Gitiles
Code Review Sign In
gerrit.openbmc.org / stefanberger / openbmc / 7b16ede1f421831d428078d2998561d5f921c71d / . / poky / bitbake / lib / bs4 / tests / test_tree.py
blob: 6d3e67f3110038eae32cab7f1995d275d8d057e8 [file] [log] [blame]
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1# -*- coding: utf-8 -*-
2"""Tests for Beautiful Soup's tree traversal methods.
3
4The tree traversal methods are the main advantage of using Beautiful
5Soup over just using a parser.
6
7Different parsers will build different Beautiful Soup trees given the
8same markup, but all Beautiful Soup trees can be traversed with the
9methods tested here.
10"""
11
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]12from pdb import set_trace
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]13import copy
14import pickle
15import re
16import warnings
17from bs4 import BeautifulSoup
18from bs4.builder import (
19 builder_registry,
20 HTMLParserTreeBuilder,
21)
22from bs4.element import (
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]23 PY3K,
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]24 CData,
25 Comment,
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]26 Declaration,
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]27 Doctype,
28 NavigableString,
29 SoupStrainer,
30 Tag,
31)
32from bs4.testing import (
33 SoupTest,
34 skipIf,
35)
36
37XML_BUILDER_PRESENT = (builder_registry.lookup("xml") is not None)
38LXML_PRESENT = (builder_registry.lookup("lxml") is not None)
39
40class TreeTest(SoupTest):
41
42 def assertSelects(self, tags, should_match):
43 """Make sure that the given tags have the correct text.
44
45 This is used in tests that define a bunch of tags, each
46 containing a single string, and then select certain strings by
47 some mechanism.
48 """
49 self.assertEqual([tag.string for tag in tags], should_match)
50
51 def assertSelectsIDs(self, tags, should_match):
52 """Make sure that the given tags have the correct IDs.
53
54 This is used in tests that define a bunch of tags, each
55 containing a single string, and then select certain strings by
56 some mechanism.
57 """
58 self.assertEqual([tag['id'] for tag in tags], should_match)
59
60
61class TestFind(TreeTest):
62 """Basic tests of the find() method.
63
64 find() just calls find_all() with limit=1, so it's not tested all
65 that thouroughly here.
66 """
67
68 def test_find_tag(self):
69 soup = self.soup("<a>1</a><b>2</b><a>3</a><b>4</b>")
70 self.assertEqual(soup.find("b").string, "2")
71
72 def test_unicode_text_find(self):
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]73 soup = self.soup('<h1>Räksmörgås</h1>')
74 self.assertEqual(soup.find(string='Räksmörgås'), 'Räksmörgås')
75
76 def test_unicode_attribute_find(self):
77 soup = self.soup('<h1 id="Räksmörgås">here it is</h1>')
78 str(soup)
79 self.assertEqual("here it is", soup.find(id='Räksmörgås').text)
80
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]81
82 def test_find_everything(self):
83 """Test an optimization that finds all tags."""
84 soup = self.soup("<a>foo</a><b>bar</b>")
85 self.assertEqual(2, len(soup.find_all()))
86
87 def test_find_everything_with_name(self):
88 """Test an optimization that finds all tags with a given name."""
89 soup = self.soup("<a>foo</a><b>bar</b><a>baz</a>")
90 self.assertEqual(2, len(soup.find_all('a')))
91
92class TestFindAll(TreeTest):
93 """Basic tests of the find_all() method."""
94
95 def test_find_all_text_nodes(self):
96 """You can search the tree for text nodes."""
97 soup = self.soup("<html>Foo<b>bar</b>\xbb</html>")
98 # Exact match.
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]99 self.assertEqual(soup.find_all(string="bar"), ["bar"])
100 self.assertEqual(soup.find_all(text="bar"), ["bar"])
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]101 # Match any of a number of strings.
102 self.assertEqual(
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]103 soup.find_all(text=["Foo", "bar"]), ["Foo", "bar"])
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]104 # Match a regular expression.
105 self.assertEqual(soup.find_all(text=re.compile('.*')),
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]106 ["Foo", "bar", '\xbb'])
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]107 # Match anything.
108 self.assertEqual(soup.find_all(text=True),
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]109 ["Foo", "bar", '\xbb'])
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]110
111 def test_find_all_limit(self):
112 """You can limit the number of items returned by find_all."""
113 soup = self.soup("<a>1</a><a>2</a><a>3</a><a>4</a><a>5</a>")
114 self.assertSelects(soup.find_all('a', limit=3), ["1", "2", "3"])
115 self.assertSelects(soup.find_all('a', limit=1), ["1"])
116 self.assertSelects(
117 soup.find_all('a', limit=10), ["1", "2", "3", "4", "5"])
118
119 # A limit of 0 means no limit.
120 self.assertSelects(
121 soup.find_all('a', limit=0), ["1", "2", "3", "4", "5"])
122
123 def test_calling_a_tag_is_calling_findall(self):
124 soup = self.soup("<a>1</a><b>2<a id='foo'>3</a></b>")
125 self.assertSelects(soup('a', limit=1), ["1"])
126 self.assertSelects(soup.b(id="foo"), ["3"])
127
128 def test_find_all_with_self_referential_data_structure_does_not_cause_infinite_recursion(self):
129 soup = self.soup("<a></a>")
130 # Create a self-referential list.
131 l = []
132 l.append(l)
133
134 # Without special code in _normalize_search_value, this would cause infinite
135 # recursion.
136 self.assertEqual([], soup.find_all(l))
137
138 def test_find_all_resultset(self):
139 """All find_all calls return a ResultSet"""
140 soup = self.soup("<a></a>")
141 result = soup.find_all("a")
142 self.assertTrue(hasattr(result, "source"))
143
144 result = soup.find_all(True)
145 self.assertTrue(hasattr(result, "source"))
146
147 result = soup.find_all(text="foo")
148 self.assertTrue(hasattr(result, "source"))
149
150
151class TestFindAllBasicNamespaces(TreeTest):
152
153 def test_find_by_namespaced_name(self):
154 soup = self.soup('<mathml:msqrt>4</mathml:msqrt><a svg:fill="red">')
155 self.assertEqual("4", soup.find("mathml:msqrt").string)
156 self.assertEqual("a", soup.find(attrs= { "svg:fill" : "red" }).name)
157
158
159class TestFindAllByName(TreeTest):
160 """Test ways of finding tags by tag name."""
161
162 def setUp(self):
163 super(TreeTest, self).setUp()
164 self.tree = self.soup("""<a>First tag.</a>
165 <b>Second tag.</b>
166 <c>Third <a>Nested tag.</a> tag.</c>""")
167
168 def test_find_all_by_tag_name(self):
169 # Find all the <a> tags.
170 self.assertSelects(
171 self.tree.find_all('a'), ['First tag.', 'Nested tag.'])
172
173 def test_find_all_by_name_and_text(self):
174 self.assertSelects(
175 self.tree.find_all('a', text='First tag.'), ['First tag.'])
176
177 self.assertSelects(
178 self.tree.find_all('a', text=True), ['First tag.', 'Nested tag.'])
179
180 self.assertSelects(
181 self.tree.find_all('a', text=re.compile("tag")),
182 ['First tag.', 'Nested tag.'])
183
184
185 def test_find_all_on_non_root_element(self):
186 # You can call find_all on any node, not just the root.
187 self.assertSelects(self.tree.c.find_all('a'), ['Nested tag.'])
188
189 def test_calling_element_invokes_find_all(self):
190 self.assertSelects(self.tree('a'), ['First tag.', 'Nested tag.'])
191
192 def test_find_all_by_tag_strainer(self):
193 self.assertSelects(
194 self.tree.find_all(SoupStrainer('a')),
195 ['First tag.', 'Nested tag.'])
196
197 def test_find_all_by_tag_names(self):
198 self.assertSelects(
199 self.tree.find_all(['a', 'b']),
200 ['First tag.', 'Second tag.', 'Nested tag.'])
201
202 def test_find_all_by_tag_dict(self):
203 self.assertSelects(
204 self.tree.find_all({'a' : True, 'b' : True}),
205 ['First tag.', 'Second tag.', 'Nested tag.'])
206
207 def test_find_all_by_tag_re(self):
208 self.assertSelects(
209 self.tree.find_all(re.compile('^[ab]$')),
210 ['First tag.', 'Second tag.', 'Nested tag.'])
211
212 def test_find_all_with_tags_matching_method(self):
213 # You can define an oracle method that determines whether
214 # a tag matches the search.
215 def id_matches_name(tag):
216 return tag.name == tag.get('id')
217
218 tree = self.soup("""<a id="a">Match 1.</a>
219 <a id="1">Does not match.</a>
220 <b id="b">Match 2.</a>""")
221
222 self.assertSelects(
223 tree.find_all(id_matches_name), ["Match 1.", "Match 2."])
224
225
226class TestFindAllByAttribute(TreeTest):
227
228 def test_find_all_by_attribute_name(self):
229 # You can pass in keyword arguments to find_all to search by
230 # attribute.
231 tree = self.soup("""
232 <a id="first">Matching a.</a>
233 <a id="second">
234 Non-matching <b id="first">Matching b.</b>a.
235 </a>""")
236 self.assertSelects(tree.find_all(id='first'),
237 ["Matching a.", "Matching b."])
238
239 def test_find_all_by_utf8_attribute_value(self):
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]240 peace = "םולש".encode("utf8")
241 data = '<a title="םולש"></a>'.encode("utf8")
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]242 soup = self.soup(data)
243 self.assertEqual([soup.a], soup.find_all(title=peace))
244 self.assertEqual([soup.a], soup.find_all(title=peace.decode("utf8")))
245 self.assertEqual([soup.a], soup.find_all(title=[peace, "something else"]))
246
247 def test_find_all_by_attribute_dict(self):
248 # You can pass in a dictionary as the argument 'attrs'. This
249 # lets you search for attributes like 'name' (a fixed argument
250 # to find_all) and 'class' (a reserved word in Python.)
251 tree = self.soup("""
252 <a name="name1" class="class1">Name match.</a>
253 <a name="name2" class="class2">Class match.</a>
254 <a name="name3" class="class3">Non-match.</a>
255 <name1>A tag called 'name1'.</name1>
256 """)
257
258 # This doesn't do what you want.
259 self.assertSelects(tree.find_all(name='name1'),
260 ["A tag called 'name1'."])
261 # This does what you want.
262 self.assertSelects(tree.find_all(attrs={'name' : 'name1'}),
263 ["Name match."])
264
265 self.assertSelects(tree.find_all(attrs={'class' : 'class2'}),
266 ["Class match."])
267
268 def test_find_all_by_class(self):
269 tree = self.soup("""
270 <a class="1">Class 1.</a>
271 <a class="2">Class 2.</a>
272 <b class="1">Class 1.</b>
273 <c class="3 4">Class 3 and 4.</c>
274 """)
275
276 # Passing in the class_ keyword argument will search against
277 # the 'class' attribute.
