diff options
| author | Leonard Richardson <leonardr@segfault.org> | 2016-07-17 11:50:48 -0400 |
|---|---|---|
| committer | Leonard Richardson <leonardr@segfault.org> | 2016-07-17 11:50:48 -0400 |
| commit | 0b2445d80c360980c948821c287da56f63629257 (patch) | |
| tree | 8d8b04f40f03fdd26b3bbcdc58ce191756270336 /bs4/tests/test_html5lib.py | |
| parent | dfe49a14ef02589d96953434fb9aeab7da5ccb30 (diff) | |
Fixed a bug in the html5lib treebuilder that deranged the tree
when a whitespace element was reparented into a tag that contained
an identical whitespace element. [bug=1505351]
Diffstat (limited to 'bs4/tests/test_html5lib.py')
| -rw-r--r-- | bs4/tests/test_html5lib.py | 11 |
1 files changed, 11 insertions, 0 deletions
diff --git a/bs4/tests/test_html5lib.py b/bs4/tests/test_html5lib.py index 65536c2..8e3cba6 100644 --- a/bs4/tests/test_html5lib.py +++ b/bs4/tests/test_html5lib.py @@ -84,6 +84,17 @@ class HTML5LibBuilderSmokeTest(SoupTest, HTML5TreeBuilderSmokeTest): self.assertEqual(u"<body><p><em>foo</em></p><em>\n</em><p><em>bar<a></a></em></p>\n</body>", soup.body.decode()) self.assertEqual(2, len(soup.find_all('p'))) + def test_reparented_markup_containing_identical_whitespace_nodes(self): + """Verify that we keep the two whitespace nodes in this + document distinct when reparenting the adjacent <tbody> tags. + """ + markup = '<table> <tbody><tbody><ims></tbody> </table>' + soup = self.soup(markup) + space1, space2 = soup.find_all(string=' ') + tbody1, tbody2 = soup.find_all('tbody') + assert space1.next_element is tbody1 + assert tbody2.next_element is space2 + def test_processing_instruction(self): """Processing instructions become comments.""" markup = b"""<?PITarget PIContent?>""" |