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
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
use std::borrow::Cow;
use std::cell::{Cell, Ref, RefCell};
#[allow(unused_imports)]
use html5ever::namespace_url;
use html5ever::parse_document;
use html5ever::tree_builder;
use html5ever::tree_builder::{ElementFlags, NodeOrText, QuirksMode, TreeSink};
use html5ever::ParseOpts;
use html5ever::{local_name, ns};
use html5ever::{Attribute, QualName};
use tendril::{StrTendril, TendrilSink};
use crate::dom_tree::Tree;
use crate::entities::wrap_tendril;
use crate::matcher::{DescendantMatches, Matcher};
use crate::node::{Element, NodeData, NodeId, NodeRef, TreeNode};
use crate::selection::Selection;
/// Document represents an HTML document to be manipulated.
#[derive(Clone)]
pub struct Document {
/// The document's dom tree.
pub tree: Tree,
/// Errors that occurred during parsing.
pub errors: RefCell<Vec<Cow<'static, str>>>,
/// The document's quirks mode.
pub quirks_mode: Cell<QuirksMode>,
}
impl Default for Document {
fn default() -> Self {
Self {
tree: Tree::new(NodeData::Document),
errors: RefCell::new(vec![]),
quirks_mode: Cell::new(tree_builder::NoQuirks),
}
}
}
impl<T: Into<StrTendril>> From<T> for Document {
fn from(html: T) -> Self {
let opts = ParseOpts {
tokenizer: Default::default(),
tree_builder: tree_builder::TreeBuilderOpts {
scripting_enabled: false,
..Default::default()
},
};
parse_document(Document::default(), opts).one(html)
}
}
// fragment
impl Document {
/// Creates a new HTML document fragment.
pub fn fragment<T: Into<StrTendril>>(html: T) -> Self {
// Note: The `body` context element is somehow ignored during parsing,
// so the `html` element becomes the first child of the root node,
// rather than being nested inside a `body` element as expected.
html5ever::parse_fragment(
Document::fragment_sink(),
ParseOpts {
tokenizer: Default::default(),
tree_builder: tree_builder::TreeBuilderOpts {
scripting_enabled: false,
drop_doctype: true,
..Default::default()
},
},
QualName::new(None, ns!(html), local_name!("body")),
Vec::new(),
false,
)
.one(html)
}
/// Create a new sink for a html document fragment
pub fn fragment_sink() -> Self {
Self {
tree: Tree::new(NodeData::Fragment),
errors: RefCell::new(vec![]),
quirks_mode: Cell::new(tree_builder::NoQuirks),
}
}
}
// property methods
impl Document {
/// Return the underlying root document node.
#[inline]
pub fn root(&self) -> NodeRef<'_> {
self.tree.root()
}
/// Returns the root element node (`<html>`) of the document.
pub fn html_root(&self) -> NodeRef<'_> {
self.tree.html_root()
}
/// Gets the HTML contents of the document. It includes
/// the text and comment nodes.
pub fn html(&self) -> StrTendril {
self.root().html()
}
/// Gets the HTML contents of the document.
/// It includes only children nodes.
pub fn inner_html(&self) -> StrTendril {
self.root().inner_html()
}
/// Gets the HTML contents of the document.
/// It includes its children nodes.
pub fn try_html(&self) -> Option<StrTendril> {
self.root().try_html()
}
/// Gets the HTML contents of the document.
/// It includes only children nodes.
pub fn try_inner_html(&self) -> Option<StrTendril> {
self.root().try_inner_html()
}
/// Gets the text content of the document.
pub fn text(&self) -> StrTendril {
self.root().text()
}
/// Returns the formatted text of the document and its descendants. This is the same as
/// the `text()` method, but with a few differences:
///
/// - Whitespace is normalized so that there is only one space between words.
/// - All whitespace is removed from the beginning and end of the text.
/// - After block elements, a double newline is added.