278 self.assertSelects(tree.find_all('a', class_='1'), ['Class 1.'])
279 self.assertSelects(tree.find_all('c', class_='3'), ['Class 3 and 4.'])
280 self.assertSelects(tree.find_all('c', class_='4'), ['Class 3 and 4.'])
281
282 # Passing in a string to 'attrs' will also search the CSS class.
283 self.assertSelects(tree.find_all('a', '1'), ['Class 1.'])
284 self.assertSelects(tree.find_all(attrs='1'), ['Class 1.', 'Class 1.'])
285 self.assertSelects(tree.find_all('c', '3'), ['Class 3 and 4.'])
286 self.assertSelects(tree.find_all('c', '4'), ['Class 3 and 4.'])
287
288 def test_find_by_class_when_multiple_classes_present(self):
289 tree = self.soup("<gar class='foo bar'>Found it</gar>")
290
291 f = tree.find_all("gar", class_=re.compile("o"))
292 self.assertSelects(f, ["Found it"])
293
294 f = tree.find_all("gar", class_=re.compile("a"))
295 self.assertSelects(f, ["Found it"])
296
297 # Since the class is not the string "foo bar", but the two
298 # strings "foo" and "bar", this will not find anything.
299 f = tree.find_all("gar", class_=re.compile("o b"))
300 self.assertSelects(f, [])
301
302 def test_find_all_with_non_dictionary_for_attrs_finds_by_class(self):
303 soup = self.soup("<a class='bar'>Found it</a>")
304
305 self.assertSelects(soup.find_all("a", re.compile("ba")), ["Found it"])
306
307 def big_attribute_value(value):
308 return len(value) > 3
309
310 self.assertSelects(soup.find_all("a", big_attribute_value), [])
311
312 def small_attribute_value(value):
313 return len(value) <= 3
314
315 self.assertSelects(
316 soup.find_all("a", small_attribute_value), ["Found it"])
317
318 def test_find_all_with_string_for_attrs_finds_multiple_classes(self):
319 soup = self.soup('<a class="foo bar"></a><a class="foo"></a>')
320 a, a2 = soup.find_all("a")
321 self.assertEqual([a, a2], soup.find_all("a", "foo"))
322 self.assertEqual([a], soup.find_all("a", "bar"))
323
324 # If you specify the class as a string that contains a
325 # space, only that specific value will be found.
326 self.assertEqual([a], soup.find_all("a", class_="foo bar"))
327 self.assertEqual([a], soup.find_all("a", "foo bar"))
328 self.assertEqual([], soup.find_all("a", "bar foo"))
329
330 def test_find_all_by_attribute_soupstrainer(self):
331 tree = self.soup("""
332 <a id="first">Match.</a>
333 <a id="second">Non-match.</a>""")
334
335 strainer = SoupStrainer(attrs={'id' : 'first'})
336 self.assertSelects(tree.find_all(strainer), ['Match.'])
337
338 def test_find_all_with_missing_atribute(self):
339 # You can pass in None as the value of an attribute to find_all.
340 # This will match tags that do not have that attribute set.
341 tree = self.soup("""<a id="1">ID present.</a>
342 <a>No ID present.</a>
343 <a id="">ID is empty.</a>""")
344 self.assertSelects(tree.find_all('a', id=None), ["No ID present."])
345
346 def test_find_all_with_defined_attribute(self):
347 # You can pass in None as the value of an attribute to find_all.
348 # This will match tags that have that attribute set to any value.
349 tree = self.soup("""<a id="1">ID present.</a>
350 <a>No ID present.</a>
351 <a id="">ID is empty.</a>""")
352 self.assertSelects(
353 tree.find_all(id=True), ["ID present.", "ID is empty."])
354
355 def test_find_all_with_numeric_attribute(self):
356 # If you search for a number, it's treated as a string.
357 tree = self.soup("""<a id=1>Unquoted attribute.</a>
358 <a id="1">Quoted attribute.</a>""")
359
360 expected = ["Unquoted attribute.", "Quoted attribute."]
361 self.assertSelects(tree.find_all(id=1), expected)
362 self.assertSelects(tree.find_all(id="1"), expected)
363
364 def test_find_all_with_list_attribute_values(self):
365 # You can pass a list of attribute values instead of just one,
366 # and you'll get tags that match any of the values.
367 tree = self.soup("""<a id="1">1</a>
368 <a id="2">2</a>
369 <a id="3">3</a>
370 <a>No ID.</a>""")
371 self.assertSelects(tree.find_all(id=["1", "3", "4"]),
372 ["1", "3"])
373
374 def test_find_all_with_regular_expression_attribute_value(self):
375 # You can pass a regular expression as an attribute value, and
376 # you'll get tags whose values for that attribute match the
377 # regular expression.
378 tree = self.soup("""<a id="a">One a.</a>
379 <a id="aa">Two as.</a>
380 <a id="ab">Mixed as and bs.</a>
381 <a id="b">One b.</a>
382 <a>No ID.</a>""")
383
384 self.assertSelects(tree.find_all(id=re.compile("^a+$")),
385 ["One a.", "Two as."])
386
387 def test_find_by_name_and_containing_string(self):
388 soup = self.soup("<b>foo</b><b>bar</b><a>foo</a>")
389 a = soup.a
390
391 self.assertEqual([a], soup.find_all("a", text="foo"))
392 self.assertEqual([], soup.find_all("a", text="bar"))
393 self.assertEqual([], soup.find_all("a", text="bar"))
394
395 def test_find_by_name_and_containing_string_when_string_is_buried(self):
396 soup = self.soup("<a>foo</a><a><b><c>foo</c></b></a>")
397 self.assertEqual(soup.find_all("a"), soup.find_all("a", text="foo"))
398
399 def test_find_by_attribute_and_containing_string(self):
400 soup = self.soup('<b id="1">foo</b><a id="2">foo</a>')
401 a = soup.a
402
403 self.assertEqual([a], soup.find_all(id=2, text="foo"))
404 self.assertEqual([], soup.find_all(id=1, text="bar"))
405
406
407
408
409class TestIndex(TreeTest):
410 """Test Tag.index"""
411 def test_index(self):
412 tree = self.soup("""<div>
413 <a>Identical</a>
414 <b>Not identical</b>
415 <a>Identical</a>
416
417 <c><d>Identical with child</d></c>
418 <b>Also not identical</b>
419 <c><d>Identical with child</d></c>
420 </div>""")
421 div = tree.div
422 for i, element in enumerate(div.contents):
423 self.assertEqual(i, div.index(element))
424 self.assertRaises(ValueError, tree.index, 1)
425
426
427class TestParentOperations(TreeTest):
428 """Test navigation and searching through an element's parents."""
429
430 def setUp(self):
431 super(TestParentOperations, self).setUp()
432 self.tree = self.soup('''<ul id="empty"></ul>
433 <ul id="top">
434 <ul id="middle">
435 <ul id="bottom">
436 <b>Start here</b>
437 </ul>
438 </ul>''')
439 self.start = self.tree.b
440
441
442 def test_parent(self):
443 self.assertEqual(self.start.parent['id'], 'bottom')
444 self.assertEqual(self.start.parent.parent['id'], 'middle')
445 self.assertEqual(self.start.parent.parent.parent['id'], 'top')
446
447 def test_parent_of_top_tag_is_soup_object(self):
448 top_tag = self.tree.contents[0]
449 self.assertEqual(top_tag.parent, self.tree)
450
451 def test_soup_object_has_no_parent(self):
452 self.assertEqual(None, self.tree.parent)
453
454 def test_find_parents(self):
455 self.assertSelectsIDs(
456 self.start.find_parents('ul'), ['bottom', 'middle', 'top'])
457 self.assertSelectsIDs(
458 self.start.find_parents('ul', id="middle"), ['middle'])
459
460 def test_find_parent(self):
461 self.assertEqual(self.start.find_parent('ul')['id'], 'bottom')
462 self.assertEqual(self.start.find_parent('ul', id='top')['id'], 'top')
463
464 def test_parent_of_text_element(self):
465 text = self.tree.find(text="Start here")
466 self.assertEqual(text.parent.name, 'b')
467
468 def test_text_element_find_parent(self):
469 text = self.tree.find(text="Start here")
470 self.assertEqual(text.find_parent('ul')['id'], 'bottom')
471
472 def test_parent_generator(self):
473 parents = [parent['id'] for parent in self.start.parents
474 if parent is not None and 'id' in parent.attrs]
475 self.assertEqual(parents, ['bottom', 'middle', 'top'])
476
477
478class ProximityTest(TreeTest):
479
480 def setUp(self):
481 super(TreeTest, self).setUp()
482 self.tree = self.soup(
483 '<html id="start"><head></head><body><b id="1">One</b><b id="2">Two</b><b id="3">Three</b></body></html>')
484
485
486class TestNextOperations(ProximityTest):
487
488 def setUp(self):
489 super(TestNextOperations, self).setUp()
490 self.start = self.tree.b
491
492 def test_next(self):
493 self.assertEqual(self.start.next_element, "One")
494 self.assertEqual(self.start.next_element.next_element['id'], "2")
495
496 def test_next_of_last_item_is_none(self):
497 last = self.tree.find(text="Three")
498 self.assertEqual(last.next_element, None)
499
500 def test_next_of_root_is_none(self):
501 # The document root is outside the next/previous chain.
502 self.assertEqual(self.tree.next_element, None)
503
504 def test_find_all_next(self):
505 self.assertSelects(self.start.find_all_next('b'), ["Two", "Three"])
506 self.start.find_all_next(id=3)
507 self.assertSelects(self.start.find_all_next(id=3), ["Three"])
508
509 def test_find_next(self):
510 self.assertEqual(self.start.find_next('b')['id'], '2')
511 self.assertEqual(self.start.find_next(text="Three"), "Three")
512
513 def test_find_next_for_text_element(self):
514 text = self.tree.find(text="One")
515 self.assertEqual(text.find_next("b").string, "Two")
516 self.assertSelects(text.find_all_next("b"), ["Two", "Three"])
517
518 def test_next_generator(self):
519 start = self.tree.find(text="Two")
520 successors = [node for node in start.next_elements]
521 # There are two successors: the final <b> tag and its text contents.
522 tag, contents = successors
523 self.assertEqual(tag['id'], '3')
524 self.assertEqual(contents, "Three")
525
526class TestPreviousOperations(ProximityTest):
527
528 def setUp(self):
529 super(TestPreviousOperations, self).setUp()
530 self.end = self.tree.find(text="Three")
531
532 def test_previous(self):
533 self.assertEqual(self.end.previous_element['id'], "3")
534 self.assertEqual(self.end.previous_element.previous_element, "Two")
535
536 def test_previous_of_first_item_is_none(self):
537 first = self.tree.find('html')
538 self.assertEqual(first.previous_element, None)
539
540 def test_previous_of_root_is_none(self):
541 # The document root is outside the next/previous chain.
