rusteerio 0.0.0-stage → 0.1.0

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 (206) hide show
  1. package/CONTRIBUTING.md +73 -0
  2. package/Cargo.lock +1051 -0
  3. package/Cargo.toml +36 -0
  4. package/LICENSE +21 -0
  5. package/README.md +252 -2
  6. package/RELEASING.md +138 -0
  7. package/THIRD_PARTY_NOTICES.md +7443 -0
  8. package/build.rs +3 -0
  9. package/dist/chunk-DKULX6VO.mjs +3185 -0
  10. package/dist/chunk-DKULX6VO.mjs.map +1 -0
  11. package/dist/chunk-OHPGEY3E.mjs +32 -0
  12. package/dist/chunk-OHPGEY3E.mjs.map +1 -0
  13. package/dist/index.cjs +3230 -0
  14. package/dist/index.cjs.map +1 -0
  15. package/dist/index.d.mts +113 -0
  16. package/dist/index.d.ts +113 -0
  17. package/dist/index.mjs +38 -0
  18. package/dist/index.mjs.map +1 -0
  19. package/dist/slim.cjs +2441 -0
  20. package/dist/slim.cjs.map +1 -0
  21. package/dist/slim.d.mts +11 -0
  22. package/dist/slim.d.ts +11 -0
  23. package/dist/slim.mjs +19 -0
  24. package/dist/slim.mjs.map +1 -0
  25. package/dist/types-BP6TpxE5.d.mts +532 -0
  26. package/dist/types-BP6TpxE5.d.ts +532 -0
  27. package/dist/utils.cjs +61 -0
  28. package/dist/utils.cjs.map +1 -0
  29. package/dist/utils.d.mts +13 -0
  30. package/dist/utils.d.ts +13 -0
  31. package/dist/utils.mjs +17 -0
  32. package/dist/utils.mjs.map +1 -0
  33. package/native/src/arena.rs +1781 -0
  34. package/native/src/attributes.rs +923 -0
  35. package/native/src/codec.rs +172 -0
  36. package/native/src/diagnostics.rs +1249 -0
  37. package/native/src/extraction.rs +226 -0
  38. package/native/src/gc.rs +166 -0
  39. package/native/src/hook_sink.rs +727 -0
  40. package/native/src/hook_stream.rs +513 -0
  41. package/native/src/lib.rs +2095 -0
  42. package/native/src/location.rs +392 -0
  43. package/native/src/runtime.rs +189 -0
  44. package/native/src/selector.rs +1287 -0
  45. package/native/src/sink.rs +261 -0
  46. package/native/src/stream.rs +334 -0
  47. package/native/src/tokenizer.rs +355 -0
  48. package/native/src/tracker.rs +232 -0
  49. package/native/vendor/README.md +25 -0
  50. package/native/vendor/html5ever-0.35.0/.cargo_vcs_info.json +6 -0
  51. package/native/vendor/html5ever-0.35.0/Cargo.lock +828 -0
  52. package/native/vendor/html5ever-0.35.0/Cargo.toml +85 -0
  53. package/native/vendor/html5ever-0.35.0/LICENSE-APACHE +201 -0
  54. package/native/vendor/html5ever-0.35.0/LICENSE-MIT +25 -0
  55. package/native/vendor/html5ever-0.35.0/README.md +60 -0
  56. package/native/vendor/html5ever-0.35.0/benches/html5ever.rs +86 -0
  57. package/native/vendor/html5ever-0.35.0/data/bench/lipsum-zh.html +19 -0
  58. package/native/vendor/html5ever-0.35.0/data/bench/lipsum.html +40 -0
  59. package/native/vendor/html5ever-0.35.0/data/bench/medium-fragment.html +24 -0
  60. package/native/vendor/html5ever-0.35.0/data/bench/small-fragment.html +7 -0
  61. package/native/vendor/html5ever-0.35.0/data/bench/strong.html +1 -0
  62. package/native/vendor/html5ever-0.35.0/data/bench/tiny-fragment.html +1 -0
  63. package/native/vendor/html5ever-0.35.0/examples/arena.rs +356 -0
  64. package/native/vendor/html5ever-0.35.0/examples/capi/tokenize.c +74 -0
  65. package/native/vendor/html5ever-0.35.0/examples/noop-tokenize.rs +47 -0
  66. package/native/vendor/html5ever-0.35.0/examples/noop-tree-builder.rs +132 -0
  67. package/native/vendor/html5ever-0.35.0/examples/print-tree-actions.rs +185 -0
  68. package/native/vendor/html5ever-0.35.0/examples/tokenize.rs +111 -0
  69. package/native/vendor/html5ever-0.35.0/src/driver.rs +148 -0
  70. package/native/vendor/html5ever-0.35.0/src/lib.rs +30 -0
  71. package/native/vendor/html5ever-0.35.0/src/macros.rs +34 -0
  72. package/native/vendor/html5ever-0.35.0/src/serialize/mod.rs +255 -0
  73. package/native/vendor/html5ever-0.35.0/src/tokenizer/char_ref/mod.rs +451 -0
  74. package/native/vendor/html5ever-0.35.0/src/tokenizer/interface.rs +99 -0
  75. package/native/vendor/html5ever-0.35.0/src/tokenizer/mod.rs +2305 -0
  76. package/native/vendor/html5ever-0.35.0/src/tokenizer/states.rs +97 -0
