html-to-markdown 3.4.0 → 3.6.0.pre.rc.24

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (188) hide show
  1. checksums.yaml +4 -4
  2. data/LICENSE +21 -0
  3. data/README.md +347 -0
  4. data/Steepfile +10 -2
  5. data/ext/html_to_markdown_rb/Cargo.toml +3 -2
  6. data/ext/html_to_markdown_rb/extconf.rb +5 -5
  7. data/ext/html_to_markdown_rb/native/Cargo.lock +962 -0
  8. data/ext/html_to_markdown_rb/native/Cargo.toml +6 -11
  9. data/ext/html_to_markdown_rb/native/extconf.rb +14 -0
  10. data/ext/html_to_markdown_rb/src/lib.rs +1715 -646
  11. data/lib/html_to_markdown/native.rb +913 -37
  12. data/lib/html_to_markdown/version.rb +3 -3
  13. data/lib/html_to_markdown.rb +9 -4
  14. data/lib/html_to_markdown_rb.so +0 -0
  15. data/sig/types.rbs +59 -292
  16. metadata +32 -179
  17. data/ext/html_to_markdown_rb/Makefile +0 -592
  18. data/lib/bin/html-to-markdown +0 -0
  19. data/vendor/Cargo.toml +0 -33
  20. data/vendor/html-to-markdown-rs/Cargo.toml +0 -54
  21. data/vendor/html-to-markdown-rs/README.md +0 -278
  22. data/vendor/html-to-markdown-rs/examples/basic.rs +0 -24
  23. data/vendor/html-to-markdown-rs/examples/table.rs +0 -25
  24. data/vendor/html-to-markdown-rs/examples/test_deser.rs +0 -12
  25. data/vendor/html-to-markdown-rs/examples/test_escape.rs +0 -58
  26. data/vendor/html-to-markdown-rs/examples/test_inline_formatting.rs +0 -113
  27. data/vendor/html-to-markdown-rs/examples/test_lists.rs +0 -39
  28. data/vendor/html-to-markdown-rs/examples/test_semantic_tags.rs +0 -89
  29. data/vendor/html-to-markdown-rs/examples/test_tables.rs +0 -100
  30. data/vendor/html-to-markdown-rs/examples/test_task_lists.rs +0 -61
  31. data/vendor/html-to-markdown-rs/examples/test_whitespace.rs +0 -34
  32. data/vendor/html-to-markdown-rs/src/convert_api.rs +0 -349
  33. data/vendor/html-to-markdown-rs/src/converter/block/blockquote.rs +0 -178
  34. data/vendor/html-to-markdown-rs/src/converter/block/container.rs +0 -114
  35. data/vendor/html-to-markdown-rs/src/converter/block/div.rs +0 -149
  36. data/vendor/html-to-markdown-rs/src/converter/block/heading.rs +0 -428
  37. data/vendor/html-to-markdown-rs/src/converter/block/horizontal_rule.rs +0 -103
  38. data/vendor/html-to-markdown-rs/src/converter/block/line_break.rs +0 -89
  39. data/vendor/html-to-markdown-rs/src/converter/block/mod.rs +0 -10
  40. data/vendor/html-to-markdown-rs/src/converter/block/paragraph.rs +0 -140
  41. data/vendor/html-to-markdown-rs/src/converter/block/preformatted.rs +0 -298
  42. data/vendor/html-to-markdown-rs/src/converter/block/table/builder.rs +0 -453
  43. data/vendor/html-to-markdown-rs/src/converter/block/table/caption.rs +0 -44
  44. data/vendor/html-to-markdown-rs/src/converter/block/table/cell.rs +0 -276
  45. data/vendor/html-to-markdown-rs/src/converter/block/table/cells.rs +0 -336
  46. data/vendor/html-to-markdown-rs/src/converter/block/table/layout.rs +0 -58
  47. data/vendor/html-to-markdown-rs/src/converter/block/table/mod.rs +0 -266
  48. data/vendor/html-to-markdown-rs/src/converter/block/table/scanner.rs +0 -146
  49. data/vendor/html-to-markdown-rs/src/converter/block/table/utils.rs +0 -34
  50. data/vendor/html-to-markdown-rs/src/converter/block/unknown.rs +0 -138
  51. data/vendor/html-to-markdown-rs/src/converter/context.rs +0 -208
  52. data/vendor/html-to-markdown-rs/src/converter/dom_context.rs +0 -337
  53. data/vendor/html-to-markdown-rs/src/converter/form/elements.rs +0 -770
  54. data/vendor/html-to-markdown-rs/src/converter/form/mod.rs +0 -82
  55. data/vendor/html-to-markdown-rs/src/converter/format/djot.rs +0 -64
  56. data/vendor/html-to-markdown-rs/src/converter/format/markdown.rs +0 -59
  57. data/vendor/html-to-markdown-rs/src/converter/format/mod.rs +0 -43
  58. data/vendor/html-to-markdown-rs/src/converter/handlers/blockquote.rs +0 -173
  59. data/vendor/html-to-markdown-rs/src/converter/handlers/code_block.rs +0 -434
  60. data/vendor/html-to-markdown-rs/src/converter/handlers/graphic.rs +0 -234
  61. data/vendor/html-to-markdown-rs/src/converter/handlers/image.rs +0 -282
  62. data/vendor/html-to-markdown-rs/src/converter/handlers/link.rs +0 -316
  63. data/vendor/html-to-markdown-rs/src/converter/handlers/mod.rs +0 -26
  64. data/vendor/html-to-markdown-rs/src/converter/inline/code.rs +0 -306
  65. data/vendor/html-to-markdown-rs/src/converter/inline/emphasis.rs +0 -345