542 # XXX This is broken!
543 #self.assertEqual(self.tree.previous_element, None)
544 pass
545
546 def test_find_all_previous(self):
547 # The <b> tag containing the "Three" node is the predecessor
548 # of the "Three" node itself, which is why "Three" shows up
549 # here.
550 self.assertSelects(
551 self.end.find_all_previous('b'), ["Three", "Two", "One"])
552 self.assertSelects(self.end.find_all_previous(id=1), ["One"])
553
554 def test_find_previous(self):
555 self.assertEqual(self.end.find_previous('b')['id'], '3')
556 self.assertEqual(self.end.find_previous(text="One"), "One")
557
558 def test_find_previous_for_text_element(self):
559 text = self.tree.find(text="Three")
560 self.assertEqual(text.find_previous("b").string, "Three")
561 self.assertSelects(
562 text.find_all_previous("b"), ["Three", "Two", "One"])
563
564 def test_previous_generator(self):
565 start = self.tree.find(text="One")
566 predecessors = [node for node in start.previous_elements]
567
568 # There are four predecessors: the <b> tag containing "One"
569 # the <body> tag, the <head> tag, and the <html> tag.
570 b, body, head, html = predecessors
571 self.assertEqual(b['id'], '1')
572 self.assertEqual(body.name, "body")
573 self.assertEqual(head.name, "head")
574 self.assertEqual(html.name, "html")
575
576
577class SiblingTest(TreeTest):
578
579 def setUp(self):
580 super(SiblingTest, self).setUp()
581 markup = '''<html>
582 <span id="1">
583 <span id="1.1"></span>
584 </span>
585 <span id="2">
586 <span id="2.1"></span>
587 </span>
588 <span id="3">
589 <span id="3.1"></span>
590 </span>
591 <span id="4"></span>
592 </html>'''
593 # All that whitespace looks good but makes the tests more
594 # difficult. Get rid of it.
595 markup = re.compile("\n\s*").sub("", markup)
596 self.tree = self.soup(markup)
597
598
599class TestNextSibling(SiblingTest):
600
601 def setUp(self):
602 super(TestNextSibling, self).setUp()
603 self.start = self.tree.find(id="1")
604
605 def test_next_sibling_of_root_is_none(self):
606 self.assertEqual(self.tree.next_sibling, None)
607
608 def test_next_sibling(self):
609 self.assertEqual(self.start.next_sibling['id'], '2')
610 self.assertEqual(self.start.next_sibling.next_sibling['id'], '3')
611
612 # Note the difference between next_sibling and next_element.
613 self.assertEqual(self.start.next_element['id'], '1.1')
614
615 def test_next_sibling_may_not_exist(self):
616 self.assertEqual(self.tree.html.next_sibling, None)
617
618 nested_span = self.tree.find(id="1.1")
619 self.assertEqual(nested_span.next_sibling, None)
620
621 last_span = self.tree.find(id="4")
622 self.assertEqual(last_span.next_sibling, None)
623
624 def test_find_next_sibling(self):
625 self.assertEqual(self.start.find_next_sibling('span')['id'], '2')
626
627 def test_next_siblings(self):
628 self.assertSelectsIDs(self.start.find_next_siblings("span"),
629 ['2', '3', '4'])
630
631 self.assertSelectsIDs(self.start.find_next_siblings(id='3'), ['3'])
632
633 def test_next_sibling_for_text_element(self):
634 soup = self.soup("Foo<b>bar</b>baz")
635 start = soup.find(text="Foo")
636 self.assertEqual(start.next_sibling.name, 'b')
637 self.assertEqual(start.next_sibling.next_sibling, 'baz')
638
639 self.assertSelects(start.find_next_siblings('b'), ['bar'])
640 self.assertEqual(start.find_next_sibling(text="baz"), "baz")
641 self.assertEqual(start.find_next_sibling(text="nonesuch"), None)
642
643
644class TestPreviousSibling(SiblingTest):
645
646 def setUp(self):
647 super(TestPreviousSibling, self).setUp()
648 self.end = self.tree.find(id="4")
649
650 def test_previous_sibling_of_root_is_none(self):
651 self.assertEqual(self.tree.previous_sibling, None)
652
653 def test_previous_sibling(self):
654 self.assertEqual(self.end.previous_sibling['id'], '3')
655 self.assertEqual(self.end.previous_sibling.previous_sibling['id'], '2')
656
657 # Note the difference between previous_sibling and previous_element.
658 self.assertEqual(self.end.previous_element['id'], '3.1')
659
660 def test_previous_sibling_may_not_exist(self):
661 self.assertEqual(self.tree.html.previous_sibling, None)
662
663 nested_span = self.tree.find(id="1.1")
664 self.assertEqual(nested_span.previous_sibling, None)
665
666 first_span = self.tree.find(id="1")
667 self.assertEqual(first_span.previous_sibling, None)
668
669 def test_find_previous_sibling(self):
670 self.assertEqual(self.end.find_previous_sibling('span')['id'], '3')
671
672 def test_previous_siblings(self):
673 self.assertSelectsIDs(self.end.find_previous_siblings("span"),
674 ['3', '2', '1'])
675
676 self.assertSelectsIDs(self.end.find_previous_siblings(id='1'), ['1'])
677
678 def test_previous_sibling_for_text_element(self):
679 soup = self.soup("Foo<b>bar</b>baz")
680 start = soup.find(text="baz")
681 self.assertEqual(start.previous_sibling.name, 'b')
682 self.assertEqual(start.previous_sibling.previous_sibling, 'Foo')
683
684 self.assertSelects(start.find_previous_siblings('b'), ['bar'])
685 self.assertEqual(start.find_previous_sibling(text="Foo"), "Foo")
686 self.assertEqual(start.find_previous_sibling(text="nonesuch"), None)
687
688
689class TestTagCreation(SoupTest):
690 """Test the ability to create new tags."""
691 def test_new_tag(self):
692 soup = self.soup("")
693 new_tag = soup.new_tag("foo", bar="baz")
694 self.assertTrue(isinstance(new_tag, Tag))
695 self.assertEqual("foo", new_tag.name)
696 self.assertEqual(dict(bar="baz"), new_tag.attrs)
697 self.assertEqual(None, new_tag.parent)
698
699 def test_tag_inherits_self_closing_rules_from_builder(self):
700 if XML_BUILDER_PRESENT:
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]701 xml_soup = BeautifulSoup("", "lxml-xml")
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]702 xml_br = xml_soup.new_tag("br")
703 xml_p = xml_soup.new_tag("p")
704
705 # Both the <br> and <p> tag are empty-element, just because
706 # they have no contents.
707 self.assertEqual(b"<br/>", xml_br.encode())
708 self.assertEqual(b"<p/>", xml_p.encode())
709
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]710 html_soup = BeautifulSoup("", "html.parser")
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]711 html_br = html_soup.new_tag("br")
712 html_p = html_soup.new_tag("p")
713
714 # The HTML builder users HTML's rules about which tags are
715 # empty-element tags, and the new tags reflect these rules.
716 self.assertEqual(b"<br/>", html_br.encode())
717 self.assertEqual(b"<p></p>", html_p.encode())
718
719 def test_new_string_creates_navigablestring(self):
720 soup = self.soup("")
721 s = soup.new_string("foo")
722 self.assertEqual("foo", s)
723 self.assertTrue(isinstance(s, NavigableString))
724
725 def test_new_string_can_create_navigablestring_subclass(self):
726 soup = self.soup("")
727 s = soup.new_string("foo", Comment)
728 self.assertEqual("foo", s)
729 self.assertTrue(isinstance(s, Comment))
730
731class TestTreeModification(SoupTest):
732
733 def test_attribute_modification(self):
734 soup = self.soup('<a id="1"></a>')
735 soup.a['id'] = 2
736 self.assertEqual(soup.decode(), self.document_for('<a id="2"></a>'))
737 del(soup.a['id'])
738 self.assertEqual(soup.decode(), self.document_for('<a></a>'))
739 soup.a['id2'] = 'foo'
740 self.assertEqual(soup.decode(), self.document_for('<a id2="foo"></a>'))
741
742 def test_new_tag_creation(self):
743 builder = builder_registry.lookup('html')()
744 soup = self.soup("<body></body>", builder=builder)
745 a = Tag(soup, builder, 'a')
746 ol = Tag(soup, builder, 'ol')
747 a['href'] = 'http://foo.com/'
748 soup.body.insert(0, a)
749 soup.body.insert(1, ol)
750 self.assertEqual(
751 soup.body.encode(),
752 b'<body><a href="http://foo.com/"></a><ol></ol></body>')
753
754 def test_append_to_contents_moves_tag(self):
755 doc = """<p id="1">Don't leave me <b>here</b>.</p>
756 <p id="2">Don\'t leave!</p>"""
757 soup = self.soup(doc)
758 second_para = soup.find(id='2')
759 bold = soup.b
760
761 # Move the <b> tag to the end of the second paragraph.
762 soup.find(id='2').append(soup.b)
763
764 # The <b> tag is now a child of the second paragraph.
765 self.assertEqual(bold.parent, second_para)
766
767 self.assertEqual(
768 soup.decode(), self.document_for(
769 '<p id="1">Don\'t leave me .</p>\n'
770 '<p id="2">Don\'t leave!<b>here</b></p>'))
771
772 def test_replace_with_returns_thing_that_was_replaced(self):
773 text = "<a></a><b><c></c></b>"
774 soup = self.soup(text)
775 a = soup.a
776 new_a = a.replace_with(soup.c)
777 self.assertEqual(a, new_a)
778
779 def test_unwrap_returns_thing_that_was_replaced(self):
780 text = "<a><b></b><c></c></a>"
781 soup = self.soup(text)
782 a = soup.a
783 new_a = a.unwrap()
784 self.assertEqual(a, new_a)
785
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]786 def test_replace_with_and_unwrap_give_useful_exception_when_tag_has_no_parent(self):
787 soup = self.soup("<a><b>Foo</b></a><c>Bar</c>")
788 a = soup.a
789 a.extract()
790 self.assertEqual(None, a.parent)
791 self.assertRaises(ValueError, a.unwrap)
792 self.assertRaises(ValueError, a.replace_with, soup.c)
793
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]794 def test_replace_tag_with_itself(self):
795 text = "<a><b></b><c>Foo<d></d></c></a><a><e></e></a>"
796 soup = self.soup(text)
797 c = soup.c
798 soup.c.replace_with(c)
799 self.assertEqual(soup.decode(), self.document_for(text))
800
801 def test_replace_tag_with_its_parent_raises_exception(self):
802 text = "<a><b></b></a>"
803 soup = self.soup(text)
804 self.assertRaises(ValueError, soup.b.replace_with, soup.a)
805
806 def test_insert_tag_into_itself_raises_exception(self):
807 text = "<a><b></b></a>"
808 soup = self.soup(text)
809 self.assertRaises(ValueError, soup.a.insert, 0, soup.a)
810
811 def test_replace_with_maintains_next_element_throughout(self):
812 soup = self.soup('<p><a>one</a><b>three</b></p>')
813 a = soup.a
814 b = a.contents[0]
815 # Make it so the <a> tag has two text children.
