html-generator 0.0.10

A robust Rust library designed for transforming Markdown into SEO-optimized, accessible HTML. Featuring front matter extraction, custom header processing, table of contents generation, and performance optimization for web projects of any scale.
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
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382

// SPDX-License-Identifier: MIT OR Apache-2.0
// Copyright (c) 2026 html-generator contributors

//! Dependency-free HTML minification.
//!
//! Replaces the `minify-html` crate, which reached
//! `rkyv 0.7` (RUSTSEC-2026-0235, out-of-bounds reads) through
//! `lightningcss` -> `parcel_sourcemap`. No upstream fix exists:
//! `parcel_sourcemap` 2.1.1 is the latest release and pins
//! `rkyv ^0.7.38`.
//!
//! # What it does
//!
//! - drops comments, except conditional comments (`<!--[if ...]>`),
//!   which are markup, not commentary
//! - drops runs of whitespace between tags where the whitespace cannot
//!   affect rendering
//! - collapses other whitespace runs to a single space
//! - leaves `<pre>`, `<textarea>`, `<script>` and `<style>` bodies byte
//!   for byte
//!
//! # What it deliberately does not do
//!
//! The previous configuration also set `minify_css` and `minify_js`.
//! Those are what pulled in `lightningcss`, and re-implementing a CSS
//! or JS minifier is a different problem with a much larger surface for
//! getting it wrong. Style and script bodies are passed through
//! unchanged. HTML structure is where the bulk of the saving is for
//! generated documents, and correctness matters more here than the last
//! few percent.
//!
//! # Why whitespace handling is element-aware
//!
//! Removing whitespace between tags is only safe when it cannot be
//! rendered. Between block-level elements it never can:
//!
//! ```text
//! <body>  <p>x</p>  </body>   ->   <body><p>x</p></body>
//! ```
//!
//! Between inline elements it can, and dropping it changes what the
//! reader sees:
//!
//! ```text
//! <span>a</span> <span>b</span>   ->   "a b", not "ab"
//! ```
//!
//! So whitespace is dropped only when the tag on at least one side is
//! block-level, and collapsed to a single space otherwise.

use crate::Result;

/// Elements whose surrounding whitespace cannot affect rendering.
///
/// Block-level and structural elements only. Anything not listed is
/// treated as inline, which is the conservative direction: an unknown
/// element keeps its whitespace.
const BLOCK_ELEMENTS: &[&str] = &[
    "address",
    "article",
    "aside",
    "blockquote",
    "body",
    "br",
    "canvas",
    "caption",
    "col",
    "colgroup",
    "dd",
    "details",
    "div",
    "dl",
    "dt",
    "fieldset",
    "figcaption",
    "figure",
    "footer",
    "form",
    "h1",
    "h2",
    "h3",
    "h4",
    "h5",
    "h6",
    "head",
    "header",
    "hr",
    "html",
    "legend",
    "li",
    "main",
    "meta",
    "nav",
    "ol",
    "option",
    "p",
    "pre",
    "script",
    "section",
    "style",
    "summary",
    "table",
    "tbody",
    "td",
    "tfoot",
    "th",
    "thead",
    "title",
    "tr",
    "ul",
];

/// Elements whose content is preserved byte for byte.
const RAW_TEXT_ELEMENTS: &[&str] =
    &["pre", "textarea", "script", "style"];

/// Minify an HTML document.
///
/// # Errors
///
/// Never fails on well-formed input; the `Result` keeps the signature
/// compatible with callers that already handle one.
pub(crate) fn minify(input: &str) -> Result<String> {
    let bytes = input.as_bytes();
    let mut out = String::with_capacity(input.len());
    let mut i = 0usize;
    // Element name of the most recent tag, for the whitespace policy.
    let mut prev_tag: Option<String> = None;

    while i < bytes.len() {
        if bytes[i] == b'<' {
            // Comment?
            if input[i..].starts_with("<!--") {
                let is_conditional = input[i..].starts_with("<!--[if");
                let end = input[i..]
                    .find("-->")
                    .map_or(bytes.len(), |e| i + e + 3);
                if is_conditional {
                    // Conditional comments are markup: keep verbatim.
                    out.push_str(&input[i..end]);
                }
                i = end;
                continue;
            }

            // A tag. Copy it through unchanged — attributes are never
            // rewritten, which is what `allow_removing_spaces_between_
            // attributes = false` and the unquoted-value setting asked
            // for.
            let Some(rel_end) = input[i..].find('>') else {
                // Unterminated tag: emit the remainder as-is rather
                // than guess.
                out.push_str(&input[i..]);
                break;
            };
            let end = i + rel_end + 1;
            let tag = &input[i..end];
            out.push_str(tag);

            let name = tag_name(tag);

            // A raw-text element's body is copied verbatim.
            if let Some(ref n) = name {
                if RAW_TEXT_ELEMENTS.contains(&n.as_str())
                    && !tag.starts_with("</")
                    && !tag.ends_with("/>")
                {
                    let close = format!("</{n}");
                    if let Some(rel) = input[end..].find(&close) {
                        out.push_str(&input[end..end + rel]);
                        i = end + rel;
                        prev_tag = name;
                        continue;
                    }
                }
            }

            prev_tag = name;
            i = end;
            continue;
        }

        // A text run, up to the next tag.
        let next = input[i..].find('<').map_or(bytes.len(), |n| i + n);
        let text = &input[i..next];
        i = next;

        if text.trim().is_empty() {
            // Pure whitespace. Droppable only when a block-level tag
            // sits on one side of it.
            let next_tag = tag_name_at(input, i);
            let droppable = is_block(prev_tag.as_deref())
                || is_block(next_tag.as_deref());
            if !droppable && !text.is_empty() {
                out.push(' ');
            }
            continue;
        }