  66. data/vendor/html-to-markdown-rs/src/converter/inline/link.rs +0 -428
  67. data/vendor/html-to-markdown-rs/src/converter/inline/mod.rs +0 -237
  68. data/vendor/html-to-markdown-rs/src/converter/inline/ruby.rs +0 -337
  69. data/vendor/html-to-markdown-rs/src/converter/inline/semantic/marks.rs +0 -566
  70. data/vendor/html-to-markdown-rs/src/converter/inline/semantic/mod.rs +0 -86
  71. data/vendor/html-to-markdown-rs/src/converter/inline/semantic/typography.rs +0 -558
  72. data/vendor/html-to-markdown-rs/src/converter/list/definition.rs +0 -232
  73. data/vendor/html-to-markdown-rs/src/converter/list/item.rs +0 -332
  74. data/vendor/html-to-markdown-rs/src/converter/list/mod.rs +0 -70
  75. data/vendor/html-to-markdown-rs/src/converter/list/ordered.rs +0 -201
  76. data/vendor/html-to-markdown-rs/src/converter/list/unordered.rs +0 -195
  77. data/vendor/html-to-markdown-rs/src/converter/list/utils.rs +0 -314
  78. data/vendor/html-to-markdown-rs/src/converter/main.rs +0 -710
  79. data/vendor/html-to-markdown-rs/src/converter/main_helpers.rs +0 -452
  80. data/vendor/html-to-markdown-rs/src/converter/media/embedded.rs +0 -393
  81. data/vendor/html-to-markdown-rs/src/converter/media/graphic.rs +0 -4
  82. data/vendor/html-to-markdown-rs/src/converter/media/image.rs +0 -183
  83. data/vendor/html-to-markdown-rs/src/converter/media/mod.rs +0 -87
  84. data/vendor/html-to-markdown-rs/src/converter/media/svg.rs +0 -280
  85. data/vendor/html-to-markdown-rs/src/converter/metadata.rs +0 -220
  86. data/vendor/html-to-markdown-rs/src/converter/mod.rs +0 -156
  87. data/vendor/html-to-markdown-rs/src/converter/plain_text.rs +0 -516
  88. data/vendor/html-to-markdown-rs/src/converter/preprocessing_helpers.rs +0 -201
  89. data/vendor/html-to-markdown-rs/src/converter/reference_collector.rs +0 -69
  90. data/vendor/html-to-markdown-rs/src/converter/semantic/attributes.rs +0 -269
  91. data/vendor/html-to-markdown-rs/src/converter/semantic/definition_list.rs +0 -266
  92. data/vendor/html-to-markdown-rs/src/converter/semantic/figure.rs +0 -391
  93. data/vendor/html-to-markdown-rs/src/converter/semantic/mod.rs +0 -112
  94. data/vendor/html-to-markdown-rs/src/converter/semantic/sectioning.rs +0 -85
  95. data/vendor/html-to-markdown-rs/src/converter/semantic/summary.rs +0 -324
  96. data/vendor/html-to-markdown-rs/src/converter/text/mod.rs +0 -8
  97. data/vendor/html-to-markdown-rs/src/converter/text/processing.rs +0 -56
  98. data/vendor/html-to-markdown-rs/src/converter/text_node.rs +0 -269
  99. data/vendor/html-to-markdown-rs/src/converter/utility/attributes.rs +0 -151
  100. data/vendor/html-to-markdown-rs/src/converter/utility/caching.rs +0 -74
  101. data/vendor/html-to-markdown-rs/src/converter/utility/content.rs +0 -271
  102. data/vendor/html-to-markdown-rs/src/converter/utility/mod.rs +0 -17
  103. data/vendor/html-to-markdown-rs/src/converter/utility/preprocessing.rs +0 -1002
  104. data/vendor/html-to-markdown-rs/src/converter/utility/serialization.rs +0 -126
  105. data/vendor/html-to-markdown-rs/src/converter/utility/siblings.rs +0 -97
  106. data/vendor/html-to-markdown-rs/src/converter/visitor_hooks.rs +0 -189
  107. data/vendor/html-to-markdown-rs/src/error.rs +0 -43
  108. data/vendor/html-to-markdown-rs/src/exports.rs +0 -24
  109. data/vendor/html-to-markdown-rs/src/inline_images.rs +0 -336
  110. data/vendor/html-to-markdown-rs/src/lib.rs +0 -139
  111. data/vendor/html-to-markdown-rs/src/metadata/collector.rs +0 -457
  112. data/vendor/html-to-markdown-rs/src/metadata/config.rs +0 -394
  113. data/vendor/html-to-markdown-rs/src/metadata/extraction.rs +0 -398
  114. data/vendor/html-to-markdown-rs/src/metadata/mod.rs +0 -288
  115. data/vendor/html-to-markdown-rs/src/metadata/types.rs +0 -477
  116. data/vendor/html-to-markdown-rs/src/options/conversion.rs +0 -559
  117. data/vendor/html-to-markdown-rs/src/options/inline_image.rs +0 -111
  118. data/vendor/html-to-markdown-rs/src/options/mod.rs +0 -20
  119. data/vendor/html-to-markdown-rs/src/options/preprocessing.rs +0 -201
  120. data/vendor/html-to-markdown-rs/src/options/validation.rs +0 -416
  121. data/vendor/html-to-markdown-rs/src/prelude.rs +0 -1
  122. data/vendor/html-to-markdown-rs/src/rcdom.rs +0 -487
  123. data/vendor/html-to-markdown-rs/src/text.rs +0 -358
  124. data/vendor/html-to-markdown-rs/src/types/document.rs +0 -191
  125. data/vendor/html-to-markdown-rs/src/types/mod.rs +0 -17
