From patchwork Fri May 31 11:41:50 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: Richard Purdie X-Patchwork-Id: 44459 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from aws-us-west-2-korg-lkml-1.web.codeaurora.org (localhost.localdomain [127.0.0.1]) by smtp.lore.kernel.org (Postfix) with ESMTP id 0C31FC25B75 for ; Fri, 31 May 2024 11:42:06 +0000 (UTC) Received: from mail-wr1-f45.google.com (mail-wr1-f45.google.com [209.85.221.45]) by mx.groups.io with SMTP id smtpd.web10.9472.1717155716242907643 for ; Fri, 31 May 2024 04:41:57 -0700 Authentication-Results: mx.groups.io; dkim=pass header.i=@linuxfoundation.org header.s=google header.b=dx24fLL4; spf=pass (domain: linuxfoundation.org, ip: 209.85.221.45, mailfrom: richard.purdie@linuxfoundation.org) Received: by mail-wr1-f45.google.com with SMTP id ffacd0b85a97d-35dceef4227so965414f8f.0 for ; Fri, 31 May 2024 04:41:56 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linuxfoundation.org; s=google; t=1717155714; x=1717760514; darn=lists.openembedded.org; h=content-transfer-encoding:mime-version:message-id:date:subject:to :from:from:to:cc:subject:date:message-id:reply-to; bh=kbfFU81E16SDViXapJZE4W5OEzwTFROVokl3nQlAyj8=; b=dx24fLL4LBS7Akk7RIUzO/9bdPyo2lLYlPu9eV4zHgI42IYcfqsZnmEbWj/mTydg/p 6jiOjMGpfIvPvqrQLyX4eDlLmm5EQjcJqwMtsaxFqCRRpnEzZR7DDkt6J+k2YhgfGUe3 JjP9d36zf+ozN6/MRf+vKv75J6rzRlbU2aEic= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1717155714; x=1717760514; h=content-transfer-encoding:mime-version:message-id:date:subject:to :from:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=kbfFU81E16SDViXapJZE4W5OEzwTFROVokl3nQlAyj8=; b=od9lTgkUqt9OZIWXELqfuFBRBDjoZGMNZvd1/sXsa/AuH+1kz/dqpN4Qqy/bnQpM5H eA1TEE1C0UPISKAygBIh+jq5/DG1koYnq7Dj1ZSpS3kktOf6wtzRevknPnebjhlm4kJz 4SAky4AqK93G9fwyIfKuH9ldYC6fdLhCxTH4UqnR3MKlUH3rt+w8+NG+jCekNckO+0b0 +Upcc+GcNBlqGQkjRGsLoo+tZTe5tAu/cGsev6rmWv4d3KI6e5+CvcBkMJpWPDEY3gMc 2MtxsSHfFD0PHB2IDRnbqhw7FawEdWVqM9t6DoeAUhsXFj5zEfHOs3kFC55HInc1r9bp UREw== X-Gm-Message-State: AOJu0YwOxiZhCUjXybNCRWjk6SS6lFNWGG64N6VpXJvSGqX0Nwoscdhw kKPk0zIBlozC29rVxc32Y+o6JCvda73+paDnf+AzMQihsM/ItJYrYE7OgTAez5uYkrDG3svyjwD I X-Google-Smtp-Source: AGHT+IE2NNUNwMyJZbEfnvZO2pGsNU5/rlepYfM8XcIlrHzSGonOHffpCL77ZFCV0BGY7fY7oia4fQ== X-Received: by 2002:adf:ea11:0:b0:354:f2b0:ebd6 with SMTP id ffacd0b85a97d-35e0f25e2b9mr1134881f8f.9.1717155712584; Fri, 31 May 2024 04:41:52 -0700 (PDT) Received: from max.int.rpsys.net ([2001:8b0:aba:5f3c:9173:4fa2:d1b3:c8be]) by smtp.gmail.com with ESMTPSA id ffacd0b85a97d-35dd04d9c68sm1661078f8f.56.2024.05.31.04.41.51 for (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Fri, 31 May 2024 04:41:51 -0700 (PDT) From: Richard Purdie To: bitbake-devel@lists.openembedded.org Subject: [PATCH] bs4: Update to 4.12.3 from 4.4.1 Date: Fri, 31 May 2024 12:41:50 +0100 Message-Id: <20240531114150.3255109-1-richard.purdie@linuxfoundation.org> X-Mailer: git-send-email 2.40.1 MIME-Version: 1.0 List-Id: X-Webhook-Received: from li982-79.members.linode.com [45.33.32.79] by aws-us-west-2-korg-lkml-1.web.codeaurora.org with HTTPS for ; Fri, 31 May 2024 11:42:06 -0000 X-Groupsio-URL: https://lists.openembedded.org/g/bitbake-devel/message/16297 It makes sense to switch to a more recent version and keep up to date with upstream changes and things like new python version support. Signed-off-by: Richard Purdie --- lib/bs4/AUTHORS | 49 + lib/bs4/AUTHORS.txt | 43 - lib/bs4/{NEWS.txt => CHANGELOG} | 779 ++++++++- lib/bs4/{COPYING.txt => LICENSE} | 11 +- lib/bs4/__init__.py | 680 ++++++-- lib/bs4/builder/__init__.py | 382 +++- lib/bs4/builder/_html5lib.py | 251 ++- lib/bs4/builder/_htmlparser.py | 433 +++-- lib/bs4/builder/_lxml.py | 212 ++- lib/bs4/css.py | 280 +++ lib/bs4/dammit.py | 411 ++++- lib/bs4/diagnose.py | 83 +- lib/bs4/element.py | 2219 ++++++++++++++++-------- lib/bs4/formatter.py | 185 ++ lib/bs4/testing.py | 686 -------- lib/bs4/tests/__init__.py | 1 - lib/bs4/tests/test_builder_registry.py | 147 -- lib/bs4/tests/test_docs.py | 32 - lib/bs4/tests/test_html5lib.py | 98 -- lib/bs4/tests/test_htmlparser.py | 31 - lib/bs4/tests/test_lxml.py | 70 - lib/bs4/tests/test_soup.py | 479 ----- lib/bs4/tests/test_tree.py | 2004 --------------------- 23 files changed, 4675 insertions(+), 4891 deletions(-) create mode 100644 lib/bs4/AUTHORS delete mode 100644 lib/bs4/AUTHORS.txt rename lib/bs4/{NEWS.txt => CHANGELOG} (56%) rename lib/bs4/{COPYING.txt => LICENSE} (80%) create mode 100644 lib/bs4/css.py create mode 100644 lib/bs4/formatter.py delete mode 100644 lib/bs4/testing.py delete mode 100644 lib/bs4/tests/__init__.py delete mode 100644 lib/bs4/tests/test_builder_registry.py delete mode 100644 lib/bs4/tests/test_docs.py delete mode 100644 lib/bs4/tests/test_html5lib.py delete mode 100644 lib/bs4/tests/test_htmlparser.py delete mode 100644 lib/bs4/tests/test_lxml.py delete mode 100644 lib/bs4/tests/test_soup.py delete mode 100644 lib/bs4/tests/test_tree.py diff --git a/lib/bs4/AUTHORS b/lib/bs4/AUTHORS new file mode 100644 index 0000000000..1f14fe07de --- /dev/null +++ b/lib/bs4/AUTHORS @@ -0,0 +1,49 @@ +Behold, mortal, the origins of Beautiful Soup... +================================================ + +Leonard Richardson is the primary maintainer. + +Aaron DeVore and Isaac Muse have made significant contributions to the +code base. + +Mark Pilgrim provided the encoding detection code that forms the base +of UnicodeDammit. + +Thomas Kluyver and Ezio Melotti finished the work of getting Beautiful +Soup 4 working under Python 3. + +Simon Willison wrote soupselect, which was used to make Beautiful Soup +support CSS selectors. Isaac Muse wrote SoupSieve, which made it +possible to _remove_ the CSS selector code from Beautiful Soup. + +Sam Ruby helped with a lot of edge cases. + +Jonathan Ellis was awarded the prestigious Beau Potage D'Or for his +work in solving the nestable tags conundrum. + +An incomplete list of people have contributed patches to Beautiful +Soup: + + Istvan Albert, Andrew Lin, Anthony Baxter, Oliver Beattie, Andrew +Boyko, Tony Chang, Francisco Canas, "Delong", Zephyr Fang, Fuzzy, +Roman Gaufman, Yoni Gilad, Richie Hindle, Toshihiro Kamiya, Peteris +Krumins, Kent Johnson, Marek Kapolka, Andreas Kostyrka, Roel Kramer, +Ben Last, Robert Leftwich, Stefaan Lippens, "liquider", Staffan +Malmgren, Ksenia Marasanova, JP Moins, Adam Monsen, John Nagle, "Jon", +Ed Oskiewicz, Martijn Peters, Greg Phillips, Giles Radford, Stefano +Revera, Arthur Rudolph, Marko Samastur, James Salter, Jouni Sepp�nen, +Alexander Schmolck, Tim Shirley, Geoffrey Sneddon, Ville Skytt�, +"Vikas", Jens Svalgaard, Andy Theyers, Eric Weiser, Glyn Webster, John +Wiseman, Paul Wright, Danny Yoo + +An incomplete list of people who made suggestions or found bugs or +found ways to break Beautiful Soup: + + Hanno B�ck, Matteo Bertini, Chris Curvey, Simon Cusack, Bruce Eckel, + Matt Ernst, Michael Foord, Tom Harris, Bill de hOra, Donald Howes, + Matt Patterson, Scott Roberts, Steve Strassmann, Mike Williams, + warchild at redho dot com, Sami Kuisma, Carlos Rocha, Bob Hutchison, + Joren Mc, Michal Migurski, John Kleven, Tim Heaney, Tripp Lilley, Ed + Summers, Dennis Sutch, Chris Smith, Aaron Swartz, Stuart + Turner, Greg Edwards, Kevin J Kalupson, Nikos Kouremenos, Artur de + Sousa Rocha, Yichun Wei, Per Vognsen diff --git a/lib/bs4/AUTHORS.txt b/lib/bs4/AUTHORS.txt deleted file mode 100644 index 2ac8fcc8cc..0000000000 --- a/lib/bs4/AUTHORS.txt +++ /dev/null @@ -1,43 +0,0 @@ -Behold, mortal, the origins of Beautiful Soup... -================================================ - -Leonard Richardson is the primary programmer. - -Aaron DeVore is awesome. - -Mark Pilgrim provided the encoding detection code that forms the base -of UnicodeDammit. - -Thomas Kluyver and Ezio Melotti finished the work of getting Beautiful -Soup 4 working under Python 3. - -Simon Willison wrote soupselect, which was used to make Beautiful Soup -support CSS selectors. - -Sam Ruby helped with a lot of edge cases. - -Jonathan Ellis was awarded the prestigous Beau Potage D'Or for his -work in solving the nestable tags conundrum. - -An incomplete list of people have contributed patches to Beautiful -Soup: - - Istvan Albert, Andrew Lin, Anthony Baxter, Andrew Boyko, Tony Chang, - Zephyr Fang, Fuzzy, Roman Gaufman, Yoni Gilad, Richie Hindle, Peteris - Krumins, Kent Johnson, Ben Last, Robert Leftwich, Staffan Malmgren, - Ksenia Marasanova, JP Moins, Adam Monsen, John Nagle, "Jon", Ed - Oskiewicz, Greg Phillips, Giles Radford, Arthur Rudolph, Marko - Samastur, Jouni Sepp�nen, Alexander Schmolck, Andy Theyers, Glyn - Webster, Paul Wright, Danny Yoo - -An incomplete list of people who made suggestions or found bugs or -found ways to break Beautiful Soup: - - Hanno B�ck, Matteo Bertini, Chris Curvey, Simon Cusack, Bruce Eckel, - Matt Ernst, Michael Foord, Tom Harris, Bill de hOra, Donald Howes, - Matt Patterson, Scott Roberts, Steve Strassmann, Mike Williams, - warchild at redho dot com, Sami Kuisma, Carlos Rocha, Bob Hutchison, - Joren Mc, Michal Migurski, John Kleven, Tim Heaney, Tripp Lilley, Ed - Summers, Dennis Sutch, Chris Smith, Aaron Sweep^W Swartz, Stuart - Turner, Greg Edwards, Kevin J Kalupson, Nikos Kouremenos, Artur de - Sousa Rocha, Yichun Wei, Per Vognsen diff --git a/lib/bs4/NEWS.txt b/lib/bs4/CHANGELOG similarity index 56% rename from lib/bs4/NEWS.txt rename to lib/bs4/CHANGELOG index 88a60a2458..2701446a6d 100644 --- a/lib/bs4/NEWS.txt +++ b/lib/bs4/CHANGELOG @@ -1,3 +1,776 @@ += 4.12.3 (20240117) + +* The Beautiful Soup documentation now has a Spanish translation, thanks + to Carlos Romero. Delong Wang's Chinese translation has been updated + to cover Beautiful Soup 4.12.0. + +* Fixed a regression such that if you set .hidden on a tag, the tag + becomes invisible but its contents are still visible. User manipulation + of .hidden is not a documented or supported feature, so don't do this, + but it wasn't too difficult to keep the old behavior working. + +* Fixed a case found by Mengyuhan where html.parser giving up on + markup would result in an AssertionError instead of a + ParserRejectedMarkup exception. + +* Added the correct stacklevel to instances of the XMLParsedAsHTMLWarning. + [bug=2034451] + +* Corrected the syntax of the license definition in pyproject.toml. Patch + by Louis Maddox. [bug=2032848] + +* Corrected a typo in a test that was causing test failures when run against + libxml2 2.12.1. [bug=2045481] + += 4.12.2 (20230407) + +* Fixed an unhandled exception in BeautifulSoup.decode_contents + and methods that call it. [bug=2015545] + += 4.12.1 (20230405) + +NOTE: the following things are likely to be dropped in the next +feature release of Beautiful Soup: + + Official support for Python 3.6. + Inclusion of unit tests and test data in the wheel file. + Two scripts: demonstrate_parser_differences.py and test-all-versions. + +Changes: + +* This version of Beautiful Soup replaces setup.py and setup.cfg + with pyproject.toml. Beautiful Soup now uses tox as its test backend + and hatch to do builds. + +* The main functional improvement in this version is a nonrecursive technique + for regenerating a tree. This technique is used to avoid situations where, + in previous versions, doing something to a very deeply nested tree + would overflow the Python interpreter stack: + + 1. Outputting a tree as a string, e.g. with + BeautifulSoup.encode() [bug=1471755] + + 2. Making copies of trees (copy.copy() and + copy.deepcopy() from the Python standard library). [bug=1709837] + + 3. Pickling a BeautifulSoup object. (Note that pickling a Tag + object can still cause an overflow.) + +* Making a copy of a BeautifulSoup object no longer parses the + document again, which should improve performance significantly. + +* When a BeautifulSoup object is unpickled, Beautiful Soup now + tries to associate an appropriate TreeBuilder object with it. + +* Tag.prettify() will now consistently end prettified markup with + a newline. + +* Added unit tests for fuzz test cases created by third + parties. Some of these tests are skipped since they point + to problems outside of Beautiful Soup, but this change + puts them all in one convenient place. + +* PageElement now implements the known_xml attribute. (This was technically + a bug, but it shouldn't be an issue in normal use.) [bug=2007895] + +* The demonstrate_parser_differences.py script was still written in + Python 2. I've converted it to Python 3, but since no one has + mentioned this over the years, it's a sign that no one uses this + script and it's not serving its purpose. + += 4.12.0 (20230320) + +* Introduced the .css property, which centralizes all access to + the Soup Sieve API. This allows Beautiful Soup to give direct + access to as much of Soup Sieve that makes sense, without cluttering + the BeautifulSoup and Tag classes with a lot of new methods. + + This does mean one addition to the BeautifulSoup and Tag classes + (the .css property itself), so this might be a breaking change if you + happen to use Beautiful Soup to parse XML that includes a tag called + . In particular, code like this will stop working in 4.12.0: + + soup.css['id'] + + Code like this will work just as before: + + soup.find_one('css')['id'] + + The Soup Sieve methods supported through the .css property are + select(), select_one(), iselect(), closest(), match(), filter(), + escape(), and compile(). The BeautifulSoup and Tag classes still + support the select() and select_one() methods; they have not been + deprecated, but they have been demoted to convenience methods. + + [bug=2003677] + +* When the html.parser parser decides it can't parse a document, Beautiful + Soup now consistently propagates this fact by raising a + ParserRejectedMarkup error. [bug=2007343] + +* Removed some error checking code from diagnose(), which is redundant with + similar (but more Pythonic) code in the BeautifulSoup constructor. + [bug=2007344] + +* Added intersphinx references to the documentation so that other + projects have a target to point to when they reference Beautiful + Soup classes. [bug=1453370] + += 4.11.2 (20230131) + +* Fixed test failures caused by nondeterministic behavior of + UnicodeDammit's character detection, depending on the platform setup. + [bug=1973072] + +* Fixed another crash when overriding multi_valued_attributes and using the + html5lib parser. [bug=1948488] + +* The HTMLFormatter and XMLFormatter constructors no longer return a + value. [bug=1992693] + +* Tag.interesting_string_types is now propagated when a tag is + copied. [bug=1990400] + +* Warnings now do their best to provide an appropriate stacklevel, + improving the usefulness of the message. [bug=1978744] + +* Passing a Tag's .contents into PageElement.extend() now works the + same way as passing the Tag itself. + +* Soup Sieve tests will be skipped if the library is not installed. + += 4.11.1 (20220408) + +This release was done to ensure that the unit tests are packaged along +with the released source. There are no functionality changes in this +release, but there are a few other packaging changes: + +* The Japanese and Korean translations of the documentation are included. +* The changelog is now packaged as CHANGELOG, and the license file is + packaged as LICENSE. NEWS.txt and COPYING.txt are still present, + but may be removed in the future. +* TODO.txt is no longer packaged, since a TODO is not relevant for released + code. + += 4.11.0 (20220407) + +* Ported unit tests to use pytest. + +* Added special string classes, RubyParenthesisString and RubyTextString, + to make it possible to treat ruby text specially in get_text() calls. + [bug=1941980] + +* It's now possible to customize the way output is indented by + providing a value for the 'indent' argument to the Formatter + constructor. The 'indent' argument works very similarly to the + argument of the same name in the Python standard library's + json.dump() function. [bug=1955497] + +* If the charset-normalizer Python module + (https://pypi.org/project/charset-normalizer/) is installed, Beautiful + Soup will use it to detect the character sets of incoming documents. + This is also the module used by newer versions of the Requests library. + For the sake of backwards compatibility, chardet and cchardet both take + precedence if installed. [bug=1955346] + +* Added a workaround for an lxml bug + (https://bugs.launchpad.net/lxml/+bug/1948551) that causes + problems when parsing a Unicode string beginning with BYTE ORDER MARK. + [bug=1947768] + +* Issue a warning when an HTML parser is used to parse a document that + looks like XML but not XHTML. [bug=1939121] + +* Do a better job of keeping track of namespaces as an XML document is + parsed, so that CSS selectors that use namespaces will do the right + thing more often. [bug=1946243] + +* Some time ago, the misleadingly named "text" argument to find-type + methods was renamed to the more accurate "string." But this supposed + "renaming" didn't make it into important places like the method + signatures or the docstrings. That's corrected in this + version. "text" still works, but will give a DeprecationWarning. + [bug=1947038] + +* Fixed a crash when pickling a BeautifulSoup object that has no + tree builder. [bug=1934003] + +* Fixed a crash when overriding multi_valued_attributes and using the + html5lib parser. [bug=1948488] + +* Standardized the wording of the MarkupResemblesLocatorWarning + warnings to omit untrusted input and make the warnings less + judgmental about what you ought to be doing. [bug=1955450] + +* Removed support for the iconv_codec library, which doesn't seem + to exist anymore and was never put up on PyPI. (The closest + replacement on PyPI, iconv_codecs, is GPL-licensed, so we can't use + it--it's also quite old.) + += 4.10.0 (20210907) + +* This is the first release of Beautiful Soup to only support Python + 3. I dropped Python 2 support to maintain support for newer versions + (58 and up) of setuptools. See: + https://github.com/pypa/setuptools/issues/2769 [bug=1942919] + +* The behavior of methods like .get_text() and .strings now differs + depending on the type of tag. The change is visible with HTML tags + like - -Hello, world! - - -''' - soup = self.soup(html) - self.assertEqual("text/javascript", soup.find('script')['type']) - - def test_comment(self): - # Comments are represented as Comment objects. - markup = "

