file
    raw: &'s str,
    /// The `#!/usr/bin/env cargo` line, if present
    shebang: Option<Span>,
    /// The code fence opener (`---`)
    open: Option<Span>,
    /// Trailing text after `ScriptSource::open` that identifies the meaning of
    /// `ScriptSource::frontmatter`
    info: Option<Span>,
    /// The lines between `ScriptSource::open` and `ScriptSource::close`
    frontmatter: Option<Span>,
    /// The code fence closer (`---`)
    close: Option<Span>,
    /// All content after the frontmatter and shebang
    content: Span,
}

impl<'s> ScriptSource<'s> {
    pub fn parse(raw: &'s str) -> Result<Self, FrontmatterError> {
        use winnow::stream::FindSlice as _;
        use winnow::stream::Location as _;
        use winnow::stream::Offset as _;
        use winnow::stream::Stream as _;

        let content_end = raw.len();
        let mut source = Self {
            raw,
            shebang: None,
            open: None,
            info: None,
            frontmatter: None,
            close: None,
            content: 0..content_end,
        };

        let mut input = winnow::stream::LocatingSlice::new(raw);

        if let Some(shebang_end) = strip_shebang(input.as_ref()) {
            let shebang_start = input.current_token_start();
            let _ = input.next_slice(shebang_end);
            let shebang_end = input.current_token_start();
            source.shebang = Some(shebang_start..shebang_end);
            source.content = shebang_end..content_end;
        }

        // Whitespace may precede a frontmatter but must end with a newline
        if let Some(nl_end) = strip_ws_lines(input.as_ref()) {
            let _ = input.next_slice(nl_end);
        }

        // Opens with a line that starts with 3 or more `-` followed by an optional identifier
        const FENCE_CHAR: char = '-';
        let fence_length = input
            .as_ref()
            .char_indices()
            .find_map(|(i, c)| (c != FENCE_CHAR).then_some(i))
            .unwrap_or_else(|| input.eof_offset());
        let open_start = input.current_token_start();
        let fence_pattern = input.next_slice(fence_length);
        let open_end = input.current_token_start();
        match fence_length {
            0 => {
                return Ok(source);
            }
            1 | 2 => {
                // either not a frontmatter or invalid frontmatter opening
                return Err(FrontmatterError::new(
                    format!(
                        "found {fence_length} `{FENCE_CHAR}` in rust frontmatter, expected at least 3"
                    ),
                    raw.len()..raw.len(),
                ).push_visible_span(open_start..open_end));
            }
            _ => {}
        }
        source.open = Some(open_start..open_end);
        let Some(info_nl) = input.find_slice("\n") else {
            return Err(FrontmatterError::new(
                format!("unclosed frontmatter; expected `{fence_pattern}`"),
                raw.len()..raw.len(),
            )
            .push_visible_span(open_start..open_end));
        };
        let info = input.next_slice(info_nl.start);
        let info = info.strip_suffix('\r').unwrap_or(info); // already excludes `\n`
        let info = info.trim_matches(is_horizontal_whitespace);
        if !info.is_empty() {
            let info_start = info.offset_from(&raw);
            let info_end = info_start + info.len();
            source.info = Some(info_start..info_end);
        }