  126. data/vendor/html-to-markdown-rs/src/types/result.rs +0 -54
  127. data/vendor/html-to-markdown-rs/src/types/structure_builder.rs +0 -791
  128. data/vendor/html-to-markdown-rs/src/types/structure_collector.rs +0 -483
  129. data/vendor/html-to-markdown-rs/src/types/tables.rs +0 -52
  130. data/vendor/html-to-markdown-rs/src/types/warnings.rs +0 -33
  131. data/vendor/html-to-markdown-rs/src/validation.rs +0 -158
  132. data/vendor/html-to-markdown-rs/src/visitor/default_impl.rs +0 -63
  133. data/vendor/html-to-markdown-rs/src/visitor/mod.rs +0 -41
  134. data/vendor/html-to-markdown-rs/src/visitor/traits.rs +0 -370
  135. data/vendor/html-to-markdown-rs/src/visitor/types.rs +0 -319
  136. data/vendor/html-to-markdown-rs/src/visitor_helpers/helpers/callbacks/mod.rs +0 -1
  137. data/vendor/html-to-markdown-rs/src/visitor_helpers/helpers/content.rs +0 -126
  138. data/vendor/html-to-markdown-rs/src/visitor_helpers/helpers/mod.rs +0 -27
  139. data/vendor/html-to-markdown-rs/src/visitor_helpers/helpers/state.rs +0 -110
  140. data/vendor/html-to-markdown-rs/src/visitor_helpers/helpers/traversal.rs +0 -250
  141. data/vendor/html-to-markdown-rs/src/visitor_helpers.rs +0 -597
  142. data/vendor/html-to-markdown-rs/src/wrapper/sync.rs +0 -413
  143. data/vendor/html-to-markdown-rs/src/wrapper/utils.rs +0 -290
  144. data/vendor/html-to-markdown-rs/src/wrapper.rs +0 -9
  145. data/vendor/html-to-markdown-rs/tests/br_in_inline_test.rs +0 -87
  146. data/vendor/html-to-markdown-rs/tests/commonmark_compliance_test.rs +0 -297
  147. data/vendor/html-to-markdown-rs/tests/djot_output_test.rs +0 -153
  148. data/vendor/html-to-markdown-rs/tests/exclude_selectors_test.rs +0 -132
  149. data/vendor/html-to-markdown-rs/tests/integration_test.rs +0 -631
  150. data/vendor/html-to-markdown-rs/tests/issue_121_regressions.rs +0 -49
  151. data/vendor/html-to-markdown-rs/tests/issue_127_regressions.rs +0 -58
  152. data/vendor/html-to-markdown-rs/tests/issue_128_regressions.rs +0 -17
  153. data/vendor/html-to-markdown-rs/tests/issue_131_regressions.rs +0 -41
  154. data/vendor/html-to-markdown-rs/tests/issue_134_regressions.rs +0 -40
  155. data/vendor/html-to-markdown-rs/tests/issue_139_regressions.rs +0 -26
  156. data/vendor/html-to-markdown-rs/tests/issue_140_regressions.rs +0 -185
  157. data/vendor/html-to-markdown-rs/tests/issue_143_regressions.rs +0 -100
  158. data/vendor/html-to-markdown-rs/tests/issue_145_regressions.rs +0 -133
  159. data/vendor/html-to-markdown-rs/tests/issue_146_regressions.rs +0 -144
  160. data/vendor/html-to-markdown-rs/tests/issue_176_regressions.rs +0 -62
  161. data/vendor/html-to-markdown-rs/tests/issue_190_regressions.rs +0 -128
  162. data/vendor/html-to-markdown-rs/tests/issue_199_regressions.rs +0 -20
  163. data/vendor/html-to-markdown-rs/tests/issue_200_regressions.rs +0 -62
  164. data/vendor/html-to-markdown-rs/tests/issue_212_regressions.rs +0 -68
  165. data/vendor/html-to-markdown-rs/tests/issue_216_217_regressions.rs +0 -87
  166. data/vendor/html-to-markdown-rs/tests/issue_336_regressions.rs +0 -74
  167. data/vendor/html-to-markdown-rs/tests/issue_339_regressions.rs +0 -92
  168. data/vendor/html-to-markdown-rs/tests/issue_347_regressions.rs +0 -154
  169. data/vendor/html-to-markdown-rs/tests/issue_348_visitor_plain.rs +0 -93
  170. data/vendor/html-to-markdown-rs/tests/json_ld_script_extraction.rs +0 -44
  171. data/vendor/html-to-markdown-rs/tests/lists_test.rs +0 -199
  172. data/vendor/html-to-markdown-rs/tests/plain_output_test.rs +0 -273
  173. data/vendor/html-to-markdown-rs/tests/preprocessing_tests.rs +0 -61
  174. data/vendor/html-to-markdown-rs/tests/reference_links_test.rs +0 -169
  175. data/vendor/html-to-markdown-rs/tests/sectioning_elements_test.rs +0 -137
  176. data/vendor/html-to-markdown-rs/tests/skip_images_test.rs +0 -522
  177. data/vendor/html-to-markdown-rs/tests/tables_test.rs +0 -743
  178. data/vendor/html-to-markdown-rs/tests/test_custom_elements.rs +0 -41
  179. data/vendor/html-to-markdown-rs/tests/test_issue_187.rs +0 -204
  180. data/vendor/html-to-markdown-rs/tests/test_issue_218.rs +0 -68
  181. data/vendor/html-to-markdown-rs/tests/test_issue_277.rs +0 -77
  182. data/vendor/html-to-markdown-rs/tests/test_max_depth.rs +0 -82
  183. data/vendor/html-to-markdown-rs/tests/test_nested_simple.rs +0 -45
