diff options
| -rw-r--r-- | CHANGELOG | 4 | ||||
| -rw-r--r-- | bs4/testing.py | 306 | ||||
| -rw-r--r-- | bs4/tests/test_builder_registry.py | 73 | ||||
| -rw-r--r-- | bs4/tests/test_dammit.py | 112 | ||||
| -rw-r--r-- | bs4/tests/test_docs.py | 2 | ||||
| -rw-r--r-- | bs4/tests/test_formatter.py | 26 | ||||
| -rw-r--r-- | bs4/tests/test_html5lib.py | 51 | ||||
| -rw-r--r-- | bs4/tests/test_htmlparser.py | 36 | ||||
| -rw-r--r-- | bs4/tests/test_lxml.py | 21 | ||||
| -rw-r--r-- | bs4/tests/test_navigablestring.py | 59 | ||||
| -rw-r--r-- | bs4/tests/test_soup.py | 193 | ||||
| -rw-r--r-- | bs4/tests/test_tree.py | 1160 | ||||
| -rw-r--r-- | setup.py | 3 |
13 files changed, 981 insertions, 1065 deletions
@@ -3,6 +3,10 @@ Beautiful Soup's official support for Python 2 ended on December 31st, 4.9.3. In the Launchpad Bazaar repository, the final revision to support Python 2 was revision 605. += 4.11.0 (Unreleased) + +* Ported unit tests to use pytest. + = 4.10.0 (20210907) * This is the first release of Beautiful Soup to only support Python diff --git a/bs4/testing.py b/bs4/testing.py index 2f9046a..97914c0 100644 --- a/bs4/testing.py +++ b/bs4/testing.py @@ -7,9 +7,8 @@ __license__ = "MIT" import pickle import copy import functools -import unittest import warnings -from unittest import TestCase +import pytest from bs4 import BeautifulSoup from bs4.element import ( CharsetMetaAttributeValue, @@ -63,7 +62,7 @@ BAD_DOCUMENT = """A bare string """ -class SoupTest(unittest.TestCase): +class SoupTest(object): @property def default_builder(self): @@ -80,15 +79,18 @@ class SoupTest(unittest.TestCase): The details depend on the builder. """ return self.default_builder(**kwargs).test_fragment_to_document(markup) - - def assertSoupEquals(self, to_parse, compare_parsed_to=None): + + def assert_soup(self, to_parse, compare_parsed_to=None): + """Parse some markup using Beautiful Soup and verify that + the output markup is as expected. + """ builder = self.default_builder obj = BeautifulSoup(to_parse, builder=builder) if compare_parsed_to is None: compare_parsed_to = to_parse # Verify that the documents come out the same. - self.assertEqual(obj.decode(), self.document_for(compare_parsed_to)) + assert obj.decode() == self.document_for(compare_parsed_to) # Also run some checks on the BeautifulSoup object itself: @@ -99,9 +101,9 @@ class SoupTest(unittest.TestCase): # The only tag in the tag stack is the one for the root # document. - self.assertEqual( - [obj.ROOT_TAG_NAME], [x.name for x in obj.tagStack] - ) + assert [obj.ROOT_TAG_NAME] == [x.name for x in obj.tagStack] + + assertSoupEquals = assert_soup def assertConnectedness(self, element): """Ensure that next_element and previous_element are properly @@ -110,8 +112,8 @@ class SoupTest(unittest.TestCase): earlier = None for e in element.descendants: if earlier: - self.assertEqual(e, earlier.next_element) - self.assertEqual(earlier, e.previous_element) + assert e == earlier.next_element + assert earlier == e.previous_element earlier = e def linkage_validator(self, el, _recursive_call=False): @@ -283,7 +285,7 @@ class HTMLTreeBuilderSmokeTest(TreeBuilderSmokeTest): ]: soup = self.soup("") new_tag = soup.new_tag(name) - self.assertEqual(True, new_tag.is_empty_element) + assert new_tag.is_empty_element == True def test_special_string_containers(self): soup = self.soup( @@ -298,7 +300,7 @@ class HTMLTreeBuilderSmokeTest(TreeBuilderSmokeTest): assert isinstance(soup.style.string, Stylesheet) # The contents of the style tag resemble an HTML comment, but # it's not treated as a comment. - self.assertEqual("<!--Some CSS-->", soup.style.string) + assert soup.style.string == "<!--Some CSS-->" assert isinstance(soup.style.string, Stylesheet) def test_pickle_and_unpickle_identity(self): @@ -307,8 +309,8 @@ class HTMLTreeBuilderSmokeTest(TreeBuilderSmokeTest): tree = self.soup("<a><b>foo</a>") dumped = pickle.dumps(tree, 2) loaded = pickle.loads(dumped) - self.assertEqual(loaded.__class__, BeautifulSoup) - self.assertEqual(loaded.decode(), tree.decode()) + assert loaded.__class__ == BeautifulSoup + assert loaded.decode() == tree.decode() def assertDoctypeHandled(self, doctype_fragment): """Assert that a given doctype string is handled correctly.""" @@ -316,16 +318,13 @@ class HTMLTreeBuilderSmokeTest(TreeBuilderSmokeTest): # Make sure a Doctype object was created. doctype = soup.contents[0] - self.assertEqual(doctype.__class__, Doctype) - self.assertEqual(doctype, doctype_fragment) - self.assertEqual( - soup.encode("utf8")[:len(doctype_str)], - doctype_str - ) + assert doctype.__class__ == Doctype + assert doctype == doctype_fragment + assert soup.encode("utf8")[:len(doctype_str)] == doctype_str # Make sure that the doctype was correctly associated with the # parse tree and that the rest of the document parsed. - self.assertEqual(soup.p.contents[0], 'foo') + assert soup.p.contents[0] == 'foo' def _document_with_doctype(self, doctype_fragment, doctype_string="DOCTYPE"): """Generate and parse a document with the given doctype.""" @@ -343,7 +342,7 @@ class HTMLTreeBuilderSmokeTest(TreeBuilderSmokeTest): def test_empty_doctype(self): soup = self.soup("<!DOCTYPE>") doctype = soup.contents[0] - self.assertEqual("", doctype.strip()) + assert "" == doctype.strip() def test_mixed_case_doctype(self): # A lowercase or mixed-case doctype becomes a Doctype. @@ -355,16 +354,13 @@ class HTMLTreeBuilderSmokeTest(TreeBuilderSmokeTest): # Make sure a Doctype object was created and that the DOCTYPE # is uppercase. doctype = soup.contents[0] - self.assertEqual(doctype.__class__, Doctype) - self.assertEqual(doctype, "html") - self.assertEqual( - soup.encode("utf8")[:len(doctype_str)], - b"<!DOCTYPE html>" - ) + assert doctype.__class__ == Doctype + assert doctype == "html" + assert soup.encode("utf8")[:len(doctype_str)] == b"<!DOCTYPE html>" # Make sure that the doctype was correctly associated with the # parse tree and that the rest of the document parsed. - self.assertEqual(soup.p.contents[0], 'foo') + assert soup.p.contents[0] == 'foo' def test_public_doctype_with_url(self): doctype = 'html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"' @@ -390,9 +386,7 @@ class HTMLTreeBuilderSmokeTest(TreeBuilderSmokeTest): <body>Goodbye.