816 a.insert(1, "two")
817
818 # Now replace each one with the empty string.
819 left, right = a.contents
820 left.replaceWith('')
821 right.replaceWith('')
822
823 # The <b> tag is still connected to the tree.
824 self.assertEqual("three", soup.b.string)
825
826 def test_replace_final_node(self):
827 soup = self.soup("<b>Argh!</b>")
828 soup.find(text="Argh!").replace_with("Hooray!")
829 new_text = soup.find(text="Hooray!")
830 b = soup.b
831 self.assertEqual(new_text.previous_element, b)
832 self.assertEqual(new_text.parent, b)
833 self.assertEqual(new_text.previous_element.next_element, new_text)
834 self.assertEqual(new_text.next_element, None)
835
836 def test_consecutive_text_nodes(self):
837 # A builder should never create two consecutive text nodes,
838 # but if you insert one next to another, Beautiful Soup will
839 # handle it correctly.
840 soup = self.soup("<a><b>Argh!</b><c></c></a>")
841 soup.b.insert(1, "Hooray!")
842
843 self.assertEqual(
844 soup.decode(), self.document_for(
845 "<a><b>Argh!Hooray!</b><c></c></a>"))
846
847 new_text = soup.find(text="Hooray!")
848 self.assertEqual(new_text.previous_element, "Argh!")
849 self.assertEqual(new_text.previous_element.next_element, new_text)
850
851 self.assertEqual(new_text.previous_sibling, "Argh!")
852 self.assertEqual(new_text.previous_sibling.next_sibling, new_text)
853
854 self.assertEqual(new_text.next_sibling, None)
855 self.assertEqual(new_text.next_element, soup.c)
856
857 def test_insert_string(self):
858 soup = self.soup("<a></a>")
859 soup.a.insert(0, "bar")
860 soup.a.insert(0, "foo")
861 # The string were added to the tag.
862 self.assertEqual(["foo", "bar"], soup.a.contents)
863 # And they were converted to NavigableStrings.
864 self.assertEqual(soup.a.contents[0].next_element, "bar")
865
866 def test_insert_tag(self):
867 builder = self.default_builder
868 soup = self.soup(
869 "<a><b>Find</b><c>lady!</c><d></d></a>", builder=builder)
870 magic_tag = Tag(soup, builder, 'magictag')
871 magic_tag.insert(0, "the")
872 soup.a.insert(1, magic_tag)
873
874 self.assertEqual(
875 soup.decode(), self.document_for(
876 "<a><b>Find</b><magictag>the</magictag><c>lady!</c><d></d></a>"))
877
878 # Make sure all the relationships are hooked up correctly.
879 b_tag = soup.b
880 self.assertEqual(b_tag.next_sibling, magic_tag)
881 self.assertEqual(magic_tag.previous_sibling, b_tag)
882
883 find = b_tag.find(text="Find")
884 self.assertEqual(find.next_element, magic_tag)
885 self.assertEqual(magic_tag.previous_element, find)
886
887 c_tag = soup.c
888 self.assertEqual(magic_tag.next_sibling, c_tag)
889 self.assertEqual(c_tag.previous_sibling, magic_tag)
890
891 the = magic_tag.find(text="the")
892 self.assertEqual(the.parent, magic_tag)
893 self.assertEqual(the.next_element, c_tag)
894 self.assertEqual(c_tag.previous_element, the)
895
896 def test_append_child_thats_already_at_the_end(self):
897 data = "<a><b></b></a>"
898 soup = self.soup(data)
899 soup.a.append(soup.b)
900 self.assertEqual(data, soup.decode())
901
902 def test_move_tag_to_beginning_of_parent(self):
903 data = "<a><b></b><c></c><d></d></a>"
904 soup = self.soup(data)
905 soup.a.insert(0, soup.d)
906 self.assertEqual("<a><d></d><b></b><c></c></a>", soup.decode())
907
908 def test_insert_works_on_empty_element_tag(self):
909 # This is a little strange, since most HTML parsers don't allow
910 # markup like this to come through. But in general, we don't
911 # know what the parser would or wouldn't have allowed, so
912 # I'm letting this succeed for now.
913 soup = self.soup("<br/>")
914 soup.br.insert(1, "Contents")
915 self.assertEqual(str(soup.br), "<br>Contents</br>")
916
917 def test_insert_before(self):
918 soup = self.soup("<a>foo</a><b>bar</b>")
919 soup.b.insert_before("BAZ")
920 soup.a.insert_before("QUUX")
921 self.assertEqual(
922 soup.decode(), self.document_for("QUUX<a>foo</a>BAZ<b>bar</b>"))
923
924 soup.a.insert_before(soup.b)
925 self.assertEqual(
926 soup.decode(), self.document_for("QUUX<b>bar</b><a>foo</a>BAZ"))
927
928 def test_insert_after(self):
929 soup = self.soup("<a>foo</a><b>bar</b>")
930 soup.b.insert_after("BAZ")
931 soup.a.insert_after("QUUX")
932 self.assertEqual(
933 soup.decode(), self.document_for("<a>foo</a>QUUX<b>bar</b>BAZ"))
934 soup.b.insert_after(soup.a)
935 self.assertEqual(
936 soup.decode(), self.document_for("QUUX<b>bar</b><a>foo</a>BAZ"))
937
938 def test_insert_after_raises_exception_if_after_has_no_meaning(self):
939 soup = self.soup("")
940 tag = soup.new_tag("a")
941 string = soup.new_string("")
942 self.assertRaises(ValueError, string.insert_after, tag)
943 self.assertRaises(NotImplementedError, soup.insert_after, tag)
944 self.assertRaises(ValueError, tag.insert_after, tag)
945
946 def test_insert_before_raises_notimplementederror_if_before_has_no_meaning(self):
947 soup = self.soup("")
948 tag = soup.new_tag("a")
949 string = soup.new_string("")
950 self.assertRaises(ValueError, string.insert_before, tag)
951 self.assertRaises(NotImplementedError, soup.insert_before, tag)
952 self.assertRaises(ValueError, tag.insert_before, tag)
953
954 def test_replace_with(self):
955 soup = self.soup(
956 "<p>There's <b>no</b> business like <b>show</b> business</p>")
957 no, show = soup.find_all('b')
958 show.replace_with(no)
959 self.assertEqual(
960 soup.decode(),
961 self.document_for(
962 "<p>There's business like <b>no</b> business</p>"))
963
964 self.assertEqual(show.parent, None)
965 self.assertEqual(no.parent, soup.p)
966 self.assertEqual(no.next_element, "no")
967 self.assertEqual(no.next_sibling, " business")
968
969 def test_replace_first_child(self):
970 data = "<a><b></b><c></c></a>"
971 soup = self.soup(data)
972 soup.b.replace_with(soup.c)
973 self.assertEqual("<a><c></c></a>", soup.decode())
974
975 def test_replace_last_child(self):
976 data = "<a><b></b><c></c></a>"
977 soup = self.soup(data)
978 soup.c.replace_with(soup.b)
979 self.assertEqual("<a><b></b></a>", soup.decode())
980
981 def test_nested_tag_replace_with(self):
982 soup = self.soup(
983 """<a>We<b>reserve<c>the</c><d>right</d></b></a><e>to<f>refuse</f><g>service</g></e>""")
984
985 # Replace the entire <b> tag and its contents ("reserve the
986 # right") with the <f> tag ("refuse").
987 remove_tag = soup.b
988 move_tag = soup.f
989 remove_tag.replace_with(move_tag)
990
991 self.assertEqual(
992 soup.decode(), self.document_for(
993 "<a>We<f>refuse</f></a><e>to<g>service</g></e>"))
994
995 # The <b> tag is now an orphan.
996 self.assertEqual(remove_tag.parent, None)
997 self.assertEqual(remove_tag.find(text="right").next_element, None)
998 self.assertEqual(remove_tag.previous_element, None)
999 self.assertEqual(remove_tag.next_sibling, None)
1000 self.assertEqual(remove_tag.previous_sibling, None)
1001
1002 # The <f> tag is now connected to the <a> tag.
1003 self.assertEqual(move_tag.parent, soup.a)
1004 self.assertEqual(move_tag.previous_element, "We")
1005 self.assertEqual(move_tag.next_element.next_element, soup.e)
1006 self.assertEqual(move_tag.next_sibling, None)
1007
1008 # The gap where the <f> tag used to be has been mended, and
1009 # the word "to" is now connected to the <g> tag.
1010 to_text = soup.find(text="to")
1011 g_tag = soup.g
1012 self.assertEqual(to_text.next_element, g_tag)
1013 self.assertEqual(to_text.next_sibling, g_tag)
1014 self.assertEqual(g_tag.previous_element, to_text)
1015 self.assertEqual(g_tag.previous_sibling, to_text)
1016
1017 def test_unwrap(self):
1018 tree = self.soup("""
1019 <p>Unneeded <em>formatting</em> is unneeded</p>
1020 """)
1021 tree.em.unwrap()
1022 self.assertEqual(tree.em, None)
1023 self.assertEqual(tree.p.text, "Unneeded formatting is unneeded")
1024
1025 def test_wrap(self):
1026 soup = self.soup("I wish I was bold.")
1027 value = soup.string.wrap(soup.new_tag("b"))
1028 self.assertEqual(value.decode(), "<b>I wish I was bold.</b>")
1029 self.assertEqual(
1030 soup.decode(), self.document_for("<b>I wish I was bold.</b>"))
1031
1032 def test_wrap_extracts_tag_from_elsewhere(self):
1033 soup = self.soup("<b></b>I wish I was bold.")
1034 soup.b.next_sibling.wrap(soup.b)
1035 self.assertEqual(
1036 soup.decode(), self.document_for("<b>I wish I was bold.</b>"))
1037
1038 def test_wrap_puts_new_contents_at_the_end(self):
1039 soup = self.soup("<b>I like being bold.</b>I wish I was bold.")