  184. data/vendor/html-to-markdown-rs/tests/test_script_style_stripping.rs +0 -396
  185. data/vendor/html-to-markdown-rs/tests/test_spa_bisect.rs +0 -34
  186. data/vendor/html-to-markdown-rs/tests/visitor_code_integration_test.rs +0 -121
  187. data/vendor/html-to-markdown-rs/tests/visitor_integration_test.rs +0 -1190
  188. data/vendor/html-to-markdown-rs/tests/xml_tables_test.rs +0 -372
@@ -1,434 +0,0 @@
1
- //! Code and pre element handlers for HTML to Markdown conversion.
2
- //!
3
- //! Handles `<code>` and `<pre>` elements including:
4
- //! - Inline code with backtick formatting
5
- //! - Code block formatting (indented or fenced)
6
- //! - Language detection from class attributes
7
- //! - Whitespace normalization and dedenting
8
- //! - Visitor callback integration
9
-
10
- use crate::converter::Context;
11
- use crate::converter::dom_context::DomContext;
12
- use crate::converter::main::walk_node;
13
- use crate::converter::text::dedent_code_block;
14
- use crate::options::ConversionOptions;
15
-
16
- #[cfg(feature = "visitor")]
17
- #[cfg(feature = "visitor")]
18
- use crate::converter::utility::content::collect_tag_attributes;
19
- #[cfg(feature = "visitor")]
20
- use std::collections::BTreeMap;
21
-
22
- #[cfg(feature = "visitor")]
23
- use crate::converter::utility::serialization::serialize_node;
24
-
25
- /// Handle an inline `<code>` element and convert to Markdown.
26
- ///
27
- /// This handler processes inline code elements including:
28
- /// - Extracting code content and applying backtick delimiters
29
- /// - Handling backticks in content by using multiple delimiters
30
- /// - Invoking visitor callbacks when the visitor feature is enabled
31
- /// - Generating appropriate markdown output with proper escaping
32
- #[allow(clippy::too_many_arguments)]
33
- #[allow(clippy::too_many_lines)]
34
- #[cfg_attr(not(feature = "visitor"), allow(unused_variables))]
35
- pub fn handle_code(
36
- node_handle: &tl::NodeHandle,
37
- tag: &tl::HTMLTag,
38
- parser: &tl::Parser,
39
- output: &mut String,
40
- options: &ConversionOptions,
41
- ctx: &Context,
42
- depth: usize,
43
- dom_ctx: &DomContext,
44
- ) {
45
- let code_ctx = Context {
46
- in_code: true,
47
- ..ctx.clone()
48
- };
49
-
50
- if ctx.in_code {
51
- let children = tag.children();
52
- {
53
- for child_handle in children.top().iter() {
54
- walk_node(child_handle, parser, output, options, &code_ctx, depth + 1, dom_ctx);
55
- }
56
- }
57
- } else {
58
- let mut content = String::with_capacity(32);
59
- let children = tag.children();
60
- {
61
- for child_handle in children.top().iter() {
62
- walk_node(
63
- child_handle,
64
- parser,
65
- &mut content,
66
- options,
67
- &code_ctx,
68
- depth + 1,
69
- dom_ctx,
70
- );
71
- }
72
- }
73
-
74
- let trimmed = &content;
75
-
76
- if !content.trim().is_empty() {
77
- #[cfg(feature = "visitor")]
78
- let code_output = if let Some(ref visitor_handle) = ctx.visitor {
79
- use crate::visitor::{NodeContext, NodeType, VisitResult};
80
-
81
- let attributes: BTreeMap<String, String> = collect_tag_attributes(tag);
82
-
83
- let node_id = node_handle.get_inner();
84
- let parent_tag = dom_ctx.parent_tag_name(node_id, parser);
85
- let index_in_parent = dom_ctx.get_sibling_index(node_id).unwrap_or(0);
86
-
87
- let node_ctx = NodeContext {
88
- node_type: NodeType::Code,
89
- tag_name: "code".to_string(),
90
- attributes,
91
- depth,
92
- index_in_parent,
93
- parent_tag,
94
- is_inline: true,
95
- };
96
-
97
- let visit_result = {
98
- let mut visitor = visitor_handle.borrow_mut();
99
- visitor.visit_code_inline(&node_ctx, trimmed)
100
- };
101
- match visit_result {
102
- VisitResult::Continue => None,
103
- VisitResult::Custom(custom) => Some(custom),
104
- VisitResult::Skip => Some(String::new()),
105
- VisitResult::PreserveHtml => Some(serialize_node(node_handle, parser)),
106
- VisitResult::Error(err) => {
107
- if ctx.visitor_error.borrow().is_none() {
108
- *ctx.visitor_error.borrow_mut() = Some(err);
109
- }
110
- None
111
- }
112
- }
113
- } else {
114
- None
115
- };
116
-
117
- #[cfg(feature = "visitor")]
118
- if let Some(custom_output) = code_output {
119
- output.push_str(&custom_output);
120
- } else {
121
- format_inline_code(trimmed, output);
122
- }
123
-
124
- #[cfg(not(feature = "visitor"))]
125
- {
126
- format_inline_code(trimmed, output);
127
- }
128
- }
129
- }
130
- }
131
-
132
- /// Handle a `<pre>` element and convert to Markdown.