</body> </html>""" soup = self.soup(markup) - self.assertEqual( - soup.encode("utf-8").replace(b"\n", b""), - markup.replace(b"\n", b"")) + assert soup.encode("utf-8").replace(b"\n", b"") == markup.replace(b"\n", b"") def test_namespaced_html(self): """When a namespaced XML document is parsed as HTML it should @@ -400,7 +394,7 @@ class HTMLTreeBuilderSmokeTest(TreeBuilderSmokeTest): """ markup = b"""<ns1:foo>content</ns1:foo><ns1:foo/><ns2:foo/>""" soup = self.soup(markup) - self.assertEqual(2, len(soup.find_all("ns1:foo"))) + assert 2 == len(soup.find_all("ns1:foo")) def test_processing_instruction(self): # We test both Unicode and bytestring to verify that @@ -409,11 +403,11 @@ class HTMLTreeBuilderSmokeTest(TreeBuilderSmokeTest): # need to process anything. markup = """<?PITarget PIContent?>""" soup = self.soup(markup) - self.assertEqual(markup, soup.decode()) + assert markup == soup.decode() markup = b"""<?PITarget PIContent?>""" soup = self.soup(markup) - self.assertEqual(markup, soup.encode("utf8")) + assert markup == soup.encode("utf8") def test_deepcopy(self): """Make sure you can copy the tree builder. @@ -430,18 +424,18 @@ class HTMLTreeBuilderSmokeTest(TreeBuilderSmokeTest): shouldn't be presented that way. """ soup = self.soup("<p/>") - self.assertFalse(soup.p.is_empty_element) - self.assertEqual(str(soup.p), "<p></p>") + assert not soup.p.is_empty_element + assert str(soup.p) == "<p></p>" def test_unclosed_tags_get_closed(self): """A tag that's not closed by the end of the document should be closed. This applies to all tags except empty-element tags. """ - self.assertSoupEquals("<p>", "<p></p>") - self.assertSoupEquals("<b>", "<b></b>") + self.assert_soup("<p>", "<p></p>") + self.assert_soup("<b>", "<b></b>") - self.assertSoupEquals("<br>", "<br/>") + self.assert_soup("<br>", "<br/>") def test_br_is_always_empty_element_tag(self): """A <br> tag is designated as an empty-element tag. @@ -450,11 +444,11 @@ class HTMLTreeBuilderSmokeTest(TreeBuilderSmokeTest): two tags, but it should always be an empty-element tag. """ soup = self.soup("<br></br>") - self.assertTrue(soup.br.is_empty_element) - self.assertEqual(str(soup.br), "<br/>") + assert soup.br.is_empty_element + assert str(soup.br) == "<br/>" def test_nested_formatting_elements(self): - self.assertSoupEquals("<em><em></em></em>") + self.assert_soup("<em><em></em></em>") def test_double_head(self): html = '''<!DOCTYPE html> @@ -471,22 +465,22 @@ Hello, world! </html> ''' soup = self.soup(html) - self.assertEqual("text/javascript", soup.find('script')['type']) + assert "text/javascript" == soup.find('script')['type'] def test_comment(self): # Comments are represented as Comment objects. markup = "<p>foo<!--foobar-->baz</p>" - self.assertSoupEquals(markup) + self.assert_soup(markup) soup = self.soup(markup) comment = soup.find(text="foobar") - self.assertEqual(comment.__class__, Comment) + assert comment.__class__ == Comment # The comment is properly integrated into the tree. foo = soup.find(text="foo") - self.assertEqual(comment, foo.next_element) + assert comment == foo.next_element baz = soup.find(text="baz") - self.assertEqual(comment, baz.previous_element) + assert comment == baz.previous_element def test_preserved_whitespace_in_pre_and_textarea(self): """Whitespace must be preserved in <pre> and <textarea> tags, @@ -494,35 +488,35 @@ Hello, world! """ pre_markup = "<pre> </pre>" textarea_markup = "<textarea> woo\nwoo </textarea>" - self.assertSoupEquals(pre_markup) - self.assertSoupEquals(textarea_markup) + self.assert_soup(pre_markup) + self.assert_soup(textarea_markup) soup = self.soup(pre_markup) - self.assertEqual(soup.pre.prettify(), pre_markup) + assert soup.pre.prettify() == pre_markup soup = self.soup(textarea_markup) - self.assertEqual(soup.textarea.prettify(), textarea_markup) + assert soup.textarea.prettify() == textarea_markup soup = self.soup("<textarea></textarea>") - self.assertEqual(soup.textarea.prettify(), "<textarea></textarea>") + assert soup.textarea.prettify() == "<textarea></textarea>" def test_nested_inline_elements(self): """Inline elements can be nested indefinitely.""" b_tag = "<b>Inside a B tag</b>" - self.assertSoupEquals(b_tag) + self.assert_soup(b_tag) nested_b_tag = "<p>A <i>nested <b>tag</b></i></p>" - self.assertSoupEquals(nested_b_tag) + self.assert_soup(nested_b_tag) double_nested_b_tag = "<p>A <a>doubly <i>nested <b>tag</b></i></a></p>" - self.assertSoupEquals(nested_b_tag) + self.assert_soup(nested_b_tag) def test_nested_block_level_elements(self): """Block elements can be nested.""" soup = self.soup('<blockquote><p><b>Foo</b></p></blockquote>') blockquote = soup.blockquote - self.assertEqual(blockquote.p.b.string, 'Foo') - self.assertEqual(blockquote.b.string, 'Foo') + assert blockquote.p.b.string == 'Foo' + assert blockquote.b.string == 'Foo' def test_correctly_nested_tables(self): """One table can go inside another one.""" @@ -533,13 +527,13 @@ Hello, world! '<tr><td>foo</td></tr>' '</table></td>') - self.assertSoupEquals( + self.assert_soup( markup, '<table id="1"><tr><td>Here\'s another table:' '<table id="2"><tr><td>foo</td></tr></table>' '</td></tr></table>') - self.assertSoupEquals( + self.assert_soup( "<table><thead><tr><td>Foo</td></tr></thead>" "<tbody><tr><td>Bar</td></tr></tbody>" "<tfoot><tr><td>Baz</td></tr></tfoot></table>") @@ -550,11 +544,11 @@ Hello, world! markup = '<div class=" foo bar "></a>' soup = self.soup(markup) - self.assertEqual(['foo', 'bar'], soup.div['class']) + assert ['foo', 'bar'] == soup.div['class'] # If you search by the literal name of the class it's like the whitespace # wasn't there. - self.assertEqual(soup.div, soup.find('div', class_="foo bar")) + assert soup.div == soup.find('div', class_="foo bar") def test_deeply_nested_multivalued_attribute(self): # html5lib can set the attributes of the same tag many times @@ -562,7 +556,7 @@ Hello, world! # multivalued attributes. markup = '<table><div><div class="css"></div></div></table>' soup = self.soup(markup) - self.assertEqual(["css"], soup.div.div['class']) + assert ["css"] == soup.div.div['class'] def test_multivalued_attribute_on_html(self): # html5lib uses a different API to set the attributes ot the @@ -570,21 +564,21 @@ Hello, world! # attributes. markup = '<html class="a b"></html>' soup = self.soup(markup) - self.assertEqual(["a", "b"], soup.html['class']) + assert ["a", "b"] == soup.html['class'] def test_angle_brackets_in_attribute_values_are_escaped(self): - self.assertSoupEquals('<a