1040 soup.b.next_sibling.wrap(soup.b)
1041 self.assertEqual(2, len(soup.b.contents))
1042 self.assertEqual(
1043 soup.decode(), self.document_for(
1044 "<b>I like being bold.I wish I was bold.</b>"))
1045
1046 def test_extract(self):
1047 soup = self.soup(
1048 '<html><body>Some content. <div id="nav">Nav crap</div> More content.</body></html>')
1049
1050 self.assertEqual(len(soup.body.contents), 3)
1051 extracted = soup.find(id="nav").extract()
1052
1053 self.assertEqual(
1054 soup.decode(), "<html><body>Some content. More content.</body></html>")
1055 self.assertEqual(extracted.decode(), '<div id="nav">Nav crap</div>')
1056
1057 # The extracted tag is now an orphan.
1058 self.assertEqual(len(soup.body.contents), 2)
1059 self.assertEqual(extracted.parent, None)
1060 self.assertEqual(extracted.previous_element, None)
1061 self.assertEqual(extracted.next_element.next_element, None)
1062
1063 # The gap where the extracted tag used to be has been mended.
1064 content_1 = soup.find(text="Some content. ")
1065 content_2 = soup.find(text=" More content.")
1066 self.assertEqual(content_1.next_element, content_2)
1067 self.assertEqual(content_1.next_sibling, content_2)
1068 self.assertEqual(content_2.previous_element, content_1)
1069 self.assertEqual(content_2.previous_sibling, content_1)
1070
1071 def test_extract_distinguishes_between_identical_strings(self):
1072 soup = self.soup("<a>foo</a><b>bar</b>")
1073 foo_1 = soup.a.string
1074 bar_1 = soup.b.string
1075 foo_2 = soup.new_string("foo")
1076 bar_2 = soup.new_string("bar")
1077 soup.a.append(foo_2)
1078 soup.b.append(bar_2)
1079
1080 # Now there are two identical strings in the <a> tag, and two
1081 # in the <b> tag. Let's remove the first "foo" and the second
1082 # "bar".
1083 foo_1.extract()
1084 bar_2.extract()
1085 self.assertEqual(foo_2, soup.a.string)
1086 self.assertEqual(bar_2, soup.b.string)
1087
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1088 def test_extract_multiples_of_same_tag(self):
1089 soup = self.soup("""
1090<html>
1091<head>
1092<script>foo</script>
1093</head>
1094<body>
1095 <script>bar</script>
1096 <a></a>
1097</body>
1098<script>baz</script>
1099</html>""")
1100 [soup.script.extract() for i in soup.find_all("script")]
1101 self.assertEqual("<body>\n\n<a></a>\n</body>", str(soup.body))
1102
1103
1104 def test_extract_works_when_element_is_surrounded_by_identical_strings(self):
1105 soup = self.soup(
1106 '<html>\n'
1107 '<body>hi</body>\n'
1108 '</html>')
1109 soup.find('body').extract()
1110 self.assertEqual(None, soup.find('body'))
1111
1112
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1113 def test_clear(self):
1114 """Tag.clear()"""
1115 soup = self.soup("<p><a>String <em>Italicized</em></a> and another</p>")
1116 # clear using extract()
1117 a = soup.a
1118 soup.p.clear()
1119 self.assertEqual(len(soup.p.contents), 0)
1120 self.assertTrue(hasattr(a, "contents"))
1121
1122 # clear using decompose()
1123 em = a.em
1124 a.clear(decompose=True)
1125 self.assertEqual(0, len(em.contents))
1126
1127 def test_string_set(self):
1128 """Tag.string = 'string'"""
1129 soup = self.soup("<a></a> <b><c></c></b>")
1130 soup.a.string = "foo"
1131 self.assertEqual(soup.a.contents, ["foo"])
1132 soup.b.string = "bar"
1133 self.assertEqual(soup.b.contents, ["bar"])
1134
1135 def test_string_set_does_not_affect_original_string(self):
1136 soup = self.soup("<a><b>foo</b><c>bar</c>")
1137 soup.b.string = soup.c.string
1138 self.assertEqual(soup.a.encode(), b"<a><b>bar</b><c>bar</c></a>")
1139
1140 def test_set_string_preserves_class_of_string(self):
1141 soup = self.soup("<a></a>")
1142 cdata = CData("foo")
1143 soup.a.string = cdata
1144 self.assertTrue(isinstance(soup.a.string, CData))
1145
1146class TestElementObjects(SoupTest):
1147 """Test various features of element objects."""
1148
1149 def test_len(self):
1150 """The length of an element is its number of children."""
1151 soup = self.soup("<top>1<b>2</b>3</top>")
1152
1153 # The BeautifulSoup object itself contains one element: the
1154 # <top> tag.
1155 self.assertEqual(len(soup.contents), 1)
1156 self.assertEqual(len(soup), 1)
1157
1158 # The <top> tag contains three elements: the text node "1", the
1159 # <b> tag, and the text node "3".
1160 self.assertEqual(len(soup.top), 3)
1161 self.assertEqual(len(soup.top.contents), 3)
1162
1163 def test_member_access_invokes_find(self):
1164 """Accessing a Python member .foo invokes find('foo')"""
1165 soup = self.soup('<b><i></i></b>')
1166 self.assertEqual(soup.b, soup.find('b'))
1167 self.assertEqual(soup.b.i, soup.find('b').find('i'))
1168 self.assertEqual(soup.a, None)
1169
1170 def test_deprecated_member_access(self):
1171 soup = self.soup('<b><i></i></b>')
1172 with warnings.catch_warnings(record=True) as w:
1173 tag = soup.bTag
1174 self.assertEqual(soup.b, tag)
1175 self.assertEqual(
1176 '.bTag is deprecated, use .find("b") instead.',
1177 str(w[0].message))
1178
1179 def test_has_attr(self):
1180 """has_attr() checks for the presence of an attribute.
1181
1182 Please note note: has_attr() is different from
1183 __in__. has_attr() checks the tag's attributes and __in__
1184 checks the tag's chidlren.
1185 """
1186 soup = self.soup("<foo attr='bar'>")
1187 self.assertTrue(soup.foo.has_attr('attr'))
1188 self.assertFalse(soup.foo.has_attr('attr2'))
1189
1190
1191 def test_attributes_come_out_in_alphabetical_order(self):
1192 markup = '<b a="1" z="5" m="3" f="2" y="4"></b>'
1193 self.assertSoupEquals(markup, '<b a="1" f="2" m="3" y="4" z="5"></b>')
1194
1195 def test_string(self):
1196 # A tag that contains only a text node makes that node
1197 # available as .string.
1198 soup = self.soup("<b>foo</b>")
1199 self.assertEqual(soup.b.string, 'foo')
1200
1201 def test_empty_tag_has_no_string(self):
1202 # A tag with no children has no .stirng.
1203 soup = self.soup("<b></b>")
1204 self.assertEqual(soup.b.string, None)
1205
1206 def test_tag_with_multiple_children_has_no_string(self):
1207 # A tag with no children has no .string.
1208 soup = self.soup("<a>foo<b></b><b></b></b>")
1209 self.assertEqual(soup.b.string, None)
1210
1211 soup = self.soup("<a>foo<b></b>bar</b>")
1212 self.assertEqual(soup.b.string, None)
1213
1214 # Even if all the children are strings, due to trickery,
1215 # it won't work--but this would be a good optimization.
1216 soup = self.soup("<a>foo</b>")
1217 soup.a.insert(1, "bar")
1218 self.assertEqual(soup.a.string, None)
1219
1220 def test_tag_with_recursive_string_has_string(self):
1221 # A tag with a single child which has a .string inherits that
1222 # .string.
1223 soup = self.soup("<a><b>foo</b></a>")
1224 self.assertEqual(soup.a.string, "foo")
1225 self.assertEqual(soup.string, "foo")
1226
1227 def test_lack_of_string(self):
1228 """Only a tag containing a single text node has a .string."""
1229 soup = self.soup("<b>f<i>e</i>o</b>")
1230 self.assertFalse(soup.b.string)
1231
1232 soup = self.soup("<b></b>")
1233 self.assertFalse(soup.b.string)
1234
1235 def test_all_text(self):
1236 """Tag.text and Tag.get_text(sep=u"") -> all child text, concatenated"""
1237 soup = self.soup("<a>a<b>r</b> <r> t </r></a>")
1238 self.assertEqual(soup.a.text, "ar t ")
1239 self.assertEqual(soup.a.get_text(strip=True), "art")
1240 self.assertEqual(soup.a.get_text(","), "a,r, , t ")
1241 self.assertEqual(soup.a.get_text(",", strip=True), "a,r,t")
1242
1243 def test_get_text_ignores_comments(self):
1244 soup = self.soup("foo<!--IGNORE-->bar")
1245 self.assertEqual(soup.get_text(), "foobar")
1246
1247 self.assertEqual(
1248 soup.get_text(types=(NavigableString, Comment)), "fooIGNOREbar")
1249 self.assertEqual(
1250 soup.get_text(types=None), "fooIGNOREbar")
1251
1252 def test_all_strings_ignores_comments(self):
1253 soup = self.soup("foo<!--IGNORE-->bar")
1254 self.assertEqual(['foo', 'bar'], list(soup.strings))
1255
1256class TestCDAtaListAttributes(SoupTest):
1257
1258 """Testing cdata-list attributes like 'class'.
1259 """
1260 def test_single_value_becomes_list(self):
1261 soup = self.soup("<a class='foo'>")
1262 self.assertEqual(["foo"],soup.a['class'])
1263
1264 def test_multiple_values_becomes_list(self):
1265 soup = self.soup("<a class='foo bar'>")
1266 self.assertEqual(["foo", "bar"], soup.a['class'])
1267
1268 def test_multiple_values_separated_by_weird_whitespace(self):
1269 soup = self.soup("<a class='foo\tbar\nbaz'>")
1270 self.assertEqual(["foo", "bar", "baz"],soup.a['class'])
1271
1272 def test_attributes_joined_into_string_on_output(self):
1273 soup = self.soup("<a class='foo\tbar'>")
1274 self.assertEqual(b'<a class="foo bar"></a>', soup.a.encode())
1275
1276 def test_accept_charset(self):
1277 soup = self.soup('<form accept-charset="ISO-8859-1 UTF-8">')
1278 self.assertEqual(['ISO-8859-1', 'UTF-8'], soup.form['accept-charset'])
1279
1280 def test_cdata_attribute_applying_only_to_one_tag(self):
1281 data = '<a accept-charset="ISO-8859-1 UTF-8"></a>'
1282 soup = self.soup(data)
1283 # We saw in another test that accept-charset is a cdata-list
1284 # attribute for the <form> tag. But it's not a cdata-list
1285 # attribute for any other tag.
1286 self.assertEqual('ISO-8859-1 UTF-8', soup.a['accept-charset'])
1287
1288 def test_string_has_immutable_name_property(self):
1289 string = self.soup("s").string
1290 self.assertEqual(None, string.name)
1291 def t():
1292 string.name = 'foo'
1293 self.assertRaises(AttributeError, t)
1294
1295class TestPersistence(SoupTest):
1296 "Testing features like pickle and deepcopy."