        // Ends with a line that starts with a matching number of `-` only followed by whitespace
        let nl_fence_pattern = format!("\n{fence_pattern}");
        let Some(frontmatter_nl) = input.find_slice(nl_fence_pattern.as_str()) else {
            for len in (2..(nl_fence_pattern.len() - 1)).rev() {
                let Some(frontmatter_nl) = input.find_slice(&nl_fence_pattern[0..len]) else {
                    continue;
                };
                let _ = input.next_slice(frontmatter_nl.start + 1);
                let close_start = input.current_token_start();
                let _ = input.next_slice(len);
                let close_end = input.current_token_start();
                let fewer_dashes = fence_length - len;
                return Err(FrontmatterError::new(
                    format!(
                        "closing code fence has {fewer_dashes} less `-` than the opening fence"
                    ),
                    close_start..close_end,
                )
                .push_visible_span(open_start..open_end));
            }
            return Err(FrontmatterError::new(
                format!("unclosed frontmatter; expected `{fence_pattern}`"),
                raw.len()..raw.len(),
            )
            .push_visible_span(open_start..open_end));
        };
        let frontmatter_start = input.current_token_start() + 1; // skip nl from infostring
        let _ = input.next_slice(frontmatter_nl.start + 1);
        let frontmatter_end = input.current_token_start();
        source.frontmatter = Some(frontmatter_start..frontmatter_end);
        let close_start = input.current_token_start();
        let _ = input.next_slice(fence_length);
        let close_end = input.current_token_start();
        source.close = Some(close_start..close_end);

        let nl = input.find_slice("\n");
        let after_closing_fence =
            input.next_slice(nl.map(|span| span.end).unwrap_or_else(|| input.eof_offset()));
        let content_start = input.current_token_start();
        let extra_dashes = after_closing_fence.chars().take_while(|b| *b == FENCE_CHAR).count();
        if 0 < extra_dashes {
            let extra_start = close_end;
            let extra_end = extra_start + extra_dashes;
            return Err(FrontmatterError::new(
                format!("closing code fence has {extra_dashes} more `-` than the opening fence"),
                extra_start..extra_end,
            )
            .push_visible_span(open_start..open_end));
        } else {
            let after_closing_fence = strip_newline(after_closing_fence);
            let after_closing_fence = after_closing_fence.trim_matches(is_horizontal_whitespace);
            if !after_closing_fence.is_empty() {
                // extra characters beyond the original fence pattern
                let after_start = after_closing_fence.offset_from(&raw);
                let after_end = after_start + after_closing_fence.len();
                return Err(FrontmatterError::new(
                    format!("unexpected characters after frontmatter close"),
                    after_start..after_end,
                )
                .push_visible_span(open_start..open_end));
            }
        }

        source.content = content_start..content_end;

        if let Some(nl_end) = strip_ws_lines(input.as_ref()) {
            let _ = input.next_slice(nl_end);
        }
        let fence_length = input
            .as_ref()
            .char_indices()
            .find_map(|(i, c)| (c != FENCE_CHAR).then_some(i))
            .unwrap_or_else(|| input.eof_offset());
        if 0 < fence_length {
            let fence_start = input.current_token_start();
            let fence_end = fence_start + fence_length;
            return Err(FrontmatterError::new(
                format!("only one frontmatter is supported"),
                fence_start..fence_end,
            )
            .push_visible_span(open_start..open_end)
            .push_visible_span(close_start..close_end));
        }

        Ok(source)
    }

    pub fn shebang(&self) -> Option<&'s str> {
        self.shebang.clone().map(|span| &self.raw[span])
    }

    pub fn shebang_span(&self) -> Option<Span> {
        self.shebang.clone()
    }

    pub fn open_span(&self) -> Option<Span> {
        self.open.clone()
    }

    pub fn info(&self) -> Option<&'s str> {
        self.info.clone().map(|span| &self.raw[span])
    }

    pub fn info_span(&self) -> Option<Span> {
        self.info.clone()
    }

    pub fn frontmatter(&self) -> Option<&'s str> {
        self.frontmatter.clone().map(|span| &self.raw[span])
    }

    pub fn frontmatter_span(&self) -> Option<Span> {
        self.frontmatter.clone()
    }