b="<a>"></a>', '<a b="<a>"></a>') + self.assert_soup('<a b="<a>"></a>', '<a b="<a>"></a>') def test_strings_resembling_character_entity_references(self): # "&T" and "&p" look like incomplete character entities, but they are # not. - self.assertSoupEquals( + self.assert_soup( "<p>• AT&T is in the s&p 500</p>", "<p>\u2022 AT&T is in the s&p 500</p>" ) def test_apos_entity(self): - self.assertSoupEquals( + self.assert_soup( "<p>Bob's Bar</p>", "<p>Bob's Bar</p>", ) @@ -599,45 +593,45 @@ Hello, world! # characters. markup = "<p>“Hello” -☃</p>" soup = self.soup(markup) - self.assertEqual("“Hello” -☃", soup.p.string) + assert "“Hello” -☃" == soup.p.string def test_entities_in_attributes_converted_to_unicode(self): expect = '<p id="pi\N{LATIN SMALL LETTER N WITH TILDE}ata"></p>' - self.assertSoupEquals('<p id="piñata"></p>', expect) - self.assertSoupEquals('<p id="piñata"></p>', expect) - self.assertSoupEquals('<p id="piñata"></p>', expect) - self.assertSoupEquals('<p id="piñata"></p>', expect) + self.assert_soup('<p id="piñata"></p>', expect) + self.assert_soup('<p id="piñata"></p>', expect) + self.assert_soup('<p id="piñata"></p>', expect) + self.assert_soup('<p id="piñata"></p>', expect) def test_entities_in_text_converted_to_unicode(self): expect = '<p>pi\N{LATIN SMALL LETTER N WITH TILDE}ata</p>' - self.assertSoupEquals("<p>piñata</p>", expect) - self.assertSoupEquals("<p>piñata</p>", expect) - self.assertSoupEquals("<p>piñata</p>", expect) - self.assertSoupEquals("<p>piñata</p>", expect) + self.assert_soup("<p>piñata</p>", expect) + self.assert_soup("<p>piñata</p>", expect) + self.assert_soup("<p>piñata</p>", expect) + self.assert_soup("<p>piñata</p>", expect) def test_quot_entity_converted_to_quotation_mark(self): - self.assertSoupEquals("<p>I said "good day!"</p>", + self.assert_soup("<p>I said "good day!"</p>", '<p>I said "good day!"</p>') def test_out_of_range_entity(self): expect = "\N{REPLACEMENT CHARACTER}" - self.assertSoupEquals("�", expect) - self.assertSoupEquals("�", expect) - self.assertSoupEquals("�", expect) + self.assert_soup("�", expect) + self.assert_soup("�", expect) + self.assert_soup("�", expect) def test_multipart_strings(self): "Mostly to prevent a recurrence of a bug in the html5lib treebuilder." soup = self.soup("<html><h2>\nfoo</h2><p></p></html>") - self.assertEqual("p", soup.h2.string.next_element.name) - self.assertEqual("p", soup.p.name) + assert "p" == soup.h2.string.next_element.name + assert "p" == soup.p.name self.assertConnectedness(soup) def test_empty_element_tags(self): """Verify consistent handling of empty-element tags, no matter how they come in through the markup. """ - self.assertSoupEquals('<br/><br/><br/>', "<br/><br/><br/>") - self.assertSoupEquals('<br /><br /><br />', "<br/><br/><br/>") + self.assert_soup('<br/><br/><br/>', "<br/><br/><br/>") + self.assert_soup('<br /><br /><br />', "<br/><br/><br/>") def test_head_tag_between_head_and_body(self): "Prevent recurrence of a bug in the html5lib treebuilder." @@ -647,7 +641,7 @@ Hello, world! </html> """ soup = self.soup(content) - self.assertNotEqual(None, soup.html.body) + assert soup.html.body is not None self.assertConnectedness(soup) def test_multiple_copies_of_a_tag(self): @@ -674,18 +668,16 @@ Hello, world! markup = b'<html xmlns="http://www.w3.org/1999/xhtml" xmlns:mathml="http://www.w3.org/1998/Math/MathML" xmlns:svg="http://www.w3.org/2000/svg"><head></head><body><mathml:msqrt>4</mathml:msqrt><b svg:fill="red"></b></body></html>' soup = self.soup(markup) - self.assertEqual(markup, soup.encode()) + assert 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']) + assert 'http://www.w3.org/1999/xhtml' == soup.html['xmlns'] + assert 'http://www.w3.org/1998/Math/MathML' == soup.html['xmlns:mathml'] + assert 'http://www.w3.org/2000/svg' == soup.html['xmlns:svg'] def test_multivalued_attribute_value_becomes_list(self): markup = b'<a class="foo bar">' soup = self.soup(markup) - self.assertEqual(['foo', 'bar'], soup.a['class']) + assert ['foo', 'bar'] == soup.a['class'] # # Generally speaking, tests below this point are more tests of @@ -700,67 +692,65 @@ Hello, world! # encoding found in the declaration! The horror! markup = '<html><head><meta encoding="euc-jp"></head><body>Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!</body>' soup = self.soup(markup) - self.assertEqual('Sacr\xe9 bleu!', soup.body.string) + assert '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 <b>bold</b> <meta/> <i>statement</i>", parse_only=strainer) - self.assertEqual(soup.decode(), "<b>bold</b>") + assert soup.decode() == "<b>bold</b>" def test_single_quote_attribute_values_become_double_quotes(self): - self.assertSoupEquals("<foo attr='bar'></foo>", + self.assert_soup("<foo attr='bar'></foo>", '<foo attr="bar"></foo>') def test_attribute_values_with_nested_quotes_are_left_alone(self): text = """<foo attr='bar "brawls" happen'>a</foo>""" - self.assertSoupEquals(text) + self.assert_soup(text) def test_attribute_values_with_double_nested_quotes_get_quoted(self): text = """<foo attr='bar "brawls" happen'>a</foo>""" soup = self.soup(text) soup.foo['attr'] = 'Brawls happen at "Bob\'s Bar"' - self.assertSoupEquals( + self.assert_soup( soup.foo.decode(), """<foo attr="Brawls happen at "Bob\'s Bar"">a</foo>""") def test_ampersand_in_attribute_value_gets_escaped(self): - self.assertSoupEquals('<this is="really messed up & stuff"></this>', + self.assert_soup('<this is="really messed up & stuff"></this>', '<this is="really messed up & stuff"></this>') - self.assertSoupEquals( + self.assert_soup( '<a href="http://example.org?a=1&b=2;3">foo</a>', '<a href="http://example.org?a=1&b=2;3">foo</a>') def test_escaped_ampersand_in_attribute_value_is_left_alone(self): - self.assertSoupEquals('<a href="http://example.org?a=1&b=2;3"></a>') + self.assert_soup('<a href="http://example.org?a=1&b=2;3"></a>') def test_entities_in_strings_converted_during_parsing(self): # Both XML and HTML entities are converted to Unicode characters # during parsing. text = "<p><<sacré bleu!