summaryrefslogtreecommitdiff
path: root/bs4/tests
AgeCommit message (Collapse)Author
2012-02-20Changd the class structure so that the default parser test class uses ↵Leonard Richardson
html.parser.
2012-02-20Added code from 2.7's standard library so that the tests will run on Python 2.6.Leonard Richardson
2012-02-17Doctests don't work because the package name is wrong and README.txt isn't ↵Leonard Richardson
in any package, so comment them out for now.
2012-02-16By default, turn unrecognized characters into numeric XML entity refs.Leonard Richardson
2012-02-16Issue a warning if characters were replaced with REPLACEMENT CHARACTER ↵Leonard Richardson
during Unicode conversion.
2012-02-15Added a kind of hacky way to interpret the restriction class='foo bar'. Stop ↵Leonard Richardson
generating a space before the slash that closes an empty-element tag.
2012-02-15The value of multi-valued attributes like class are always turned into a ↵Leonard Richardson
list, even if there's only one value.
2012-02-15Some cdata-list attributes are only cdata lists for certain tags.Leonard Richardson
2012-02-15Better defined behavior when the user wants to search for a combination of ↵Leonard Richardson
text and tag-specific arguments. [bug=695312]
2012-02-15Tested and cleaned up html5lib insertBefore.Leonard Richardson
2012-02-15Tested improvements to html5lib treebuilder.Leonard Richardson
2012-02-15Tested that extract() distinguishes between idientical strings.Leonard Richardson
2012-02-09As a last-ditch attempt to turn data into Unicode, use errors=replace ↵Leonard Richardson
instead of errors=strict.
2012-02-09Unicode, Dammit now detects the encoding in HTML 5-style <meta> tags like ↵Leonard Richardson
<meta charset="utf-8" />. [bug=837268]
2012-02-09Fixed a bug that wrecked the tree if you replaced an element with an empty ↵Leonard Richardson
string. [bug=728697]
2012-02-08Rationalized the treatment of multi-valued HTML attributes such as 'class'Leonard Richardson
2012-02-08Switched around the implementation of insert_before and insert_after to ↵Leonard Richardson
match jQuery.
2012-02-08Renamed move_* back to insert_*Leonard Richardson
2012-02-07Newly created tags use the same empty-element rules as the builder used to ↵Leonard Richardson
originally create the soup.
2012-02-07BeautifulSoup objects don't support move_before() or move_after.Leonard Richardson
2012-02-07Renamed insert_before to move_before, to avoid confusion with the way ↵Leonard Richardson
insert() works.
2012-02-07Added and tested insert_before and insert_after.Leonard Richardson
2012-02-07Made the formatter something you pass in rather than a variable you set.Leonard Richardson
2012-02-07On output, always convert special XML characters to entities.Leonard Richardson
2012-02-06Monkeypatch Python 3.2 versions prior to 3.2.3 to solve some major ↵Leonard Richardson
HTMLParser bugs.
2012-02-06You can't replace a tag with its parent, or insert a tag into itself. ↵Leonard Richardson
replace operations return the thing that was replaced.
2012-02-01Removed test that we can't really test anymore.Leonard Richardson
2012-02-01Cleaned up the TODO.Leonard Richardson
2012-02-01Generators no longer yield None after going off the end of the iteration.Leonard Richardson
2012-02-01Got the new-tag test to pass.Leonard Richardson
2012-02-01Added tag creation and string generators, both created while writing the docs.Leonard Richardson
2012-01-20Tests now work in both versions, and it's possible to test both versions by ↵Leonard Richardson
running one command.
2012-01-20Made it easier to convert BS3 code to BS4.Leonard Richardson
2012-01-20Replaced assertEquals with assertEqual to get rid of deprecation notice.Leonard Richardson
2012-01-20Actually fixed the test failure.Leonard Richardson
2012-01-20Got the test suite to pass on Python 3.2 (skipping the html5lib stuff, which ↵Leonard Richardson
doesn't seem to have Python 3 support yet.)
2011-07-07Small fix for test.Thomas Kluyver
2011-06-29Various changes so most tests pass on Python 3.Thomas Kluyver