    pub fn close_span(&self) -> Option<Span> {
        self.close.clone()
    }

    pub fn content(&self) -> &'s str {
        &self.raw[self.content.clone()]
    }

    pub fn content_span(&self) -> Span {
        self.content.clone()
    }
}

/// Returns the index after the shebang line, if present
pub fn strip_shebang(input: &str) -> Option<usize> {
    // See rust-lang/rust's compiler/rustc_lexer/src/lib.rs's `strip_shebang`
    // Shebang must start with `#!` literally, without any preceding whitespace.
    // For simplicity we consider any line starting with `#!` a shebang,
    // regardless of restrictions put on shebangs by specific platforms.
    if let Some(rest) = input.strip_prefix("#!") {
        // Ok, this is a shebang but if the next non-whitespace token is `[`,
        // then it may be valid Rust code, so consider it Rust code.
        //
        // NOTE: rustc considers line and block comments to be whitespace but to avoid
        // any more awareness of Rust grammar, we are excluding it.
        if !rest.trim_start().starts_with('[') {
            // No other choice than to consider this a shebang.
            let newline_end = input.find('\n').map(|pos| pos + 1).unwrap_or(input.len());
            return Some(newline_end);
        }
    }
    None
}

/// Returns the index after any lines with only whitespace, if present
pub fn strip_ws_lines(input: &str) -> Option<usize> {
    let ws_end = input.find(|c| !is_whitespace(c)).unwrap_or(input.len());
    if ws_end == 0 {
        return None;
    }

    let nl_start = input[0..ws_end].rfind('\n')?;
    let nl_end = nl_start + 1;
    Some(nl_end)
}

/// True if `c` is considered a whitespace according to Rust language definition.
/// See [Rust language reference](https://doc.rust-lang.org/reference/whitespace.html)
/// for definitions of these classes.
pub(crate) fn is_whitespace(c: char) -> bool {
    // This is Pattern_White_Space.
    //
    // Note that this set is stable (ie, it doesn't change with different
    // Unicode versions), so it's ok to just hard-code the values.

    matches!(
        c,
        // End-of-line characters
        | '\u{000A}' // line feed (\n)
        | '\u{000B}' // vertical tab
        | '\u{000C}' // form feed
        | '\u{000D}' // carriage return (\r)
        | '\u{0085}' // next line (from latin1)
        | '\u{2028}' // LINE SEPARATOR
        | '\u{2029}' // PARAGRAPH SEPARATOR

        // `Default_Ignorable_Code_Point` characters
        | '\u{200E}' // LEFT-TO-RIGHT MARK
        | '\u{200F}' // RIGHT-TO-LEFT MARK

        // Horizontal space characters
        | '\u{0009}'   // tab (\t)
        | '\u{0020}' // space
    )
}

/// True if `c` is considered horizontal whitespace according to Rust language definition.
fn is_horizontal_whitespace(c: char) -> bool {
    // This is Pattern_White_Space.
    //
    // Note that this set is stable (ie, it doesn't change with different
    // Unicode versions), so it's ok to just hard-code the values.

    matches!(
        c,
        // Horizontal space characters
        '\u{0009}'   // tab (\t)
        | '\u{0020}' // space
    )
}

fn strip_newline(text: &str) -> &str {
    text.strip_suffix("\r\n").or_else(|| text.strip_suffix('\n')).unwrap_or(text)
}

#[derive(Debug)]
pub struct FrontmatterError {
    message: String,
    primary_span: Span,
    visible_spans: Vec<Span>,
}

impl FrontmatterError {
    pub fn new(message: impl Into<String>, span: Span) -> Self {
        Self { message: message.into(), primary_span: span, visible_spans: Vec::new() }
    }

    pub fn push_visible_span(mut self, span: Span) -> Self {
        self.visible_spans.push(span);
        self
    }

    pub fn message(&self) -> &str {
        self.message.as_str()
    }

    pub fn primary_span(&self) -> Span {
        self.primary_span.clone()
    }

    pub fn visible_spans(&self) -> &[Span] {
        &self.visible_spans
    }
}

impl std::fmt::Display for FrontmatterError {
    fn fmt(&self, fmt: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
        self.message.fmt(fmt)
    }
}

impl std::error::Error for FrontmatterError {}