foobaz

" - self.assertSoupEquals(markup) - - soup = self.soup(markup) - comment = soup.find(text="foobar") - self.assertEqual(comment.__class__, Comment) - - # The comment is properly integrated into the tree. - foo = soup.find(text="foo") - self.assertEqual(comment, foo.next_element) - baz = soup.find(text="baz") - self.assertEqual(comment, baz.previous_element) - - def test_preserved_whitespace_in_pre_and_textarea(self): - """Whitespace must be preserved in 
 and ")
-
-    def test_nested_inline_elements(self):
-        """Inline elements can be nested indefinitely."""
-        b_tag = "Inside a B tag"
-        self.assertSoupEquals(b_tag)
-
-        nested_b_tag = "
A nested tag
"
-        self.assertSoupEquals(nested_b_tag)
-
-        double_nested_b_tag = "
A doubly nested tag
"
-        self.assertSoupEquals(nested_b_tag)
-
-    def test_nested_block_level_elements(self):
-        """Block elements can be nested."""
-        soup = self.soup('
Foo
')
-        blockquote = soup.blockquote
-        self.assertEqual(blockquote.p.b.string, 'Foo')
-        self.assertEqual(blockquote.b.string, 'Foo')
-
-    def test_correctly_nested_tables(self):
-        """One table can go inside another one."""
-        markup = (''
-                  ''
-                  "')
-
-        self.assertSoupEquals(
-            markup,
-            '
Here's another table:"
-                  ''
-                  ''
-                  '
foo
Here\'s another table:'
-            '
foo
'
-            '
')
-
-        self.assertSoupEquals(
-            ""
-            ""
-            "
Foo
Bar
Baz
")
-
-    def test_deeply_nested_multivalued_attribute(self):
-        # html5lib can set the attributes of the same tag many times
-        # as it rearranges the tree. This has caused problems with
-        # multivalued attributes.
-        markup = '
'
-        soup = self.soup(markup)
-        self.assertEqual(["css"], soup.div.div['class'])
-
-    def test_multivalued_attribute_on_html(self):
-        # html5lib uses a different API to set the attributes ot the
-        #  tag. This has caused problems with multivalued
-        # attributes.
-        markup = ''
-        soup = self.soup(markup)
-        self.assertEqual(["a", "b"], soup.html['class'])
-
-    def test_angle_brackets_in_attribute_values_are_escaped(self):
-        self.assertSoupEquals('', '')
-
-    def test_entities_in_attributes_converted_to_unicode(self):
-        expect = '
'
-        self.assertSoupEquals('
', expect)
-        self.assertSoupEquals('
', expect)
-        self.assertSoupEquals('
', expect)
-        self.assertSoupEquals('
', expect)
-
-    def test_entities_in_text_converted_to_unicode(self):
-        expect = '
pi\N{LATIN SMALL LETTER N WITH TILDE}ata
'
-        self.assertSoupEquals("
piñata
", expect)
-        self.assertSoupEquals("
piñata
", expect)
-        self.assertSoupEquals("
piñata
", expect)
-        self.assertSoupEquals("
piñata
", expect)
-
-    def test_quot_entity_converted_to_quotation_mark(self):
-        self.assertSoupEquals("
I said "good day!"
",
-                              '
I said "good day!"
')
-
-    def test_out_of_range_entity(self):
-        expect = "\N{REPLACEMENT CHARACTER}"
-        self.assertSoupEquals("�", expect)
-        self.assertSoupEquals("�", expect)
-        self.assertSoupEquals("�", expect)
-
-    def test_multipart_strings(self):
-        "Mostly to prevent a recurrence of a bug in the html5lib treebuilder."
-        soup = self.soup("
\nfoo
")
-        self.assertEqual("p", soup.h2.string.next_element.name)
-        self.assertEqual("p", soup.p.name)
-        self.assertConnectedness(soup)
-
-    def test_head_tag_between_head_and_body(self):
-        "Prevent recurrence of a bug in the html5lib treebuilder."
-        content = """
-  
-  foo
-
-"""
-        soup = self.soup(content)
-        self.assertNotEqual(None, soup.html.body)
-        self.assertConnectedness(soup)
-
-    def test_multiple_copies_of_a_tag(self):
-        "Prevent recurrence of a bug in the html5lib treebuilder."
-        content = """
-
- 
-   
-  
-
-"""
-        soup = self.soup(content)
-        self.assertConnectedness(soup.article)
-
-    def test_basic_namespaces(self):
-        """Parsers don't need to *understand* namespaces, but at the
-        very least they should not choke on namespaces or lose
-        data."""
-
-        markup = b'4'
-        soup = self.soup(markup)
-        self.assertEqual(markup, soup.encode())
-        html = soup.html
-        self.assertEqual('http://www.w3.org/1999/xhtml', soup.html['xmlns'])
-        self.assertEqual(
-            'http://www.w3.org/1998/Math/MathML', soup.html['xmlns:mathml'])
-        self.assertEqual(
-            'http://www.w3.org/2000/svg', soup.html['xmlns:svg'])
-
-    def test_multivalued_attribute_value_becomes_list(self):
-        markup = b''
-        soup = self.soup(markup)
-        self.assertEqual(['foo', 'bar'], soup.a['class'])
-
-    #
-    # Generally speaking, tests below this point are more tests of
-    # Beautiful Soup than tests of the tree builders. But parsers are
-    # weird, so we run these tests separately for every tree builder
-    # to detect any differences between them.
-    #
-
-    def test_can_parse_unicode_document(self):
-        # A seemingly innocuous document... but it's in Unicode! And
-        # it contains characters that can't be represented in the
-        # encoding found in the  declaration! The horror!
-        markup = 'Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!'
-        soup = self.soup(markup)
-        self.assertEqual('Sacr\xe9 bleu!', soup.body.string)
-
-    def test_soupstrainer(self):
-        """Parsers should be able to work with SoupStrainers."""
-        strainer = SoupStrainer("b")
-        soup = self.soup("A bold  statement",
-                         parse_only=strainer)
-        self.assertEqual(soup.decode(), "bold")
-
-    def test_single_quote_attribute_values_become_double_quotes(self):
-        self.assertSoupEquals("",
-                              '')
-
-    def test_attribute_values_with_nested_quotes_are_left_alone(self):
-        text = """a"""
-        self.assertSoupEquals(text)
-
-    def test_attribute_values_with_double_nested_quotes_get_quoted(self):
-        text = """a"""
-        soup = self.soup(text)
-        soup.foo['attr'] = 'Brawls happen at "Bob\'s Bar"'
-        self.assertSoupEquals(
-            soup.foo.decode(),
-            """a""")
-
-    def test_ampersand_in_attribute_value_gets_escaped(self):
-        self.assertSoupEquals('',
-                              '')
-
-        self.assertSoupEquals(
-            'foo',
-            'foo')
-
-    def test_escaped_ampersand_in_attribute_value_is_left_alone(self):
-        self.assertSoupEquals('')
-
-    def test_entities_in_strings_converted_during_parsing(self):
-        # Both XML and HTML entities are converted to Unicode characters
-        # during parsing.
-        text = "
<<sacré bleu!>>
"
-        expected = "
<<sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>>
"
-        self.assertSoupEquals(text, expected)
-
-    def test_smart_quotes_converted_on_the_way_in(self):
-        # Microsoft smart quotes are converted to Unicode characters during
-        # parsing.
-        quote = b"
\x91Foo\x92
"
-        soup = self.soup(quote)
-        self.assertEqual(
-            soup.p.string,
-            "\N{LEFT SINGLE QUOTATION MARK}Foo\N{RIGHT SINGLE QUOTATION MARK}")
-
-    def test_non_breaking_spaces_converted_on_the_way_in(self):
-        soup = self.soup("  ")
-        self.assertEqual(soup.a.string, "\N{NO-BREAK SPACE}" * 2)
-
-    def test_entities_converted_on_the_way_out(self):
-        text = "
<<sacré bleu!>>
"
-        expected = "
<<sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>>
".encode("utf-8")
-        soup = self.soup(text)
-        self.assertEqual(soup.p.encode("utf-8"), expected)
-
-    def test_real_iso_latin_document(self):
-        # Smoke test of interrelated functionality, using an
-        # easy-to-understand document.
-
-        # Here it is in Unicode. Note that it claims to be in ISO-Latin-1.
-        unicode_html = '
Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!
'
-
-        # That's because we're going to encode it into ISO-Latin-1, and use
-        # that to test.
-        iso_latin_html = unicode_html.encode("iso-8859-1")
-
-        # Parse the ISO-Latin-1 HTML.
-        soup = self.soup(iso_latin_html)
-        # Encode it to UTF-8.
-        result = soup.encode("utf-8")
-
-        # What do we expect the result to look like? Well, it would
-        # look like unicode_html, except that the META tag would say
-        # UTF-8 instead of ISO-Latin-1.
-        expected = unicode_html.replace("ISO-Latin-1", "utf-8")
-
-        # And, of course, it would be in UTF-8, not Unicode.
-        expected = expected.encode("utf-8")
-
-        # Ta-da!
-        self.assertEqual(result, expected)
-
-    def test_real_shift_jis_document(self):
-        # Smoke test to make sure the parser can handle a document in
-        # Shift-JIS encoding, without choking.
-        shift_jis_html = (
-            b'
'
-            b'\x82\xb1\x82\xea\x82\xcdShift-JIS\x82\xc5\x83R\x81[\x83f'
-            b'\x83B\x83\x93\x83O\x82\xb3\x82\xea\x82\xbd\x93\xfa\x96{\x8c'
-            b'\xea\x82\xcc\x83t\x83@\x83C\x83\x8b\x82\xc5\x82\xb7\x81B'
-            b'
')
-        unicode_html = shift_jis_html.decode("shift-jis")
-        soup = self.soup(unicode_html)
-
-        # Make sure the parse tree is correctly encoded to various
-        # encodings.
-        self.assertEqual(soup.encode("utf-8"), unicode_html.encode("utf-8"))
-        self.assertEqual(soup.encode("euc_jp"), unicode_html.encode("euc_jp"))
-
-    def test_real_hebrew_document(self):
-        # A real-world test to make sure we can convert ISO-8859-9 (a
-        # Hebrew encoding) to UTF-8.
-        hebrew_document = b'Hebrew (ISO 8859-8) in Visual Directionality
Hebrew (ISO 8859-8) in Visual Directionality
\xed\xe5\xec\xf9'
-        soup = self.soup(
-            hebrew_document, from_encoding="iso8859-8")
-        self.assertEqual(soup.original_encoding, 'iso8859-8')
-        self.assertEqual(
-            soup.encode('utf-8'),
-            hebrew_document.decode("iso8859-8").encode("utf-8"))
-
-    def test_meta_tag_reflects_current_encoding(self):
-        # Here's the  tag saying that a document is
-        # encoded in Shift-JIS.
-        meta_tag = ('')
-
-        # Here's a document incorporating that meta tag.
-        shift_jis_html = (
-            '\n%s\n'
-            ''
-            'Shift-JIS markup goes here.') % meta_tag
-        soup = self.soup(shift_jis_html)
-
-        # Parse the document, and the charset is seemingly unaffected.
-        parsed_meta = soup.find('meta', {'http-equiv': 'Content-type'})
-        content = parsed_meta['content']
-        self.assertEqual('text/html; charset=x-sjis', content)
-
-        # But that value is actually a ContentMetaAttributeValue object.
-        self.assertTrue(isinstance(content, ContentMetaAttributeValue))
-
-        # And it will take on a value that reflects its current
-        # encoding.
-        self.assertEqual('text/html; charset=utf8', content.encode("utf8"))
-
-        # For the rest of the story, see TestSubstitutions in
-        # test_tree.py.
-
-    def test_html5_style_meta_tag_reflects_current_encoding(self):
-        # Here's the  tag saying that a document is
-        # encoded in Shift-JIS.
-        meta_tag = ('')
-
-        # Here's a document incorporating that meta tag.
-        shift_jis_html = (
-            '\n%s\n'
-            ''
-            'Shift-JIS markup goes here.') % meta_tag
-        soup = self.soup(shift_jis_html)
-
-        # Parse the document, and the charset is seemingly unaffected.
-        parsed_meta = soup.find('meta', id="encoding")
-        charset = parsed_meta['charset']
-        self.assertEqual('x-sjis', charset)
-
-        # But that value is actually a CharsetMetaAttributeValue object.
-        self.assertTrue(isinstance(charset, CharsetMetaAttributeValue))
-
-        # And it will take on a value that reflects its current
-        # encoding.
-        self.assertEqual('utf8', charset.encode("utf8"))
-
-    def test_tag_with_no_attributes_can_have_attributes_added(self):
-        data = self.soup("text")
-        data.a['foo'] = 'bar'
-        self.assertEqual('text', data.a.decode())
-
-class XMLTreeBuilderSmokeTest(SoupTest):
-
-    def test_pickle_and_unpickle_identity(self):
-        # Pickling a tree, then unpickling it, yields a tree identical
-        # to the original.
-        tree = self.soup("foo")
-        dumped = pickle.dumps(tree, 2)
-        loaded = pickle.loads(dumped)
-        self.assertEqual(loaded.__class__, BeautifulSoup)
-        self.assertEqual(loaded.decode(), tree.decode())
-
-    def test_docstring_generated(self):
-        soup = self.soup("")
-        self.assertEqual(
-            soup.encode(), b'\n')
-
-    def test_xml_declaration(self):
-        markup = b"""\n"""
-        soup = self.soup(markup)
-        self.assertEqual(markup, soup.encode("utf8"))
-
-    def test_real_xhtml_document(self):
-        """A real XHTML document should come out *exactly* the same as it went in."""