  77. package/native/vendor/html5ever-0.35.0/src/tree_builder/data.rs +170 -0
  78. package/native/vendor/html5ever-0.35.0/src/tree_builder/mod.rs +1848 -0
  79. package/native/vendor/html5ever-0.35.0/src/tree_builder/rules.rs +1494 -0
  80. package/native/vendor/html5ever-0.35.0/src/tree_builder/tag_sets.rs +114 -0
  81. package/native/vendor/html5ever-0.35.0/src/tree_builder/types.rs +91 -0
  82. package/native/vendor/html5ever-0.35.0/src/util/str.rs +52 -0
  83. package/native/vendor/html5gum-0.8.4/Cargo.lock +1006 -0
  84. package/native/vendor/html5gum-0.8.4/Cargo.toml +109 -0
  85. package/native/vendor/html5gum-0.8.4/LICENSE +19 -0
  86. package/native/vendor/html5gum-0.8.4/README.md +110 -0
  87. package/native/vendor/html5gum-0.8.4/benches/files/0050.html +2131 -0
  88. package/native/vendor/html5gum-0.8.4/benches/files/6D30.html +1004 -0
  89. package/native/vendor/html5gum-0.8.4/benches/files.rs +32 -0
  90. package/native/vendor/html5gum-0.8.4/benches/patterns.rs +31 -0
  91. package/native/vendor/html5gum-0.8.4/src/arrayvec.rs +29 -0
  92. package/native/vendor/html5gum-0.8.4/src/char_validator.rs +134 -0
  93. package/native/vendor/html5gum-0.8.4/src/emitters/callback.rs +676 -0
  94. package/native/vendor/html5gum-0.8.4/src/emitters/default.rs +207 -0
  95. package/native/vendor/html5gum-0.8.4/src/emitters/emitter.rs +344 -0
  96. package/native/vendor/html5gum-0.8.4/src/emitters/html5ever.rs +306 -0
  97. package/native/vendor/html5gum-0.8.4/src/emitters/mod.rs +25 -0
  98. package/native/vendor/html5gum-0.8.4/src/entities.rs +3342 -0
  99. package/native/vendor/html5gum-0.8.4/src/error.rs +96 -0
  100. package/native/vendor/html5gum-0.8.4/src/htmlstring.rs +130 -0
  101. package/native/vendor/html5gum-0.8.4/src/lib.rs +66 -0
  102. package/native/vendor/html5gum-0.8.4/src/machine.rs +2136 -0
  103. package/native/vendor/html5gum-0.8.4/src/machine_helper.rs +277 -0
  104. package/native/vendor/html5gum-0.8.4/src/read_helper.rs +244 -0
  105. package/native/vendor/html5gum-0.8.4/src/reader.rs +440 -0
  106. package/native/vendor/html5gum-0.8.4/src/span.rs +98 -0
  107. package/native/vendor/html5gum-0.8.4/src/state.rs +16 -0
  108. package/native/vendor/html5gum-0.8.4/src/testutils.rs +71 -0
  109. package/native/vendor/html5gum-0.8.4/src/tokenizer.rs +189 -0
  110. package/native/vendor/html5gum-0.8.4/src/utils.rs +88 -0
  111. package/native/vendor/markup5ever-0.35.0/.cargo_vcs_info.json +6 -0
  112. package/native/vendor/markup5ever-0.35.0/Cargo.toml +45 -0
  113. package/native/vendor/markup5ever-0.35.0/LICENSE-APACHE +201 -0
  114. package/native/vendor/markup5ever-0.35.0/LICENSE-MIT +25 -0
  115. package/native/vendor/markup5ever-0.35.0/interface/mod.rs +422 -0
  116. package/native/vendor/markup5ever-0.35.0/interface/tree_builder.rs +280 -0
  117. package/native/vendor/markup5ever-0.35.0/lib.rs +59 -0
  118. package/native/vendor/markup5ever-0.35.0/serialize.rs +77 -0
  119. package/native/vendor/markup5ever-0.35.0/util/buffer_queue.rs +333 -0
  120. package/native/vendor/markup5ever-0.35.0/util/smallcharset.rs +88 -0
  121. package/package.json +137 -4
  122. package/rusteerio.darwin-arm64.node +0 -0
  123. package/rusteerio.darwin-x64.node +0 -0
  124. package/rusteerio.linux-arm64-gnu.node +0 -0
  125. package/rusteerio.linux-arm64-musl.node +0 -0
  126. package/rusteerio.linux-x64-gnu.node +0 -0
  127. package/rusteerio.linux-x64-musl.node +0 -0
  128. package/rusteerio.win32-arm64-msvc.node +0 -0
  129. package/rusteerio.win32-x64-msvc.node +0 -0
  130. package/scripts/check-native.mjs +15 -0
  131. package/scripts/check-release.mjs +235 -0
  132. package/scripts/fix-declarations.mjs +27 -0
  133. package/scripts/generate-notices.mjs +176 -0
  134. package/scripts/notices/README.md +13 -0
  135. package/scripts/notices/cheerio-MIT.txt +21 -0
  136. package/scripts/notices/html5ever-AUTHORS.txt +20 -0
  137. package/scripts/notices/html5ever-COPYRIGHT.txt +8 -0
  138. package/scripts/notices/html5ever-LICENSE-APACHE.txt +201 -0
  139. package/scripts/notices/html5ever-LICENSE-MIT.txt +25 -0