/// - For elements like `br`, 'hr', 'li', 'tr' a single newline is added.
pub fn formatted_text(&self) -> StrTendril {
self.root().formatted_text()
}
/// Finds the base URI of the tree by looking for `<base>` tags in document's head.
///
/// The base URI is the value of the `href` attribute of the first
/// `<base>` tag in the document's head. If no such tag is found,
/// the method returns `None`.
///
pub fn base_uri(&self) -> Option<StrTendril> {
self.tree.base_uri()
}
/// Returns the document's `<body>` element, or `None` if absent.
/// For fragments ([crate::NodeData::Fragment]), this typically returns `None`.
pub fn body(&self) -> Option<NodeRef<'_>> {
self.tree.body()
}
/// Returns the document's `<head>` element, or `None` if absent.
/// For fragments ([crate::NodeData::Fragment]), this typically returns `None`.
pub fn head(&self) -> Option<NodeRef<'_>> {
self.tree.head()
}
/// Merges adjacent text nodes and removes empty text nodes.
///
/// Normalization is necessary to ensure that adjacent text nodes are merged into one text node.
///
/// # Example
///
/// ```
/// use dom_query::Document;
///
/// let doc = Document::from("<div>Hello</div>");
/// let sel = doc.select("div");
/// let div = sel.nodes().first().unwrap();
/// let text1 = doc.tree.new_text(" ");
/// let text2 = doc.tree.new_text("World");
/// let text3 = doc.tree.new_text("");
/// div.append_child(&text1);
/// div.append_child(&text2);
/// div.append_child(&text3);
/// assert_eq!(div.children().len(), 4);
/// doc.normalize();
/// assert_eq!(div.children().len(), 1);
/// assert_eq!(&div.text(), "Hello World");
/// ```
pub fn normalize(&self) {
self.root().normalize();
}
}
// traversal methods
impl Document {
/// Gets the descendants of the root document node in the current, filter by a selector.
/// It returns a new selection object containing these matched elements.
///
/// # Panics
///
/// Panics if failed to parse the given CSS selector.
pub fn select(&self, sel: &str) -> Selection<'_> {
let matcher = Matcher::new(sel).expect("Invalid CSS selector");
self.select_matcher(&matcher)
}
/// Alias for `select`, it gets the descendants of the root document node in the current, filter by a selector.
/// It returns a new selection object containing these matched elements.
///
/// # Panics
///
/// Panics if failed to parse the given CSS selector.
pub fn nip(&self, sel: &str) -> Selection<'_> {
self.select(sel)
}
/// Gets the descendants of the root document node in the current, filter by a selector.
/// It returns a new selection object containing these matched elements.
pub fn try_select(&self, sel: &str) -> Option<Selection<'_>> {
Matcher::new(sel).ok().and_then(|matcher| {
let selection = self.select_matcher(&matcher);
if !selection.is_empty() {
Some(selection)
} else {
None
}
})
}
/// Gets the descendants of the root document node in the current, filter by a matcher.
/// It returns a new selection object containing these matched elements.
pub fn select_matcher(&self, matcher: &Matcher) -> Selection<'_> {
let root = self.tree.root();
let nodes = DescendantMatches::new(root, matcher).collect();
Selection { nodes }
}
/// Gets the descendants of the root document node in the current, filter by a matcher.
/// It returns a new selection object containing elements of the single (first) match.
pub fn select_single_matcher(&self, matcher: &Matcher) -> Selection<'_> {
let node = DescendantMatches::new(self.tree.root(), matcher).next();
match node {
Some(node) => node.into(),
None => Default::default(),
}
}
/// Gets the descendants of the root document node in the current, filter by a selector.
/// It returns a new selection object containing elements of the single (first) match.