>></p>" expected = "<p><<sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>></p>" - self.assertSoupEquals(text, expected) + self.assert_soup(text, expected) def test_smart_quotes_converted_on_the_way_in(self): # Microsoft smart quotes are converted to Unicode characters during # parsing. quote = b"<p>\x91Foo\x92</p>" soup = self.soup(quote) - self.assertEqual( - soup.p.string, - "\N{LEFT SINGLE QUOTATION MARK}Foo\N{RIGHT SINGLE QUOTATION MARK}") + assert 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("<a> </a>") - self.assertEqual(soup.a.string, "\N{NO-BREAK SPACE}" * 2) + assert soup.a.string == "\N{NO-BREAK SPACE}" * 2 def test_entities_converted_on_the_way_out(self): text = "<p><<sacré bleu!>></p>" expected = "<p><<sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>></p>".encode("utf-8") soup = self.soup(text) - self.assertEqual(soup.p.encode("utf-8"), expected) + assert soup.p.encode("utf-8") == expected def test_real_iso_latin_document(self): # Smoke test of interrelated functionality, using an @@ -787,7 +777,7 @@ Hello, world! expected = expected.encode("utf-8") # Ta-da! - self.assertEqual(result, expected) + assert result == expected def test_real_shift_jis_document(self): # Smoke test to make sure the parser can handle a document in @@ -803,8 +793,8 @@ Hello, world! # 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")) + assert soup.encode("utf-8") == unicode_html.encode("utf-8") + assert 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 @@ -815,9 +805,9 @@ Hello, world! # Some tree builders call it iso8859-8, others call it iso-8859-9. # That's not a difference we really care about. assert soup.original_encoding in ('iso8859-8', 'iso-8859-8') - self.assertEqual( - soup.encode('utf-8'), - hebrew_document.decode("iso8859-8").encode("utf-8")) + assert soup.encode('utf-8') == ( + hebrew_document.decode("iso8859-8").encode("utf-8") + ) def test_meta_tag_reflects_current_encoding(self): # Here's the <meta> tag saying that a document is @@ -835,14 +825,14 @@ Hello, world! # 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) + assert 'text/html; charset=x-sjis' == content # But that value is actually a ContentMetaAttributeValue object. - self.assertTrue(isinstance(content, ContentMetaAttributeValue)) + assert isinstance(content, ContentMetaAttributeValue) # And it will take on a value that reflects its current # encoding. - self.assertEqual('text/html; charset=utf8', content.encode("utf8")) + assert 'text/html; charset=utf8' == content.encode("utf8") # For the rest of the story, see TestSubstitutions in # test_tree.py. @@ -862,14 +852,14 @@ Hello, world! # 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) + assert 'x-sjis' == charset # But that value is actually a CharsetMetaAttributeValue object. - self.assertTrue(isinstance(charset, CharsetMetaAttributeValue)) + assert isinstance(charset, CharsetMetaAttributeValue) # And it will take on a value that reflects its current # encoding. - self.assertEqual('utf8', charset.encode("utf8")) + assert 'utf8' == charset.encode("utf8") def test_python_specific_encodings_not_used_in_charset(self): # You can encode an HTML document using a Python-specific @@ -897,7 +887,7 @@ Hello, world! def test_tag_with_no_attributes_can_have_attributes_added(self): data = self.soup("<a>text</a>") data.a['foo'] = 'bar' - self.assertEqual('<a foo="bar">text</a>', data.a.decode()) + assert '<a foo="bar">text</a>' == data.a.decode() def test_closing_tag_with_no_opening_tag(self): # Without BeautifulSoup.open_tag_counter, the </span> tag will @@ -905,9 +895,7 @@ Hello, world! # for a <span> tag that wasn't there. The result is that 'text2' # will show up outside the body of the document. soup = self.soup("<body><div><p>text1</p></span>text2</div></body>") - self.assertEqual( - "<body><div><p>text1</p>text2</div></body>", soup.body.decode() - ) + assert "<body><div><p>text1</p>text2</div></body>" == soup.body.decode() def test_worst_case(self): """Test the worst case (currently) for linking issues.""" @@ -924,18 +912,17 @@ class XMLTreeBuilderSmokeTest(TreeBuilderSmokeTest): tree = self.soup("<a><b>foo</a>") dumped = pickle.dumps(tree, 2) loaded = pickle.loads(dumped) - self.assertEqual(loaded.__class__, BeautifulSoup) - self.assertEqual(loaded.decode(), tree.decode()) + assert loaded.__class__ == BeautifulSoup + assert loaded.decode() == tree.decode() def test_docstring_generated(self): soup = self.soup("<root/>") - self.assertEqual( - soup.encode(), b'<?xml version="1.0" encoding="utf-8"?>\n<root/>') + assert soup.encode() == b'<?xml version="1.0" encoding="utf-8"?>\n<root/>' def test_xml_declaration(self): markup = b"""<?xml version="1.0" encoding="utf8"?>\n<foo/>""" soup = self.soup(markup) - self.assertEqual(markup, soup.encode("utf8")) + assert markup == soup.encode("utf8") def test_python_specific_encodings_not_used_in_xml_declaration(self): # You can encode an XML document using a Python-specific @@ -959,7 +946,7 @@ class XMLTreeBuilderSmokeTest(TreeBuilderSmokeTest): def test_processing_instruction(self): markup = b"""<?xml version="1.0" encoding="utf8"?>\n<?PITarget PIContent?>""" soup = self.soup(markup) - self.assertEqual(markup, soup.encode("utf8")) + assert markup == soup.encode("utf8") def test_real_xhtml_document(self): """A real XHTML document should come out *exactly* the same as it went in.""" @@ -970,8 +957,7 @@ class XMLTreeBuilderSmokeTest(TreeBuilderSmokeTest): <body>Goodbye.</body> </html>""" soup = self.soup(markup) - self.assertEqual( - soup.encode("utf-8"), markup) + assert soup.encode("utf-8") == markup def test_nested_namespaces(self): doc = b"""<?xml version="1.0" encoding="utf-8"?> @@ -982,7 +968,7 @@ class XMLTreeBuilderSmokeTest(TreeBuilderSmokeTest): </child> </parent>""" soup = self.soup(doc) - self.assertEqual(doc, soup.encode()) + assert doc == soup.encode() def test_formatter_processes_script_tag_for_xml_documents(self): doc = """ @@ -994,24 +980,21 @@ class XMLTreeBuilderSmokeTest(TreeBuilderSmokeTest): # it later. soup.script.string = 'console.log("< < hey > > ");' encoded = soup.encode() - self.assertTrue(b"< < hey > >" in encoded) + assert b"< < hey > >" in encoded def test_can_parse_unicode_document(self): markup = '<?xml version="1.0" encoding="euc-jp"><root>Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!</root>' soup = self.soup(markup) - self.assertEqual('Sacr\xe9 bleu!', soup.root.string) + assert 'Sacr\xe9 bleu!' == soup.root.string def test_popping_namespaced_tag(self): markup = '<rss xmlns:dc="foo"><dc:creator>b</dc:creator><dc:date>2012-07-02T20:33:42Z</dc:date><dc:rights>c</dc:rights><image>d</image></rss>' soup = self.soup(markup) - self.assertEqual( - str(soup.rss), markup) + assert str(soup.rss) == markup def test_docstring_includes_correct_encoding(self): soup = self.soup("<root/>") - self.assertEqual( - soup.encode("latin1"), - b'<?xml version="1.0" encoding="latin1"?