  140. package/scripts/notices/napi-rs-MIT.txt +43 -0
  141. package/scripts/notices/neon-MIT.txt +19 -0
  142. package/scripts/notices/runtime-01c266bced4a434da005.txt +3 -0
  143. package/scripts/notices/runtime-0a54e647fe54104658b5.txt +14 -0
  144. package/scripts/notices/runtime-0b74dfa0bcee5c420c6b.txt +25 -0
  145. package/scripts/notices/runtime-0d542e0c8804e39aa7f3.txt +177 -0
  146. package/scripts/notices/runtime-0f96a83840e146e43c0e.txt +21 -0
  147. package/scripts/notices/runtime-123a331b5dbf04c30097.txt +25 -0
  148. package/scripts/notices/runtime-172020dbfd5b53a226df.txt +34 -0
  149. package/scripts/notices/runtime-209fbbe0ad52d9235e37.txt +26 -0
  150. package/scripts/notices/runtime-23f18e03dc49df91622f.txt +23 -0
  151. package/scripts/notices/runtime-268872b9816f90fd8e85.txt +220 -0
  152. package/scripts/notices/runtime-35242e7a83f69875e6ed.txt +176 -0
  153. package/scripts/notices/runtime-378f5840b258e2779c39.txt +25 -0
  154. package/scripts/notices/runtime-4108245a1f2df9d4e94d.txt +25 -0
  155. package/scripts/notices/runtime-4455bf75a91154108304.txt +21 -0
  156. package/scripts/notices/runtime-553fffcd9b1cb158bc3e.txt +13 -0
  157. package/scripts/notices/runtime-62c7a1e35f56406896d7.txt +176 -0
  158. package/scripts/notices/runtime-6485b8ed310d3f0340bf.txt +25 -0
  159. package/scripts/notices/runtime-69fef7b0f322a6555415.txt +201 -0
  160. package/scripts/notices/runtime-6df43f6f4b5d4587f3d8.txt +187 -0
  161. package/scripts/notices/runtime-799e9ca9d179295ef372.txt +24 -0
  162. package/scripts/notices/runtime-7b63ecd5f1902af1b637.txt +25 -0
  163. package/scripts/notices/runtime-7bc7d5de2a0b793d4e1a.txt +21 -0
  164. package/scripts/notices/runtime-7e12e5df4bae12cb2158.txt +24 -0
  165. package/scripts/notices/runtime-861399f8c21c042b1105.txt +12 -0
  166. package/scripts/notices/runtime-8ada45cd9f843acf64e4.txt +201 -0
  167. package/scripts/notices/runtime-90eb64f0279b0d9432ac.txt +12 -0
  168. package/scripts/notices/runtime-a60eea817514531668d7.txt +201 -0
  169. package/scripts/notices/runtime-a7c936ff1ed8fa340172.txt +25 -0
  170. package/scripts/notices/runtime-ab6eec6caf0fa5775e41.txt +275 -0
  171. package/scripts/notices/runtime-b1181a40b2a7b25cf66f.txt +19 -0
  172. package/scripts/notices/runtime-b71bd43a069ca0641a9e.txt +25 -0
  173. package/scripts/notices/runtime-c16f8dcf1a368b83be78.txt +201 -0
  174. package/scripts/notices/runtime-c2cfccb812fe482101a8.txt +21 -0
  175. package/scripts/notices/runtime-e99d88d232bf57d70f0f.txt +25 -0
  176. package/scripts/notices/runtime-f32fb3b417a194167cfa.txt +9 -0
  177. package/scripts/notices/runtime-f5062c9a188d81dfe66b.txt +39 -0
  178. package/scripts/notices/runtime-f9070a832833fa7488fd.txt +4 -0
  179. package/scripts/notices/runtime-fab3dd6bdab226f1c086.txt +373 -0
  180. package/scripts/notices/runtime-fba3318d0c600380177e.txt +62 -0
  181. package/scripts/notices/runtime-ff8f68cb076caf8cefe7.txt +25 -0
  182. package/scripts/notices/runtime-ff92bed461f50338dd70.txt +74 -0
  183. package/scripts/notices/runtime.json +1632 -0
  184. package/scripts/notices/sources.json +555 -0
  185. package/scripts/notices/xray-MIT.txt +21 -0
  186. package/src/attributes.ts +271 -0
  187. package/src/codec.ts +74 -0
  188. package/src/dom.ts +87 -0
  189. package/src/extract-plan.ts +97 -0
  190. package/src/extract.ts +130 -0
  191. package/src/hooks.ts +220 -0
  192. package/src/index.ts +223 -0
  193. package/src/io.ts +491 -0
  194. package/src/native.ts +135 -0
  195. package/src/nodes.ts +290 -0
  196. package/src/parse.ts +53 -0
  197. package/src/parser-types.ts +157 -0
  198. package/src/router.ts +327 -0
  199. package/src/selection.ts +566 -0
  200. package/src/slim.ts +12 -0
  201. package/src/transport.ts +104 -0
  202. package/src/types.ts +186 -0
  203. package/src/utils.ts +29 -0
  204. package/src/workers.ts +21 -0
  205. package/tsconfig.json +15 -0
  206. package/tsup.config.ts +16 -0
@@ -0,0 +1,676 @@
1
+ //! Consume the parsed HTML as a series of events through a callback.
