diff --git a/lib/bs4/tests/__init__.py b/lib/bs4/tests/__init__.py deleted file mode 100644 index 142c8cc3..00000000 --- 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 90cad829..00000000 --- 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 5b9f6770..00000000 --- a/lib/bs4/tests/test_docs.py +++ /dev/null @@ -1,36 +0,0 @@ -"Test harness for doctests." - -# pylint: disable-msg=E0611,W0142 - -__metaclass__ = type -__all__ = [ - 'additional_tests', - ] - -import atexit -import doctest -import os -#from pkg_resources import ( -# resource_filename, resource_exists, resource_listdir, cleanup_resources) -import unittest - -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 f8902ad7..00000000 --- a/lib/bs4/tests/test_html5lib.py +++ /dev/null @@ -1,226 +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_reparented_markup_containing_identical_whitespace_nodes(self): - """Verify that we keep the two whitespace nodes in this - document distinct when reparenting the adjacent tags. - """ - markup = '
' - soup = self.soup(markup) - space1, space2 = soup.find_all(string=' ') - tbody1, tbody2 = soup.find_all('tbody') - assert space1.next_element is tbody1 - assert tbody2.next_element is space2 - - def test_reparented_markup_containing_children(self): - markup = '
aftermath

aftermath

' - soup = self.soup(markup) - noscript = soup.noscript - self.assertEqual("target", noscript.next_element) - target = soup.find(string='target') - - # The 'aftermath' string was duplicated; we want the second one. - final_aftermath = soup.find_all(string='aftermath')[-1] - - # The