brik 0.10.0

HTML tree manipulation library - a building block for HTML parsing and manipulation
Documentation 
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
297
298
299
300
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

use super::preamble::{HtmlPreamble, Rule};
use super::tag_info::{HtmlTagInfo, XmlnsPositions};
use crate::ns::{NsError, NsResult};
use pest::iterators::Pair;
use pest::Parser;

/// Parses the top of an HTML document to locate and analyze the `<html>` tag.
///
/// This function uses the Pest parser to efficiently parse only the beginning
/// of the HTML document (typically ~1KB) to find the `<html>` tag and extract
/// information about its attributes, particularly existing `xmlns:*` declarations.
///
/// The function walks the parse tree to:
/// - Locate the `<html>` tag position
/// - Extract all existing `xmlns:*` attributes
/// - Record the injection point (before the tag close)
///
/// # Arguments
///
/// * `html` - The HTML string to parse
///
/// # Returns
///
/// Returns `Ok(HtmlTagInfo)` with information about the html tag location and
/// existing xmlns attributes, or `Err(NsError)` if parsing fails.
///
/// # Errors
///
/// Returns `NsError::ParseError` if:
/// - The `<html>` tag is not found in the document
/// - The `<html>` tag attributes are malformed
/// - The HTML structure cannot be parsed by the Pest grammar
///
/// Note: The parser is forgiving with malformed preambles (comments, DOCTYPEs, etc.)
/// and will skip over them to find the `<html>` tag. However, the `<html>` tag
/// itself must be well-formed
///
/// # Examples
///
/// ```ignore
/// use brik::ns::defaults::parse::parse_preamble;
///
/// let html = r#"<!DOCTYPE html><html lang="en"><body>Hello</body></html>"#;
/// let info = parse_preamble(html).unwrap();
/// assert!(info.existing_xmlns.is_empty());
/// ```
pub fn parse_preamble(html: impl AsRef<str>) -> NsResult<HtmlTagInfo> {
    let html = html.as_ref();

    // Parse the document using Pest.
    let pairs = HtmlPreamble::parse(Rule::document, html)
        .map_err(|e| NsError::ParseError(format!("Failed to parse HTML: {e}")))?;

    // Walk the parse tree to find the html_tag and extract information.
    for pair in pairs {
        if pair.as_rule() == Rule::document {
            for inner in pair.into_inner() {
                if inner.as_rule() == Rule::html_tag {
                    return extract_html_tag_info(inner);
                }
            }
        }
    }

    Err(NsError::ParseError(
        "No <html> tag found in document".to_string(),
    ))
}

/// Extracts tag information from an html_tag parse node.
///
/// Processes the html_tag's children to extract tag positions and xmlns attributes.
#[inline]
fn extract_html_tag_info(html_tag: Pair<Rule>) -> NsResult<HtmlTagInfo> {
    let tag_start = html_tag.as_span().start();
    let tag_end = html_tag.as_span().end();
    let mut tag_close_start = 0;
    let mut existing_xmlns = Vec::new();

    // Process the html_tag's children.
    for tag_part in html_tag.into_inner() {
        match tag_part.as_rule() {
            Rule::attributes => {
                extract_xmlns_attributes(tag_part, &mut existing_xmlns);
            }
            Rule::tag_close => {
                // Mark where the tag close starts (position of `>` or `/>`).
                tag_close_start = tag_part.as_span().start();
            }
            _ => {}
        }
    }

    Ok(HtmlTagInfo {
        tag_start,
        tag_close_start,
        tag_end,
        existing_xmlns,
    })
}

/// Extracts xmlns:* attributes from an attributes parse node.
///
/// Processes all attributes and stores the positions of xmlns:* declarations.
#[inline]
fn extract_xmlns_attributes(attributes: Pair<Rule>, existing_xmlns: &mut Vec<XmlnsPositions>) {
    for attr in attributes.into_inner() {
        if attr.as_rule() == Rule::attribute {
            if let Some(xmlns_positions) = extract_xmlns_from_attribute(attr) {
                existing_xmlns.push(xmlns_positions);
            }
        }
    }
}

/// Extracts xmlns namespace positions from a single attribute.
///
/// Returns positions for the prefix and URI if this is an xmlns:* attribute.
#[inline]
fn extract_xmlns_from_attribute(attr: Pair<Rule>) -> Option<XmlnsPositions> {
    let mut attr_name_span = None;
    let mut attr_value_span = None;

    for attr_part in attr.into_inner() {
        match attr_part.as_rule() {
            Rule::attr_name => {
                attr_name_span = Some(attr_part.as_span());
            }
            Rule::attr_value => {
                attr_value_span = Some(extract_value_positions(attr_part));
            }
            _ => {}
        }
    }

    // If this is an xmlns:* attribute, return its positions.
    if let (Some(name_span), Some(value_span)) = (attr_name_span, attr_value_span) {
        let name = name_span.as_str();
        if name.starts_with("xmlns:") {
            let prefix_offset = "xmlns:".len();
            let prefix_start = name_span.start() + prefix_offset;
            let prefix_end = name_span.end();

            return Some(((prefix_start, prefix_end), value_span));
        }
    }

    None
}

/// Extracts value positions from an attribute value, excluding quotes.
///
/// Calculates the start and end positions of the attribute value,
/// removing surrounding quotes if present.
#[inline]
fn extract_value_positions(value_pair: Pair<Rule>) -> (usize, usize) {
    let span = value_pair.as_span();
    let value = span.as_str();