2
+ //!
3
+ //! While using the [crate::DefaultEmitter] provides an easy-to-use API with low performance, and
4
+ //! implementing your own [crate::Emitter] brings maximal performance and maximal pain, this is a middle
5
+ //! ground. All strings are borrowed from some intermediate buffer instead of individually
6
+ //! allocated.
7
+ //!
8
+ //! ```
9
+ //! // Extract all text between span tags, in a naive (but fast) way. Does not handle tags inside of the span. See `examples/` as well.
10
+ //! use html5gum::{Span, Tokenizer};
11
+ //! use html5gum::emitters::callback::{CallbackEvent, CallbackEmitter};
12
+ //!
13
+ //! let mut is_in_span = false;
14
+ //! let emitter = CallbackEmitter::new(move |event: CallbackEvent<'_>, _span: Span<()>| -> Option<Vec<u8>> {
15
+ //! match event {
16
+ //! CallbackEvent::OpenStartTag { name } => {
17
+ //! is_in_span = name == b"span";
18
+ //! },
19
+ //! CallbackEvent::String { value } if is_in_span => {
20
+ //! return Some(value.to_vec());
21
+ //! }
22
+ //! CallbackEvent::EndTag { .. } => {
23
+ //! is_in_span = false;
24
+ //! }
25
+ //! _ => {}
26
+ //! }
27
+ //!
28
+ //! None
29
+ //! });
30
+ //!
31
+ //! let input = r#"<h1><span class=hello>Hello</span> world!</h1>"#;
32
+ //! let Ok(text_fragments) = Tokenizer::new_with_emitter(input, emitter)
33
+ //! .collect::<Result<Vec<_>, _>>();
34
+ //!
35
+ //! assert_eq!(text_fragments, vec![b"Hello".to_vec()]);
36
+ //! ```
37
+
38
+ use std::collections::VecDeque;
39
+ use std::convert::Infallible;
40
+ use std::marker::PhantomData;
41
+ use std::mem::swap;
42
+
43
+ use crate::utils::trace_log;
44
+ use crate::{naive_next_state, Emitter, Error, Span, SpanBound, State};
45
+
46
+ /// Events used by [CallbackEmitter].
47
+ ///
48
+ /// This operates at a slightly lower level than [crate::Token], as start tags are split up into multiple
49
+ /// events.
50
+ #[derive(Debug)]
51
+ pub enum CallbackEvent<'a> {
52
+ /// Visit the `"<mytag"` in `"<mytag mykey=myvalue>"`. Signifies the beginning of a new start
53
+ /// tag.
54
+ ///
55
+ /// Attributes have not yet been read.
56
+ OpenStartTag {
57
+ /// The name of the start tag.
58
+ name: &'a [u8],
59
+ },
60
+
61
+ /// Visit an attribute name, for example `"mykey"` in `"<mytag mykey=myvalue>"`.
62
+ ///
63
+ /// The attribute value has not yet been read.
64
+ AttributeName {
65
+ /// The name of the attribute.
66
+ name: &'a [u8],
67
+ },
68
+
69
+ /// Visit an attribute value, for example `"myvalue"` in `"<mytag mykey=myvalue>"`.
70
+ ///
71
+ /// Things like whitespace, quote handling is taken care of.
72
+ ///
73
+ /// After this event, the start tag may be closed using `CloseStartTag`, or another
74
+ /// `AttributeName` may follow.
75
+ AttributeValue {
76
+ /// The value of the attribute.
77
+ value: &'a [u8],
78
+ },
79
+
80
+ /// Visit the end of the start tag, for example `">"` in `"<mytag mykey=myvalue>"`.
81
+ ///
82
+ CloseStartTag {
83
+ /// Whether the tag ended with `"/>"`.
84
+ ///
85
+ /// Note that in HTML5 this difference is largely ignored, and tags are considered
86
+ /// self-closing based on a hardcoded list of names, not based on syntax.
87
+ self_closing: bool,
88
+ },
89
+
90
+ /// Visit `"</mytag>"`.
91
+ ///
92
+ /// Note: Because of strangeness in the HTML spec, attributes may be observed outside of start
93
+ /// tags, before this event. It's best to ignore them as they are not valid HTML, but can still
94
+ /// be observed through most HTML parsers.
95
+ EndTag {
96
+ /// The name of the end tag.
97
+ name: &'a [u8],
98
+ },
99
+
100
+ /// Visit a string, as in, the actual text between tags. The content. Remember actual content
101
+ /// in HTML, before SPAs took over? I remember.
102
+ ///
103
+ /// It's guaranteed that all consecutive "character tokens" (as the spec calls them) are folded
104
+ /// into one string event.
105
+ String {
106
+ /// A series of character tokens.
107
+ value: &'a [u8],
108
+ },
109
+
110
+ /// Visit a comment, like `<!-- DON'T HACK THIS WEBSITE -->`
111
+ Comment {
112
+ /// The contents of the comment.
113
+ value: &'a [u8],
114
+ },
115
+
116
+ /// Visit `<!DOCTYPE html>`.
117
+ Doctype {
118
+ /// Name of the docstring.
119
+ name: &'a [u8],
120
+ /// Public identifier (see spec)
121
+ public_identifier: Option<&'a [u8]>,
122
+ /// System identifier (see spec)
123
+ system_identifier: Option<&'a [u8]>,
124
+ /// Enable quirksmode
125
+ force_quirks: bool,
126
+ },
127
+
128
+ /// Visit a parsing error.