-        markup = b"""
-
-
-Hello.
-Goodbye.
-"""
-        soup = self.soup(markup)
-        self.assertEqual(
-            soup.encode("utf-8"), markup)
-
-    def test_formatter_processes_script_tag_for_xml_documents(self):
-        doc = """
-  
-"""
-        soup = BeautifulSoup(doc, "lxml-xml")
-        # lxml would have stripped this while parsing, but we can add
-        # it later.
-        soup.script.string = 'console.log("< < hey > > ");'
-        encoded = soup.encode()
-        self.assertTrue(b"< < hey > >" in encoded)
-
-    def test_can_parse_unicode_document(self):
-        markup = 'Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!'
-        soup = self.soup(markup)
-        self.assertEqual('Sacr\xe9 bleu!', soup.root.string)
-
-    def test_popping_namespaced_tag(self):
-        markup = 'b2012-07-02T20:33:42Zcd'
-        soup = self.soup(markup)
-        self.assertEqual(
-            str(soup.rss), markup)
-
-    def test_docstring_includes_correct_encoding(self):
-        soup = self.soup("")
-        self.assertEqual(
-            soup.encode("latin1"),
-            b'\n')
-
-    def test_large_xml_document(self):
-        """A large XML document should come out the same as it went in."""
-        markup = (b'\n'
-                  + b'0' * (2**12)
-                  + b'')
-        soup = self.soup(markup)
-        self.assertEqual(soup.encode("utf-8"), markup)
-
-
-    def test_tags_are_empty_element_if_and_only_if_they_are_empty(self):
-        self.assertSoupEquals("
", "
")
-        self.assertSoupEquals("
foo
")
-
-    def test_namespaces_are_preserved(self):
-        markup = 'This tag is in the a namespaceThis tag is in the b namespace'
-        soup = self.soup(markup)
-        root = soup.root
-        self.assertEqual("http://example.com/", root['xmlns:a'])
-        self.assertEqual("http://example.net/", root['xmlns:b'])
-
-    def test_closing_namespaced_tag(self):
-        markup = '
20010504
'
-        soup = self.soup(markup)
-        self.assertEqual(str(soup.p), markup)
-
-    def test_namespaced_attributes(self):
-        markup = ''
-        soup = self.soup(markup)
-        self.assertEqual(str(soup.foo), markup)
-
-    def test_namespaced_attributes_xml_namespace(self):
-        markup = 'bar'
-        soup = self.soup(markup)
-        self.assertEqual(str(soup.foo), markup)
-
-class HTML5TreeBuilderSmokeTest(HTMLTreeBuilderSmokeTest):
-    """Smoke test for a tree builder that supports HTML5."""
-
-    def test_real_xhtml_document(self):
-        # Since XHTML is not HTML5, HTML5 parsers are not tested to handle
-        # XHTML documents in any particular way.
-        pass
-
-    def test_html_tags_have_namespace(self):
-        markup = ""
-        soup = self.soup(markup)
-        self.assertEqual("http://www.w3.org/1999/xhtml", soup.a.namespace)
-
-    def test_svg_tags_have_namespace(self):
-        markup = ''
-        soup = self.soup(markup)
-        namespace = "http://www.w3.org/2000/svg"
-        self.assertEqual(namespace, soup.svg.namespace)
-        self.assertEqual(namespace, soup.circle.namespace)
-
-
-    def test_mathml_tags_have_namespace(self):
-        markup = '5'
-        soup = self.soup(markup)
-        namespace = 'http://www.w3.org/1998/Math/MathML'
-        self.assertEqual(namespace, soup.math.namespace)
-        self.assertEqual(namespace, soup.msqrt.namespace)
-
-    def test_xml_declaration_becomes_comment(self):
-        markup = ''
-        soup = self.soup(markup)
-        self.assertTrue(isinstance(soup.contents[0], Comment))
-        self.assertEqual(soup.contents[0], '?xml version="1.0" encoding="utf-8"?')
-        self.assertEqual("html", soup.contents[0].next_element.name)
-
-def skipIf(condition, reason):
-   def nothing(test, *args, **kwargs):
-       return None
-
-   def decorator(test_item):
-       if condition:
-           return nothing
-       else:
-           return test_item
-
-   return decorator
diff --git a/lib/bs4/tests/__init__.py b/lib/bs4/tests/__init__.py
deleted file mode 100644
index 142c8cc3f1..0000000000
--- a/lib/bs4/tests/__init__.py
+++ /dev/null
@@ -1 +0,0 @@
-"The beautifulsoup tests."
diff --git a/lib/bs4/tests/test_builder_registry.py b/lib/bs4/tests/test_builder_registry.py
deleted file mode 100644
index 90cad82933..0000000000
--- a/lib/bs4/tests/test_builder_registry.py
+++ /dev/null
@@ -1,147 +0,0 @@
-"""Tests of the builder registry."""
-
-import unittest
-import warnings
-
-from bs4 import BeautifulSoup
-from bs4.builder import (
-    builder_registry as registry,
-    HTMLParserTreeBuilder,
-    TreeBuilderRegistry,
-)
-
-try:
-    from bs4.builder import HTML5TreeBuilder
-    HTML5LIB_PRESENT = True
-except ImportError:
-    HTML5LIB_PRESENT = False
-
-try:
-    from bs4.builder import (
-        LXMLTreeBuilderForXML,
-        LXMLTreeBuilder,
-        )
-    LXML_PRESENT = True
-except ImportError:
-    LXML_PRESENT = False
-
-
-class BuiltInRegistryTest(unittest.TestCase):
-    """Test the built-in registry with the default builders registered."""
-
-    def test_combination(self):
-        if LXML_PRESENT:
-            self.assertEqual(registry.lookup('fast', 'html'),
-                             LXMLTreeBuilder)
-
-        if LXML_PRESENT:
-            self.assertEqual(registry.lookup('permissive', 'xml'),
-                             LXMLTreeBuilderForXML)
-        self.assertEqual(registry.lookup('strict', 'html'),
-                          HTMLParserTreeBuilder)
-        if HTML5LIB_PRESENT:
-            self.assertEqual(registry.lookup('html5lib', 'html'),
-                              HTML5TreeBuilder)
-
-    def test_lookup_by_markup_type(self):
-        if LXML_PRESENT:
-            self.assertEqual(registry.lookup('html'), LXMLTreeBuilder)
-            self.assertEqual(registry.lookup('xml'), LXMLTreeBuilderForXML)
-        else:
-            self.assertEqual(registry.lookup('xml'), None)
-            if HTML5LIB_PRESENT:
-                self.assertEqual(registry.lookup('html'), HTML5TreeBuilder)
-            else:
-                self.assertEqual(registry.lookup('html'), HTMLParserTreeBuilder)
-
-    def test_named_library(self):
-        if LXML_PRESENT:
-            self.assertEqual(registry.lookup('lxml', 'xml'),
-                             LXMLTreeBuilderForXML)
-            self.assertEqual(registry.lookup('lxml', 'html'),
-                             LXMLTreeBuilder)
-        if HTML5LIB_PRESENT:
-            self.assertEqual(registry.lookup('html5lib'),
-                              HTML5TreeBuilder)
-
-        self.assertEqual(registry.lookup('html.parser'),
-                          HTMLParserTreeBuilder)
-
-    def test_beautifulsoup_constructor_does_lookup(self):
-
-        with warnings.catch_warnings(record=True) as w:
-            # This will create a warning about not explicitly
-            # specifying a parser, but we'll ignore it.
-
-            # You can pass in a string.
-            BeautifulSoup("", features="html")
-            # Or a list of strings.
-            BeautifulSoup("", features=["html", "fast"])
-
-        # You'll get an exception if BS can't find an appropriate
-        # builder.
-        self.assertRaises(ValueError, BeautifulSoup,
-                          "", features="no-such-feature")
-
-class RegistryTest(unittest.TestCase):
-    """Test the TreeBuilderRegistry class in general."""
-
-    def setUp(self):
-        self.registry = TreeBuilderRegistry()
-
-    def builder_for_features(self, *feature_list):
-        cls = type('Builder_' + '_'.join(feature_list),
-                   (object,), {'features' : feature_list})
-
-        self.registry.register(cls)
-        return cls
-
-    def test_register_with_no_features(self):
-        builder = self.builder_for_features()
-
-        # Since the builder advertises no features, you can't find it
-        # by looking up features.
-        self.assertEqual(self.registry.lookup('foo'), None)
-
-        # But you can find it by doing a lookup with no features, if
-        # this happens to be the only registered builder.
-        self.assertEqual(self.registry.lookup(), builder)
-
-    def test_register_with_features_makes_lookup_succeed(self):
-        builder = self.builder_for_features('foo', 'bar')
-        self.assertEqual(self.registry.lookup('foo'), builder)
-        self.assertEqual(self.registry.lookup('bar'), builder)
-
-    def test_lookup_fails_when_no_builder_implements_feature(self):
-        builder = self.builder_for_features('foo', 'bar')
-        self.assertEqual(self.registry.lookup('baz'), None)
-
-    def test_lookup_gets_most_recent_registration_when_no_feature_specified(self):
-        builder1 = self.builder_for_features('foo')
-        builder2 = self.builder_for_features('bar')
-        self.assertEqual(self.registry.lookup(), builder2)
-
-    def test_lookup_fails_when_no_tree_builders_registered(self):
-        self.assertEqual(self.registry.lookup(), None)
-
-    def test_lookup_gets_most_recent_builder_supporting_all_features(self):
-        has_one = self.builder_for_features('foo')
-        has_the_other = self.builder_for_features('bar')
-        has_both_early = self.builder_for_features('foo', 'bar', 'baz')
-        has_both_late = self.builder_for_features('foo', 'bar', 'quux')
-        lacks_one = self.builder_for_features('bar')
-        has_the_other = self.builder_for_features('foo')
-
-        # There are two builders featuring 'foo' and 'bar', but
-        # the one that also features 'quux' was registered later.
-        self.assertEqual(self.registry.lookup('foo', 'bar'),
-                          has_both_late)
-
-        # There is only one builder featuring 'foo', 'bar', and 'baz'.
-        self.assertEqual(self.registry.lookup('foo', 'bar', 'baz'),
-                          has_both_early)
-
-    def test_lookup_fails_when_cannot_reconcile_requested_features(self):
-        builder1 = self.builder_for_features('foo', 'bar')
-        builder2 = self.builder_for_features('foo', 'baz')
-        self.assertEqual(self.registry.lookup('bar', 'baz'), None)
diff --git a/lib/bs4/tests/test_docs.py b/lib/bs4/tests/test_docs.py
deleted file mode 100644
index d1d76a33bf..0000000000
--- a/lib/bs4/tests/test_docs.py
+++ /dev/null
@@ -1,32 +0,0 @@
-"Test harness for doctests."
-
-# pylint: disable-msg=E0611,W0142
-
-__metaclass__ = type
-__all__ = [
-    'additional_tests',
-    ]
-
-import doctest
-#from pkg_resources import (
-#    resource_filename, resource_exists, resource_listdir, cleanup_resources)
-
-DOCTEST_FLAGS = (
-    doctest.ELLIPSIS |
-    doctest.NORMALIZE_WHITESPACE |
-    doctest.REPORT_NDIFF)
-
-# def additional_tests():
-#     "Run the doc tests (README.txt and docs/*, if any exist)"
-#     doctest_files = [
-#         os.path.abspath(resource_filename('bs4', 'README.txt'))]
-#     if resource_exists('bs4', 'docs'):
-#         for name in resource_listdir('bs4', 'docs'):
-#             if name.endswith('.txt'):
-#                 doctest_files.append(
-#                     os.path.abspath(
-#                         resource_filename('bs4', 'docs/%s' % name)))
-#     kwargs = dict(module_relative=False, optionflags=DOCTEST_FLAGS)
-#     atexit.register(cleanup_resources)
-#     return unittest.TestSuite((
-#         doctest.DocFileSuite(*doctest_files, **kwargs)))
diff --git a/lib/bs4/tests/test_html5lib.py b/lib/bs4/tests/test_html5lib.py
deleted file mode 100644
index a7494ca5ba..0000000000
--- a/lib/bs4/tests/test_html5lib.py
+++ /dev/null
@@ -1,98 +0,0 @@
-"""Tests to ensure that the html5lib tree builder generates good trees."""
-
-import warnings
-
-try:
-    from bs4.builder import HTML5TreeBuilder
-    HTML5LIB_PRESENT = True
-except ImportError as e:
-    HTML5LIB_PRESENT = False
-from bs4.element import SoupStrainer
-from bs4.testing import (
-    HTML5TreeBuilderSmokeTest,
-    SoupTest,
-    skipIf,
-)
-
-@skipIf(
-    not HTML5LIB_PRESENT,
-    "html5lib seems not to be present, not testing its tree builder.")
-class HTML5LibBuilderSmokeTest(SoupTest, HTML5TreeBuilderSmokeTest):
-    """See ``HTML5TreeBuilderSmokeTest``."""
-
-    @property
-    def default_builder(self):
-        return HTML5TreeBuilder()
-
-    def test_soupstrainer(self):
-        # The html5lib tree builder does not support SoupStrainers.
-        strainer = SoupStrainer("b")
-        markup = "
A bold statement.
"
-        with warnings.catch_warnings(record=True) as w:
-            soup = self.soup(markup, parse_only=strainer)
-        self.assertEqual(
-            soup.decode(), self.document_for(markup))
-
-        self.assertTrue(
-            "the html5lib tree builder doesn't support parse_only" in
-            str(w[0].message))
-
-    def test_correctly_nested_tables(self):
-        """html5lib inserts  tags where other parsers don't."""
-        markup = (''
-                  ''
-                  "')
-
-        self.assertSoupEquals(
-            markup,
-            '
Here's another table:"
-                  ''
-                  ''
-                  '
foo
Here\'s another table:'
-            '
foo
'
-            '
')
-
-        self.assertSoupEquals(
-            ""
-            ""
-            "
Foo
Bar
Baz
")
-
-    def test_xml_declaration_followed_by_doctype(self):
-        markup = '''
-
-
-  
-  
-  
-   
foo