133
- ///
134
- /// This handler processes code block elements including:
135
- /// - Extracting language information from class attributes
136
- /// - Processing whitespace and dedenting code content
137
- /// - Supporting multiple code block styles (indented, backticks, tildes)
138
- /// - Invoking visitor callbacks when the visitor feature is enabled
139
- /// - Generating appropriate markdown output
140
- #[allow(clippy::too_many_arguments)]
141
- #[allow(clippy::too_many_lines)]
142
- #[cfg_attr(not(feature = "visitor"), allow(unused_variables))]
143
- pub fn handle_pre(
144
- node_handle: &tl::NodeHandle,
145
- tag: &tl::HTMLTag,
146
- parser: &tl::Parser,
147
- output: &mut String,
148
- options: &ConversionOptions,
149
- ctx: &Context,
150
- depth: usize,
151
- dom_ctx: &DomContext,
152
- ) {
153
- let code_ctx = Context {
154
- in_code: true,
155
- ..ctx.clone()
156
- };
157
-
158
- #[cfg_attr(not(feature = "visitor"), allow(unused_variables))]
159
- let language: Option<String> = {
160
- let mut lang: Option<String> = None;
161
-
162
- // First, try to extract language from <pre> tag's class attribute
163
- if let Some(class_attr) = tag.attributes().get("class") {
164
- if let Some(class_bytes) = class_attr {
165
- let class_str = class_bytes.as_utf8_str();
166
- for cls in class_str.split_whitespace() {
167
- if let Some(stripped) = cls.strip_prefix("language-") {
168
- lang = Some(String::from(stripped));
169
- break;
170
- } else if let Some(stripped) = cls.strip_prefix("lang-") {
171
- lang = Some(String::from(stripped));
172
- break;
173
- }
174
- }
175
- }
176
- }
177
-
178
- // If not found on <pre>, try to extract from nested <code> tag's class attribute
179
- if lang.is_none() {
180
- let children = tag.children();
181
- for child_handle in children.top().iter() {
182
- if let Some(tl::Node::Tag(child_tag)) = child_handle.get(parser) {
183
- if child_tag.name() == "code" {
184
- if let Some(class_attr) = child_tag.attributes().get("class") {
185
- if let Some(class_bytes) = class_attr {
186
- let class_str = class_bytes.as_utf8_str();
187
- for cls in class_str.split_whitespace() {
188
- if let Some(stripped) = cls.strip_prefix("language-") {
189
- lang = Some(String::from(stripped));
190
- break;
191
- } else if let Some(stripped) = cls.strip_prefix("lang-") {
192
- lang = Some(String::from(stripped));
193
- break;
194
- }
195
- }
196
- }
197
- }
198
- break;
199
- }
200
- }
201
- }
202
- }
203
-
204
- lang
205
- };
206
-
207
- let mut content = String::with_capacity(256);
208
- let children = tag.children();
209
- {
210
- for child_handle in children.top().iter() {
211
- walk_node(
212
- child_handle,
213
- parser,
214
- &mut content,
215
- options,
216
- &code_ctx,
217
- depth + 1,
218
- dom_ctx,
219
- );
220
- }
221
- }
222
-
223
- if !content.is_empty() {
224
- let leading_newlines = content.chars().take_while(|&c| c == '\n').count();
225
- let trailing_newlines = content.chars().rev().take_while(|&c| c == '\n').count();
226
- let core = content.trim_matches('\n');
227
- let is_whitespace_only = core.trim().is_empty();
228
-
229
- let processed_content = if options.whitespace_mode == crate::options::WhitespaceMode::Strict {
230
- content
231
- } else {
232
- // Always dedent code blocks to remove common leading whitespace
233
- let mut core_text = dedent_code_block(core);
234
-
235
- if is_whitespace_only {
236
- let mut rebuilt = String::new();
237
- for _ in 0..leading_newlines {
238
- rebuilt.push('\n');
239
- }
240
- rebuilt.push_str(&core_text);
241
- for _ in 0..trailing_newlines {
242
- rebuilt.push('\n');
243
- }
244
- rebuilt
245
- } else {
246
- for _ in 0..trailing_newlines {
247
- core_text.push('\n');