129
+ Error(Error),
130
+ }
131
+
132
+ #[derive(Debug, Clone, Copy)]
133
+ enum CurrentTag {
134
+ Start,
135
+ End,
136
+ }
137
+
138
+ #[derive(Debug)]
139
+ struct CallbackState<F, T, S> {
140
+ callback: F,
141
+ emitted_tokens: VecDeque<T>,
142
+ phantom: PhantomData<S>,
143
+ }
144
+
145
+ /// This trait is implemented for all functions that have the same signature as
146
+ /// [Callback::handle_event]. The trait only exists in case you want to implement it on a nameable
147
+ /// type.
148
+ pub trait Callback<T, S: SpanBound> {
149
+ /// Perform some action on a parsing event, and, optionally, return a value that can be yielded
150
+ /// from the [crate::Tokenizer] iterator.
151
+ fn handle_event(&mut self, event: CallbackEvent<'_>, span: Span<S>) -> Option<T>;
152
+ }
153
+
154
+ impl<F, T, S: SpanBound> Callback<T, S> for F
155
+ where
156
+ F: FnMut(CallbackEvent<'_>, Span<S>) -> Option<T>,
157
+ {
158
+ fn handle_event(&mut self, event: CallbackEvent<'_>, span: Span<S>) -> Option<T> {
159
+ self(event, span)
160
+ }
161
+ }
162
+
163
+ impl<F, T, S: SpanBound> CallbackState<F, T, S>
164
+ where
165
+ F: Callback<T, S>,
166
+ {
167
+ fn emit_event(&mut self, event: CallbackEvent<'_>, span: Span<S>) {
168
+ let res = self.callback.handle_event(event, span);
169
+ if let Some(token) = res {
170
+ self.emitted_tokens.push_front(token);
171
+ }
172
+ }
173
+ }
174
+
175
+ impl<F, T, S> Default for CallbackState<F, T, S>
176
+ where
177
+ F: Default,
178
+ {
179
+ fn default() -> Self {
180
+ CallbackState {
181
+ callback: F::default(),
182
+ emitted_tokens: VecDeque::default(),
183
+ phantom: PhantomData,
184
+ }
185
+ }
186
+ }
187
+
188
+ #[derive(Debug, Default)]
189
+ struct EmitterState<S: SpanBound> {
190
+ naively_switch_states: bool,
191
+
192
+ current_characters: Vec<u8>,
193
+ current_characters_start: S,
194
+ current_characters_end: S,
195
+ current_comment: Vec<u8>,
196
+
197
+ last_start_tag: Vec<u8>,
198
+ current_tag_had_attributes: bool,
199
+ current_tag_type: Option<CurrentTag>,
200
+ current_tag_self_closing: bool,
201
+ current_tag_name: Vec<u8>,
202
+ current_attribute_name: Vec<u8>,
203
+ current_attribute_value: Vec<u8>,
204
+ current_attribute_name_start: S,
205
+ current_attribute_name_end: S,
206
+ current_attribute_value_start: S,
207
+ current_attribute_value_end: S,
208
+
209
+ // strings related to doctype
210
+ doctype_name: Vec<u8>,
211
+ doctype_has_public_identifier: bool,
212
+ doctype_has_system_identifier: bool,
213
+ doctype_public_identifier: Vec<u8>,
214
+ doctype_system_identifier: Vec<u8>,
215
+ doctype_force_quirks: bool,
216
+
217
+ current_taglike_span: S,
218
+ position: S,
219
+ }
220
+
221
+ /// The emitter class to pass to [crate::Tokenizer::new_with_emitter]. Please refer to the
222
+ /// module-level documentation on [crate::emitters::callback] for usage.
223
+ #[derive(Debug)]
224
+ pub struct CallbackEmitter<F, T = Infallible, S: SpanBound = ()>
225
+ where
226
+ // add this requirement, so that users get better errors, if the callback does not implement `Callback`.
227
+ F: Callback<T, S>,
228
+ {
229
+ // this struct is only split out so [CallbackState::emit_event] can borrow things concurrently
230
+ // with other attributes.
231
+ callback_state: CallbackState<F, T, S>,
232
+ emitter_state: EmitterState<S>,
233
+ }
234
+
235
+ impl<F, T, S: SpanBound> Default for CallbackEmitter<F, T, S>
236
+ where
237
+ F: Default + Callback<T, S>,
238
+ {
239
+ fn default() -> Self {
240
+ CallbackEmitter {
241
+ callback_state: CallbackState::default(),
242
+ emitter_state: EmitterState::default(),
243
+ }
244
+ }
245
+ }
246
+
247
+ impl<F, T, S: SpanBound> CallbackEmitter<F, T, S>
248
+ where
249
+ F: Callback<T, S>,
250
+ {
251
+ /// Create a new emitter.
252
+ ///
253
+ /// The given callback may return optional tokens that then become available through the
254
+ /// [crate::Tokenizer]'s iterator. If that's not used, return `Option<Infallible>`.
255
+ pub fn new(callback: F) -> Self {
256
+ CallbackEmitter {
257
+ callback_state: CallbackState {
258
+ callback,
259
+ emitted_tokens: VecDeque::new(),
260
+ phantom: PhantomData,
261
+ },
262
+ emitter_state: EmitterState::default(),
263
+ }
264
+ }
265
+
266
+ /// Get mutable access to the inner callback.