>\n<root/>') + assert soup.encode("latin1") == b'<?xml version="1.0" encoding="latin1"?>\n<root/>' def test_large_xml_document(self): """A large XML document should come out the same as it went in.""" @@ -1019,34 +1002,33 @@ class XMLTreeBuilderSmokeTest(TreeBuilderSmokeTest): + b'0' * (2**12) + b'</root>') soup = self.soup(markup) - self.assertEqual(soup.encode("utf-8"), markup) - + assert soup.encode("utf-8") == markup def test_tags_are_empty_element_if_and_only_if_they_are_empty(self): - self.assertSoupEquals("<p>", "<p/>") - self.assertSoupEquals("<p>foo</p>") + self.assert_soup("<p>", "<p/>") + self.assert_soup("<p>foo</p>") def test_namespaces_are_preserved(self): markup = '<root xmlns:a="http://example.com/" xmlns:b="http://example.net/"><a:foo>This tag is in the a namespace</a:foo><b:foo>This tag is in the b namespace</b:foo></root>' soup = self.soup(markup) root = soup.root - self.assertEqual("http://example.com/", root['xmlns:a']) - self.assertEqual("http://example.net/", root['xmlns:b']) + assert "http://example.com/" == root['xmlns:a'] + assert "http://example.net/" == root['xmlns:b'] def test_closing_namespaced_tag(self): markup = '<p xmlns:dc="http://purl.org/dc/elements/1.1/"><dc:date>20010504</dc:date></p>' soup = self.soup(markup) - self.assertEqual(str(soup.p), markup) + assert str(soup.p) == markup def test_namespaced_attributes(self): markup = '<foo xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"><bar xsi:schemaLocation="http://www.example.com"/></foo>' soup = self.soup(markup) - self.assertEqual(str(soup.foo), markup) + assert str(soup.foo) == markup def test_namespaced_attributes_xml_namespace(self): markup = '<foo xml:lang="fr">bar</foo>' soup = self.soup(markup) - self.assertEqual(str(soup.foo), markup) + assert str(soup.foo) == markup def test_find_by_prefixed_name(self): doc = """<?xml version="1.0" encoding="utf-8"?> @@ -1061,14 +1043,14 @@ class XMLTreeBuilderSmokeTest(TreeBuilderSmokeTest): soup = self.soup(doc) # There are three <tag> tags. - self.assertEqual(3, len(soup.find_all('tag'))) + assert 3 == len(soup.find_all('tag')) # But two of them are ns1:tag and one of them is ns2:tag. - self.assertEqual(2, len(soup.find_all('ns1:tag'))) - self.assertEqual(1, len(soup.find_all('ns2:tag'))) + assert 2 == len(soup.find_all('ns1:tag')) + assert 1 == len(soup.find_all('ns2:tag')) - self.assertEqual(1, len(soup.find_all('ns2:tag', key='value'))) - self.assertEqual(3, len(soup.find_all(['ns1:tag', 'ns2:tag']))) + assert 1, len(soup.find_all('ns2:tag', key='value')) + assert 3, len(soup.find_all(['ns1:tag', 'ns2:tag'])) def test_copy_tag_preserves_namespace(self): xml = """<?xml version="1.0" encoding="UTF-8" standalone="yes"?> @@ -1079,7 +1061,7 @@ class XMLTreeBuilderSmokeTest(TreeBuilderSmokeTest): duplicate = copy.copy(tag) # The two tags have the same namespace prefix. - self.assertEqual(tag.prefix, duplicate.prefix) + assert tag.prefix == duplicate.prefix def test_worst_case(self): """Test the worst case (currently) for linking issues.""" @@ -1099,29 +1081,29 @@ class HTML5TreeBuilderSmokeTest(HTMLTreeBuilderSmokeTest): def test_html_tags_have_namespace(self): markup = "<a>" soup = self.soup(markup) - self.assertEqual("http://www.w3.org/1999/xhtml", soup.a.namespace) + assert "http://www.w3.org/1999/xhtml" == soup.a.namespace def test_svg_tags_have_namespace(self): markup = '<svg><circle/></svg>' soup = self.soup(markup) namespace = "http://www.w3.org/2000/svg" - self.assertEqual(namespace, soup.svg.namespace) - self.assertEqual(namespace, soup.circle.namespace) + assert namespace == soup.svg.namespace + assert namespace == soup.circle.namespace def test_mathml_tags_have_namespace(self): markup = '<math><msqrt>5</msqrt></math>' soup = self.soup(markup) namespace = 'http://www.w3.org/1998/Math/MathML' - self.assertEqual(namespace, soup.math.namespace) - self.assertEqual(namespace, soup.msqrt.namespace) + assert namespace == soup.math.namespace + assert namespace == soup.msqrt.namespace def test_xml_declaration_becomes_comment(self): markup = '<?xml version="1.0" encoding="utf-8"?><html></html>' 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) + assert isinstance(soup.contents[0], Comment) + assert soup.contents[0] == '?xml version="1.0" encoding="utf-8"?' + assert "html" == soup.contents[0].next_element.name def skipIf(condition, reason): def nothing(test, *args, **kwargs): diff --git a/bs4/tests/test_builder_registry.py b/bs4/tests/test_builder_registry.py index 90cad82..5fa874c 100644 --- a/bs4/tests/test_builder_registry.py +++ b/bs4/tests/test_builder_registry.py @@ -1,6 +1,6 @@ """Tests of the builder registry.""" -import unittest +import pytest import warnings from bs4 import BeautifulSoup @@ -26,46 +26,36 @@ except ImportError: LXML_PRESENT = False -class BuiltInRegistryTest(unittest.TestCase): +class TestBuiltInRegistry(object): """Test the built-in registry with the default builders registered.""" def test_combination(self): + assert registry.lookup('strict', 'html') == HTMLParserTreeBuilder 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) + assert registry.lookup('fast', 'html') == LXMLTreeBuilder + assert registry.lookup('permissive', 'xml') == LXMLTreeBuilderForXML if HTML5LIB_PRESENT: - self.assertEqual(registry.lookup('html5lib', 'html'), - HTML5TreeBuilder) + assert 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) + assert registry.lookup('html') == LXMLTreeBuilder + assert registry.lookup('xml') == LXMLTreeBuilderForXML else: - self.assertEqual(registry.lookup('xml'), None) + assert registry.lookup('xml') == None if HTML5LIB_PRESENT: - self.assertEqual(registry.lookup('html'), HTML5TreeBuilder) + assert registry.lookup('html') == HTML5TreeBuilder else: - self.assertEqual(registry.lookup('html'), HTMLParserTreeBuilder) + assert 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) + assert registry.lookup('lxml', 'xml') == LXMLTreeBuilderForXML + assert registry.lookup('lxml', 'html') == LXMLTreeBuilder if HTML5LIB_PRESENT: - self.assertEqual(registry.lookup('html5lib'), - HTML5TreeBuilder) + assert registry.lookup('html5lib') == HTML5TreeBuilder - self.assertEqual(registry.lookup('html.parser'), - HTMLParserTreeBuilder) + assert