-  
-'''
-        soup = self.soup(markup)
-        # Verify that we can reach the 
 tag; this means the tree is connected.
-        self.assertEqual(b"
foo
", soup.p.encode())
-
-    def test_reparented_markup(self):
-        markup = '
foo
\n
bar
'
-        soup = self.soup(markup)
-        self.assertEqual("
foo
\n
bar
", soup.body.decode())
-        self.assertEqual(2, len(soup.find_all('p')))
-
-
-    def test_reparented_markup_ends_with_whitespace(self):
-        markup = '
foo
\n
bar
\n'
-        soup = self.soup(markup)
-        self.assertEqual("
foo
\n
bar
\n", soup.body.decode())
-        self.assertEqual(2, len(soup.find_all('p')))
-
-    def test_processing_instruction(self):
-        """Processing instructions become comments."""
-        markup = b""""""
-        soup = self.soup(markup)
-        assert str(soup).startswith("")
-
-    def test_cloned_multivalue_node(self):
-        markup = b"""
"""
-        soup = self.soup(markup)
-        a1, a2 = soup.find_all('a')
-        self.assertEqual(a1, a2)
-        assert a1 is not a2
diff --git a/lib/bs4/tests/test_htmlparser.py b/lib/bs4/tests/test_htmlparser.py
deleted file mode 100644
index 30a25e6709..0000000000
--- a/lib/bs4/tests/test_htmlparser.py
+++ /dev/null
@@ -1,31 +0,0 @@
-"""Tests to ensure that the html.parser tree builder generates good
-trees."""
-
-import pickle
-from bs4.testing import SoupTest, HTMLTreeBuilderSmokeTest
-from bs4.builder import HTMLParserTreeBuilder
-
-class HTMLParserTreeBuilderSmokeTest(SoupTest, HTMLTreeBuilderSmokeTest):
-
-    @property
-    def default_builder(self):
-        return HTMLParserTreeBuilder()
-
-    def test_namespaced_system_doctype(self):
-        # html.parser can't handle namespaced doctypes, so skip this one.
-        pass
-
-    def test_namespaced_public_doctype(self):
-        # html.parser can't handle namespaced doctypes, so skip this one.
-        pass
-
-    def test_builder_is_pickled(self):
-        """Unlike most tree builders, HTMLParserTreeBuilder and will
-        be restored after pickling.
-        """
-        tree = self.soup("foo")
-        dumped = pickle.dumps(tree, 2)
-        loaded = pickle.loads(dumped)
-        self.assertTrue(isinstance(loaded.builder, type(tree.builder)))
-
-
diff --git a/lib/bs4/tests/test_lxml.py b/lib/bs4/tests/test_lxml.py
deleted file mode 100644
index 6b6cdd07cb..0000000000
--- a/lib/bs4/tests/test_lxml.py
+++ /dev/null
@@ -1,70 +0,0 @@
-"""Tests to ensure that the lxml tree builder generates good trees."""
-
-import warnings
-
-try:
-    import lxml.etree
-    LXML_PRESENT = True
-    LXML_VERSION = lxml.etree.LXML_VERSION
-except ImportError as e:
-    LXML_PRESENT = False
-    LXML_VERSION = (0,)
-
-if LXML_PRESENT:
-    from bs4.builder import LXMLTreeBuilder, LXMLTreeBuilderForXML
-
-from bs4 import BeautifulStoneSoup
-from bs4.testing import skipIf
-from bs4.testing import (
-    HTMLTreeBuilderSmokeTest,
-    XMLTreeBuilderSmokeTest,
-    SoupTest,
-    skipIf,
-)
-
-@skipIf(
-    not LXML_PRESENT,
-    "lxml seems not to be present, not testing its tree builder.")
-class LXMLTreeBuilderSmokeTest(SoupTest, HTMLTreeBuilderSmokeTest):
-    """See ``HTMLTreeBuilderSmokeTest``."""
-
-    @property
-    def default_builder(self):
-        return LXMLTreeBuilder()
-
-    def test_out_of_range_entity(self):
-        self.assertSoupEquals(
-            "
foo�bar
", "
foobar
")
-        self.assertSoupEquals(
-            "
foo�bar
", "
foobar
")
-        self.assertSoupEquals(
-            "
foo�bar
", "
foobar
")
-
-    # In lxml < 2.3.5, an empty doctype causes a segfault. Skip this
-    # test if an old version of lxml is installed.
-
-    @skipIf(
-        not LXML_PRESENT or LXML_VERSION < (2,3,5,0),
-        "Skipping doctype test for old version of lxml to avoid segfault.")
-    def test_empty_doctype(self):
-        soup = self.soup("")
-        doctype = soup.contents[0]
-        self.assertEqual("", doctype.strip())
-
-    def test_beautifulstonesoup_is_xml_parser(self):
-        # Make sure that the deprecated BSS class uses an xml builder
-        # if one is installed.
-        with warnings.catch_warnings(record=True) as w:
-            soup = BeautifulStoneSoup("")
-        self.assertEqual("", str(soup.b))
-        self.assertTrue("BeautifulStoneSoup class is deprecated" in str(w[0].message))
-
-@skipIf(
-    not LXML_PRESENT,
-    "lxml seems not to be present, not testing its XML tree builder.")
-class LXMLXMLTreeBuilderSmokeTest(SoupTest, XMLTreeBuilderSmokeTest):
-    """See ``HTMLTreeBuilderSmokeTest``."""
-
-    @property
-    def default_builder(self):
-        return LXMLTreeBuilderForXML()
diff --git a/lib/bs4/tests/test_soup.py b/lib/bs4/tests/test_soup.py
deleted file mode 100644
index 6ad3cb3765..0000000000
--- a/lib/bs4/tests/test_soup.py
+++ /dev/null
@@ -1,479 +0,0 @@
-# -*- coding: utf-8 -*-
-"""Tests of Beautiful Soup as a whole."""
-
-import logging
-import unittest
-import sys
-import tempfile
-
-from bs4 import BeautifulSoup
-from bs4.element import (
-    CharsetMetaAttributeValue,
-    ContentMetaAttributeValue,
-    SoupStrainer,
-    NamespacedAttribute,
-    )
-import bs4.dammit
-from bs4.dammit import (
-    EntitySubstitution,
-    UnicodeDammit,
-    EncodingDetector,
-)
-from bs4.testing import (
-    SoupTest,
-    skipIf,
-)
-import warnings
-
-try:
-    from bs4.builder import LXMLTreeBuilder, LXMLTreeBuilderForXML
-    LXML_PRESENT = True
-except ImportError as e:
-    LXML_PRESENT = False
-
-PYTHON_2_PRE_2_7 = (sys.version_info < (2,7))
-PYTHON_3_PRE_3_2 = (sys.version_info[0] == 3 and sys.version_info < (3,2))
-
-class TestConstructor(SoupTest):
-
-    def test_short_unicode_input(self):
-        data = "
éé
"
-        soup = self.soup(data)
-        self.assertEqual("éé", soup.h1.string)
-
-    def test_embedded_null(self):
-        data = "
foo\0bar
"
-        soup = self.soup(data)
-        self.assertEqual("foo\0bar", soup.h1.string)
-
-    def test_exclude_encodings(self):
-        utf8_data = "Räksmörgås".encode("utf-8")
-        soup = self.soup(utf8_data, exclude_encodings=["utf-8"])
-        self.assertEqual("windows-1252", soup.original_encoding)
-
-
-class TestWarnings(SoupTest):
-
-    def _no_parser_specified(self, s, is_there=True):
-        v = s.startswith(BeautifulSoup.NO_PARSER_SPECIFIED_WARNING[:80])
-        self.assertTrue(v)
-
-    def test_warning_if_no_parser_specified(self):
-        with warnings.catch_warnings(record=True) as w:
-            soup = self.soup("")
-        msg = str(w[0].message)
-        self._assert_no_parser_specified(msg)
-
-    def test_warning_if_parser_specified_too_vague(self):
-        with warnings.catch_warnings(record=True) as w:
-            soup = self.soup("", "html")
-        msg = str(w[0].message)
-        self._assert_no_parser_specified(msg)
-
-    def test_no_warning_if_explicit_parser_specified(self):
-        with warnings.catch_warnings(record=True) as w:
-            soup = self.soup("", "html.parser")
-        self.assertEqual([], w)
-
-    def test_parseOnlyThese_renamed_to_parse_only(self):
-        with warnings.catch_warnings(record=True) as w:
-            soup = self.soup("", parseOnlyThese=SoupStrainer("b"))
-        msg = str(w[0].message)
-        self.assertTrue("parseOnlyThese" in msg)
-        self.assertTrue("parse_only" in msg)
-        self.assertEqual(b"", soup.encode())
-
-    def test_fromEncoding_renamed_to_from_encoding(self):
-        with warnings.catch_warnings(record=True) as w:
-            utf8 = b"\xc3\xa9"
-            soup = self.soup(utf8, fromEncoding="utf8")
-        msg = str(w[0].message)
-        self.assertTrue("fromEncoding" in msg)
-        self.assertTrue("from_encoding" in msg)
-        self.assertEqual("utf8", soup.original_encoding)
-
-    def test_unrecognized_keyword_argument(self):
-        self.assertRaises(
-            TypeError, self.soup, "", no_such_argument=True)
-
-class TestWarnings(SoupTest):
-
-    def test_disk_file_warning(self):
-        filehandle = tempfile.NamedTemporaryFile()
-        filename = filehandle.name
-        try:
-            with warnings.catch_warnings(record=True) as w:
-                soup = self.soup(filename)
-            msg = str(w[0].message)
-            self.assertTrue("looks like a filename" in msg)
-        finally:
-            filehandle.close()
-
-        # The file no longer exists, so Beautiful Soup will no longer issue the warning.
-        with warnings.catch_warnings(record=True) as w:
-            soup = self.soup(filename)
-        self.assertEqual(0, len(w))
-
-    def test_url_warning(self):
-        with warnings.catch_warnings(record=True) as w:
-            soup = self.soup("http://www.crummy.com/")
-        msg = str(w[0].message)
-        self.assertTrue("looks like a URL" in msg)
-
-        with warnings.catch_warnings(record=True) as w:
-            soup = self.soup("http://www.crummy.com/ is great")
-        self.assertEqual(0, len(w))
-
-class TestSelectiveParsing(SoupTest):
-
-    def test_parse_with_soupstrainer(self):
-        markup = "NoYesNoYes Yes"
-        strainer = SoupStrainer("b")
-        soup = self.soup(markup, parse_only=strainer)
-        self.assertEqual(soup.encode(), b"YesYes Yes")
-
-
-class TestEntitySubstitution(unittest.TestCase):
-    """Standalone tests of the EntitySubstitution class."""
-    def setUp(self):
-        self.sub = EntitySubstitution
-
-    def test_simple_html_substitution(self):
-        # Unicode characters corresponding to named HTML entites
-        # are substituted, and no others.
-        s = "foo\u2200\N{SNOWMAN}\u00f5bar"
-        self.assertEqual(self.sub.substitute_html(s),
-                          "foo∀\N{SNOWMAN}õbar")
-
-    def test_smart_quote_substitution(self):
-        # MS smart quotes are a common source of frustration, so we
-        # give them a special test.
-        quotes = b"\x91\x92foo\x93\x94"
-        dammit = UnicodeDammit(quotes)
-        self.assertEqual(self.sub.substitute_html(dammit.markup),
-                          "‘’foo“”")
-
-    def test_xml_converstion_includes_no_quotes_if_make_quoted_attribute_is_false(self):
-        s = 'Welcome to "my bar"'
-        self.assertEqual(self.sub.substitute_xml(s, False), s)
-
-    def test_xml_attribute_quoting_normally_uses_double_quotes(self):
-        self.assertEqual(self.sub.substitute_xml("Welcome", True),
-                          '"Welcome"')
-        self.assertEqual(self.sub.substitute_xml("Bob's Bar", True),
-                          '"Bob\'s Bar"')
-
-    def test_xml_attribute_quoting_uses_single_quotes_when_value_contains_double_quotes(self):
-        s = 'Welcome to "my bar"'
-        self.assertEqual(self.sub.substitute_xml(s, True),
-                          "'Welcome to \"my bar\"'")
-
-    def test_xml_attribute_quoting_escapes_single_quotes_when_value_contains_both_single_and_double_quotes(self):
-        s = 'Welcome to "Bob\'s Bar"'
-        self.assertEqual(
-            self.sub.substitute_xml(s, True),
-            '"Welcome to "Bob\'s Bar""')
-
-    def test_xml_quotes_arent_escaped_when_value_is_not_being_quoted(self):
-        quoted = 'Welcome to "Bob\'s Bar"'
-        self.assertEqual(self.sub.substitute_xml(quoted), quoted)
-
-    def test_xml_quoting_handles_angle_brackets(self):
-        self.assertEqual(
-            self.sub.substitute_xml("foo"),
-            "foo<bar>")
-
-    def test_xml_quoting_handles_ampersands(self):
-        self.assertEqual(self.sub.substitute_xml("AT&T"), "AT&T")
-
-    def test_xml_quoting_including_ampersands_when_they_are_part_of_an_entity(self):
-        self.assertEqual(
-            self.sub.substitute_xml("ÁT&T"),
-            "&Aacute;T&T")
-
-    def test_xml_quoting_ignoring_ampersands_when_they_are_part_of_an_entity(self):
-        self.assertEqual(
-            self.sub.substitute_xml_containing_entities("ÁT&T"),
-            "ÁT&T")
-
-    def test_quotes_not_html_substituted(self):
-        """There's no need to do this except inside attribute values."""
-        text = 'Bob\'s "bar"'
-        self.assertEqual(self.sub.substitute_html(text), text)
-
-
-class TestEncodingConversion(SoupTest):
-    # Test Beautiful Soup's ability to decode and encode from various
-    # encodings.
-
-    def setUp(self):
-        super(TestEncodingConversion, self).setUp()
-        self.unicode_data = 'Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!'
-        self.utf8_data = self.unicode_data.encode("utf-8")
-        # Just so you know what it looks like.
-        self.assertEqual(
-            self.utf8_data,
-            b'Sacr\xc3\xa9 bleu!')
-
-    def test_ascii_in_unicode_out(self):
-        # ASCII input is converted to Unicode. The original_encoding
-        # attribute is set to 'utf-8', a superset of ASCII.
-        chardet = bs4.dammit.chardet_dammit
-        logging.disable(logging.WARNING)
-        try:
-            def noop(str):
-                return None
-            # Disable chardet, which will realize that the ASCII is ASCII.
-            bs4.dammit.chardet_dammit = noop
-            ascii = b"a"
-            soup_from_ascii = self.soup(ascii)
-            unicode_output = soup_from_ascii.decode()
-            self.assertTrue(isinstance(unicode_output, str))
-            self.assertEqual(unicode_output, self.document_for(ascii.decode()))
-            self.assertEqual(soup_from_ascii.original_encoding.lower(), "utf-8")
-        finally:
-            logging.disable(logging.NOTSET)
-            bs4.dammit.chardet_dammit = chardet
-
-    def test_unicode_in_unicode_out(self):
-        # Unicode input is left alone. The original_encoding attribute
-        # is not set.
-        soup_from_unicode = self.soup(self.unicode_data)
-        self.assertEqual(soup_from_unicode.decode(), self.unicode_data)
-        self.assertEqual(soup_from_unicode.foo.string, 'Sacr\xe9 bleu!')
-        self.assertEqual(soup_from_unicode.original_encoding, None)
-
-    def test_utf8_in_unicode_out(self):
-        # UTF-8 input is converted to Unicode. The original_encoding
-        # attribute is set.
-        soup_from_utf8 = self.soup(self.utf8_data)
-        self.assertEqual(soup_from_utf8.decode(), self.unicode_data)
-        self.assertEqual(soup_from_utf8.foo.string, 'Sacr\xe9 bleu!')
-
-    def test_utf8_out(self):
-        # The internal data structures can be encoded as UTF-8.
-        soup_from_unicode = self.soup(self.unicode_data)
-        self.assertEqual(soup_from_unicode.encode('utf-8'), self.utf8_data)
-
-    @skipIf(
-        PYTHON_2_PRE_2_7 or PYTHON_3_PRE_3_2,
-        "Bad HTMLParser detected; skipping test of non-ASCII characters in attribute name.")
-    def test_attribute_name_containing_unicode_characters(self):
-        markup = '
'
-        self.assertEqual(self.soup(markup).div.encode("utf8"), markup.encode("utf8"))
-
-class TestUnicodeDammit(unittest.TestCase):
-    """Standalone tests of UnicodeDammit."""
-
-    def test_unicode_input(self):
-        markup = "I'm already Unicode! \N{SNOWMAN}"
-        dammit = UnicodeDammit(markup)
-        self.assertEqual(dammit.unicode_markup, markup)
-
-    def test_smart_quotes_to_unicode(self):
-        markup = b"\x91\x92\x93\x94"
-        dammit = UnicodeDammit(markup)
-        self.assertEqual(
-            dammit.unicode_markup, "\u2018\u2019\u201c\u201d")
-
-    def test_smart_quotes_to_xml_entities(self):
-        markup = b"\x91\x92\x93\x94"
-        dammit = UnicodeDammit(markup, smart_quotes_to="xml")
-        self.assertEqual(
-            dammit.unicode_markup, "‘’“”")
-
-    def test_smart_quotes_to_html_entities(self):
-        markup = b"\x91\x92\x93\x94"
-        dammit = UnicodeDammit(markup, smart_quotes_to="html")
-        self.assertEqual(
-            dammit.unicode_markup, "‘’“”")
-
-    def test_smart_quotes_to_ascii(self):
-        markup = b"\x91\x92\x93\x94"
-        dammit = UnicodeDammit(markup, smart_quotes_to="ascii")
-        self.assertEqual(
-            dammit.unicode_markup, """''""""")
-
-    def test_detect_utf8(self):
-        utf8 = b"Sacr\xc3\xa9 bleu! \xe2\x98\x83"
-        dammit = UnicodeDammit(utf8)
-        self.assertEqual(dammit.original_encoding.lower(), 'utf-8')
-        self.assertEqual(dammit.unicode_markup, 'Sacr\xe9 bleu! \N{SNOWMAN}')
-
-
-    def test_convert_hebrew(self):
-        hebrew = b"\xed\xe5\xec\xf9"
-        dammit = UnicodeDammit(hebrew, ["iso-8859-8"])
-        self.assertEqual(dammit.original_encoding.lower(), 'iso-8859-8')
-        self.assertEqual(dammit.unicode_markup, '\u05dd\u05d5\u05dc\u05e9')
-
-    def test_dont_see_smart_quotes_where_there_are_none(self):
-        utf_8 = b"\343\202\261\343\203\274\343\202\277\343\202\244 Watch"
-        dammit = UnicodeDammit(utf_8)
-        self.assertEqual(dammit.original_encoding.lower(), 'utf-8')
-        self.assertEqual(dammit.unicode_markup.encode("utf-8"), utf_8)
-
-    def test_ignore_inappropriate_codecs(self):
-        utf8_data = "Räksmörgås".encode("utf-8")
-        dammit = UnicodeDammit(utf8_data, ["iso-8859-8"])
-        self.assertEqual(dammit.original_encoding.lower(), 'utf-8')
-
-    def test_ignore_invalid_codecs(self):
-        utf8_data = "Räksmörgås".encode("utf-8")
-        for bad_encoding in ['.utf8', '...', 'utF---16.!']:
-            dammit = UnicodeDammit(utf8_data, [bad_encoding])
-            self.assertEqual(dammit.original_encoding.lower(), 'utf-8')
-
-    def test_exclude_encodings(self):
-        # This is UTF-8.
-        utf8_data = "Räksmörgås".encode("utf-8")
-
-        # But if we exclude UTF-8 from consideration, the guess is
-        # Windows-1252.
-        dammit = UnicodeDammit(utf8_data, exclude_encodings=["utf-8"])
-        self.assertEqual(dammit.original_encoding.lower(), 'windows-1252')
-
-        # And if we exclude that, there is no valid guess at all.
-        dammit = UnicodeDammit(
-            utf8_data, exclude_encodings=["utf-8", "windows-1252"])
-        self.assertEqual(dammit.original_encoding, None)
-
-    def test_encoding_detector_replaces_junk_in_encoding_name_with_replacement_character(self):
-        detected = EncodingDetector(
-            b'')
-        encodings = list(detected.encodings)
-        assert 'utf-\N{REPLACEMENT CHARACTER}' in encodings
-
-    def test_detect_html5_style_meta_tag(self):
-
-        for data in (
-            b'',
-            b"",
-            b"",
-            b""):
-            dammit = UnicodeDammit(data, is_html=True)
-            self.assertEqual(
-                "euc-jp", dammit.original_encoding)
-
-    def test_last_ditch_entity_replacement(self):
-        # This is a UTF-8 document that contains bytestrings
-        # completely incompatible with UTF-8 (ie. encoded with some other
-        # encoding).
-        #
-        # Since there is no consistent encoding for the document,
-        # Unicode, Dammit will eventually encode the document as UTF-8
-        # and encode the incompatible characters as REPLACEMENT
-        # CHARACTER.
-        #
-        # If chardet is installed, it will detect that the document
-        # can be converted into ISO-8859-1 without errors. This happens
-        # to be the wrong encoding, but it is a consistent encoding, so the
-        # code we're testing here won't run.
-        #
-        # So we temporarily disable chardet if it's present.
-        doc = b"""\357\273\277
-\330\250\330\252\330\261
-\310\322\321\220\312\321\355\344"""
-        chardet = bs4.dammit.chardet_dammit
-        logging.disable(logging.WARNING)
-        try:
-            def noop(str):