267
+ pub fn callback_mut(&mut self) -> &mut F {
268
+ &mut self.callback_state.callback
269
+ }
270
+
271
+ /// Whether to use [`naive_next_state`] to switch states automatically.
272
+ ///
273
+ /// The default is off.
274
+ pub fn naively_switch_states(&mut self, yes: bool) {
275
+ self.emitter_state.naively_switch_states = yes;
276
+ }
277
+
278
+ fn flush_attribute_name(&mut self) {
279
+ self.flush_open_start_tag();
280
+
281
+ if !self.emitter_state.current_attribute_name.is_empty() {
282
+ self.callback_state.emit_event(
283
+ CallbackEvent::AttributeName {
284
+ name: &self.emitter_state.current_attribute_name,
285
+ },
286
+ Span {
287
+ start: self.emitter_state.current_attribute_name_start,
288
+ end: self.emitter_state.current_attribute_name_end,
289
+ },
290
+ );
291
+ self.emitter_state.current_attribute_name.clear();
292
+ }
293
+ }
294
+
295
+ fn flush_attribute(&mut self) {
296
+ self.flush_attribute_name();
297
+
298
+ if !self.emitter_state.current_attribute_value.is_empty() {
299
+ self.callback_state.emit_event(
300
+ CallbackEvent::AttributeValue {
301
+ value: &self.emitter_state.current_attribute_value,
302
+ },
303
+ Span {
304
+ start: self.emitter_state.current_attribute_value_start,
305
+ end: self.emitter_state.current_attribute_value_end,
306
+ },
307
+ );
308
+ self.emitter_state.current_attribute_value.clear();
309
+ }
310
+ }
311
+
312
+ fn flush_open_start_tag(&mut self) {
313
+ self.flush_current_characters();
314
+
315
+ if matches!(self.emitter_state.current_tag_type, Some(CurrentTag::Start))
316
+ && !self.emitter_state.current_tag_name.is_empty()
317
+ {
318
+ self.callback_state.emit_event(
319
+ CallbackEvent::OpenStartTag {
320
+ name: &self.emitter_state.current_tag_name,
321
+ },
322
+ Span {
323
+ start: self.emitter_state.current_taglike_span,
324
+ end: self.emitter_state.position.offset(-1),
325
+ },
326
+ );
327
+
328
+ self.emitter_state.last_start_tag.clear();
329
+ swap(
330
+ &mut self.emitter_state.last_start_tag,
331
+ &mut self.emitter_state.current_tag_name,
332
+ );
333
+ }
334
+ }
335
+
336
+ fn flush_current_characters(&mut self) {
337
+ if self.emitter_state.current_characters.is_empty() {
338
+ return;
339
+ }
340
+
341
+ self.callback_state.emit_event(
342
+ CallbackEvent::String {
343
+ value: &self.emitter_state.current_characters,
344
+ },
345
+ Span {
346
+ start: self.emitter_state.current_characters_start,
347
+ end: self.emitter_state.current_characters_end,
348
+ },
349
+ );
350
+ self.emitter_state.current_characters.clear();
351
+ }
352
+ }
353
+
354
+ impl<F, T, S: SpanBound> Emitter for CallbackEmitter<F, T, S>
355
+ where
356
+ F: Callback<T, S>,
357
+ {
358
+ type Token = T;
359
+
360
+ #[inline]
361
+ fn move_position(&mut self, offset: isize) {
362
+ self.emitter_state.position = self.emitter_state.position.offset(offset);
363
+ trace_log!(
364
+ "callbacks: move_position, offset={}, now={:?}",
365
+ offset,
366
+ self.emitter_state.position
367
+ );
368
+ }
369
+
370
+ fn set_last_start_tag(&mut self, last_start_tag: Option<&[u8]>) {
371
+ self.emitter_state.last_start_tag.clear();
372
+ self.emitter_state
373
+ .last_start_tag
374
+ .extend(last_start_tag.unwrap_or_default());
375
+ }
376
+
377
+ fn emit_eof(&mut self) {
378
+ self.flush_current_characters();
379
+ }
380
+
381
+ fn emit_error(&mut self, error: Error) {
382
+ self.callback_state.emit_event(
383
+ CallbackEvent::Error(error),
384
+ Span {
385
+ start: self.emitter_state.position,
386
+ end: self.emitter_state.position,
387
+ },
388
+ );
389
+ }
390
+
391
+ fn pop_token(&mut self) -> Option<Self::Token> {
392
+ self.callback_state.emitted_tokens.pop_back()
393
+ }
394
+
395
+ fn init_string(&mut self) {
396
+ // Only reset the start position if we're not already accumulating characters
397
+ // This prevents overwriting the start position when returning from character
398
+ // reference states that have already emitted buffered content
399
+ if self.emitter_state.current_characters.is_empty() {
400
+ self.emitter_state.current_characters_start = self.emitter_state.position;
401
+ }
402
+ }
403
+
404
+ fn emit_string(&mut self, s: &[u8]) {
405
+ self.emitter_state.current_characters_end = self.emitter_state.position;
406
+ crate::utils::trace_log!(
407
+ "callbacks: emit_string, len={}, start={:?}, end={:?}",
408