1297
1298 def setUp(self):
1299 super(TestPersistence, self).setUp()
1300 self.page = """<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"
1301"http://www.w3.org/TR/REC-html40/transitional.dtd">
1302<html>
1303<head>
1304<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
1305<title>Beautiful Soup: We called him Tortoise because he taught us.</title>
1306<link rev="made" href="mailto:leonardr@segfault.org">
1307<meta name="Description" content="Beautiful Soup: an HTML parser optimized for screen-scraping.">
1308<meta name="generator" content="Markov Approximation 1.4 (module: leonardr)">
1309<meta name="author" content="Leonard Richardson">
1310</head>
1311<body>
1312<a href="foo">foo</a>
1313<a href="foo"><b>bar</b></a>
1314</body>
1315</html>"""
1316 self.tree = self.soup(self.page)
1317
1318 def test_pickle_and_unpickle_identity(self):
1319 # Pickling a tree, then unpickling it, yields a tree identical
1320 # to the original.
1321 dumped = pickle.dumps(self.tree, 2)
1322 loaded = pickle.loads(dumped)
1323 self.assertEqual(loaded.__class__, BeautifulSoup)
1324 self.assertEqual(loaded.decode(), self.tree.decode())
1325
1326 def test_deepcopy_identity(self):
1327 # Making a deepcopy of a tree yields an identical tree.
1328 copied = copy.deepcopy(self.tree)
1329 self.assertEqual(copied.decode(), self.tree.decode())
1330
1331 def test_unicode_pickle(self):
1332 # A tree containing Unicode characters can be pickled.
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1333 html = "<b>\N{SNOWMAN}</b>"
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1334 soup = self.soup(html)
1335 dumped = pickle.dumps(soup, pickle.HIGHEST_PROTOCOL)
1336 loaded = pickle.loads(dumped)
1337 self.assertEqual(loaded.decode(), soup.decode())
1338
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1339 def test_copy_navigablestring_is_not_attached_to_tree(self):
1340 html = "<b>Foo<a></a></b><b>Bar</b>"
1341 soup = self.soup(html)
1342 s1 = soup.find(string="Foo")
1343 s2 = copy.copy(s1)
1344 self.assertEqual(s1, s2)
1345 self.assertEqual(None, s2.parent)
1346 self.assertEqual(None, s2.next_element)
1347 self.assertNotEqual(None, s1.next_sibling)
1348 self.assertEqual(None, s2.next_sibling)
1349 self.assertEqual(None, s2.previous_element)
1350
1351 def test_copy_navigablestring_subclass_has_same_type(self):
1352 html = "<b><!--Foo--></b>"
1353 soup = self.soup(html)
1354 s1 = soup.string
1355 s2 = copy.copy(s1)
1356 self.assertEqual(s1, s2)
1357 self.assertTrue(isinstance(s2, Comment))
1358
1359 def test_copy_entire_soup(self):
1360 html = "<div><b>Foo<a></a></b><b>Bar</b></div>end"
1361 soup = self.soup(html)
1362 soup_copy = copy.copy(soup)
1363 self.assertEqual(soup, soup_copy)
1364
1365 def test_copy_tag_copies_contents(self):
1366 html = "<div><b>Foo<a></a></b><b>Bar</b></div>end"
1367 soup = self.soup(html)
1368 div = soup.div
1369 div_copy = copy.copy(div)
1370
1371 # The two tags look the same, and evaluate to equal.
1372 self.assertEqual(str(div), str(div_copy))
1373 self.assertEqual(div, div_copy)
1374
1375 # But they're not the same object.
1376 self.assertFalse(div is div_copy)
1377
1378 # And they don't have the same relation to the parse tree. The
1379 # copy is not associated with a parse tree at all.
1380 self.assertEqual(None, div_copy.parent)
1381 self.assertEqual(None, div_copy.previous_element)
1382 self.assertEqual(None, div_copy.find(string='Bar').next_element)
1383 self.assertNotEqual(None, div.find(string='Bar').next_element)
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1384
1385class TestSubstitutions(SoupTest):
1386
1387 def test_default_formatter_is_minimal(self):
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1388 markup = "<b>&lt;&lt;Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!&gt;&gt;</b>"
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1389 soup = self.soup(markup)
1390 decoded = soup.decode(formatter="minimal")
1391 # The < is converted back into &lt; but the e-with-acute is left alone.
1392 self.assertEqual(
1393 decoded,
1394 self.document_for(
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1395 "<b>&lt;&lt;Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!&gt;&gt;</b>"))
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1396
1397 def test_formatter_html(self):
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1398 markup = "<b>&lt;&lt;Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!&gt;&gt;</b>"
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1399 soup = self.soup(markup)
1400 decoded = soup.decode(formatter="html")
1401 self.assertEqual(
1402 decoded,
1403 self.document_for("<b>&lt;&lt;Sacr&eacute; bleu!&gt;&gt;</b>"))
1404
1405 def test_formatter_minimal(self):
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1406 markup = "<b>&lt;&lt;Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!&gt;&gt;</b>"
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1407 soup = self.soup(markup)
1408 decoded = soup.decode(formatter="minimal")
1409 # The < is converted back into &lt; but the e-with-acute is left alone.
1410 self.assertEqual(
1411 decoded,
1412 self.document_for(
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1413 "<b>&lt;&lt;Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!&gt;&gt;</b>"))
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1414
1415 def test_formatter_null(self):
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1416 markup = "<b>&lt;&lt;Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!&gt;&gt;</b>"
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1417 soup = self.soup(markup)
1418 decoded = soup.decode(formatter=None)
1419 # Neither the angle brackets nor the e-with-acute are converted.
1420 # This is not valid HTML, but it's what the user wanted.
1421 self.assertEqual(decoded,
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1422 self.document_for("<b><<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>></b>"))
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1423
1424 def test_formatter_custom(self):
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1425 markup = "<b>&lt;foo&gt;</b><b>bar</b>"
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1426 soup = self.soup(markup)
1427 decoded = soup.decode(formatter = lambda x: x.upper())
1428 # Instead of normal entity conversion code, the custom
1429 # callable is called on every string.
1430 self.assertEqual(
1431 decoded,
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1432 self.document_for("<b><FOO></b><b>BAR</b>"))
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1433
1434 def test_formatter_is_run_on_attribute_values(self):
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1435 markup = '<a href="http://a.com?a=b&c=é">e</a>'
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1436 soup = self.soup(markup)
1437 a = soup.a
1438
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1439 expect_minimal = '<a href="http://a.com?a=b&amp;c=é">e</a>'
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1440
1441 self.assertEqual(expect_minimal, a.decode())
1442 self.assertEqual(expect_minimal, a.decode(formatter="minimal"))
1443
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1444 expect_html = '<a href="http://a.com?a=b&amp;c=&eacute;">e</a>'
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1445 self.assertEqual(expect_html, a.decode(formatter="html"))
1446
1447 self.assertEqual(markup, a.decode(formatter=None))
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1448 expect_upper = '<a href="HTTP://A.COM?A=B&C=É">E</a>'
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1449 self.assertEqual(expect_upper, a.decode(formatter=lambda x: x.upper()))
1450
1451 def test_formatter_skips_script_tag_for_html_documents(self):
1452 doc = """
1453 <script type="text/javascript">
1454 console.log("< < hey > > ");
1455 </script>
1456"""
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1457 encoded = BeautifulSoup(doc, 'html.parser').encode()
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1458 self.assertTrue(b"< < hey > >" in encoded)
1459
1460 def test_formatter_skips_style_tag_for_html_documents(self):
1461 doc = """
1462 <style type="text/css">
1463 console.log("< < hey > > ");
1464 </style>
1465"""
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1466 encoded = BeautifulSoup(doc, 'html.parser').encode()
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1467 self.assertTrue(b"< < hey > >" in encoded)
1468
1469 def test_prettify_leaves_preformatted_text_alone(self):
1470 soup = self.soup("<div> foo <pre> \tbar\n \n </pre> baz ")
1471 # Everything outside the <pre> tag is reformatted, but everything
1472 # inside is left alone.
1473 self.assertEqual(
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1474 '<div>\n foo\n <pre> \tbar\n \n </pre>\n baz\n</div>',
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1475 soup.div.prettify())
1476
1477 def test_prettify_accepts_formatter(self):
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1478 soup = BeautifulSoup("<html><body>foo</body></html>", 'html.parser')
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1479 pretty = soup.prettify(formatter = lambda x: x.upper())
1480 self.assertTrue("FOO" in pretty)
1481
1482 def test_prettify_outputs_unicode_by_default(self):
1483 soup = self.soup("<a></a>")
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1484 self.assertEqual(str, type(soup.prettify()))
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1485
1486 def test_prettify_can_encode_data(self):
1487 soup = self.soup("<a></a>")
1488 self.assertEqual(bytes, type(soup.prettify("utf-8")))
1489
1490 def test_html_entity_substitution_off_by_default(self):
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1491 markup = "<b>Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!</b>"
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1492 soup = self.soup(markup)
1493 encoded = soup.b.encode("utf-8")
1494 self.assertEqual(encoded, markup.encode('utf-8'))
1495
1496 def test_encoding_substitution(self):
1497 # Here's the <meta> tag saying that a document is
1498 # encoded in Shift-JIS.
1499 meta_tag = ('<meta content="text/html; charset=x-sjis" '
1500 'http-equiv="Content-type"/>')
1501 soup = self.soup(meta_tag)
1502
1503 # Parse the document, and the charset apprears unchanged.
1504 self.assertEqual(soup.meta['content'], 'text/html; charset=x-sjis')
1505
1506 # Encode the document into some encoding, and the encoding is
1507 # substituted into the meta tag.
1508 utf_8 = soup.encode("utf-8")
1509 self.assertTrue(b"charset=utf-8" in utf_8)
1510
1511 euc_jp = soup.encode("euc_jp")
1512 self.assertTrue(b"charset=euc_jp" in euc_jp)
1513
1514 shift_jis = soup.encode("shift-jis")
1515 self.assertTrue(b"charset=shift-jis" in shift_jis)
1516
1517 utf_16_u = soup.encode("utf-16").decode("utf-16")
1518 self.assertTrue("charset=utf-16" in utf_16_u)
1519
1520 def test_encoding_substitution_doesnt_happen_if_tag_is_strained(self):
1521 markup = ('<head><meta content="text/html; charset=x-sjis" '
1522 'http-equiv="Content-type"/></head><pre>foo</pre>')
1523
1524 # Beautiful Soup used to try to rewrite the meta tag even if the
1525 # meta tag got filtered out by the strainer. This test makes
1526 # sure that doesn't happen.
1527 strainer = SoupStrainer('pre')
1528 soup = self.soup(markup, parse_only=strainer)
1529 self.assertEqual(soup.contents[0].name, 'pre')
1530
1531class TestEncoding(SoupTest):
1532 """Test the ability to encode objects into strings."""