-                return None
-            bs4.dammit.chardet_dammit = noop
-            dammit = UnicodeDammit(doc)
-            self.assertEqual(True, dammit.contains_replacement_characters)
-            self.assertTrue("\ufffd" in dammit.unicode_markup)
-
-            soup = BeautifulSoup(doc, "html.parser")
-            self.assertTrue(soup.contains_replacement_characters)
-        finally:
-            logging.disable(logging.NOTSET)
-            bs4.dammit.chardet_dammit = chardet
-
-    def test_byte_order_mark_removed(self):
-        # A document written in UTF-16LE will have its byte order marker stripped.
-        data = b'\xff\xfe<\x00a\x00>\x00\xe1\x00\xe9\x00<\x00/\x00a\x00>\x00'
-        dammit = UnicodeDammit(data)
-        self.assertEqual("áé", dammit.unicode_markup)
-        self.assertEqual("utf-16le", dammit.original_encoding)
-
-    def test_detwingle(self):
-        # Here's a UTF8 document.
-        utf8 = ("\N{SNOWMAN}" * 3).encode("utf8")
-
-        # Here's a Windows-1252 document.
-        windows_1252 = (
-            "\N{LEFT DOUBLE QUOTATION MARK}Hi, I like Windows!"
-            "\N{RIGHT DOUBLE QUOTATION MARK}").encode("windows_1252")
-
-        # Through some unholy alchemy, they've been stuck together.
-        doc = utf8 + windows_1252 + utf8
-
-        # The document can't be turned into UTF-8:
-        self.assertRaises(UnicodeDecodeError, doc.decode, "utf8")
-
-        # Unicode, Dammit thinks the whole document is Windows-1252,
-        # and decodes it into "☃☃☃“Hi, I like Windows!”☃☃☃"
-
-        # But if we run it through fix_embedded_windows_1252, it's fixed:
-
-        fixed = UnicodeDammit.detwingle(doc)
-        self.assertEqual(
-            "☃☃☃“Hi, I like Windows!”☃☃☃", fixed.decode("utf8"))
-
-    def test_detwingle_ignores_multibyte_characters(self):
-        # Each of these characters has a UTF-8 representation ending
-        # in \x93. \x93 is a smart quote if interpreted as
-        # Windows-1252. But our code knows to skip over multibyte
-        # UTF-8 characters, so they'll survive the process unscathed.
-        for tricky_unicode_char in (
-            "\N{LATIN SMALL LIGATURE OE}", # 2-byte char '\xc5\x93'
-            "\N{LATIN SUBSCRIPT SMALL LETTER X}", # 3-byte char '\xe2\x82\x93'
-            "\xf0\x90\x90\x93", # This is a CJK character, not sure which one.
-            ):
-            input = tricky_unicode_char.encode("utf8")
-            self.assertTrue(input.endswith(b'\x93'))
-            output = UnicodeDammit.detwingle(input)
-            self.assertEqual(output, input)
-
-class TestNamedspacedAttribute(SoupTest):
-
-    def test_name_may_be_none(self):
-        a = NamespacedAttribute("xmlns", None)
-        self.assertEqual(a, "xmlns")
-
-    def test_attribute_is_equivalent_to_colon_separated_string(self):
-        a = NamespacedAttribute("a", "b")
-        self.assertEqual("a:b", a)
-
-    def test_attributes_are_equivalent_if_prefix_and_name_identical(self):
-        a = NamespacedAttribute("a", "b", "c")
-        b = NamespacedAttribute("a", "b", "c")
-        self.assertEqual(a, b)
-
-        # The actual namespace is not considered.
-        c = NamespacedAttribute("a", "b", None)
-        self.assertEqual(a, c)
-
-        # But name and prefix are important.
-        d = NamespacedAttribute("a", "z", "c")
-        self.assertNotEqual(a, d)
-
-        e = NamespacedAttribute("z", "b", "c")
-        self.assertNotEqual(a, e)
-
-
-class TestAttributeValueWithCharsetSubstitution(unittest.TestCase):
-
-    def test_content_meta_attribute_value(self):
-        value = CharsetMetaAttributeValue("euc-jp")
-        self.assertEqual("euc-jp", value)
-        self.assertEqual("euc-jp", value.original_value)
-        self.assertEqual("utf8", value.encode("utf8"))
-
-
-    def test_content_meta_attribute_value(self):
-        value = ContentMetaAttributeValue("text/html; charset=euc-jp")
-        self.assertEqual("text/html; charset=euc-jp", value)
-        self.assertEqual("text/html; charset=euc-jp", value.original_value)
-        self.assertEqual("text/html; charset=utf8", value.encode("utf8"))
diff --git a/lib/bs4/tests/test_tree.py b/lib/bs4/tests/test_tree.py
deleted file mode 100644
index cf0f1abe0c..0000000000
--- a/lib/bs4/tests/test_tree.py
+++ /dev/null
@@ -1,2004 +0,0 @@
-# -*- coding: utf-8 -*-
-"""Tests for Beautiful Soup's tree traversal methods.
-
-The tree traversal methods are the main advantage of using Beautiful
-Soup over just using a parser.
-
-Different parsers will build different Beautiful Soup trees given the
-same markup, but all Beautiful Soup trees can be traversed with the
-methods tested here.
-"""
-
-import copy
-import pickle
-import re
-import warnings
-from bs4 import BeautifulSoup
-from bs4.builder import builder_registry
-from bs4.element import (
-    PY3K,
-    CData,
-    Comment,
-    Declaration,
-    Doctype,
-    NavigableString,
-    SoupStrainer,
-    Tag,
-)
-from bs4.testing import SoupTest
-
-XML_BUILDER_PRESENT = (builder_registry.lookup("xml") is not None)
-LXML_PRESENT = (builder_registry.lookup("lxml") is not None)
-
-class TreeTest(SoupTest):
-
-    def assertSelects(self, tags, should_match):
-        """Make sure that the given tags have the correct text.
-
-        This is used in tests that define a bunch of tags, each
-        containing a single string, and then select certain strings by
-        some mechanism.
-        """
-        self.assertEqual([tag.string for tag in tags], should_match)
-
-    def assertSelectsIDs(self, tags, should_match):
-        """Make sure that the given tags have the correct IDs.
-
-        This is used in tests that define a bunch of tags, each
-        containing a single string, and then select certain strings by
-        some mechanism.
-        """
-        self.assertEqual([tag['id'] for tag in tags], should_match)
-
-
-class TestFind(TreeTest):
-    """Basic tests of the find() method.
-
-    find() just calls find_all() with limit=1, so it's not tested all
-    that thouroughly here.
-    """
-
-    def test_find_tag(self):
-        soup = self.soup("1234")
-        self.assertEqual(soup.find("b").string, "2")
-
-    def test_unicode_text_find(self):
-        soup = self.soup('
Räksmörgås
')
-        self.assertEqual(soup.find(string='Räksmörgås'), 'Räksmörgås')
-
-    def test_unicode_attribute_find(self):
-        soup = self.soup('
here it is
')
-        str(soup)
-        self.assertEqual("here it is", soup.find(id='Räksmörgås').text)
-
-
-    def test_find_everything(self):
-        """Test an optimization that finds all tags."""
-        soup = self.soup("foobar")
-        self.assertEqual(2, len(soup.find_all()))
-
-    def test_find_everything_with_name(self):
-        """Test an optimization that finds all tags with a given name."""
-        soup = self.soup("foobarbaz")
-        self.assertEqual(2, len(soup.find_all('a')))
-
-class TestFindAll(TreeTest):
-    """Basic tests of the find_all() method."""
-
-    def test_find_all_text_nodes(self):
-        """You can search the tree for text nodes."""
-        soup = self.soup("Foobar\xbb")
-        # Exact match.
-        self.assertEqual(soup.find_all(string="bar"), ["bar"])
-        self.assertEqual(soup.find_all(text="bar"), ["bar"])
-        # Match any of a number of strings.
-        self.assertEqual(
-            soup.find_all(text=["Foo", "bar"]), ["Foo", "bar"])
-        # Match a regular expression.
-        self.assertEqual(soup.find_all(text=re.compile('.*')),
-                         ["Foo", "bar", '\xbb'])
-        # Match anything.
-        self.assertEqual(soup.find_all(text=True),
-                         ["Foo", "bar", '\xbb'])
-
-    def test_find_all_limit(self):
-        """You can limit the number of items returned by find_all."""
-        soup = self.soup("12345")
-        self.assertSelects(soup.find_all('a', limit=3), ["1", "2", "3"])
-        self.assertSelects(soup.find_all('a', limit=1), ["1"])
-        self.assertSelects(
-            soup.find_all('a', limit=10), ["1", "2", "3", "4", "5"])
-
-        # A limit of 0 means no limit.
-        self.assertSelects(
-            soup.find_all('a', limit=0), ["1", "2", "3", "4", "5"])
-
-    def test_calling_a_tag_is_calling_findall(self):
-        soup = self.soup("123")
-        self.assertSelects(soup('a', limit=1), ["1"])
-        self.assertSelects(soup.b(id="foo"), ["3"])
-
-    def test_find_all_with_self_referential_data_structure_does_not_cause_infinite_recursion(self):
-        soup = self.soup("")
-        # Create a self-referential list.
-        l = []
-        l.append(l)
-
-        # Without special code in _normalize_search_value, this would cause infinite
-        # recursion.
-        self.assertEqual([], soup.find_all(l))
-
-    def test_find_all_resultset(self):
-        """All find_all calls return a ResultSet"""
-        soup = self.soup("")
-        result = soup.find_all("a")
-        self.assertTrue(hasattr(result, "source"))
-
-        result = soup.find_all(True)
-        self.assertTrue(hasattr(result, "source"))
-
-        result = soup.find_all(text="foo")
-        self.assertTrue(hasattr(result, "source"))
-
-
-class TestFindAllBasicNamespaces(TreeTest):
-
-    def test_find_by_namespaced_name(self):
-        soup = self.soup('4')
-        self.assertEqual("4", soup.find("mathml:msqrt").string)
-        self.assertEqual("a", soup.find(attrs= { "svg:fill" : "red" }).name)
-
-
-class TestFindAllByName(TreeTest):
-    """Test ways of finding tags by tag name."""
-
-    def setUp(self):
-        super(TreeTest, self).setUp()
-        self.tree =  self.soup("""First tag.
-                                  Second tag.
-                                  Third Nested tag. tag.""")
-
-    def test_find_all_by_tag_name(self):
-        # Find all the  tags.
-        self.assertSelects(
-            self.tree.find_all('a'), ['First tag.', 'Nested tag.'])
-
-    def test_find_all_by_name_and_text(self):
-        self.assertSelects(
-            self.tree.find_all('a', text='First tag.'), ['First tag.'])
-
-        self.assertSelects(
-            self.tree.find_all('a', text=True), ['First tag.', 'Nested tag.'])
-
-        self.assertSelects(
-            self.tree.find_all('a', text=re.compile("tag")),
-            ['First tag.', 'Nested tag.'])
-
-
-    def test_find_all_on_non_root_element(self):
-        # You can call find_all on any node, not just the root.
-        self.assertSelects(self.tree.c.find_all('a'), ['Nested tag.'])
-
-    def test_calling_element_invokes_find_all(self):
-        self.assertSelects(self.tree('a'), ['First tag.', 'Nested tag.'])
-
-    def test_find_all_by_tag_strainer(self):
-        self.assertSelects(
-            self.tree.find_all(SoupStrainer('a')),
-            ['First tag.', 'Nested tag.'])
-
-    def test_find_all_by_tag_names(self):
-        self.assertSelects(
-            self.tree.find_all(['a', 'b']),
-            ['First tag.', 'Second tag.', 'Nested tag.'])
-
-    def test_find_all_by_tag_dict(self):
-        self.assertSelects(
-            self.tree.find_all({'a' : True, 'b' : True}),
-            ['First tag.', 'Second tag.', 'Nested tag.'])
-
-    def test_find_all_by_tag_re(self):
-        self.assertSelects(
-            self.tree.find_all(re.compile('^[ab]$')),
-            ['First tag.', 'Second tag.', 'Nested tag.'])
-
-    def test_find_all_with_tags_matching_method(self):
-        # You can define an oracle method that determines whether
-        # a tag matches the search.
-        def id_matches_name(tag):
-            return tag.name == tag.get('id')
-
-        tree = self.soup("""Match 1.
-                            Does not match.
-                            Match 2.""")
-
-        self.assertSelects(
-            tree.find_all(id_matches_name), ["Match 1.", "Match 2."])
-
-
-class TestFindAllByAttribute(TreeTest):
-
-    def test_find_all_by_attribute_name(self):
-        # You can pass in keyword arguments to find_all to search by
-        # attribute.
-        tree = self.soup("""
-                         Matching a.
-                         
-                          Non-matching Matching b.a.
-                         """)
-        self.assertSelects(tree.find_all(id='first'),
-                           ["Matching a.", "Matching b."])
-
-    def test_find_all_by_utf8_attribute_value(self):
-        peace = "םולש".encode("utf8")
-        data = ''.encode("utf8")
-        soup = self.soup(data)
-        self.assertEqual([soup.a], soup.find_all(title=peace))
-        self.assertEqual([soup.a], soup.find_all(title=peace.decode("utf8")))
-        self.assertEqual([soup.a], soup.find_all(title=[peace, "something else"]))
-
-    def test_find_all_by_attribute_dict(self):
-        # You can pass in a dictionary as the argument 'attrs'. This
-        # lets you search for attributes like 'name' (a fixed argument
-        # to find_all) and 'class' (a reserved word in Python.)
-        tree = self.soup("""
-                         Name match.
-                         Class match.
-                         Non-match.
-                         A tag called 'name1'.
-                         """)
-
-        # This doesn't do what you want.
-        self.assertSelects(tree.find_all(name='name1'),
-                           ["A tag called 'name1'."])
-        # This does what you want.
-        self.assertSelects(tree.find_all(attrs={'name' : 'name1'}),
-                           ["Name match."])
-
-        self.assertSelects(tree.find_all(attrs={'class' : 'class2'}),
-                           ["Class match."])
-
-    def test_find_all_by_class(self):
-        tree = self.soup("""
-                         Class 1.
-                         Class 2.
-                         Class 1.
-                         Class 3 and 4.
-                         """)
-
-        # Passing in the class_ keyword argument will search against
-        # the 'class' attribute.
-        self.assertSelects(tree.find_all('a', class_='1'), ['Class 1.'])
-        self.assertSelects(tree.find_all('c', class_='3'), ['Class 3 and 4.'])
-        self.assertSelects(tree.find_all('c', class_='4'), ['Class 3 and 4.'])
-
-        # Passing in a string to 'attrs' will also search the CSS class.
-        self.assertSelects(tree.find_all('a', '1'), ['Class 1.'])
-        self.assertSelects(tree.find_all(attrs='1'), ['Class 1.', 'Class 1.'])
-        self.assertSelects(tree.find_all('c', '3'), ['Class 3 and 4.'])
-        self.assertSelects(tree.find_all('c', '4'), ['Class 3 and 4.'])
-
-    def test_find_by_class_when_multiple_classes_present(self):
-        tree = self.soup("Found it")
-
-        f = tree.find_all("gar", class_=re.compile("o"))
-        self.assertSelects(f, ["Found it"])
-
-        f = tree.find_all("gar", class_=re.compile("a"))
-        self.assertSelects(f, ["Found it"])
-
-        # Since the class is not the string "foo bar", but the two
-        # strings "foo" and "bar", this will not find anything.
-        f = tree.find_all("gar", class_=re.compile("o b"))
-        self.assertSelects(f, [])
-
-    def test_find_all_with_non_dictionary_for_attrs_finds_by_class(self):
-        soup = self.soup("Found it")
-
-        self.assertSelects(soup.find_all("a", re.compile("ba")), ["Found it"])
-
-        def big_attribute_value(value):
-            return len(value) > 3
-
-        self.assertSelects(soup.find_all("a", big_attribute_value), [])
-
-        def small_attribute_value(value):
-            return len(value) <= 3
-
-        self.assertSelects(
-            soup.find_all("a", small_attribute_value), ["Found it"])
-
-    def test_find_all_with_string_for_attrs_finds_multiple_classes(self):
-        soup = self.soup('')
-        a, a2 = soup.find_all("a")
-        self.assertEqual([a, a2], soup.find_all("a", "foo"))
-        self.assertEqual([a], soup.find_all("a", "bar"))
-
-        # If you specify the class as a string that contains a
-        # space, only that specific value will be found.
-        self.assertEqual([a], soup.find_all("a", class_="foo bar"))
-        self.assertEqual([a], soup.find_all("a", "foo bar"))
-        self.assertEqual([], soup.find_all("a", "bar foo"))
-
-    def test_find_all_by_attribute_soupstrainer(self):
-        tree = self.soup("""
-                         Match.
-                         Non-match.""")
-
-        strainer = SoupStrainer(attrs={'id' : 'first'})
-        self.assertSelects(tree.find_all(strainer), ['Match.'])
-
-    def test_find_all_with_missing_atribute(self):
-        # You can pass in None as the value of an attribute to find_all.
-        # This will match tags that do not have that attribute set.
-        tree = self.soup("""ID present.
-                            No ID present.
-                            ID is empty.""")
-        self.assertSelects(tree.find_all('a', id=None), ["No ID present."])
-
-    def test_find_all_with_defined_attribute(self):
-        # You can pass in None as the value of an attribute to find_all.
-        # This will match tags that have that attribute set to any value.
-        tree = self.soup("""ID present.
-                            No ID present.
-                            ID is empty.""")
-        self.assertSelects(
-            tree.find_all(id=True), ["ID present.", "ID is empty."])
-
-    def test_find_all_with_numeric_attribute(self):
-        # If you search for a number, it's treated as a string.
-        tree = self.soup("""Unquoted attribute.
-                            Quoted attribute.""")
-
-        expected = ["Unquoted attribute.", "Quoted attribute."]
-        self.assertSelects(tree.find_all(id=1), expected)
-        self.assertSelects(tree.find_all(id="1"), expected)
-
-    def test_find_all_with_list_attribute_values(self):
-        # You can pass a list of attribute values instead of just one,
-        # and you'll get tags that match any of the values.
-        tree = self.soup("""1
-                            2
-                            3
-                            No ID.""")
-        self.assertSelects(tree.find_all(id=["1", "3", "4"]),
-                           ["1", "3"])
-
-    def test_find_all_with_regular_expression_attribute_value(self):
-        # You can pass a regular expression as an attribute value, and
-        # you'll get tags whose values for that attribute match the
-        # regular expression.
-        tree = self.soup("""One a.
-                            Two as.
-                            Mixed as and bs.
-                            One b.
-                            No ID.""")
-
-        self.assertSelects(tree.find_all(id=re.compile("^a+$")),
-                           ["One a.", "Two as."])
-
-    def test_find_by_name_and_containing_string(self):
-        soup = self.soup("foobarfoo")
-        a = soup.a
-
-        self.assertEqual([a], soup.find_all("a", text="foo"))
-        self.assertEqual([], soup.find_all("a", text="bar"))
-        self.assertEqual([], soup.find_all("a", text="bar"))
-
-    def test_find_by_name_and_containing_string_when_string_is_buried(self):
-        soup = self.soup("foofoo")
-        self.assertEqual(soup.find_all("a"), soup.find_all("a", text="foo"))
-
-    def test_find_by_attribute_and_containing_string(self):
-        soup = self.soup('foofoo')
-        a = soup.a
-
-        self.assertEqual([a], soup.find_all(id=2, text="foo"))
-        self.assertEqual([], soup.find_all(id=1, text="bar"))
-
-
-
-
-class TestIndex(TreeTest):
-    """Test Tag.index"""
-    def test_index(self):
-        tree = self.soup("""