248
- }
249
- core_text
250
- }
251
- };
252
-
253
- #[cfg(feature = "visitor")]
254
- let code_block_output = if let Some(ref visitor_handle) = ctx.visitor {
255
- use crate::visitor::{NodeContext, NodeType, VisitResult};
256
-
257
- let attributes: BTreeMap<String, String> = collect_tag_attributes(tag);
258
-
259
- let node_id = node_handle.get_inner();
260
- let parent_tag = dom_ctx.parent_tag_name(node_id, parser);
261
- let index_in_parent = dom_ctx.get_sibling_index(node_id).unwrap_or(0);
262
-
263
- let node_ctx = NodeContext {
264
- node_type: NodeType::Pre,
265
- tag_name: "pre".to_string(),
266
- attributes,
267
- depth,
268
- index_in_parent,
269
- parent_tag,
270
- is_inline: false,
271
- };
272
-
273
- let visit_result = {
274
- let mut visitor = visitor_handle.borrow_mut();
275
- visitor.visit_code_block(&node_ctx, language.as_deref(), &processed_content)
276
- };
277
- match visit_result {
278
- VisitResult::Continue => None,
279
- VisitResult::Custom(custom) => Some(custom),
280
- VisitResult::Skip => Some(String::new()),
281
- VisitResult::PreserveHtml => Some(serialize_node(node_handle, parser)),
282
- VisitResult::Error(err) => {
283
- if ctx.visitor_error.borrow().is_none() {
284
- *ctx.visitor_error.borrow_mut() = Some(err);
285
- }
286
- None
287
- }
288
- }
289
- } else {
290
- None
291
- };
292
-
293
- #[cfg(feature = "visitor")]
294
- if let Some(custom_output) = code_block_output {
295
- output.push_str(&custom_output);
296
- } else {
297
- format_code_block(&processed_content, language.as_deref(), output, options, ctx);
298
- }
299
-
300
- #[cfg(not(feature = "visitor"))]
301
- {
302
- format_code_block(&processed_content, language.as_deref(), output, options, ctx);
303
- }
304
-
305
- if let Some(ref sc) = ctx.structure_collector {
306
- sc.borrow_mut().push_code(&processed_content, language.as_deref());
307
- }
308
- }
309
- }
310
-
311
- /// Format inline code with appropriate backtick delimiters.
312
- ///
313
- /// Handles:
314
- /// - Single backticks for normal content
315
- /// - Double backticks when content contains backticks
316
- /// - Space padding when needed to avoid backtick adjacency
317
- fn format_inline_code(content: &str, output: &mut String) {
318
- let contains_backtick = content.contains('`');
319
-
320
- let needs_delimiter_spaces = {
321
- let first_char = content.chars().next();
322
- let last_char = content.chars().last();
323
- let starts_with_space = first_char == Some(' ');
324
- let ends_with_space = last_char == Some(' ');
325
- let starts_with_backtick = first_char == Some('`');
326
- let ends_with_backtick = last_char == Some('`');
327
- let all_spaces = content.chars().all(|c| c == ' ');
328
-
329
- all_spaces
330
- || starts_with_backtick
331
- || ends_with_backtick
332
- || (starts_with_space && ends_with_space && contains_backtick)
333
- };
334
-
335
- let (num_backticks, needs_spaces) = if contains_backtick {
336
- let max_consecutive = content
337
- .chars()
338
- .fold((0, 0), |(max, current), c| {
339
- if c == '`' {
340
- let new_current = current + 1;
341
- (max.max(new_current), new_current)
342
- } else {
343
- (max, 0)
344
- }
345
- })
346
- .0;
347
- let num = if max_consecutive == 1 { 2 } else { 1 };
348
- (num, needs_delimiter_spaces)
349
- } else {
350
- (1, needs_delimiter_spaces)
351
- };
352
-
353
- for _ in 0..num_backticks {
354
- output.push('`');
355
- }
356
- if needs_spaces {
357
- output.push(' ');
358
- }
359
- output.push_str(content);
360
- if needs_spaces {
361
- output.push(' ');
362
- }
363
- for _ in 0..num_backticks {
364
- output.push('`');
365
- }
366
- }
367
-
368
- /// Format a code block with the specified style and language.