///
/// # Panics
///
/// Panics if failed to parse the given CSS selector.
pub fn select_single(&self, sel: &str) -> Selection<'_> {
let matcher = Matcher::new(sel).expect("Invalid CSS selector");
self.select_single_matcher(&matcher)
}
}
impl TreeSink for Document {
type ElemName<'a> = Ref<'a, QualName>;
/// The overall result of parsing.
type Output = Self;
/// Handle is a reference to a DOM node. The tree builder requires that a `Handle` implements `Clone` to get
/// another reference to the same node.
type Handle = NodeId;
/// Consume this sink and return the overall result of parsing.
#[inline]
fn finish(self) -> Self {
self
}
/// Signal a parse error.
#[inline]
fn parse_error(&self, msg: Cow<'static, str>) {
let mut errors = self.errors.borrow_mut();
errors.push(msg);
}
/// Get a handle to the `Document` node.
#[inline]
fn get_document(&self) -> Self::Handle {
self.tree.root_id()
}
/// Get a handle to a template's template contents. The tree builder promises this will never be called with
/// something else than a template element.
#[inline]
fn get_template_contents(&self, target: &Self::Handle) -> Self::Handle {
self.tree
.query_node_or(target, None, |node| {
node.as_element().and_then(|elem| elem.template_contents)
})
.expect("target node is not a template element!")
}
/// Set the document's quirks mode.
#[inline]
fn set_quirks_mode(&self, mode: QuirksMode) {
self.quirks_mode.set(mode);
}
/// Do two handles refer to the same node?.
#[inline]
fn same_node(&self, x: &Self::Handle, y: &Self::Handle) -> bool {
*x == *y
}
/// What is the name of the element?
/// Should never be called on a non-element node; Feel free to `panic!`.
#[inline]
fn elem_name(&self, target: &Self::Handle) -> Self::ElemName<'_> {
self.tree
.get_name(target)
.expect("target node is not an element!")
}
/// Create an element.
/// When creating a template element (`name.ns.expanded() == expanded_name!(html"template")`), an
/// associated document fragment called the "template contents" should also be created. Later calls to
/// self.get_template_contents() with that given element return it. See `the template element in the whatwg spec`,
#[inline]
fn create_element(
&self,
name: QualName,
attrs: Vec<Attribute>,
flags: ElementFlags,
) -> Self::Handle {
let mut nodes = self.tree.nodes.borrow_mut();
let new_elem_id = NodeId::new(nodes.len());
let template_contents = if flags.template {
Some(NodeId::new(nodes.len() + 1))
} else {
None
};
let data = NodeData::Element(Element::new(
name,
attrs,
template_contents,
flags.mathml_annotation_xml_integration_point,
));
nodes.push(TreeNode::new(new_elem_id, data));
if let Some(fragment_id) = template_contents {
nodes.push(TreeNode::new(fragment_id, NodeData::Fragment));
// The template's content is considered outside of the main document,
// so its DocumentFragment remains parentless.
}
new_elem_id
}
/// Create a comment node.
#[inline]
fn create_comment(&self, text: StrTendril) -> Self::Handle {
self.tree.create_node(NodeData::Comment {
contents: wrap_tendril(text),
})
}
/// Create a Processing Instruction node.
#[inline]
fn create_pi(&self, target: StrTendril, data: StrTendril) -> Self::Handle {
self.tree.create_node(NodeData::ProcessingInstruction {
target: wrap_tendril(target),
contents: wrap_tendril(data),
})
}
/// Append a node as the last child of the given node. If this would produce adjacent sibling text nodes, it
/// should concatenate the text instead.
/// The child node will not already have a parent.
fn append(&self, parent: &Self::Handle, child: NodeOrText<Self::Handle>) {
// Append to an existing Text node if we have one.
match child {
NodeOrText::AppendNode(node_id) => self.tree.append_child_of(parent, &node_id),
NodeOrText::AppendText(text) => {
let last_child = self.tree.last_child_of(parent);
let merged = last_child
.map(|child| append_to_existing_text(&child, &text))
.unwrap_or(false);
if merged {
return;
}
self.tree.append_child_data_of(
parent,
NodeData::Text {
contents: wrap_tendril(text),
},
)
}
}
}
/// Append a node as the sibling immediately before the given node.