-                            Identical
-                            Not identical
-                            Identical
-
-                            Identical with child
-                            Also not identical
-                            Identical with child
-                            
""")
-        div = tree.div
-        for i, element in enumerate(div.contents):
-            self.assertEqual(i, div.index(element))
-        self.assertRaises(ValueError, tree.index, 1)
-
-
-class TestParentOperations(TreeTest):
-    """Test navigation and searching through an element's parents."""
-
-    def setUp(self):
-        super(TestParentOperations, self).setUp()
-        self.tree = self.soup('''
    
-                                 
      
-                                  
        
-                                   
          
-                                    Start here
-                                   
        
-                                  
      ''')
-        self.start = self.tree.b
-
-
-    def test_parent(self):
-        self.assertEqual(self.start.parent['id'], 'bottom')
-        self.assertEqual(self.start.parent.parent['id'], 'middle')
-        self.assertEqual(self.start.parent.parent.parent['id'], 'top')
-
-    def test_parent_of_top_tag_is_soup_object(self):
-        top_tag = self.tree.contents[0]
-        self.assertEqual(top_tag.parent, self.tree)
-
-    def test_soup_object_has_no_parent(self):
-        self.assertEqual(None, self.tree.parent)
-
-    def test_find_parents(self):
-        self.assertSelectsIDs(
-            self.start.find_parents('ul'), ['bottom', 'middle', 'top'])
-        self.assertSelectsIDs(
-            self.start.find_parents('ul', id="middle"), ['middle'])
-
-    def test_find_parent(self):
-        self.assertEqual(self.start.find_parent('ul')['id'], 'bottom')
-        self.assertEqual(self.start.find_parent('ul', id='top')['id'], 'top')
-
-    def test_parent_of_text_element(self):
-        text = self.tree.find(text="Start here")
-        self.assertEqual(text.parent.name, 'b')
-
-    def test_text_element_find_parent(self):
-        text = self.tree.find(text="Start here")
-        self.assertEqual(text.find_parent('ul')['id'], 'bottom')
-
-    def test_parent_generator(self):
-        parents = [parent['id'] for parent in self.start.parents
-                   if parent is not None and 'id' in parent.attrs]
-        self.assertEqual(parents, ['bottom', 'middle', 'top'])
-
-
-class ProximityTest(TreeTest):
-
-    def setUp(self):
-        super(TreeTest, self).setUp()
-        self.tree = self.soup(
-            'OneTwoThree')
-
-
-class TestNextOperations(ProximityTest):
-
-    def setUp(self):
-        super(TestNextOperations, self).setUp()
-        self.start = self.tree.b
-
-    def test_next(self):
-        self.assertEqual(self.start.next_element, "One")
-        self.assertEqual(self.start.next_element.next_element['id'], "2")
-
-    def test_next_of_last_item_is_none(self):
-        last = self.tree.find(text="Three")
-        self.assertEqual(last.next_element, None)
-
-    def test_next_of_root_is_none(self):
-        # The document root is outside the next/previous chain.
-        self.assertEqual(self.tree.next_element, None)
-
-    def test_find_all_next(self):
-        self.assertSelects(self.start.find_all_next('b'), ["Two", "Three"])
-        self.start.find_all_next(id=3)
-        self.assertSelects(self.start.find_all_next(id=3), ["Three"])
-
-    def test_find_next(self):
-        self.assertEqual(self.start.find_next('b')['id'], '2')
-        self.assertEqual(self.start.find_next(text="Three"), "Three")
-
-    def test_find_next_for_text_element(self):
-        text = self.tree.find(text="One")
-        self.assertEqual(text.find_next("b").string, "Two")
-        self.assertSelects(text.find_all_next("b"), ["Two", "Three"])
-
-    def test_next_generator(self):
-        start = self.tree.find(text="Two")
-        successors = [node for node in start.next_elements]
-        # There are two successors: the final  tag and its text contents.
-        tag, contents = successors
-        self.assertEqual(tag['id'], '3')
-        self.assertEqual(contents, "Three")
-
-class TestPreviousOperations(ProximityTest):
-
-    def setUp(self):
-        super(TestPreviousOperations, self).setUp()
-        self.end = self.tree.find(text="Three")
-
-    def test_previous(self):
-        self.assertEqual(self.end.previous_element['id'], "3")
-        self.assertEqual(self.end.previous_element.previous_element, "Two")
-
-    def test_previous_of_first_item_is_none(self):
-        first = self.tree.find('html')
-        self.assertEqual(first.previous_element, None)
-
-    def test_previous_of_root_is_none(self):
-        # The document root is outside the next/previous chain.
-        # XXX This is broken!
-        #self.assertEqual(self.tree.previous_element, None)
-        pass
-
-    def test_find_all_previous(self):
-        # The  tag containing the "Three" node is the predecessor
-        # of the "Three" node itself, which is why "Three" shows up
-        # here.
-        self.assertSelects(
-            self.end.find_all_previous('b'), ["Three", "Two", "One"])
-        self.assertSelects(self.end.find_all_previous(id=1), ["One"])
-
-    def test_find_previous(self):
-        self.assertEqual(self.end.find_previous('b')['id'], '3')
-        self.assertEqual(self.end.find_previous(text="One"), "One")
-
-    def test_find_previous_for_text_element(self):
-        text = self.tree.find(text="Three")
-        self.assertEqual(text.find_previous("b").string, "Three")
-        self.assertSelects(
-            text.find_all_previous("b"), ["Three", "Two", "One"])
-
-    def test_previous_generator(self):
-        start = self.tree.find(text="One")
-        predecessors = [node for node in start.previous_elements]
-
-        # There are four predecessors: the  tag containing "One"
-        # the  tag, the  tag, and the  tag.
-        b, body, head, html = predecessors
-        self.assertEqual(b['id'], '1')
-        self.assertEqual(body.name, "body")
-        self.assertEqual(head.name, "head")
-        self.assertEqual(html.name, "html")
-
-
-class SiblingTest(TreeTest):
-
-    def setUp(self):
-        super(SiblingTest, self).setUp()
-        markup = '''
-                    
-                     
-                    
-                    
-                     
-                    
-                    
-                     
-                    
-                    
-                    '''
-        # All that whitespace looks good but makes the tests more
-        # difficult. Get rid of it.
-        markup = re.compile(r"\n\s*").sub("", markup)
-        self.tree = self.soup(markup)
-
-
-class TestNextSibling(SiblingTest):
-
-    def setUp(self):
-        super(TestNextSibling, self).setUp()
-        self.start = self.tree.find(id="1")
-
-    def test_next_sibling_of_root_is_none(self):
-        self.assertEqual(self.tree.next_sibling, None)
-
-    def test_next_sibling(self):
-        self.assertEqual(self.start.next_sibling['id'], '2')
-        self.assertEqual(self.start.next_sibling.next_sibling['id'], '3')
-
-        # Note the difference between next_sibling and next_element.
-        self.assertEqual(self.start.next_element['id'], '1.1')
-
-    def test_next_sibling_may_not_exist(self):
-        self.assertEqual(self.tree.html.next_sibling, None)
-
-        nested_span = self.tree.find(id="1.1")
-        self.assertEqual(nested_span.next_sibling, None)
-
-        last_span = self.tree.find(id="4")
-        self.assertEqual(last_span.next_sibling, None)
-
-    def test_find_next_sibling(self):
-        self.assertEqual(self.start.find_next_sibling('span')['id'], '2')
-
-    def test_next_siblings(self):
-        self.assertSelectsIDs(self.start.find_next_siblings("span"),
-                              ['2', '3', '4'])
-
-        self.assertSelectsIDs(self.start.find_next_siblings(id='3'), ['3'])
-
-    def test_next_sibling_for_text_element(self):
-        soup = self.soup("Foobarbaz")
-        start = soup.find(text="Foo")
-        self.assertEqual(start.next_sibling.name, 'b')
-        self.assertEqual(start.next_sibling.next_sibling, 'baz')
-
-        self.assertSelects(start.find_next_siblings('b'), ['bar'])
-        self.assertEqual(start.find_next_sibling(text="baz"), "baz")
-        self.assertEqual(start.find_next_sibling(text="nonesuch"), None)
-
-
-class TestPreviousSibling(SiblingTest):
-
-    def setUp(self):
-        super(TestPreviousSibling, self).setUp()
-        self.end = self.tree.find(id="4")
-
-    def test_previous_sibling_of_root_is_none(self):
-        self.assertEqual(self.tree.previous_sibling, None)
-
-    def test_previous_sibling(self):
-        self.assertEqual(self.end.previous_sibling['id'], '3')
-        self.assertEqual(self.end.previous_sibling.previous_sibling['id'], '2')
-
-        # Note the difference between previous_sibling and previous_element.
-        self.assertEqual(self.end.previous_element['id'], '3.1')
-
-    def test_previous_sibling_may_not_exist(self):
-        self.assertEqual(self.tree.html.previous_sibling, None)
-
-        nested_span = self.tree.find(id="1.1")
-        self.assertEqual(nested_span.previous_sibling, None)
-
-        first_span = self.tree.find(id="1")
-        self.assertEqual(first_span.previous_sibling, None)
-
-    def test_find_previous_sibling(self):
-        self.assertEqual(self.end.find_previous_sibling('span')['id'], '3')
-
-    def test_previous_siblings(self):
-        self.assertSelectsIDs(self.end.find_previous_siblings("span"),
-                              ['3', '2', '1'])
-
-        self.assertSelectsIDs(self.end.find_previous_siblings(id='1'), ['1'])
-
-    def test_previous_sibling_for_text_element(self):
-        soup = self.soup("Foobarbaz")
-        start = soup.find(text="baz")
-        self.assertEqual(start.previous_sibling.name, 'b')
-        self.assertEqual(start.previous_sibling.previous_sibling, 'Foo')
-
-        self.assertSelects(start.find_previous_siblings('b'), ['bar'])
-        self.assertEqual(start.find_previous_sibling(text="Foo"), "Foo")
-        self.assertEqual(start.find_previous_sibling(text="nonesuch"), None)
-
-
-class TestTagCreation(SoupTest):
-    """Test the ability to create new tags."""
-    def test_new_tag(self):
-        soup = self.soup("")
-        new_tag = soup.new_tag("foo", bar="baz")
-        self.assertTrue(isinstance(new_tag, Tag))
-        self.assertEqual("foo", new_tag.name)
-        self.assertEqual(dict(bar="baz"), new_tag.attrs)
-        self.assertEqual(None, new_tag.parent)
-
-    def test_tag_inherits_self_closing_rules_from_builder(self):
-        if XML_BUILDER_PRESENT:
-            xml_soup = BeautifulSoup("", "lxml-xml")
-            xml_br = xml_soup.new_tag("br")
-            xml_p = xml_soup.new_tag("p")
-
-            # Both the 
       and 
       tag are empty-element, just because
-            # they have no contents.
-            self.assertEqual(b"
      ", xml_br.encode())
-            self.assertEqual(b"
      ", xml_p.encode())
-
-        html_soup = BeautifulSoup("", "html.parser")
-        html_br = html_soup.new_tag("br")
-        html_p = html_soup.new_tag("p")
-
-        # The HTML builder users HTML's rules about which tags are
-        # empty-element tags, and the new tags reflect these rules.
-        self.assertEqual(b"
      ", html_br.encode())
-        self.assertEqual(b"
      ", html_p.encode())
-
-    def test_new_string_creates_navigablestring(self):
-        soup = self.soup("")
-        s = soup.new_string("foo")
-        self.assertEqual("foo", s)
-        self.assertTrue(isinstance(s, NavigableString))
-
-    def test_new_string_can_create_navigablestring_subclass(self):
-        soup = self.soup("")
-        s = soup.new_string("foo", Comment)
-        self.assertEqual("foo", s)
-        self.assertTrue(isinstance(s, Comment))
-
-class TestTreeModification(SoupTest):
-
-    def test_attribute_modification(self):
-        soup = self.soup('')
-        soup.a['id'] = 2
-        self.assertEqual(soup.decode(), self.document_for(''))
-        del(soup.a['id'])
-        self.assertEqual(soup.decode(), self.document_for(''))
-        soup.a['id2'] = 'foo'
-        self.assertEqual(soup.decode(), self.document_for(''))
-
-    def test_new_tag_creation(self):
-        builder = builder_registry.lookup('html')()
-        soup = self.soup("", builder=builder)
-        a = Tag(soup, builder, 'a')
-        ol = Tag(soup, builder, 'ol')
-        a['href'] = 'http://foo.com/'
-        soup.body.insert(0, a)
-        soup.body.insert(1, ol)
-        self.assertEqual(
-            soup.body.encode(),
-            b'
        ')
-
-    def test_append_to_contents_moves_tag(self):
-        doc = """
        Don't leave me here.
        
-                
        Don\'t leave!
        """
-        soup = self.soup(doc)
-        second_para = soup.find(id='2')
-        bold = soup.b
-
-        # Move the  tag to the end of the second paragraph.
-        soup.find(id='2').append(soup.b)
-
-        # The  tag is now a child of the second paragraph.
-        self.assertEqual(bold.parent, second_para)
-
-        self.assertEqual(
-            soup.decode(), self.document_for(
-                '
        Don\'t leave me .
        \n'
-                '
        Don\'t leave!here
        '))
-
-    def test_replace_with_returns_thing_that_was_replaced(self):
-        text = ""
-        soup = self.soup(text)
-        a = soup.a
-        new_a = a.replace_with(soup.c)
-        self.assertEqual(a, new_a)
-
-    def test_unwrap_returns_thing_that_was_replaced(self):
-        text = ""
-        soup = self.soup(text)
-        a = soup.a
-        new_a = a.unwrap()
-        self.assertEqual(a, new_a)
-
-    def test_replace_with_and_unwrap_give_useful_exception_when_tag_has_no_parent(self):
-        soup = self.soup("FooBar")
-        a = soup.a
-        a.extract()
-        self.assertEqual(None, a.parent)
-        self.assertRaises(ValueError, a.unwrap)
-        self.assertRaises(ValueError, a.replace_with, soup.c)
-
-    def test_replace_tag_with_itself(self):
-        text = "Foo"
-        soup = self.soup(text)
-        c = soup.c
-        soup.c.replace_with(c)
-        self.assertEqual(soup.decode(), self.document_for(text))
-
-    def test_replace_tag_with_its_parent_raises_exception(self):
-        text = ""
-        soup = self.soup(text)
-        self.assertRaises(ValueError, soup.b.replace_with, soup.a)
-
-    def test_insert_tag_into_itself_raises_exception(self):
-        text = ""
-        soup = self.soup(text)
-        self.assertRaises(ValueError, soup.a.insert, 0, soup.a)
-
-    def test_replace_with_maintains_next_element_throughout(self):
-        soup = self.soup('
        onethree
        ')
-        a = soup.a
-        b = a.contents[0]
-        # Make it so the  tag has two text children.
-        a.insert(1, "two")
-
-        # Now replace each one with the empty string.
-        left, right = a.contents
-        left.replaceWith('')
-        right.replaceWith('')
-
-        # The  tag is still connected to the tree.
-        self.assertEqual("three", soup.b.string)
-
-    def test_replace_final_node(self):
-        soup = self.soup("Argh!")
-        soup.find(text="Argh!").replace_with("Hooray!")
-        new_text = soup.find(text="Hooray!")
-        b = soup.b
-        self.assertEqual(new_text.previous_element, b)
-        self.assertEqual(new_text.parent, b)
-        self.assertEqual(new_text.previous_element.next_element, new_text)
-        self.assertEqual(new_text.next_element, None)
-
-    def test_consecutive_text_nodes(self):
-        # A builder should never create two consecutive text nodes,
-        # but if you insert one next to another, Beautiful Soup will
-        # handle it correctly.
-        soup = self.soup("Argh!")
-        soup.b.insert(1, "Hooray!")
-
-        self.assertEqual(
-            soup.decode(), self.document_for(
-                "Argh!Hooray!"))
-
-        new_text = soup.find(text="Hooray!")
-        self.assertEqual(new_text.previous_element, "Argh!")
-        self.assertEqual(new_text.previous_element.next_element, new_text)
-
-        self.assertEqual(new_text.previous_sibling, "Argh!")
-        self.assertEqual(new_text.previous_sibling.next_sibling, new_text)
-
-        self.assertEqual(new_text.next_sibling, None)
-        self.assertEqual(new_text.next_element, soup.c)
-
-    def test_insert_string(self):
-        soup = self.soup("")
-        soup.a.insert(0, "bar")
-        soup.a.insert(0, "foo")
-        # The string were added to the tag.
-        self.assertEqual(["foo", "bar"], soup.a.contents)
-        # And they were converted to NavigableStrings.
-        self.assertEqual(soup.a.contents[0].next_element, "bar")
-
-    def test_insert_tag(self):
-        builder = self.default_builder
-        soup = self.soup(
-            "Findlady!", builder=builder)
-        magic_tag = Tag(soup, builder, 'magictag')
-        magic_tag.insert(0, "the")
-        soup.a.insert(1, magic_tag)
-
-        self.assertEqual(
-            soup.decode(), self.document_for(
-                "Findthelady!"))
-
-        # Make sure all the relationships are hooked up correctly.
-        b_tag = soup.b
-        self.assertEqual(b_tag.next_sibling, magic_tag)
-        self.assertEqual(magic_tag.previous_sibling, b_tag)
-
-        find = b_tag.find(text="Find")
-        self.assertEqual(find.next_element, magic_tag)
-        self.assertEqual(magic_tag.previous_element, find)
-
-        c_tag = soup.c
-        self.assertEqual(magic_tag.next_sibling, c_tag)
-        self.assertEqual(c_tag.previous_sibling, magic_tag)
-
-        the = magic_tag.find(text="the")
-        self.assertEqual(the.parent, magic_tag)
-        self.assertEqual(the.next_element, c_tag)
-        self.assertEqual(c_tag.previous_element, the)
-
-    def test_append_child_thats_already_at_the_end(self):
-        data = ""
-        soup = self.soup(data)
-        soup.a.append(soup.b)
-        self.assertEqual(data, soup.decode())
-
-    def test_move_tag_to_beginning_of_parent(self):
-        data = ""
-        soup = self.soup(data)
-        soup.a.insert(0, soup.d)
-        self.assertEqual("", soup.decode())
-
-    def test_insert_works_on_empty_element_tag(self):
-        # This is a little strange, since most HTML parsers don't allow
-        # markup like this to come through. But in general, we don't
-        # know what the parser would or wouldn't have allowed, so
-        # I'm letting this succeed for now.
-        soup = self.soup("
        ")
-        soup.br.insert(1, "Contents")
-        self.assertEqual(str(soup.br), "
        Contents
        ")
-
-    def test_insert_before(self):
-        soup = self.soup("foobar")
-        soup.b.insert_before("BAZ")
-        soup.a.insert_before("QUUX")
-        self.assertEqual(
-            soup.decode(), self.document_for("QUUXfooBAZbar"))
-
-        soup.a.insert_before(soup.b)
-        self.assertEqual(
-            soup.decode(), self.document_for("QUUXbarfooBAZ"))
-
-    def test_insert_after(self):
-        soup = self.soup("foobar")
-        soup.b.insert_after("BAZ")
-        soup.a.insert_after("QUUX")
-        self.assertEqual(
-            soup.decode(), self.document_for("fooQUUXbarBAZ"))
-        soup.b.insert_after(soup.a)
-        self.assertEqual(
-            soup.decode(), self.document_for("QUUXbarfooBAZ"))
-
-    def test_insert_after_raises_exception_if_after_has_no_meaning(self):
-        soup = self.soup("")
-        tag = soup.new_tag("a")
-        string = soup.new_string("")
-        self.assertRaises(ValueError, string.insert_after, tag)
-        self.assertRaises(NotImplementedError, soup.insert_after, tag)
-        self.assertRaises(ValueError, tag.insert_after, tag)
-
-    def test_insert_before_raises_notimplementederror_if_before_has_no_meaning(self):
-        soup = self.soup("")
-        tag = soup.new_tag("a")
-        string = soup.new_string("")
-        self.assertRaises(ValueError, string.insert_before, tag)
-        self.assertRaises(NotImplementedError, soup.insert_before, tag)
-        self.assertRaises(ValueError, tag.insert_before, tag)
-
-    def test_replace_with(self):
-        soup = self.soup(
-                "
        There's no business like show business
        ")
-        no, show = soup.find_all('b')
-        show.replace_with(no)
-        self.assertEqual(
-            soup.decode(),
-            self.document_for(
-                "
        There's  business like no business
        "))
-
-        self.assertEqual(show.parent, None)
-        self.assertEqual(no.parent, soup.p)
-        self.assertEqual(no.next_element, "no")
-        self.assertEqual(no.next_sibling, " business")
-
-    def test_replace_first_child(self):
-        data = ""
-        soup = self.soup(data)
-        soup.b.replace_with(soup.c)
-        self.assertEqual("", soup.decode())
-
-    def test_replace_last_child(self):
-        data = ""
-        soup = self.soup(data)
-        soup.c.replace_with(soup.b)
-        self.assertEqual("", soup.decode())
-
-    def test_nested_tag_replace_with(self):
-        soup = self.soup(
-            """Wereservetherighttorefuseservice""")
-
-        # Replace the entire  tag and its contents ("reserve the
-        # right") with the  tag ("refuse").
-        remove_tag = soup.b
-        move_tag = soup.f
-        remove_tag.replace_with(move_tag)
-
-        self.assertEqual(
-            soup.decode(), self.document_for(
-                "Werefusetoservice"))
-
-        # The  tag is now an orphan.
-        self.assertEqual(remove_tag.parent, None)
-        self.assertEqual(remove_tag.find(text="right").next_element, None)
-        self.assertEqual(remove_tag.previous_element, None)
-        self.assertEqual(remove_tag.next_sibling, None)
-        self.assertEqual(remove_tag.previous_sibling, None)
-
-        # The  tag is now connected to the  tag.
-        self.assertEqual(move_tag.parent, soup.a)
-        self.assertEqual(move_tag.previous_element, "We")
-        self.assertEqual(move_tag.next_element.next_element, soup.e)
-        self.assertEqual(move_tag.next_sibling, None)
-
-        # The gap where the  tag used to be has been mended, and
-        # the word "to" is now connected to the  tag.
-        to_text = soup.find(text="to")
-        g_tag = soup.g
-        self.assertEqual(to_text.next_element, g_tag)
-        self.assertEqual(to_text.next_sibling, g_tag)
-        self.assertEqual(g_tag.previous_element, to_text)
-        self.assertEqual(g_tag.previous_sibling, to_text)
-
-    def test_unwrap(self):
-        tree = self.soup("""
-            
        Unneeded formatting is unneeded
        