1533
1534 def test_unicode_string_can_be_encoded(self):
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1535 html = "<b>\N{SNOWMAN}</b>"
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1536 soup = self.soup(html)
1537 self.assertEqual(soup.b.string.encode("utf-8"),
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1538 "\N{SNOWMAN}".encode("utf-8"))
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1539
1540 def test_tag_containing_unicode_string_can_be_encoded(self):
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1541 html = "<b>\N{SNOWMAN}</b>"
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1542 soup = self.soup(html)
1543 self.assertEqual(
1544 soup.b.encode("utf-8"), html.encode("utf-8"))
1545
1546 def test_encoding_substitutes_unrecognized_characters_by_default(self):
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1547 html = "<b>\N{SNOWMAN}</b>"
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1548 soup = self.soup(html)
1549 self.assertEqual(soup.b.encode("ascii"), b"<b>&#9731;</b>")
1550
1551 def test_encoding_can_be_made_strict(self):
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1552 html = "<b>\N{SNOWMAN}</b>"
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1553 soup = self.soup(html)
1554 self.assertRaises(
1555 UnicodeEncodeError, soup.encode, "ascii", errors="strict")
1556
1557 def test_decode_contents(self):
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1558 html = "<b>\N{SNOWMAN}</b>"
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1559 soup = self.soup(html)
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1560 self.assertEqual("\N{SNOWMAN}", soup.b.decode_contents())
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1561
1562 def test_encode_contents(self):
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1563 html = "<b>\N{SNOWMAN}</b>"
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1564 soup = self.soup(html)
1565 self.assertEqual(
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1566 "\N{SNOWMAN}".encode("utf8"), soup.b.encode_contents(
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1567 encoding="utf8"))
1568
1569 def test_deprecated_renderContents(self):
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1570 html = "<b>\N{SNOWMAN}</b>"
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1571 soup = self.soup(html)
1572 self.assertEqual(
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1573 "\N{SNOWMAN}".encode("utf8"), soup.b.renderContents())
1574
1575 def test_repr(self):
1576 html = "<b>\N{SNOWMAN}</b>"
1577 soup = self.soup(html)
1578 if PY3K:
1579 self.assertEqual(html, repr(soup))
1580 else:
1581 self.assertEqual(b'<b>\\u2603</b>', repr(soup))
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1582
1583class TestNavigableStringSubclasses(SoupTest):
1584
1585 def test_cdata(self):
1586 # None of the current builders turn CDATA sections into CData
1587 # objects, but you can create them manually.
1588 soup = self.soup("")
1589 cdata = CData("foo")
1590 soup.insert(1, cdata)
1591 self.assertEqual(str(soup), "<![CDATA[foo]]>")
1592 self.assertEqual(soup.find(text="foo"), "foo")
1593 self.assertEqual(soup.contents[0], "foo")
1594
1595 def test_cdata_is_never_formatted(self):
1596 """Text inside a CData object is passed into the formatter.
1597
1598 But the return value is ignored.
1599 """
1600
1601 self.count = 0
1602 def increment(*args):
1603 self.count += 1
1604 return "BITTER FAILURE"
1605
1606 soup = self.soup("")
1607 cdata = CData("<><><>")
1608 soup.insert(1, cdata)
1609 self.assertEqual(
1610 b"<![CDATA[<><><>]]>", soup.encode(formatter=increment))
1611 self.assertEqual(1, self.count)
1612
1613 def test_doctype_ends_in_newline(self):
1614 # Unlike other NavigableString subclasses, a DOCTYPE always ends
1615 # in a newline.
1616 doctype = Doctype("foo")
1617 soup = self.soup("")
1618 soup.insert(1, doctype)
1619 self.assertEqual(soup.encode(), b"<!DOCTYPE foo>\n")
1620
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1621 def test_declaration(self):
1622 d = Declaration("foo")
1623 self.assertEqual("<?foo?>", d.output_ready())
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1624
1625class TestSoupSelector(TreeTest):
1626
1627 HTML = """
1628<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
1629"http://www.w3.org/TR/html4/strict.dtd">
1630<html>
1631<head>
1632<title>The title</title>
1633<link rel="stylesheet" href="blah.css" type="text/css" id="l1">
1634</head>
1635<body>
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1636<custom-dashed-tag class="dashed" id="dash1">Hello there.</custom-dashed-tag>
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1637<div id="main" class="fancy">
1638<div id="inner">
1639<h1 id="header1">An H1</h1>
1640<p>Some text</p>
1641<p class="onep" id="p1">Some more text</p>
1642<h2 id="header2">An H2</h2>
1643<p class="class1 class2 class3" id="pmulti">Another</p>
1644<a href="http://bob.example.org/" rel="friend met" id="bob">Bob</a>
1645<h2 id="header3">Another H2</h2>
1646<a id="me" href="http://simonwillison.net/" rel="me">me</a>
1647<span class="s1">
1648<a href="#" id="s1a1">span1a1</a>
1649<a href="#" id="s1a2">span1a2 <span id="s1a2s1">test</span></a>
1650<span class="span2">
1651<a href="#" id="s2a1">span2a1</a>
1652</span>
1653<span class="span3"></span>
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1654<custom-dashed-tag class="dashed" id="dash2"/>
1655<div data-tag="dashedvalue" id="data1"/>
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1656</span>
1657</div>
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1658<x id="xid">
1659<z id="zida"/>
1660<z id="zidab"/>
1661<z id="zidac"/>
1662</x>
1663<y id="yid">
1664<z id="zidb"/>
1665</y>
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1666<p lang="en" id="lang-en">English</p>
1667<p lang="en-gb" id="lang-en-gb">English UK</p>
1668<p lang="en-us" id="lang-en-us">English US</p>
1669<p lang="fr" id="lang-fr">French</p>
1670</div>
1671
1672<div id="footer">
1673</div>
1674"""
1675
1676 def setUp(self):
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1677 self.soup = BeautifulSoup(self.HTML, 'html.parser')
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1678
1679 def assertSelects(self, selector, expected_ids):
1680 el_ids = [el['id'] for el in self.soup.select(selector)]
1681 el_ids.sort()
1682 expected_ids.sort()
1683 self.assertEqual(expected_ids, el_ids,
1684 "Selector %s, expected [%s], got [%s]" % (
1685 selector, ', '.join(expected_ids), ', '.join(el_ids)
1686 )
1687 )
1688
1689 assertSelect = assertSelects
1690
1691 def assertSelectMultiple(self, *tests):
1692 for selector, expected_ids in tests:
1693 self.assertSelect(selector, expected_ids)
1694
1695 def test_one_tag_one(self):
1696 els = self.soup.select('title')
1697 self.assertEqual(len(els), 1)
1698 self.assertEqual(els[0].name, 'title')
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1699 self.assertEqual(els[0].contents, ['The title'])
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1700
1701 def test_one_tag_many(self):
1702 els = self.soup.select('div')
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1703 self.assertEqual(len(els), 4)
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1704 for div in els:
1705 self.assertEqual(div.name, 'div')
1706
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1707 el = self.soup.select_one('div')
1708 self.assertEqual('main', el['id'])
1709
1710 def test_select_one_returns_none_if_no_match(self):
1711 match = self.soup.select_one('nonexistenttag')
1712 self.assertEqual(None, match)
1713
1714
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1715 def test_tag_in_tag_one(self):
1716 els = self.soup.select('div div')
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1717 self.assertSelects('div div', ['inner', 'data1'])
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1718
1719 def test_tag_in_tag_many(self):
1720 for selector in ('html div', 'html body div', 'body div'):
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1721 self.assertSelects(selector, ['data1', 'main', 'inner', 'footer'])
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1722
1723 def test_tag_no_match(self):
1724 self.assertEqual(len(self.soup.select('del')), 0)
1725
1726 def test_invalid_tag(self):
1727 self.assertRaises(ValueError, self.soup.select, 'tag%t')
1728
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1729 def test_select_dashed_tag_ids(self):
1730 self.assertSelects('custom-dashed-tag', ['dash1', 'dash2'])
1731
1732 def test_select_dashed_by_id(self):
1733 dashed = self.soup.select('custom-dashed-tag[id=\"dash2\"]')
1734 self.assertEqual(dashed[0].name, 'custom-dashed-tag')
1735 self.assertEqual(dashed[0]['id'], 'dash2')
1736
1737 def test_dashed_tag_text(self):
1738 self.assertEqual(self.soup.select('body > custom-dashed-tag')[0].text, 'Hello there.')