+ s.len(),
409
+ self.emitter_state.current_characters_start,
410
+ self.emitter_state.current_characters_end
411
+ );
412
+ self.emitter_state.current_characters.extend(s);
413
+ }
414
+
415
+ fn init_start_tag(&mut self) {
416
+ self.emitter_state.current_tag_name.clear();
417
+ self.emitter_state.current_tag_type = Some(CurrentTag::Start);
418
+ self.emitter_state.current_tag_self_closing = false;
419
+ }
420
+
421
+ fn init_end_tag(&mut self) {
422
+ self.emitter_state.current_tag_name.clear();
423
+ self.emitter_state.current_tag_type = Some(CurrentTag::End);
424
+ self.emitter_state.current_tag_had_attributes = false;
425
+ }
426
+
427
+ fn init_comment(&mut self) {
428
+ self.flush_current_characters();
429
+ self.emitter_state.current_comment.clear();
430
+ }
431
+
432
+ fn emit_current_tag(&mut self) -> Option<State> {
433
+ self.flush_attribute();
434
+ match self.emitter_state.current_tag_type {
435
+ Some(CurrentTag::Start) => {
436
+ self.flush_open_start_tag();
437
+ let s = self.emitter_state.position;
438
+ self.callback_state.emit_event(
439
+ CallbackEvent::CloseStartTag {
440
+ self_closing: self.emitter_state.current_tag_self_closing,
441
+ },
442
+ Span {
443
+ start: s.offset(-1),
444
+ end: s,
445
+ },
446
+ );
447
+ }
448
+ Some(CurrentTag::End) => {
449
+ if self.emitter_state.current_tag_had_attributes {
450
+ self.emit_error(Error::EndTagWithAttributes);
451
+ }
452
+ self.emitter_state.last_start_tag.clear();
453
+ self.callback_state.emit_event(
454
+ CallbackEvent::EndTag {
455
+ name: &self.emitter_state.current_tag_name,
456
+ },
457
+ Span {
458
+ start: self.emitter_state.current_taglike_span,
459
+ end: self.emitter_state.position,
460
+ },
461
+ );
462
+ }
463
+ _ => {}
464
+ }
465
+
466
+ let next_state = if self.emitter_state.naively_switch_states {
467
+ naive_next_state(&self.emitter_state.last_start_tag)
468
+ } else {
469
+ None
470
+ };
471
+
472
+ // After flushing characters, initialize a new string span for text-accumulating states
473
+ if next_state.is_some() {
474
+ self.init_string();
475
+ }
476
+
477
+ next_state
478
+ }
479
+ fn emit_current_comment(&mut self) {
480
+ self.callback_state.emit_event(
481
+ CallbackEvent::Comment {
482
+ value: &self.emitter_state.current_comment,
483
+ },
484
+ Span {
485
+ start: self.emitter_state.current_taglike_span,
486
+ end: self.emitter_state.position,
487
+ },
488
+ );
489
+ self.emitter_state.current_comment.clear();
490
+ }
491
+
492
+ fn emit_current_doctype(&mut self) {
493
+ self.callback_state.emit_event(
494
+ CallbackEvent::Doctype {
495
+ name: &self.emitter_state.doctype_name,
496
+ public_identifier: if self.emitter_state.doctype_has_public_identifier {
497
+ Some(&self.emitter_state.doctype_public_identifier)
498
+ } else {
499
+ None
500
+ },
501
+ system_identifier: if self.emitter_state.doctype_has_system_identifier {
502
+ Some(&self.emitter_state.doctype_system_identifier)
503
+ } else {
504
+ None
505
+ },
506
+ force_quirks: self.emitter_state.doctype_force_quirks,
507
+ },
508
+ Span {
509
+ start: self.emitter_state.current_taglike_span,
510
+ end: self.emitter_state.position,
511
+ },
512
+ );
513
+ }
514
+
515
+ fn set_self_closing(&mut self) {
516
+ trace_log!("set_self_closing");
517
+ if matches!(self.emitter_state.current_tag_type, Some(CurrentTag::End)) {
518
+ self.callback_state.emit_event(
519
+ CallbackEvent::Error(Error::EndTagWithTrailingSolidus),
520
+ Span {
521
+ start: self.emitter_state.position,
522
+ end: self.emitter_state.position,
523
+ },
524
+ );
525
+ } else {
526
+ self.emitter_state.current_tag_self_closing = true;
527
+ }
528
+ }
529
+
530
+ fn set_force_quirks(&mut self) {
531
+ self.emitter_state.doctype_force_quirks = true;
532
+ }
533
+
534
+ fn push_tag_name(&mut self, s: &[u8]) {
535
+ self.emitter_state.current_tag_name.extend(s);
536
+ }
537
+
538
+ fn push_comment(&mut self, s: &[u8]) {
539
+ self.emitter_state.current_comment.extend(s);
540
+ }
541
+
542
+ fn push_doctype_name(&mut self, s: &[u8]) {
543
+ self.emitter_state.doctype_name.extend(s);
544
+ }
545
+
546
+ fn init_doctype(&mut self) {
547
+ self.flush_current_characters();
548
+ self.emitter_state.doctype_name.clear();
549
+ self.emitter_state.doctype_has_public_identifier = false;
550
+ self.emitter_state.doctype_has_system_identifier = false;
551