369
- ///
370
- /// Supports:
371
- /// - Indented style (4-space indentation)
372
- /// - Fenced style with backticks (```language)
373
- /// - Fenced style with tildes (~~~language)
374
- fn format_code_block(
375
- content: &str,
376
- language: Option<&str>,
377
- output: &mut String,
378
- options: &ConversionOptions,
379
- ctx: &Context,
380
- ) {
381
- match options.code_block_style {
382
- crate::options::CodeBlockStyle::Indented => {
383
- if !ctx.convert_as_inline && !output.is_empty() && !output.ends_with("\n\n") {
384
- if output.ends_with('\n') {
385
- output.push('\n');
386
- } else {
387
- output.push_str("\n\n");
388
- }
389
- }
390
-
391
- let indented = content
392
- .lines()
393
- .map(|line| {
394
- if line.is_empty() {
395
- String::new()
396
- } else {
397
- format!(" {line}")
398
- }
399
- })
400
- .collect::<Vec<_>>()
401
- .join("\n");
402
- output.push_str(&indented);
403
-
404
- output.push_str("\n\n");
405
- }
406
- crate::options::CodeBlockStyle::Backticks | crate::options::CodeBlockStyle::Tildes => {
407
- if !ctx.convert_as_inline && !output.is_empty() && !output.ends_with("\n\n") {
408
- if output.ends_with('\n') {
409
- output.push('\n');
410
- } else {
411
- output.push_str("\n\n");
412
- }
413
- }
414
-
415
- let fence = if options.code_block_style == crate::options::CodeBlockStyle::Backticks {
416
- "```"
417
- } else {
418
- "~~~"
419
- };
420
-
421
- output.push_str(fence);
422
- if let Some(lang) = language {
423
- output.push_str(lang);
424
- } else if !options.code_language.is_empty() {
425
- output.push_str(&options.code_language);
426
- }
427
- output.push('\n');
428
- output.push_str(content);
429
- output.push('\n');
430
- output.push_str(fence);
431
- output.push('\n');
432
- }
433
- }
434
- }
@@ -1,234 +0,0 @@
1
- //! Graphic element handler for HTML to Markdown conversion.
2
- //!
3
- //! Handles `<graphic>` elements including:
4
- //! - Alternative source attributes (url, href, xlink:href, src)
5
- //! - Fallback alt text from filename attribute
6
- //! - Metadata collection for graphic extraction
7
- //! - Visitor callback integration
8
-
9
- use std::borrow::Cow;
10
- #[cfg(any(feature = "metadata", feature = "visitor"))]
11
- use std::collections::BTreeMap;
12
-
13
- use crate::converter::Context;
14
- use crate::converter::dom_context::DomContext;
15
- #[cfg(feature = "visitor")]
16
- use crate::converter::utility::content::collect_tag_attributes;
17
- use crate::options::ConversionOptions;
18
-
19
- #[cfg(feature = "visitor")]
20
- use crate::converter::utility::serialization::serialize_node;
21
-
22
- #[cfg(feature = "metadata")]
23
- type GraphicMetadataPayload = (BTreeMap<String, String>, Option<u32>, Option<u32>);
24
-
25
- /// Handle a `<graphic>` element and convert to Markdown.
26
- ///
27
- /// This handler processes graphic elements including:
28
- /// - Extracting source from url, href, xlink:href, or src attributes
29
- /// - Using alt attribute, with fallback to filename
30
- /// - Collecting metadata when the metadata feature is enabled
31
- /// - Invoking visitor callbacks when the visitor feature is enabled
32
- /// - Generating appropriate markdown output
33
- #[allow(clippy::too_many_arguments)]
34
- #[allow(clippy::too_many_lines)]
35
- #[cfg_attr(not(feature = "visitor"), allow(unused_variables))]
36
- pub fn handle_graphic(
37
- node_handle: &tl::NodeHandle,
38
- tag: &tl::HTMLTag,
39
- parser: &tl::Parser,
40
- output: &mut String,
41
- options: &ConversionOptions,
42
- ctx: &Context,
43
- depth: usize,
44
- dom_ctx: &DomContext,
45
- ) {
46
- // Check source attributes in order: url, href, xlink:href, src
47
- let src = tag
48
- .attributes()
49
- .get("url")
50
- .flatten()
51
- .or_else(|| tag.attributes().get("href").flatten())
52
- .or_else(|| tag.attributes().get("xlink:href").flatten())
53
- .or_else(|| tag.attributes().get("src").flatten())
54
- .map_or(Cow::Borrowed(""), |v| v.as_utf8_str());
55
-
56
- // Use "alt" attribute, fallback to "filename"
57
- let alt = tag
58
- .attributes()
59
- .get("alt")
60
- .flatten()
61
- .map(|v| v.as_utf8_str())
62
- .or_else(|| tag.attributes().get("filename").flatten().map(|v| v.as_utf8_str()))
63
- .unwrap_or(Cow::Borrowed(""));
64
-
65
- let title = tag.attributes().get("title").flatten().map(|v| v.as_utf8_str());
66
-
67
- // Collect metadata payload if metadata feature is enabled
68
- #[cfg(feature = "metadata")]
69
- #[allow(clippy::useless_let_if_seq)]
70
- let mut metadata_payload: Option<GraphicMetadataPayload> = None;
71
- #[cfg(feature = "metadata")]
72
- if ctx.metadata_wants_images {
73
- let mut attributes_map = BTreeMap::new();
74
- let mut width: Option<u32> = None;
75
- let mut height: Option<u32> = None;
76
- for (key, value_opt) in tag.attributes().iter() {
77
- let key_str = key.to_string();
78