        // Text with content: collapse internal whitespace runs, and
        // keep a single leading/trailing space where one was present —
        // that space is rendered.
        out.push_str(&collapse_whitespace(text));
    }

    Ok(out)
}

/// Collapse every run of whitespace to one space, preserving whether
/// the run touched either end.
fn collapse_whitespace(text: &str) -> String {
    let mut out = String::with_capacity(text.len());
    let mut in_ws = false;
    for ch in text.chars() {
        if ch.is_whitespace() {
            if !in_ws {
                out.push(' ');
                in_ws = true;
            }
        } else {
            out.push(ch);
            in_ws = false;
        }
    }
    out
}

/// Lower-cased element name of a tag, if it has one.
fn tag_name(tag: &str) -> Option<String> {
    let body = tag
        .trim_start_matches('<')
        .trim_start_matches('/')
        .trim_end_matches('>')
        .trim_end_matches('/');
    if body.starts_with('!') || body.starts_with('?') {
        return None; // doctype or processing instruction
    }
    let name: String = body
        .chars()
        .take_while(|c| c.is_ascii_alphanumeric() || *c == '-')
        .collect();
    if name.is_empty() {
        None
    } else {
        Some(name.to_ascii_lowercase())
    }
}

/// Element name of the tag starting at `idx`, if one starts there.
fn tag_name_at(input: &str, idx: usize) -> Option<String> {
    if !input[idx..].starts_with('<') {
        return None;
    }
    let end = input[idx..].find('>')? + idx + 1;
    tag_name(&input[idx..end])
}

fn is_block(name: Option<&str>) -> bool {
    // No tag on that side means a document boundary, where whitespace
    // is not rendered either.
    // `map_or` rather than `is_none_or`: the latter is stable only
    // from 1.82 and this crate's MSRV is 1.80.
    name.map_or(true, |n| BLOCK_ELEMENTS.contains(&n))
}

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

    fn m(s: &str) -> String {
        minify(s).expect("minify")
    }

    #[test]
    fn collapses_whitespace_between_block_elements() {
        assert_eq!(
            m("<html>  <body>    <p>Test</p>  </body>  </html>"),
            "<html><body><p>Test</p></body></html>"
        );
    }

    #[test]
    fn keeps_whitespace_between_inline_elements() {
        // The reason the policy is element-aware. Dropping this space
        // renders "ab" instead of "a b".
        assert_eq!(
            m("<span>a</span> <span>b</span>"),
            "<span>a</span> <span>b</span>"
        );
        assert_eq!(
            m("<em>x</em>   <em>y</em>"),
            "<em>x</em> <em>y</em>"
        );
    }

    #[test]
    fn drops_comments_but_keeps_conditional_ones() {
        assert_eq!(
            m("<html><!-- note --><body><p>T</p></body></html>"),
            "<html><body><p>T</p></body></html>"
        );
        let cond = "<!--[if IE]><p>old</p><![endif]-->";
        assert!(
            m(cond).contains("[if IE]"),
            "conditional comments are markup and must survive"
        );
    }

    #[test]
    fn preserves_raw_text_bodies_byte_for_byte() {
        let pre = "<pre>  two  spaces\n  and a newline</pre>";
        assert_eq!(m(pre), pre, "pre content must not be touched");

        let script = "<script>const a = 1;   const b = 2;</script>";
        assert_eq!(m(script), script, "script bodies pass through");

        let style = "<style>body {  color : red ; }</style>";
        assert_eq!(m(style), style, "style bodies pass through");

        let ta = "<textarea>  keep   me  </textarea>";
        assert_eq!(m(ta), ta, "textarea content is user-visible");
    }

    #[test]
    fn preserves_utf8_and_in_text_spacing() {
        assert_eq!(
            m("<html><body><p>Test 你好 🦀</p></body></html>"),
            "<html><body><p>Test 你好 🦀</p></body></html>"
        );
    }

    #[test]
    fn collapses_runs_inside_text() {
        assert_eq!(m("<p>a     b</p>"), "<p>a b</p>");
    }

    #[test]
    fn leaves_entities_alone() {
        let s = "<div>&lt;Special&gt; &amp; Characters</div>";
        assert_eq!(m(s), s);
    }

    #[test]
    fn leaves_attributes_untouched() {
        let s = r#"<a href="x.html"  class="a b"  data-x='1'>t</a>"#;
        assert_eq!(m(s), s, "attribute text is never rewritten");
    }

    #[test]
    fn keeps_the_doctype() {
        let s = "<!DOCTYPE html><html><body><p>x</p></body></html>";
        assert_eq!(m(s), s);
    }

    #[test]
    fn handles_unterminated_tag_without_panicking() {
        // Emitting the remainder verbatim beats guessing at a fix.
        assert_eq!(m("<p>ok</p><div"), "<p>ok</p><div");
    }

    #[test]
    fn empty_and_whitespace_only_inputs() {
        assert_eq!(m(""), "");
        assert_eq!(m("   \n  "), "");
    }

    #[test]
    fn actually_reduces_size_on_a_realistic_document() {
        let src =
            "<html>\n  <head>\n    <title>T</title>\n  </head>\n  \
                   <body>\n    <!-- hi -->\n    <p>Hello</p>\n  \
                   </body>\n</html>";
        let out = m(src);
        assert!(
            out.len() < src.len(),
            "{} !< {}",
            out.len(),
            src.len()
        );
        assert!(!out.contains("<!--"), "comments removed");
    }
}