+ self.emitter_state.doctype_public_identifier.clear();
552
+ self.emitter_state.doctype_system_identifier.clear();
553
+ self.emitter_state.doctype_force_quirks = false;
554
+ }
555
+
556
+ fn init_attribute(&mut self) {
557
+ self.flush_open_start_tag();
558
+ self.flush_attribute();
559
+ self.emitter_state.current_tag_had_attributes = true;
560
+ self.emitter_state.current_attribute_name_start = self.emitter_state.position.offset(-1);
561
+ }
562
+
563
+ fn push_attribute_name(&mut self, s: &[u8]) {
564
+ self.emitter_state.current_attribute_name.extend(s);
565
+ self.emitter_state.current_attribute_name_end = self.emitter_state.position;
566
+ }
567
+
568
+ fn init_attribute_value(&mut self) {
569
+ self.emitter_state.current_attribute_value_start = self.emitter_state.position;
570
+ }
571
+
572
+ fn push_attribute_value(&mut self, s: &[u8]) {
573
+ self.flush_attribute_name();
574
+ self.emitter_state.current_attribute_value.extend(s);
575
+ self.emitter_state.current_attribute_value_end = self.emitter_state.position;
576
+ }
577
+
578
+ fn set_doctype_public_identifier(&mut self, value: &[u8]) {
579
+ self.emitter_state.doctype_has_public_identifier = true;
580
+ self.emitter_state.doctype_public_identifier.clear();
581
+ self.emitter_state.doctype_public_identifier.extend(value);
582
+ }
583
+ fn set_doctype_system_identifier(&mut self, value: &[u8]) {
584
+ self.emitter_state.doctype_has_system_identifier = true;
585
+ self.emitter_state.doctype_system_identifier.clear();
586
+ self.emitter_state.doctype_system_identifier.extend(value);
587
+ }
588
+ fn push_doctype_public_identifier(&mut self, value: &[u8]) {
589
+ self.emitter_state.doctype_public_identifier.extend(value);
590
+ }
591
+ fn push_doctype_system_identifier(&mut self, value: &[u8]) {
592
+ self.emitter_state.doctype_system_identifier.extend(value);
593
+ }
594
+
595
+ fn start_open_tag(&mut self) {
596
+ self.emitter_state.current_taglike_span = self.emitter_state.position.offset(-1);
597
+ }
598
+
599
+ fn current_is_appropriate_end_tag_token(&mut self) -> bool {
600
+ if self.emitter_state.last_start_tag.is_empty() {
601
+ crate::utils::trace_log!(
602
+ "current_is_appropriate_end_tag_token: no, because last_start_tag is empty"
603
+ );
604
+ return false;
605
+ }
606
+
607
+ if !matches!(self.emitter_state.current_tag_type, Some(CurrentTag::End)) {
608
+ crate::utils::trace_log!(
609
+ "current_is_appropriate_end_tag_token: no, because current_tag_type is not end"
610
+ );
611
+ return false;
612
+ }
613
+
614
+ crate::utils::trace_log!(
615
+ "current_is_appropriate_end_tag_token: last_start_tag = {:?}",
616
+ self.emitter_state.last_start_tag
617
+ );
618
+ crate::utils::trace_log!(
619
+ "current_is_appropriate_end_tag_token: current_tag = {:?}",
620
+ self.emitter_state.current_tag_name
621
+ );
622
+ self.emitter_state.last_start_tag == self.emitter_state.current_tag_name
623
+ }
624
+ }
625
+
626
+ #[test]
627
+ fn round_trip() {
628
+ use crate::Tokenizer;
629
+
630
+ let mut rt = Vec::new();
631
+ let emitter = CallbackEmitter::new(|event: CallbackEvent<'_>, _: crate::Span<()>| {
632
+ match event {
633
+ CallbackEvent::OpenStartTag { name } => {
634
+ rt.push(b'<');
635
+ rt.extend(name);
636
+ }
637
+ CallbackEvent::AttributeName { name } => {
638
+ rt.push(b' ');
639
+ rt.extend(name);
640
+ }
641
+ CallbackEvent::AttributeValue { value } => {
642
+ rt.push(b'=');
643
+ rt.push(b'"');
644
+ rt.extend(value);
645
+ rt.push(b'"');
646
+ }
647
+ CallbackEvent::String { value } => {
648
+ rt.extend(value);
649
+ }
650
+ CallbackEvent::CloseStartTag { self_closing } => {
651
+ if self_closing {
652
+ rt.push(b'/');
653
+ }
654
+ rt.push(b'>');
655
+ }
656
+ CallbackEvent::EndTag { name } => {
657
+ rt.extend(b"</");
658
+ rt.extend(name);
659
+ rt.push(b'>');
660
+ }
661
+ CallbackEvent::Comment { value } => {
662
+ rt.extend(b"<!--");
663
+ rt.extend(value);
664
+ rt.extend(b"-->");
665
+ }
666
+ CallbackEvent::Doctype { .. } => {}
667
+ CallbackEvent::Error(_) => {}
668
+ }
669
+
670
+ None::<core::convert::Infallible>
671
+ });
672
+
673
+ let source = " <!-- a --> <h1>Hello</h1> world <a href=\"foo\" title=\"baz\">bar</a>";
674
+ Tokenizer::new_with_emitter(source, emitter).finish();
675
+ assert_eq!(source.as_bytes(), rt, "{} != {}", source, rt.escape_ascii());
676
+ }