    // Calculate positions excluding quotes if present.
    let starts_with_quote = value.starts_with('"') || value.starts_with('\'');
    let ends_with_quote = value.ends_with('"') || value.ends_with('\'');
    let start_offset = if starts_with_quote { 1 } else { 0 };
    let end_offset = if ends_with_quote { 1 } else { 0 };

    (span.start() + start_offset, span.end() - end_offset)
}

#[cfg(test)]
mod tests {
    use super::*;

    /// Tests parsing a simple HTML document without xmlns attributes.
    ///
    /// Verifies that the parser correctly identifies the html tag positions
    /// and returns an empty xmlns map when no namespace declarations exist.
    #[test]
    fn parse_simple_html() {
        let html = r#"<!DOCTYPE html>
<html lang="en">
<body>Hello</body>
</html>"#;

        let result = parse_preamble(html);
        assert!(result.is_ok());

        let info = result.unwrap();
        assert!(info.existing_xmlns.is_empty());
        assert!(info.tag_start < info.tag_close_start);
        assert!(info.tag_close_start < info.tag_end);
    }

    /// Tests parsing HTML with existing xmlns declarations.
    ///
    /// Verifies that the parser correctly extracts xmlns:* attributes
    /// and stores them with their positions in the existing_xmlns vector.
    #[test]
    fn parse_with_xmlns() {
        let html = r#"<!DOCTYPE html>
<html xmlns:custom="http://example.com/ns" xmlns:other="http://other.com" lang="en">
<body>Hello</body>
</html>"#;

        let result = parse_preamble(html);
        assert!(result.is_ok());

        let info = result.unwrap();
        assert_eq!(info.existing_xmlns.len(), 2);

        // Verify the first xmlns attribute (custom).
        let ((prefix_start, prefix_end), (uri_start, uri_end)) = info.existing_xmlns[0];
        assert_eq!(&html[prefix_start..prefix_end], "custom");
        assert_eq!(&html[uri_start..uri_end], "http://example.com/ns");

        // Verify the second xmlns attribute (other).
        let ((prefix_start, prefix_end), (uri_start, uri_end)) = info.existing_xmlns[1];
        assert_eq!(&html[prefix_start..prefix_end], "other");
        assert_eq!(&html[uri_start..uri_end], "http://other.com");
    }

    /// Tests parsing HTML with comments in the preamble.
    ///
    /// Verifies that the parser correctly skips comments and doesn't
    /// falsely match html tags within comment content.
    #[test]
    fn parse_with_comment() {
        let html = r#"<!-- This has <html> in it -->
<!DOCTYPE html>
<html>
<body>Hello</body>
</html>"#;

        let result = parse_preamble(html);
        assert!(result.is_ok());
    }

    /// Tests that missing html tag returns an error.
    ///
    /// Verifies that documents without an html tag return an appropriate error.
    #[test]
    fn parse_missing_html_tag() {
        let html = r#"<!DOCTYPE html>
<body>Hello</body>"#;

        let result = parse_preamble(html);
        assert!(result.is_err());
    }

    /// Tests parsing HTML with processing instructions.
    ///
    /// Verifies that the parser correctly skips XML processing instructions
    /// in the preamble.
    #[test]
    fn parse_with_pi() {
        let html = r#"<?xml version="1.0"?>
<!DOCTYPE html>
<html>
<body>Hello</body>
</html>"#;

        let result = parse_preamble(html);
        assert!(result.is_ok());
    }

    /// Tests parsing HTML with multiple xmlns attributes.
    ///
    /// Verifies that the parser correctly handles multiple xmlns
    /// declarations and stores their positions correctly.
    #[test]
    fn parse_multiple_xmlns() {
        let html = r#"<html xmlns:svg="http://www.w3.org/2000/svg" xmlns:math="http://www.w3.org/1998/Math/MathML">
<body>Hello</body>
</html>"#;

        let result = parse_preamble(html);
        assert!(result.is_ok());

        let info = result.unwrap();
        assert_eq!(info.xmlns_count(), 2);

        let prefix1 = info.get_prefix(0, html).unwrap();
        let uri1 = info.get_uri(0, html).unwrap();
        assert_eq!(prefix1, "svg");
        assert_eq!(uri1, "http://www.w3.org/2000/svg");

        let prefix2 = info.get_prefix(1, html).unwrap();
        let uri2 = info.get_uri(1, html).unwrap();
        assert_eq!(prefix2, "math");
        assert_eq!(uri2, "http://www.w3.org/1998/Math/MathML");
    }

    /// Tests parsing HTML with single-quoted attribute values.
    ///
    /// Verifies that the parser correctly handles single quotes
    /// and strips them from the value positions.
    #[test]
    fn parse_single_quoted_attribute() {
        let html = r#"<html xmlns:custom='http://example.com/single' lang='en'>
<body>Hello</body>
</html>"#;

        let result = parse_preamble(html);
        assert!(result.is_ok());

        let info = result.unwrap();
        assert_eq!(info.xmlns_count(), 1);

        let uri = info.get_uri(0, html).unwrap();
        assert_eq!(uri, "http://example.com/single");
        // Ensure quotes were stripped.
        assert!(!uri.contains('\''));
    }

    /// Tests parsing malformed HTML that fails Pest grammar.
    ///
    /// Verifies that genuinely malformed HTML returns ParseError.
    #[test]
    fn parse_malformed_html_tag() {
        let html = r#"<!DOCTYPE html>
<html xmlns:broken="unclosed
<body>Hello</body>
</html>"#;

        let result = parse_preamble(html);
        assert!(result.is_err());

        match result {
            Err(NsError::ParseError(msg)) => {
                assert!(msg.contains("Failed to parse HTML"));
            }
            _ => panic!("Expected ParseError"),
        }
    }
}