/// The tree builder promises that `sibling` is not a text node. However its old previous sibling, which would
/// become the new node's previous sibling, could be a text node. If the new node is also a text node, the two
/// should be merged, as in the behavior of `append`.
fn append_before_sibling(&self, sibling: &Self::Handle, child: NodeOrText<Self::Handle>) {
match child {
NodeOrText::AppendText(text) => {
let prev_sibling = self.tree.prev_sibling_of(sibling);
let merged = prev_sibling
.map(|sibling| append_to_existing_text(&sibling, &text))
.unwrap_or(false);
if merged {
return;
}
let id = self.tree.create_node(NodeData::Text {
contents: wrap_tendril(text),
});
self.tree.insert_before_of(sibling, &id);
}
// The tree builder promises we won't have a text node after
// the insertion point.
// Any other kind of node.
NodeOrText::AppendNode(id) => self.tree.insert_before_of(sibling, &id),
};
}
/// When the insertion point is decided by the existence of a parent node of the element, we consider both
/// possibilities and send the element which will be used if a parent node exists, along with the element to be
/// used if there isn't one.
fn append_based_on_parent_node(
&self,
element: &Self::Handle,
prev_element: &Self::Handle,
child: NodeOrText<Self::Handle>,
) {
let has_parent = self
.tree
.nodes
.borrow()
.get(element.value)
.is_some_and(|node| node.parent.is_some());
if has_parent {
self.append_before_sibling(element, child);
} else {
self.append(prev_element, child);
}
}
/// Append a `DOCTYPE` element to the `Document` node.
#[inline]
fn append_doctype_to_document(
&self,
name: StrTendril,
public_id: StrTendril,
system_id: StrTendril,
) {
let root = self.tree.root_id();
self.tree.append_child_data_of(
&root,
NodeData::Doctype {
name: wrap_tendril(name),
public_id: wrap_tendril(public_id),
system_id: wrap_tendril(system_id),
},
);
}
/// Add each attribute to the given element, if no attribute with that name already exists. The tree builder
/// promises this will never be called with something else than an element.
fn add_attrs_if_missing(&self, target: &Self::Handle, attrs: Vec<Attribute>) {
self.tree.update_node(target, |node| {
if let Some(el) = node.as_element_mut() {
el.add_attrs_if_missing(attrs);
}
});
}
/// Detach the given node from its parent.
#[inline]
fn remove_from_parent(&self, target: &Self::Handle) {
self.tree.remove_from_parent(target);
}
/// Remove all the children from node and append them to new_parent.
#[inline]
fn reparent_children(&self, node: &Self::Handle, new_parent: &Self::Handle) {
self.tree.reparent_children_of(node, Some(*new_parent));
}
fn is_mathml_annotation_xml_integration_point(&self, handle: &Self::Handle) -> bool {
self.tree.is_mathml_annotation_xml_integration_point(handle)
}
}
fn append_to_existing_text(prev: &NodeRef, text: &StrTendril) -> bool {
prev.tree
.update_node(&prev.id, |tree_node| match tree_node.data {
NodeData::Text { ref mut contents } => {
#[cfg(not(feature = "atomic"))]
contents.push_tendril(text);
#[cfg(feature = "atomic")]
contents.push_slice(text);
true
}
_ => false,
})
.unwrap_or(false)
}
#[cfg(feature = "markdown")]
impl Document {
/// Produces a *Markdown* representation of the [`Document`],
/// skipping elements matching the specified `skip_tags` list along with their descendants.
///
/// - If `skip_tags` is `None`, the default list is used: `["script", "style", "meta", "head"]`.
/// - To process all elements without exclusions, pass `Some(&[])`.
pub fn md(&self, skip_tags: Option<&[&str]>) -> StrTendril {
self.root().md(skip_tags)
}
}