-            """)
-        tree.em.unwrap()
-        self.assertEqual(tree.em, None)
-        self.assertEqual(tree.p.text, "Unneeded formatting is unneeded")
-
-    def test_wrap(self):
-        soup = self.soup("I wish I was bold.")
-        value = soup.string.wrap(soup.new_tag("b"))
-        self.assertEqual(value.decode(), "I wish I was bold.")
-        self.assertEqual(
-            soup.decode(), self.document_for("I wish I was bold."))
-
-    def test_wrap_extracts_tag_from_elsewhere(self):
-        soup = self.soup("I wish I was bold.")
-        soup.b.next_sibling.wrap(soup.b)
-        self.assertEqual(
-            soup.decode(), self.document_for("I wish I was bold."))
-
-    def test_wrap_puts_new_contents_at_the_end(self):
-        soup = self.soup("I like being bold.I wish I was bold.")
-        soup.b.next_sibling.wrap(soup.b)
-        self.assertEqual(2, len(soup.b.contents))
-        self.assertEqual(
-            soup.decode(), self.document_for(
-                "I like being bold.I wish I was bold."))
-
-    def test_extract(self):
-        soup = self.soup(
-            'Some content.  More content.')
-
-        self.assertEqual(len(soup.body.contents), 3)
-        extracted = soup.find(id="nav").extract()
-
-        self.assertEqual(
-            soup.decode(), "Some content.  More content.")
-        self.assertEqual(extracted.decode(), '')
-
-        # The extracted tag is now an orphan.
-        self.assertEqual(len(soup.body.contents), 2)
-        self.assertEqual(extracted.parent, None)
-        self.assertEqual(extracted.previous_element, None)
-        self.assertEqual(extracted.next_element.next_element, None)
-
-        # The gap where the extracted tag used to be has been mended.
-        content_1 = soup.find(text="Some content. ")
-        content_2 = soup.find(text=" More content.")
-        self.assertEqual(content_1.next_element, content_2)
-        self.assertEqual(content_1.next_sibling, content_2)
-        self.assertEqual(content_2.previous_element, content_1)
-        self.assertEqual(content_2.previous_sibling, content_1)
-
-    def test_extract_distinguishes_between_identical_strings(self):
-        soup = self.soup("        foobar")
-        foo_1 = soup.a.string
-        bar_1 = soup.b.string
-        foo_2 = soup.new_string("foo")
-        bar_2 = soup.new_string("bar")
-        soup.a.append(foo_2)
-        soup.b.append(bar_2)
-
-        # Now there are two identical strings in the  tag, and two
-        # in the  tag. Let's remove the first "foo" and the second
-        # "bar".
-        foo_1.extract()
-        bar_2.extract()
-        self.assertEqual(foo_2, soup.a.string)
-        self.assertEqual(bar_2, soup.b.string)
-
-    def test_extract_multiples_of_same_tag(self):
-        soup = self.soup("""
-
-
-
-
-
- 
- 
-
-
-""")
-        [soup.script.extract() for i in soup.find_all("script")]
-        self.assertEqual("\n\n\n", str(soup.body))
-
-
-    def test_extract_works_when_element_is_surrounded_by_identical_strings(self):
-        soup = self.soup(
- '\n'
- 'hi\n'
- '')
-        soup.find('body').extract()
-        self.assertEqual(None, soup.find('body'))
-
-
-    def test_clear(self):
-        """Tag.clear()"""
-        soup = self.soup("
        String Italicized and another
        ")
-        # clear using extract()
-        a = soup.a
-        soup.p.clear()
-        self.assertEqual(len(soup.p.contents), 0)
-        self.assertTrue(hasattr(a, "contents"))
-
-        # clear using decompose()
-        em = a.em
-        a.clear(decompose=True)
-        self.assertEqual(0, len(em.contents))
-
-    def test_string_set(self):
-        """Tag.string = 'string'"""
-        soup = self.soup(" ")
-        soup.a.string = "foo"
-        self.assertEqual(soup.a.contents, ["foo"])
-        soup.b.string = "bar"
-        self.assertEqual(soup.b.contents, ["bar"])
-
-    def test_string_set_does_not_affect_original_string(self):
-        soup = self.soup("foobar")
-        soup.b.string = soup.c.string
-        self.assertEqual(soup.a.encode(), b"barbar")
-
-    def test_set_string_preserves_class_of_string(self):
-        soup = self.soup("")
-        cdata = CData("foo")
-        soup.a.string = cdata
-        self.assertTrue(isinstance(soup.a.string, CData))
-
-class TestElementObjects(SoupTest):
-    """Test various features of element objects."""
-
-    def test_len(self):
-        """The length of an element is its number of children."""
-        soup = self.soup("123")
-
-        # The BeautifulSoup object itself contains one element: the
-        #  tag.
-        self.assertEqual(len(soup.contents), 1)
-        self.assertEqual(len(soup), 1)
-
-        # The  tag contains three elements: the text node "1", the
-        #  tag, and the text node "3".
-        self.assertEqual(len(soup.top), 3)
-        self.assertEqual(len(soup.top.contents), 3)
-
-    def test_member_access_invokes_find(self):
-        """Accessing a Python member .foo invokes find('foo')"""
-        soup = self.soup('')
-        self.assertEqual(soup.b, soup.find('b'))
-        self.assertEqual(soup.b.i, soup.find('b').find('i'))
-        self.assertEqual(soup.a, None)
-
-    def test_deprecated_member_access(self):
-        soup = self.soup('')
-        with warnings.catch_warnings(record=True) as w:
-            tag = soup.bTag
-        self.assertEqual(soup.b, tag)
-        self.assertEqual(
-            '.bTag is deprecated, use .find("b") instead.',
-            str(w[0].message))
-
-    def test_has_attr(self):
-        """has_attr() checks for the presence of an attribute.
-
-        Please note note: has_attr() is different from
-        __in__. has_attr() checks the tag's attributes and __in__
-        checks the tag's chidlren.
-        """
-        soup = self.soup("")
-        self.assertTrue(soup.foo.has_attr('attr'))
-        self.assertFalse(soup.foo.has_attr('attr2'))
-
-
-    def test_attributes_come_out_in_alphabetical_order(self):
-        markup = ''
-        self.assertSoupEquals(markup, '')
-
-    def test_string(self):
-        # A tag that contains only a text node makes that node
-        # available as .string.
-        soup = self.soup("foo")
-        self.assertEqual(soup.b.string, 'foo')
-
-    def test_empty_tag_has_no_string(self):
-        # A tag with no children has no .stirng.
-        soup = self.soup("")
-        self.assertEqual(soup.b.string, None)
-
-    def test_tag_with_multiple_children_has_no_string(self):
-        # A tag with no children has no .string.
-        soup = self.soup("foo")
-        self.assertEqual(soup.b.string, None)
-
-        soup = self.soup("foobar        ")
-        self.assertEqual(soup.b.string, None)
-
-        # Even if all the children are strings, due to trickery,
-        # it won't work--but this would be a good optimization.
-        soup = self.soup("foo        ")
-        soup.a.insert(1, "bar")
-        self.assertEqual(soup.a.string, None)
-
-    def test_tag_with_recursive_string_has_string(self):
-        # A tag with a single child which has a .string inherits that
-        # .string.
-        soup = self.soup("foo")
-        self.assertEqual(soup.a.string, "foo")
-        self.assertEqual(soup.string, "foo")
-
-    def test_lack_of_string(self):
-        """Only a tag containing a single text node has a .string."""
-        soup = self.soup("feo")
-        self.assertFalse(soup.b.string)
-
-        soup = self.soup("")
-        self.assertFalse(soup.b.string)
-
-    def test_all_text(self):
-        """Tag.text and Tag.get_text(sep=u"") -> all child text, concatenated"""
-        soup = self.soup("ar    t ")
-        self.assertEqual(soup.a.text, "ar  t ")
-        self.assertEqual(soup.a.get_text(strip=True), "art")
-        self.assertEqual(soup.a.get_text(","), "a,r, , t ")
-        self.assertEqual(soup.a.get_text(",", strip=True), "a,r,t")
-
-    def test_get_text_ignores_comments(self):
-        soup = self.soup("foobar")
-        self.assertEqual(soup.get_text(), "foobar")
-
-        self.assertEqual(
-            soup.get_text(types=(NavigableString, Comment)), "fooIGNOREbar")
-        self.assertEqual(
-            soup.get_text(types=None), "fooIGNOREbar")
-
-    def test_all_strings_ignores_comments(self):
-        soup = self.soup("foobar")
-        self.assertEqual(['foo', 'bar'], list(soup.strings))
-
-class TestCDAtaListAttributes(SoupTest):
-
-    """Testing cdata-list attributes like 'class'.
-    """
-    def test_single_value_becomes_list(self):
-        soup = self.soup("")
-        self.assertEqual(["foo"],soup.a['class'])
-
-    def test_multiple_values_becomes_list(self):
-        soup = self.soup("")
-        self.assertEqual(["foo", "bar"], soup.a['class'])
-
-    def test_multiple_values_separated_by_weird_whitespace(self):
-        soup = self.soup("")
-        self.assertEqual(["foo", "bar", "baz"],soup.a['class'])
-
-    def test_attributes_joined_into_string_on_output(self):
-        soup = self.soup("")
-        self.assertEqual(b'', soup.a.encode())
-
-    def test_accept_charset(self):
-        soup = self.soup('
        ')
-        self.assertEqual(['ISO-8859-1', 'UTF-8'], soup.form['accept-charset'])
-
-    def test_cdata_attribute_applying_only_to_one_tag(self):
-        data = ''
-        soup = self.soup(data)
-        # We saw in another test that accept-charset is a cdata-list
-        # attribute for the  tag. But it's not a cdata-list
-        # attribute for any other tag.
-        self.assertEqual('ISO-8859-1 UTF-8', soup.a['accept-charset'])
-
-    def test_string_has_immutable_name_property(self):
-        string = self.soup("s").string
-        self.assertEqual(None, string.name)
-        def t():
-            string.name = 'foo'
-        self.assertRaises(AttributeError, t)
-
-class TestPersistence(SoupTest):
-    "Testing features like pickle and deepcopy."
-
-    def setUp(self):
-        super(TestPersistence, self).setUp()
-        self.page = """
-
-
-
-Beautiful Soup: We called him Tortoise because he taught us.
-
-
-
-
-
-
-foo
-bar
-
-"""
-        self.tree = self.soup(self.page)
-
-    def test_pickle_and_unpickle_identity(self):
-        # Pickling a tree, then unpickling it, yields a tree identical
-        # to the original.
-        dumped = pickle.dumps(self.tree, 2)
-        loaded = pickle.loads(dumped)
-        self.assertEqual(loaded.__class__, BeautifulSoup)
-        self.assertEqual(loaded.decode(), self.tree.decode())
-
-    def test_deepcopy_identity(self):
-        # Making a deepcopy of a tree yields an identical tree.
-        copied = copy.deepcopy(self.tree)
-        self.assertEqual(copied.decode(), self.tree.decode())
-
-    def test_unicode_pickle(self):
-        # A tree containing Unicode characters can be pickled.
-        html = "\N{SNOWMAN}"
-        soup = self.soup(html)
-        dumped = pickle.dumps(soup, pickle.HIGHEST_PROTOCOL)
-        loaded = pickle.loads(dumped)
-        self.assertEqual(loaded.decode(), soup.decode())
-
-    def test_copy_navigablestring_is_not_attached_to_tree(self):
-        html = "FooBar"
-        soup = self.soup(html)
-        s1 = soup.find(string="Foo")
-        s2 = copy.copy(s1)
-        self.assertEqual(s1, s2)
-        self.assertEqual(None, s2.parent)
-        self.assertEqual(None, s2.next_element)
-        self.assertNotEqual(None, s1.next_sibling)
-        self.assertEqual(None, s2.next_sibling)
-        self.assertEqual(None, s2.previous_element)
-
-    def test_copy_navigablestring_subclass_has_same_type(self):
-        html = ""
-        soup = self.soup(html)
-        s1 = soup.string
-        s2 = copy.copy(s1)
-        self.assertEqual(s1, s2)
-        self.assertTrue(isinstance(s2, Comment))
-
-    def test_copy_entire_soup(self):
-        html = "
        FooBar
        end"
-        soup = self.soup(html)
-        soup_copy = copy.copy(soup)
-        self.assertEqual(soup, soup_copy)
-
-    def test_copy_tag_copies_contents(self):
-        html = "
        FooBar
        end"
-        soup = self.soup(html)
-        div = soup.div
-        div_copy = copy.copy(div)
-
-        # The two tags look the same, and evaluate to equal.
-        self.assertEqual(str(div), str(div_copy))
-        self.assertEqual(div, div_copy)
-
-        # But they're not the same object.
-        self.assertFalse(div is div_copy)
-
-        # And they don't have the same relation to the parse tree. The
-        # copy is not associated with a parse tree at all.
-        self.assertEqual(None, div_copy.parent)
-        self.assertEqual(None, div_copy.previous_element)
-        self.assertEqual(None, div_copy.find(string='Bar').next_element)
-        self.assertNotEqual(None, div.find(string='Bar').next_element)
-
-class TestSubstitutions(SoupTest):
-
-    def test_default_formatter_is_minimal(self):
-        markup = "<<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>>"
-        soup = self.soup(markup)
-        decoded = soup.decode(formatter="minimal")
-        # The < is converted back into < but the e-with-acute is left alone.
-        self.assertEqual(
-            decoded,
-            self.document_for(
-                "<<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>>"))
-
-    def test_formatter_html(self):
-        markup = "<<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>>"
-        soup = self.soup(markup)
-        decoded = soup.decode(formatter="html")
-        self.assertEqual(
-            decoded,
-            self.document_for("<<Sacré bleu!>>"))
-
-    def test_formatter_minimal(self):
-        markup = "<<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>>"
-        soup = self.soup(markup)
-        decoded = soup.decode(formatter="minimal")
-        # The < is converted back into < but the e-with-acute is left alone.
-        self.assertEqual(
-            decoded,
-            self.document_for(
-                "<<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>>"))
-
-    def test_formatter_null(self):
-        markup = "<<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>>"
-        soup = self.soup(markup)
-        decoded = soup.decode(formatter=None)
-        # Neither the angle brackets nor the e-with-acute are converted.
-        # This is not valid HTML, but it's what the user wanted.
-        self.assertEqual(decoded,
-                          self.document_for("<>"))
-
-    def test_formatter_custom(self):
-        markup = "<foo>bar"
-        soup = self.soup(markup)
-        decoded = soup.decode(formatter = lambda x: x.upper())
-        # Instead of normal entity conversion code, the custom
-        # callable is called on every string.
-        self.assertEqual(
-            decoded,
-            self.document_for("BAR"))
-
-    def test_formatter_is_run_on_attribute_values(self):
-        markup = 'e'
-        soup = self.soup(markup)
-        a = soup.a
-
-        expect_minimal = 'e'
-
-        self.assertEqual(expect_minimal, a.decode())
-        self.assertEqual(expect_minimal, a.decode(formatter="minimal"))
-
-        expect_html = 'e'
-        self.assertEqual(expect_html, a.decode(formatter="html"))
-
-        self.assertEqual(markup, a.decode(formatter=None))
-        expect_upper = 'E'
-        self.assertEqual(expect_upper, a.decode(formatter=lambda x: x.upper()))
-
-    def test_formatter_skips_script_tag_for_html_documents(self):
-        doc = """
-  
-"""
-        encoded = BeautifulSoup(doc, 'html.parser').encode()
-        self.assertTrue(b"< < hey > >" in encoded)
-
-    def test_formatter_skips_style_tag_for_html_documents(self):
-        doc = """
-  
-"""
-        encoded = BeautifulSoup(doc, 'html.parser').encode()
-        self.assertTrue(b"< < hey > >" in encoded)
-
-    def test_prettify_leaves_preformatted_text_alone(self):
-        soup = self.soup("
          foo  
          \tbar\n  \n  
          baz  ")
-        # Everything outside the 
         tag is reformatted, but everything
-        # inside is left alone.
-        self.assertEqual(
-            '
        \n foo\n 
          \tbar\n  \n  
        \n baz\n
        ',
-            soup.div.prettify())
-
-    def test_prettify_accepts_formatter(self):
-        soup = BeautifulSoup("foo", 'html.parser')
-        pretty = soup.prettify(formatter = lambda x: x.upper())
-        self.assertTrue("FOO" in pretty)
-
-    def test_prettify_outputs_unicode_by_default(self):
-        soup = self.soup("")
-        self.assertEqual(str, type(soup.prettify()))
-
-    def test_prettify_can_encode_data(self):
-        soup = self.soup("")
-        self.assertEqual(bytes, type(soup.prettify("utf-8")))
-
-    def test_html_entity_substitution_off_by_default(self):
-        markup = "Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!"
-        soup = self.soup(markup)
-        encoded = soup.b.encode("utf-8")
-        self.assertEqual(encoded, markup.encode('utf-8'))
-
-    def test_encoding_substitution(self):
-        # Here's the  tag saying that a document is
-        # encoded in Shift-JIS.
-        meta_tag = ('')
-        soup = self.soup(meta_tag)
-
-        # Parse the document, and the charset apprears unchanged.
-        self.assertEqual(soup.meta['content'], 'text/html; charset=x-sjis')
-
-        # Encode the document into some encoding, and the encoding is
-        # substituted into the meta tag.
-        utf_8 = soup.encode("utf-8")
-        self.assertTrue(b"charset=utf-8" in utf_8)
-
-        euc_jp = soup.encode("euc_jp")
-        self.assertTrue(b"charset=euc_jp" in euc_jp)
-
-        shift_jis = soup.encode("shift-jis")
-        self.assertTrue(b"charset=shift-jis" in shift_jis)
-
-        utf_16_u = soup.encode("utf-16").decode("utf-16")
-        self.assertTrue("charset=utf-16" in utf_16_u)
-
-    def test_encoding_substitution_doesnt_happen_if_tag_is_strained(self):
-        markup = ('
        foo
        ')
-
-        # Beautiful Soup used to try to rewrite the meta tag even if the
-        # meta tag got filtered out by the strainer. This test makes
-        # sure that doesn't happen.
-        strainer = SoupStrainer('pre')
-        soup = self.soup(markup, parse_only=strainer)
-        self.assertEqual(soup.contents[0].name, 'pre')
-
-class TestEncoding(SoupTest):
-    """Test the ability to encode objects into strings."""
-
-    def test_unicode_string_can_be_encoded(self):
-        html = "\N{SNOWMAN}"
-        soup = self.soup(html)
-        self.assertEqual(soup.b.string.encode("utf-8"),
-                          "\N{SNOWMAN}".encode("utf-8"))
-
-    def test_tag_containing_unicode_string_can_be_encoded(self):
-        html = "\N{SNOWMAN}"
-        soup = self.soup(html)
-        self.assertEqual(
-            soup.b.encode("utf-8"), html.encode("utf-8"))
-
-    def test_encoding_substitutes_unrecognized_characters_by_default(self):
-        html = "\N{SNOWMAN}"
-        soup = self.soup(html)
-        self.assertEqual(soup.b.encode("ascii"), b"☃")
-
-    def test_encoding_can_be_made_strict(self):
-        html = "\N{SNOWMAN}"
-        soup = self.soup(html)
-        self.assertRaises(
-            UnicodeEncodeError, soup.encode, "ascii", errors="strict")
-
-    def test_decode_contents(self):
-        html = "\N{SNOWMAN}"
-        soup = self.soup(html)
-        self.assertEqual("\N{SNOWMAN}", soup.b.decode_contents())
-
-    def test_encode_contents(self):
-        html = "\N{SNOWMAN}"
-        soup = self.soup(html)
-        self.assertEqual(
-            "\N{SNOWMAN}".encode("utf8"), soup.b.encode_contents(
-                encoding="utf8"))
-
-    def test_deprecated_renderContents(self):
-        html = "\N{SNOWMAN}"
-        soup = self.soup(html)
-        self.assertEqual(
-            "\N{SNOWMAN}".encode("utf8"), soup.b.renderContents())
-
-    def test_repr(self):
-        html = "\N{SNOWMAN}"
-        soup = self.soup(html)
-        if PY3K:
-            self.assertEqual(html, repr(soup))
-        else:
-            self.assertEqual(b'\\u2603', repr(soup))
-
-class TestNavigableStringSubclasses(SoupTest):
-
-    def test_cdata(self):
-        # None of the current builders turn CDATA sections into CData
-        # objects, but you can create them manually.
-        soup = self.soup("")
-        cdata = CData("foo")
-        soup.insert(1, cdata)
-        self.assertEqual(str(soup), "")
-        self.assertEqual(soup.find(text="foo"), "foo")
-        self.assertEqual(soup.contents[0], "foo")
-
-    def test_cdata_is_never_formatted(self):
-        """Text inside a CData object is passed into the formatter.
-
-        But the return value is ignored.
-        """
-
-        self.count = 0
-        def increment(*args):
-            self.count += 1
-            return "BITTER FAILURE"
-
-        soup = self.soup("")
-        cdata = CData("<><><>")
-        soup.insert(1, cdata)
-        self.assertEqual(
-            b"<><>]]>", soup.encode(formatter=increment))
-        self.assertEqual(1, self.count)
-
-    def test_doctype_ends_in_newline(self):
-        # Unlike other NavigableString subclasses, a DOCTYPE always ends
-        # in a newline.
-        doctype = Doctype("foo")
-        soup = self.soup("")
-        soup.insert(1, doctype)
-        self.assertEqual(soup.encode(), b"\n")
-
-    def test_declaration(self):
-        d = Declaration("foo")
-        self.assertEqual("", d.output_ready())
-
-class TestSoupSelector(TreeTest):
-
-    HTML = """
-
-
-
-The title
-
-
-
-Hello there.
-
        