- if key_str == "url" || key_str == "href" || key_str == "xlink:href" || key_str == "src" {
79
- continue;
80
- }
81
- let value = value_opt.map(|v| v.to_string()).unwrap_or_default();
82
- if key_str == "width" {
83
- if let Ok(parsed) = value.parse::<u32>() {
84
- width = Some(parsed);
85
- }
86
- } else if key_str == "height" {
87
- if let Ok(parsed) = value.parse::<u32>() {
88
- height = Some(parsed);
89
- }
90
- }
91
- attributes_map.insert(key_str, value);
92
- }
93
- metadata_payload = Some((attributes_map, width, height));
94
- }
95
-
96
- let keep_as_markdown = ctx.in_heading && ctx.heading_allow_inline_images;
97
-
98
- let should_use_alt_text =
99
- !keep_as_markdown && (ctx.convert_as_inline || (ctx.in_heading && !ctx.heading_allow_inline_images));
100
-
101
- // Generate graphic output with visitor integration
102
- #[cfg(feature = "visitor")]
103
- let graphic_output = if let Some(ref visitor_handle) = ctx.visitor {
104
- use crate::visitor::{NodeContext, NodeType, VisitResult};
105
-
106
- let attributes: BTreeMap<String, String> = collect_tag_attributes(tag);
107
-
108
- let node_id = node_handle.get_inner();
109
- let parent_tag = dom_ctx.parent_tag_name(node_id, parser);
110
- let index_in_parent = dom_ctx.get_sibling_index(node_id).unwrap_or(0);
111
-
112
- let node_ctx = NodeContext {
113
- node_type: NodeType::Image,
114
- tag_name: "graphic".to_string(),
115
- attributes,
116
- depth,
117
- index_in_parent,
118
- parent_tag,
119
- is_inline: true,
120
- };
121
-
122
- let visit_result = {
123
- let mut visitor = visitor_handle.borrow_mut();
124
- visitor.visit_image(&node_ctx, &src, &alt, title.as_deref())
125
- };
126
- match visit_result {
127
- VisitResult::Continue => Some(format_graphic_markdown(
128
- &src,
129
- &alt,
130
- title.as_deref(),
131
- should_use_alt_text,
132
- options.link_style,
133
- ctx.reference_collector.as_ref(),
134
- )),
135
- VisitResult::Custom(custom) => Some(custom),
136
- VisitResult::Skip => None,
137
- VisitResult::Error(err) => {
138
- if ctx.visitor_error.borrow().is_none() {
139
- *ctx.visitor_error.borrow_mut() = Some(err);
140
- }
141
- None
142
- }
143
- VisitResult::PreserveHtml => Some(serialize_node(node_handle, parser)),
144
- }
145
- } else {
146
- Some(format_graphic_markdown(
147
- &src,
148
- &alt,
149
- title.as_deref(),
150
- should_use_alt_text,
151
- options.link_style,
152
- ctx.reference_collector.as_ref(),
153
- ))
154
- };
155
-
156
- #[cfg(not(feature = "visitor"))]
157
- let graphic_output = Some(format_graphic_markdown(
158
- &src,
159
- &alt,
160
- title.as_deref(),
161
- should_use_alt_text,
162
- options.link_style,
163
- ctx.reference_collector.as_ref(),
164
- ));
165
-
166
- if !options.skip_images {
167
- if let Some(graphic_text) = graphic_output {
168
- output.push_str(&graphic_text);
169
- }
170
- }
171
-
172
- // Add graphic to metadata collector
173
- #[cfg(feature = "metadata")]
174
- if ctx.metadata_wants_images {
175
- if let Some(ref collector) = ctx.metadata_collector {
176
- if let Some((attributes_map, width, height)) = metadata_payload {
177
- if !src.is_empty() {
178
- let dimensions = match (width, height) {
179
- (Some(w), Some(h)) => Some((w, h)),
180
- _ => None,
181
- };
182
- collector.borrow_mut().add_image(
183
- src.to_string(),
184
- if alt.is_empty() { None } else { Some(alt.to_string()) },
185
- title.as_deref().map(std::string::ToString::to_string),
186
- dimensions,
187
- attributes_map,
188
- );
189
- }
190
- }
191
- }
192
- }
193
- }
194
-
195
- /// Format a graphic as Markdown syntax.
196
- ///
197
- /// If `use_alt_only` is true, returns just the alt text.
198
- /// Otherwise returns the full `![alt](src "title")` syntax.
199
- fn format_graphic_markdown(
200
- src: &str,
201
- alt: &str,
202
- title: Option<&str>,
203
- use_alt_only: bool,
204
- link_style: crate::options::validation::LinkStyle,
205
- reference_collector: Option<&crate::converter::reference_collector::ReferenceCollectorHandle>,
206
- ) -> String {
207
- if use_alt_only {
208
- return alt.to_string();
209
- }
210
- if link_style == crate::options::validation::LinkStyle::Reference {
211
- if let Some(collector) = reference_collector {
212
- let ref_num = collector.borrow_mut().get_or_insert(src, title);
213
- let mut buf = String::with_capacity(alt.len() + 10);
214
- buf.push_str("![");
215
- buf.push_str(alt);
216
- buf.push_str("][");
217
- buf.push_str(&ref_num.to_string());
218
- buf.push(']');
219
- return buf;
220
- }
221
- }
222
- let mut buf = String::with_capacity(src.len() + alt.len() + 10);
223
- buf.push_str("![");
224
- buf.push_str(alt);
225
- buf.push_str("](");
226
- buf.push_str(src);
227
- if let Some(title_text) = title {
228
- buf.push_str(" \"");
229
- buf.push_str(title_text);
230
- buf.push('"');
231
- }
232
- buf.push(')');
233
- buf
234
- }