1739
1740 def test_select_dashed_matches_find_all(self):
1741 self.assertEqual(self.soup.select('custom-dashed-tag'), self.soup.find_all('custom-dashed-tag'))
1742
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1743 def test_header_tags(self):
1744 self.assertSelectMultiple(
1745 ('h1', ['header1']),
1746 ('h2', ['header2', 'header3']),
1747 )
1748
1749 def test_class_one(self):
1750 for selector in ('.onep', 'p.onep', 'html p.onep'):
1751 els = self.soup.select(selector)
1752 self.assertEqual(len(els), 1)
1753 self.assertEqual(els[0].name, 'p')
1754 self.assertEqual(els[0]['class'], ['onep'])
1755
1756 def test_class_mismatched_tag(self):
1757 els = self.soup.select('div.onep')
1758 self.assertEqual(len(els), 0)
1759
1760 def test_one_id(self):
1761 for selector in ('div#inner', '#inner', 'div div#inner'):
1762 self.assertSelects(selector, ['inner'])
1763
1764 def test_bad_id(self):
1765 els = self.soup.select('#doesnotexist')
1766 self.assertEqual(len(els), 0)
1767
1768 def test_items_in_id(self):
1769 els = self.soup.select('div#inner p')
1770 self.assertEqual(len(els), 3)
1771 for el in els:
1772 self.assertEqual(el.name, 'p')
1773 self.assertEqual(els[1]['class'], ['onep'])
1774 self.assertFalse(els[0].has_attr('class'))
1775
1776 def test_a_bunch_of_emptys(self):
1777 for selector in ('div#main del', 'div#main div.oops', 'div div#main'):
1778 self.assertEqual(len(self.soup.select(selector)), 0)
1779
1780 def test_multi_class_support(self):
1781 for selector in ('.class1', 'p.class1', '.class2', 'p.class2',
1782 '.class3', 'p.class3', 'html p.class2', 'div#inner .class2'):
1783 self.assertSelects(selector, ['pmulti'])
1784
1785 def test_multi_class_selection(self):
1786 for selector in ('.class1.class3', '.class3.class2',
1787 '.class1.class2.class3'):
1788 self.assertSelects(selector, ['pmulti'])
1789
1790 def test_child_selector(self):
1791 self.assertSelects('.s1 > a', ['s1a1', 's1a2'])
1792 self.assertSelects('.s1 > a span', ['s1a2s1'])
1793
1794 def test_child_selector_id(self):
1795 self.assertSelects('.s1 > a#s1a2 span', ['s1a2s1'])
1796
1797 def test_attribute_equals(self):
1798 self.assertSelectMultiple(
1799 ('p[class="onep"]', ['p1']),
1800 ('p[id="p1"]', ['p1']),
1801 ('[class="onep"]', ['p1']),
1802 ('[id="p1"]', ['p1']),
1803 ('link[rel="stylesheet"]', ['l1']),
1804 ('link[type="text/css"]', ['l1']),
1805 ('link[href="blah.css"]', ['l1']),
1806 ('link[href="no-blah.css"]', []),
1807 ('[rel="stylesheet"]', ['l1']),
1808 ('[type="text/css"]', ['l1']),
1809 ('[href="blah.css"]', ['l1']),
1810 ('[href="no-blah.css"]', []),
1811 ('p[href="no-blah.css"]', []),
1812 ('[href="no-blah.css"]', []),
1813 )
1814
1815 def test_attribute_tilde(self):
1816 self.assertSelectMultiple(
1817 ('p[class~="class1"]', ['pmulti']),
1818 ('p[class~="class2"]', ['pmulti']),
1819 ('p[class~="class3"]', ['pmulti']),
1820 ('[class~="class1"]', ['pmulti']),
1821 ('[class~="class2"]', ['pmulti']),
1822 ('[class~="class3"]', ['pmulti']),
1823 ('a[rel~="friend"]', ['bob']),
1824 ('a[rel~="met"]', ['bob']),
1825 ('[rel~="friend"]', ['bob']),
1826 ('[rel~="met"]', ['bob']),
1827 )
1828
1829 def test_attribute_startswith(self):
1830 self.assertSelectMultiple(
1831 ('[rel^="style"]', ['l1']),
1832 ('link[rel^="style"]', ['l1']),
1833 ('notlink[rel^="notstyle"]', []),
1834 ('[rel^="notstyle"]', []),
1835 ('link[rel^="notstyle"]', []),
1836 ('link[href^="bla"]', ['l1']),
1837 ('a[href^="http://"]', ['bob', 'me']),
1838 ('[href^="http://"]', ['bob', 'me']),
1839 ('[id^="p"]', ['pmulti', 'p1']),
1840 ('[id^="m"]', ['me', 'main']),
1841 ('div[id^="m"]', ['main']),
1842 ('a[id^="m"]', ['me']),
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1843 ('div[data-tag^="dashed"]', ['data1'])
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1844 )
1845
1846 def test_attribute_endswith(self):
1847 self.assertSelectMultiple(
1848 ('[href$=".css"]', ['l1']),
1849 ('link[href$=".css"]', ['l1']),
1850 ('link[id$="1"]', ['l1']),
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1851 ('[id$="1"]', ['data1', 'l1', 'p1', 'header1', 's1a1', 's2a1', 's1a2s1', 'dash1']),
1852 ('div[id$="1"]', ['data1']),
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1853 ('[id$="noending"]', []),
1854 )
1855
1856 def test_attribute_contains(self):
1857 self.assertSelectMultiple(
1858 # From test_attribute_startswith
1859 ('[rel*="style"]', ['l1']),
1860 ('link[rel*="style"]', ['l1']),
1861 ('notlink[rel*="notstyle"]', []),
1862 ('[rel*="notstyle"]', []),
1863 ('link[rel*="notstyle"]', []),
1864 ('link[href*="bla"]', ['l1']),
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1865 ('[href*="http://"]', ['bob', 'me']),
1866 ('[id*="p"]', ['pmulti', 'p1']),
1867 ('div[id*="m"]', ['main']),
1868 ('a[id*="m"]', ['me']),
1869 # From test_attribute_endswith
1870 ('[href*=".css"]', ['l1']),
1871 ('link[href*=".css"]', ['l1']),
1872 ('link[id*="1"]', ['l1']),
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1873 ('[id*="1"]', ['data1', 'l1', 'p1', 'header1', 's1a1', 's1a2', 's2a1', 's1a2s1', 'dash1']),
1874 ('div[id*="1"]', ['data1']),
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1875 ('[id*="noending"]', []),
1876 # New for this test
1877 ('[href*="."]', ['bob', 'me', 'l1']),
1878 ('a[href*="."]', ['bob', 'me']),
1879 ('link[href*="."]', ['l1']),
1880 ('div[id*="n"]', ['main', 'inner']),
1881 ('div[id*="nn"]', ['inner']),
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1882 ('div[data-tag*="edval"]', ['data1'])
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1883 )
1884
1885 def test_attribute_exact_or_hypen(self):
1886 self.assertSelectMultiple(
1887 ('p[lang|="en"]', ['lang-en', 'lang-en-gb', 'lang-en-us']),
1888 ('[lang|="en"]', ['lang-en', 'lang-en-gb', 'lang-en-us']),
1889 ('p[lang|="fr"]', ['lang-fr']),
1890 ('p[lang|="gb"]', []),
1891 )
1892
1893 def test_attribute_exists(self):
1894 self.assertSelectMultiple(
1895 ('[rel]', ['l1', 'bob', 'me']),
1896 ('link[rel]', ['l1']),
1897 ('a[rel]', ['bob', 'me']),
1898 ('[lang]', ['lang-en', 'lang-en-gb', 'lang-en-us', 'lang-fr']),
1899 ('p[class]', ['p1', 'pmulti']),
1900 ('[blah]', []),
1901 ('p[blah]', []),
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1902 ('div[data-tag]', ['data1'])
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1903 )
1904
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1905 def test_unsupported_pseudoclass(self):
1906 self.assertRaises(
1907 NotImplementedError, self.soup.select, "a:no-such-pseudoclass")
1908
1909 self.assertRaises(
1910 NotImplementedError, self.soup.select, "a:nth-of-type(a)")
1911
1912
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1913 def test_nth_of_type(self):
1914 # Try to select first paragraph
1915 els = self.soup.select('div#inner p:nth-of-type(1)')
1916 self.assertEqual(len(els), 1)
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1917 self.assertEqual(els[0].string, 'Some text')
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1918
1919 # Try to select third paragraph
1920 els = self.soup.select('div#inner p:nth-of-type(3)')
1921 self.assertEqual(len(els), 1)
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1922 self.assertEqual(els[0].string, 'Another')
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1923
1924 # Try to select (non-existent!) fourth paragraph
1925 els = self.soup.select('div#inner p:nth-of-type(4)')
1926 self.assertEqual(len(els), 0)
1927
1928 # Pass in an invalid value.
1929 self.assertRaises(
1930 ValueError, self.soup.select, 'div p:nth-of-type(0)')
1931
1932 def test_nth_of_type_direct_descendant(self):
1933 els = self.soup.select('div#inner > p:nth-of-type(1)')
1934 self.assertEqual(len(els), 1)
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1935 self.assertEqual(els[0].string, 'Some text')
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1936
1937 def test_id_child_selector_nth_of_type(self):
1938 self.assertSelects('#inner > p:nth-of-type(2)', ['p1'])
1939
1940 def test_select_on_element(self):
1941 # Other tests operate on the tree; this operates on an element
1942 # within the tree.
1943 inner = self.soup.find("div", id="main")
1944 selected = inner.select("div")
1945 # The <div id="inner"> tag was selected. The <div id="footer">
1946 # tag was not.
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1947 self.assertSelectsIDs(selected, ['inner', 'data1'])
Patrick Williamsc124f4f2015-09-15 14:41:29 -0500[diff] [blame]1948
1949 def test_overspecified_child_id(self):
1950 self.assertSelects(".fancy #inner", ['inner'])
1951 self.assertSelects(".normal #inner", [])
1952
1953 def test_adjacent_sibling_selector(self):
1954 self.assertSelects('#p1 + h2', ['header2'])
1955 self.assertSelects('#p1 + h2 + p', ['pmulti'])
1956 self.assertSelects('#p1 + #header2 + .class1', ['pmulti'])
1957 self.assertEqual([], self.soup.select('#p1 + p'))
1958
1959 def test_general_sibling_selector(self):
1960 self.assertSelects('#p1 ~ h2', ['header2', 'header3'])
1961 self.assertSelects('#p1 ~ #header2', ['header2'])
1962 self.assertSelects('#p1 ~ h2 + a', ['me'])
1963 self.assertSelects('#p1 ~ h2 + [rel="me"]', ['me'])
1964 self.assertEqual([], self.soup.select('#inner ~ h2'))
1965
1966 def test_dangling_combinator(self):
1967 self.assertRaises(ValueError, self.soup.select, 'h1 >')
1968
1969 def test_sibling_combinator_wont_select_same_tag_twice(self):
1970 self.assertSelects('p[lang] ~ p', ['lang-en-gb', 'lang-en-us', 'lang-fr'])
Patrick Williamsc0f7c042017-02-23 20:41:17 -0600[diff] [blame]1971
1972 # Test the selector grouping operator (the comma)
1973 def test_multiple_select(self):
1974 self.assertSelects('x, y', ['xid', 'yid'])
1975
1976 def test_multiple_select_with_no_space(self):
1977 self.assertSelects('x,y', ['xid', 'yid'])
1978
1979 def test_multiple_select_with_more_space(self):
1980 self.assertSelects('x, y', ['xid', 'yid'])
1981
1982 def test_multiple_select_duplicated(self):
1983 self.assertSelects('x, x', ['xid'])
1984
1985 def test_multiple_select_sibling(self):
1986 self.assertSelects('x, y ~ p[lang=fr]', ['xid', 'lang-fr'])
1987
1988 def test_multiple_select_tag_and_direct_descendant(self):
1989 self.assertSelects('x, y > z', ['xid', 'zidb'])
1990
1991 def test_multiple_select_direct_descendant_and_tags(self):
1992 self.assertSelects('div > x, y, z', ['xid', 'yid', 'zida', 'zidb', 'zidab', 'zidac'])
1993
1994 def test_multiple_select_indirect_descendant(self):
1995 self.assertSelects('div x,y, z', ['xid', 'yid', 'zida', 'zidb', 'zidab', 'zidac'])
1996
1997 def test_invalid_multiple_select(self):
1998 self.assertRaises(ValueError, self.soup.select, ',x, y')
1999 self.assertRaises(ValueError, self.soup.select, 'x,,y')
2000
2001 def test_multiple_select_attrs(self):
2002 self.assertSelects('p[lang=en], p[lang=en-gb]', ['lang-en', 'lang-en-gb'])
2003
2004 def test_multiple_select_ids(self):
2005 self.assertSelects('x, y > z[id=zida], z[id=zidab], z[id=zidb]', ['xid', 'zidb', 'zidab'])
2006
2007 def test_multiple_select_nested(self):
2008 self.assertSelects('body > div > x, y > z', ['xid', 'zidb'])
2009
2010
2011