-
        
-
        An H1
        
-
        Some text
        
-
        Some more text
        
-
        An H2
        
-
        Another
        
-Bob
-
        Another H2
        
-me
-
-span1a1
-span1a2 test
-
-span2a1
-
-
-
-
        
-
-
        
-
-
-
-
-
-
-
-
-
        English
        
-
        English UK
        
-
        English US
        
-
        French
        
-
        
-
-
-"""
-
-    def setUp(self):
-        self.soup = BeautifulSoup(self.HTML, 'html.parser')
-
-    def assertSelects(self, selector, expected_ids):
-        el_ids = [el['id'] for el in self.soup.select(selector)]
-        el_ids.sort()
-        expected_ids.sort()
-        self.assertEqual(expected_ids, el_ids,
-            "Selector %s, expected [%s], got [%s]" % (
-                selector, ', '.join(expected_ids), ', '.join(el_ids)
-            )
-        )
-
-    assertSelect = assertSelects
-
-    def assertSelectMultiple(self, *tests):
-        for selector, expected_ids in tests:
-            self.assertSelect(selector, expected_ids)
-
-    def test_one_tag_one(self):
-        els = self.soup.select('title')
-        self.assertEqual(len(els), 1)
-        self.assertEqual(els[0].name, 'title')
-        self.assertEqual(els[0].contents, ['The title'])
-
-    def test_one_tag_many(self):
-        els = self.soup.select('div')
-        self.assertEqual(len(els), 4)
-        for div in els:
-            self.assertEqual(div.name, 'div')
-
-        el = self.soup.select_one('div')
-        self.assertEqual('main', el['id'])
-
-    def test_select_one_returns_none_if_no_match(self):
-        match = self.soup.select_one('nonexistenttag')
-        self.assertEqual(None, match)
-
-
-    def test_tag_in_tag_one(self):
-        els = self.soup.select('div div')
-        self.assertSelects('div div', ['inner', 'data1'])
-
-    def test_tag_in_tag_many(self):
-        for selector in ('html div', 'html body div', 'body div'):
-            self.assertSelects(selector, ['data1', 'main', 'inner', 'footer'])
-
-    def test_tag_no_match(self):
-        self.assertEqual(len(self.soup.select('del')), 0)
-
-    def test_invalid_tag(self):
-        self.assertRaises(ValueError, self.soup.select, 'tag%t')
-
-    def test_select_dashed_tag_ids(self):
-        self.assertSelects('custom-dashed-tag', ['dash1', 'dash2'])
-
-    def test_select_dashed_by_id(self):
-        dashed = self.soup.select('custom-dashed-tag[id=\"dash2\"]')
-        self.assertEqual(dashed[0].name, 'custom-dashed-tag')
-        self.assertEqual(dashed[0]['id'], 'dash2')
-
-    def test_dashed_tag_text(self):
-        self.assertEqual(self.soup.select('body > custom-dashed-tag')[0].text, 'Hello there.')
-
-    def test_select_dashed_matches_find_all(self):
-        self.assertEqual(self.soup.select('custom-dashed-tag'), self.soup.find_all('custom-dashed-tag'))
-
-    def test_header_tags(self):
-        self.assertSelectMultiple(
-            ('h1', ['header1']),
-            ('h2', ['header2', 'header3']),
-        )
-
-    def test_class_one(self):
-        for selector in ('.onep', 'p.onep', 'html p.onep'):
-            els = self.soup.select(selector)
-            self.assertEqual(len(els), 1)
-            self.assertEqual(els[0].name, 'p')
-            self.assertEqual(els[0]['class'], ['onep'])
-
-    def test_class_mismatched_tag(self):
-        els = self.soup.select('div.onep')
-        self.assertEqual(len(els), 0)
-
-    def test_one_id(self):
-        for selector in ('div#inner', '#inner', 'div div#inner'):
-            self.assertSelects(selector, ['inner'])
-
-    def test_bad_id(self):
-        els = self.soup.select('#doesnotexist')
-        self.assertEqual(len(els), 0)
-
-    def test_items_in_id(self):
-        els = self.soup.select('div#inner p')
-        self.assertEqual(len(els), 3)
-        for el in els:
-            self.assertEqual(el.name, 'p')
-        self.assertEqual(els[1]['class'], ['onep'])
-        self.assertFalse(els[0].has_attr('class'))
-
-    def test_a_bunch_of_emptys(self):
-        for selector in ('div#main del', 'div#main div.oops', 'div div#main'):
-            self.assertEqual(len(self.soup.select(selector)), 0)
-
-    def test_multi_class_support(self):
-        for selector in ('.class1', 'p.class1', '.class2', 'p.class2',
-            '.class3', 'p.class3', 'html p.class2', 'div#inner .class2'):
-            self.assertSelects(selector, ['pmulti'])
-
-    def test_multi_class_selection(self):
-        for selector in ('.class1.class3', '.class3.class2',
-                         '.class1.class2.class3'):
-            self.assertSelects(selector, ['pmulti'])
-
-    def test_child_selector(self):
-        self.assertSelects('.s1 > a', ['s1a1', 's1a2'])
-        self.assertSelects('.s1 > a span', ['s1a2s1'])
-
-    def test_child_selector_id(self):
-        self.assertSelects('.s1 > a#s1a2 span', ['s1a2s1'])
-
-    def test_attribute_equals(self):
-        self.assertSelectMultiple(
-            ('p[class="onep"]', ['p1']),
-            ('p[id="p1"]', ['p1']),
-            ('[class="onep"]', ['p1']),
-            ('[id="p1"]', ['p1']),
-            ('link[rel="stylesheet"]', ['l1']),
-            ('link[type="text/css"]', ['l1']),
-            ('link[href="blah.css"]', ['l1']),
-            ('link[href="no-blah.css"]', []),
-            ('[rel="stylesheet"]', ['l1']),
-            ('[type="text/css"]', ['l1']),
-            ('[href="blah.css"]', ['l1']),
-            ('[href="no-blah.css"]', []),
-            ('p[href="no-blah.css"]', []),
-            ('[href="no-blah.css"]', []),
-        )
-
-    def test_attribute_tilde(self):
-        self.assertSelectMultiple(
-            ('p[class~="class1"]', ['pmulti']),
-            ('p[class~="class2"]', ['pmulti']),
-            ('p[class~="class3"]', ['pmulti']),
-            ('[class~="class1"]', ['pmulti']),
-            ('[class~="class2"]', ['pmulti']),
-            ('[class~="class3"]', ['pmulti']),
-            ('a[rel~="friend"]', ['bob']),
-            ('a[rel~="met"]', ['bob']),
-            ('[rel~="friend"]', ['bob']),
-            ('[rel~="met"]', ['bob']),
-        )
-
-    def test_attribute_startswith(self):
-        self.assertSelectMultiple(
-            ('[rel^="style"]', ['l1']),
-            ('link[rel^="style"]', ['l1']),
-            ('notlink[rel^="notstyle"]', []),
-            ('[rel^="notstyle"]', []),
-            ('link[rel^="notstyle"]', []),
-            ('link[href^="bla"]', ['l1']),
-            ('a[href^="http://"]', ['bob', 'me']),
-            ('[href^="http://"]', ['bob', 'me']),
-            ('[id^="p"]', ['pmulti', 'p1']),
-            ('[id^="m"]', ['me', 'main']),
-            ('div[id^="m"]', ['main']),
-            ('a[id^="m"]', ['me']),
-            ('div[data-tag^="dashed"]', ['data1'])
-        )
-
-    def test_attribute_endswith(self):
-        self.assertSelectMultiple(
-            ('[href$=".css"]', ['l1']),
-            ('link[href$=".css"]', ['l1']),
-            ('link[id$="1"]', ['l1']),
-            ('[id$="1"]', ['data1', 'l1', 'p1', 'header1', 's1a1', 's2a1', 's1a2s1', 'dash1']),
-            ('div[id$="1"]', ['data1']),
-            ('[id$="noending"]', []),
-        )
-
-    def test_attribute_contains(self):
-        self.assertSelectMultiple(
-            # From test_attribute_startswith
-            ('[rel*="style"]', ['l1']),
-            ('link[rel*="style"]', ['l1']),
-            ('notlink[rel*="notstyle"]', []),
-            ('[rel*="notstyle"]', []),
-            ('link[rel*="notstyle"]', []),
-            ('link[href*="bla"]', ['l1']),
-            ('[href*="http://"]', ['bob', 'me']),
-            ('[id*="p"]', ['pmulti', 'p1']),
-            ('div[id*="m"]', ['main']),
-            ('a[id*="m"]', ['me']),
-            # From test_attribute_endswith
-            ('[href*=".css"]', ['l1']),
-            ('link[href*=".css"]', ['l1']),
-            ('link[id*="1"]', ['l1']),
-            ('[id*="1"]', ['data1', 'l1', 'p1', 'header1', 's1a1', 's1a2', 's2a1', 's1a2s1', 'dash1']),
-            ('div[id*="1"]', ['data1']),
-            ('[id*="noending"]', []),
-            # New for this test
-            ('[href*="."]', ['bob', 'me', 'l1']),
-            ('a[href*="."]', ['bob', 'me']),
-            ('link[href*="."]', ['l1']),
-            ('div[id*="n"]', ['main', 'inner']),
-            ('div[id*="nn"]', ['inner']),
-            ('div[data-tag*="edval"]', ['data1'])
-        )
-
-    def test_attribute_exact_or_hypen(self):
-        self.assertSelectMultiple(
-            ('p[lang|="en"]', ['lang-en', 'lang-en-gb', 'lang-en-us']),
-            ('[lang|="en"]', ['lang-en', 'lang-en-gb', 'lang-en-us']),
-            ('p[lang|="fr"]', ['lang-fr']),
-            ('p[lang|="gb"]', []),
-        )
-
-    def test_attribute_exists(self):
-        self.assertSelectMultiple(
-            ('[rel]', ['l1', 'bob', 'me']),
-            ('link[rel]', ['l1']),
-            ('a[rel]', ['bob', 'me']),
-            ('[lang]', ['lang-en', 'lang-en-gb', 'lang-en-us', 'lang-fr']),
-            ('p[class]', ['p1', 'pmulti']),
-            ('[blah]', []),
-            ('p[blah]', []),
-            ('div[data-tag]', ['data1'])
-        )
-
-    def test_unsupported_pseudoclass(self):
-        self.assertRaises(
-            NotImplementedError, self.soup.select, "a:no-such-pseudoclass")
-
-        self.assertRaises(
-            NotImplementedError, self.soup.select, "a:nth-of-type(a)")
-
-
-    def test_nth_of_type(self):
-        # Try to select first paragraph
-        els = self.soup.select('div#inner p:nth-of-type(1)')
-        self.assertEqual(len(els), 1)
-        self.assertEqual(els[0].string, 'Some text')
-
-        # Try to select third paragraph
-        els = self.soup.select('div#inner p:nth-of-type(3)')
-        self.assertEqual(len(els), 1)
-        self.assertEqual(els[0].string, 'Another')
-
-        # Try to select (non-existent!) fourth paragraph
-        els = self.soup.select('div#inner p:nth-of-type(4)')
-        self.assertEqual(len(els), 0)
-
-        # Pass in an invalid value.
-        self.assertRaises(
-            ValueError, self.soup.select, 'div p:nth-of-type(0)')
-
-    def test_nth_of_type_direct_descendant(self):
-        els = self.soup.select('div#inner > p:nth-of-type(1)')
-        self.assertEqual(len(els), 1)
-        self.assertEqual(els[0].string, 'Some text')
-
-    def test_id_child_selector_nth_of_type(self):
-        self.assertSelects('#inner > p:nth-of-type(2)', ['p1'])
-
-    def test_select_on_element(self):
-        # Other tests operate on the tree; this operates on an element
-        # within the tree.
-        inner = self.soup.find("div", id="main")
-        selected = inner.select("div")
-        # The 
         tag was selected. The