registry.lookup('html.parser') == HTMLParserTreeBuilder def test_beautifulsoup_constructor_does_lookup(self): @@ -77,16 +67,17 @@ class BuiltInRegistryTest(unittest.TestCase): BeautifulSoup("", features="html") # Or a list of strings. BeautifulSoup("", features=["html", "fast"]) - + pass + # You'll get an exception if BS can't find an appropriate # builder. - self.assertRaises(ValueError, BeautifulSoup, - "", features="no-such-feature") + with pytest.raises(ValueError): + BeautifulSoup("", features="no-such-feature") -class RegistryTest(unittest.TestCase): +class TestRegistry(object): """Test the TreeBuilderRegistry class in general.""" - def setUp(self): + def setup_method(self): self.registry = TreeBuilderRegistry() def builder_for_features(self, *feature_list): @@ -101,28 +92,28 @@ class RegistryTest(unittest.TestCase): # Since the builder advertises no features, you can't find it # by looking up features. - self.assertEqual(self.registry.lookup('foo'), None) + assert self.registry.lookup('foo') is 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) + assert 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) + assert self.registry.lookup('foo') is builder + assert self.registry.lookup('bar') is 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) + assert self.registry.lookup('baz') is 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) + assert self.registry.lookup() == builder2 def test_lookup_fails_when_no_tree_builders_registered(self): - self.assertEqual(self.registry.lookup(), None) + assert self.registry.lookup() is None def test_lookup_gets_most_recent_builder_supporting_all_features(self): has_one = self.builder_for_features('foo') @@ -134,14 +125,12 @@ class RegistryTest(unittest.TestCase): # 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) + assert 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) + assert 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) + assert self.registry.lookup('bar', 'baz') is None diff --git a/bs4/tests/test_dammit.py b/bs4/tests/test_dammit.py index 5177503..348c600 100644 --- a/bs4/tests/test_dammit.py +++ b/bs4/tests/test_dammit.py @@ -1,6 +1,6 @@ # encoding: utf-8 +import pytest import logging -import unittest import bs4 from bs4 import BeautifulSoup from bs4.dammit import ( @@ -8,67 +8,62 @@ from bs4.dammit import ( UnicodeDammit, ) -class TestUnicodeDammit(unittest.TestCase): +class TestUnicodeDammit(object): """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) + assert dammit.unicode_markup == markup def test_smart_quotes_to_unicode(self): markup = b"<foo>\x91\x92\x93\x94</foo>" dammit = UnicodeDammit(markup) - self.assertEqual( - dammit.unicode_markup, "<foo>\u2018\u2019\u201c\u201d</foo>") + assert dammit.unicode_markup == "<foo>\u2018\u2019\u201c\u201d</foo>" def test_smart_quotes_to_xml_entities(self): markup = b"<foo>\x91\x92\x93\x94</foo>" dammit = UnicodeDammit(markup, smart_quotes_to="xml") - self.assertEqual( - dammit.unicode_markup, "<foo>‘’“”</foo>") + assert dammit.unicode_markup == "<foo>‘’“”</foo>" def test_smart_quotes_to_html_entities(self): markup = b"<foo>\x91\x92\x93\x94</foo>" dammit = UnicodeDammit(markup, smart_quotes_to="html") - self.assertEqual( - dammit.unicode_markup, "<foo>‘’“”</foo>") + assert dammit.unicode_markup == "<foo>‘’“”</foo>" def test_smart_quotes_to_ascii(self): markup = b"<foo>\x91\x92\x93\x94</foo>" dammit = UnicodeDammit(markup, smart_quotes_to="ascii") - self.assertEqual( - dammit.unicode_markup, """<foo>''""</foo>""") + assert dammit.unicode_markup == """<foo>''""</foo>""" 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}') - + assert dammit.original_encoding.lower() == 'utf-8' + assert 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') + assert dammit.original_encoding.lower() == 'iso-8859-8' + assert 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) + assert dammit.original_encoding.lower() == 'utf-8' + assert 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') + assert 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') + assert dammit.original_encoding.lower() == 'utf-8' def test_exclude_encodings(self): # This is UTF-8. @@ -77,14 +72,14 @@ class TestUnicodeDammit(unittest.TestCase): # 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') + assert 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) + assert dammit.original_encoding == None -class TestEncodingDetector(unittest.TestCase): +class TestEncodingDetector(object): def test_encoding_detector_replaces_junk_in_encoding_name_with_replacement_character(self): detected = EncodingDetector( @@ -100,8 +95,7 @@ class TestEncodingDetector(unittest.TestCase): b"<html><meta charset=euc-jp /></html>", b"<html><meta charset=euc-jp/></html>"): dammit = UnicodeDammit(data, is_html=True) - self.assertEqual( - "euc-jp", dammit.original_encoding) + assert "euc-jp" == dammit.original_encoding def test_last_ditch_entity_replacement(self): # This is a UTF-8 document that contains bytestrings @@ -129,11 +123,11 @@ class TestEncodingDetector(unittest.TestCase): return None bs4.dammit.chardet_dammit = noop dammit = UnicodeDammit(doc) - self.assertEqual(True, dammit.contains_replacement_characters) - self.assertTrue("\ufffd" in dammit.unicode_markup) + assert True == dammit.contains_replacement_characters + assert "\ufffd" in dammit.unicode_markup soup = BeautifulSoup(doc, "html.parser") - self.assertTrue(soup.contains_replacement_characters) + assert soup.contains_replacement_characters finally: logging.disable(logging.NOTSET) bs4.dammit.chardet_dammit = chardet @@ -142,8 +136,8 @@ class TestEncodingDetector(unittest.TestCase): # 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("<a>áé</a>", dammit.unicode_markup) - self.assertEqual("utf-16le", dammit.original_encoding) + assert "<a>áé</a>" == dammit.unicode_markup + assert "utf-16le" == dammit.original_encoding def test_known_definite_versus_user_encodings(self): # The known_definite_encodings are used before sniffing the @@ -156,16 +150,14 @@ class TestEncodingDetector(unittest.TestCase): # We can process it as UTF-16 by passing it in as a known # definite encoding. before = UnicodeDammit(data, known_definite_encodings=["utf-16"]) - self.assertEqual("utf-16", before.original_encoding) + assert "utf-16" == before.original_encoding # If we pass UTF-18 as a user encoding, it's not even # tried--the encoding sniffed from the byte-order mark takes # precedence. after = UnicodeDammit(data, user_encodings=["utf-8"]) - self.assertEqual("utf-16le", after.original_encoding) - self.assertEqual( - ["utf-16le"], [x[0] for x in dammit.tried_encodings] - ) + assert "utf-16le" == after.original_encoding + assert ["utf-16le"] == [x[0] for x in dammit.tried_encodings] # Here's a document in ISO-8859-8. hebrew = b"\xed\xe5\xec\xf9" @@ -175,10 +167,8 @@ class TestEncodingDetector(unittest.TestCase): # The known_definite_encodings don't work, BOM sniffing does # nothing (it only works for a few UTF encodings), but one of # the user_encodings does work. - self.assertEqual("iso-8859-8", dammit.original_encoding) - self.assertEqual( - ["utf-8", "iso-8859-8"], [x[0] for x in dammit.tried_encodings] - ) + assert "iso-8859-8" == dammit.original_encoding + assert ["utf-8", "iso-8859-8"] == [x[0] for x in dammit.tried_encodings] def test_deprecated_override_encodings(self): # override_encodings is a deprecated alias for @@ -190,12 +180,11 @@ class TestEncodingDetector(unittest.TestCase): override_encodings=["utf-8"], user_encodings=["iso-8859-8"], ) - self.assertEqual("iso-8859-8", dammit.original_encoding) + assert "iso-8859-8" == dammit.original_encoding # known_definite_encodings and override_encodings were tried # before user_encodings. - self.assertEqual( - ["shift-jis", "utf-8", "iso-8859-8"], + assert ["shift-jis", "utf-8", "iso-8859-8"] == ( [x[0] for x in dammit.tried_encodings] ) @@ -212,16 +201,15 @@ class TestEncodingDetector(unittest.TestCase): doc = utf8 + windows_1252 + utf8 # The document can't be turned into UTF-8: - self.assertRaises(UnicodeDecodeError, doc.decode, "utf8") + with pytest.raises(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")) + assert "☃☃☃“Hi, I like Windows!”☃☃☃" == fixed.decode("utf8") def test_detwingle_ignores_multibyte_characters(self): # Each of these characters has a UTF-8 representation ending @@ -234,9 +222,9 @@ class TestEncodingDetector(unittest.TestCase): "\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')) + assert input.endswith(b'\x93') output = UnicodeDammit.detwingle(input) - self.assertEqual(output, input) + assert output == input def test_find_declared_encoding(self): # Test our ability to find a declared encoding inside an @@ -253,34 +241,30 @@ class TestEncodingDetector(unittest.TestCase): xml_bytes = xml_unicode.encode("ascii") m = EncodingDetector.find_declared_encoding - self.assertEqual(None, m(html_unicode, is_html=False)) - self.assertEqual("utf-8", m(html_unicode, is_html=True)) - self.assertEqual("utf-8", m(html_bytes, is_html=True)) + assert m(html_unicode, is_html=False) is None + assert "utf-8" == m(html_unicode, is_html=True) + assert "utf-8" == m(html_bytes, is_html=True) - self.assertEqual("iso-8859-1", m(xml_unicode)) - self.assertEqual("iso-8859-1", m(xml_bytes)) + assert "iso-8859-1" == m(xml_unicode) + assert "iso-8859-1" == m(xml_bytes) # Normally, only the first few kilobytes of a document are checked for # an encoding. spacer = b' ' * 5000 - self.assertEqual(None, m(spacer + html_bytes)) - self.assertEqual(None, m(spacer + xml_bytes)) + assert m(spacer + html_bytes) is None + assert m(spacer + xml_bytes) is None # But you can tell find_declared_encoding to search an entire # HTML document. - self.assertEqual( - "utf-8", + assert ( m(spacer + html_bytes, is_html=True, search_entire_document=True) + == "utf-8" ) # The XML encoding declaration has to be the very first thing # in the document. We'll allow whitespace before the document # starts, but nothing else. - self.assertEqual( - "iso-8859-1", - m(xml_bytes, search_entire_document=True) - ) - self.assertEqual( - None, m(b'a' + xml_bytes, search_entire_document=True) - ) + assert m(xml_bytes, search_entire_document=True) == "iso-8859-1" + assert m(b' ' + xml_bytes, search_entire_document=True) == "iso-8859-1" + assert m(b'a' + xml_bytes, search_entire_document=True) is None diff --git a/bs4/tests/test_docs.py b/bs4/tests/test_docs.py index 5b9f677..0194d69 100644 --- a/bs4/tests/test_docs.py +++ b/bs4/tests/test_docs.py @@ -1,5 +1,7 @@ "Test harness for doctests." +# TODO: Pretty sure this isn't used and should be deleted. + # pylint: disable-msg=E0611,W0142 __metaclass__ = type diff --git a/bs4/tests/test_formatter.py b/bs4/tests/test_formatter.py index c188e18..e944759 100644 --- a/bs4/tests/test_formatter.py +++ b/bs4/tests/test_formatter.py @@ -18,12 +18,12 @@ class TestFormatter(SoupTest): # Attributes come out sorted by name. In Python 3, attributes # normally come out of a dictionary in the order they were # added. - self.assertEqual([('a', 2), ('b', 1)], formatter.attributes(tag)) + assert [('a', 2), ('b', 1)] == formatter.attributes(tag) # This works even if Tag.attrs is None, though this shouldn't # normally happen. tag.attrs = None - self.assertEqual([], formatter.attributes(tag)) + assert [] == formatter.attributes(tag) def test_sort_attributes(self): # Test the ability to override Formatter.attributes() to, @@ -42,8 +42,8 @@ class TestFormatter(SoupTest): # attributes() was called on the <p> tag. It filtered out one # attribute and sorted the other two. - self.assertEqual(formatter.called_with, soup.p) - self.assertEqual('<p aval="2" cval="1"></p>', decoded) + assert formatter.called_with == soup.p + assert '<p aval="2" cval="1"></p>' == decoded def test_empty_attributes_are_booleans(self): # Test the behavior of empty_attributes_are_booleans as well @@ -51,17 +51,17 @@ class TestFormatter(SoupTest): for name in ('html', 'minimal', None): formatter = HTMLFormatter.REGISTRY[name] - self.assertEqual(False, formatter.empty_attributes_are_booleans) + assert False == formatter.empty_attributes_are_booleans formatter = XMLFormatter.REGISTRY[None] - self.assertEqual(False, formatter.empty_attributes_are_booleans) + assert False == formatter.empty_attributes_are_booleans formatter = HTMLFormatter.REGISTRY['html5'] - self.assertEqual(True, formatter.empty_attributes_are_booleans) + assert True == formatter.empty_attributes_are_booleans # Verify that the constructor sets the value. formatter = Formatter(empty_attributes_are_booleans=True) - self.assertEqual(True, formatter.empty_attributes_are_booleans) + assert True == formatter.empty_attributes_are_booleans # Now demonstrate what it does to markup. for markup in ( @@ -70,12 +70,6 @@ class TestFormatter(SoupTest): ): soup = self.soup(markup) for formatter in ('html', 'minimal', 'xml', None): - self.assertEqual( - b'<option selected=""></option>', - soup.option.encode(formatter='html') - ) - self.assertEqual( - b'<option selected></option>', - soup.option.encode(formatter='html5') - ) + assert b'<option selected=""></option>' == soup.option.encode(formatter='html') + assert b'<option selected></option>' == soup.option.encode(formatter='html5') diff --git a/bs4/tests/test_html5lib.py b/bs4/tests/test_html5lib.py index f8902ad..9ac6ccc 100644 --- a/bs4/tests/test_html5lib.py +++ b/bs4/tests/test_html5lib.py @@ -17,7 +17,7 @@ from bs4.testing import ( @skipIf( not HTML5LIB_PRESENT, "html5lib seems not to be present, not testing its tree builder.") -class HTML5LibBuilderSmokeTest(SoupTest, HTML5TreeBuilderSmokeTest): +class TestHTML5LibBuilder(SoupTest, HTML5TreeBuilderSmokeTest): """See ``HTML5TreeBuilderSmokeTest``.""" @property @@ -30,12 +30,9 @@ class HTML5LibBuilderSmokeTest(SoupTest, HTML5TreeBuilderSmokeTest): markup = "<p>A <b>bold</b> statement.</p>" with warnings.catch_warnings(record=True) as w: soup = self.soup(markup, parse_only=strainer) - self.assertEqual( - soup.decode(), self.document_for(markup)) + assert soup.decode() == self.document_for(markup) - self.assertTrue( - "the html5lib tree builder doesn't support parse_only" in - str(w[0].message)) + assert "the html5lib tree builder doesn't support parse_only" in str(w[0].message) def test_correctly_nested_tables(self): """html5lib inserts <tbody> tags where other parsers don't.""" @@ -46,13 +43,13 @@ class HTML5LibBuilderSmokeTest(SoupTest, HTML5TreeBuilderSmokeTest): '<tr><td>foo</td></tr>' '</table></td>') - self.assertSoupEquals( + self.assert_soup( markup, '<table id="1"><tbody><tr><td>Here\'s another table:' '<table id="2"><tbody><tr><td>foo</td></tr></tbody></table>' '</td></tr></tbody></table>') - self.assertSoupEquals( + self.assert_soup( "<table><thead><tr><td>Foo</td></tr></thead>" "<tbody><tr><td>Bar</td></tr></tbody>" "<tfoot><tr><td>Baz</td></tr></tfoot></table>") @@ -69,20 +66,20 @@ class HTML5LibBuilderSmokeTest(SoupTest, HTML5TreeBuilderSmokeTest): </html>''' soup = self.soup(markup) # Verify that we can reach the <p> tag; this means the tree is connected. - self.assertEqual(b"<p>foo</p>", soup.p.encode()) + assert b"<p>foo</p>" == soup.p.encode() def test_reparented_markup(self): markup = '<p><em>foo</p>\n<p>bar<a></a></em></p>' soup = self.soup(markup) - self.assertEqual("<body><p><em>foo</em></p><em>\n</em><p><em>bar<a></a></em></p></body>", soup.body.decode()) - self.assertEqual(2, len(soup.find_all('p'))) + assert "<body><p><em>foo</em></p><em>\n</em><p><em>bar<a></a></em></p></body>" == soup.body.decode() + assert 2 == len(soup.find_all('p')) def test_reparented_markup_ends_with_whitespace(self): markup = '<p><em>foo</p>\n<p>bar<a></a></em></p>\n' soup = self.soup(markup) - self.assertEqual("<body><p><em>foo</em></p><em>\n</em><p><em>bar<a></a></em></p>\n</body>", soup.body.decode()) - self.assertEqual(2, len(soup.find_all('p'))) + assert "<body><p><em>foo</em></p><em>\n</em><p><em>bar<a></a></em></p>\n</body>" == soup.body.decode() + assert 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 @@ -99,7 +96,7 @@ class HTML5LibBuilderSmokeTest(SoupTest, HTML5TreeBuilderSmokeTest): markup = '<div><a>aftermath<p><noscript>target</noscript>aftermath</a></p></div>' soup = self.soup(markup) noscript = soup.noscript - self.assertEqual("target", noscript.next_element) + assert "target" == noscript.next_element target = soup.find(string='target') # The 'aftermath' string was duplicated; we want the second one. @@ -108,8 +105,8 @@ class HTML5LibBuilderSmokeTest(SoupTest, HTML5TreeBuilderSmokeTest): # The <noscript> tag was moved beneath a copy of the <a> tag, # but the 'target' string within is still connected to the # (second) 'aftermath' string. - self.assertEqual(final_aftermath, target.next_element) - self.assertEqual(target, final_aftermath.previous_element) + assert final_aftermath == target.next_element + assert target == final_aftermath.previous_element def test_processing_instruction(self): """Processing instructions become comments.""" @@ -121,13 +118,13 @@ class HTML5LibBuilderSmokeTest(SoupTest, HTML5TreeBuilderSmokeTest): markup = b"""<a class="my_class"><p></a>""" soup = self.soup(markup) a1, a2 = soup.find_all('a') - self.assertEqual(a1, a2) + assert a1 == a2 assert a1 is not a2 def test_foster_parenting(self): markup = b"""<table><td></tbody>A""" soup = self.soup(markup) - self.assertEqual("<body>A<table><tbody><tr><td></td></tr></tbody></table></body>", soup.body.decode()) + assert "<body>A<table><tbody><tr><td></td></tr></tbody></table></body>" == soup.body.decode() def test_extraction(self): """ @@ -145,7 +142,7 @@ class HTML5LibBuilderSmokeTest(SoupTest, HTML5TreeBuilderSmokeTest): [s.extract() for s in soup('script')] [s.extract() for s in soup('style')] - self.assertEqual(len(soup.find_all("p")), 1) + assert len(soup.find_all("p")) == 1 def test_empty_comment(self): """ @@ -167,21 +164,21 @@ class HTML5LibBuilderSmokeTest(SoupTest, HTML5TreeBuilderSmokeTest): inputs = [] for form in soup.find_all('form'): inputs.extend(form.find_all('input')) - self.assertEqual(len(inputs), 1) + assert len(inputs) == 1 def test_tracking_line_numbers(self): # The html.parser TreeBuilder keeps track of line number and # position of each element. markup = "\n <p>\n\n<sourceline>\n<b>text</b></sourceline><sourcepos></p>" soup = self.soup(markup) - self.assertEqual(2, soup.p.sourceline) - self.assertEqual(5, soup.p.sourcepos) - self.assertEqual("sourceline", soup.p.find('sourceline').name) + assert 2 == soup.p.sourceline + assert 5 == soup.p.sourcepos + assert "sourceline" == soup.p.find('sourceline').name # You can deactivate this behavior. soup = self.soup(markup, store_line_numbers=False) - self.assertEqual("sourceline", soup.p.sourceline.name) - self.assertEqual("sourcepos", soup.p.sourcepos.name) + assert "sourceline" == soup.p.sourceline.name + assert "sourcepos" == soup.p.sourcepos.name def test_special_string_containers(self): # The html5lib tree builder doesn't support this standard feature, @@ -219,8 +216,8 @@ class HTML5LibBuilderSmokeTest(SoupTest, HTML5TreeBuilderSmokeTest): div = self.soup(markup).div without_element = div.encode() expect = b"<div>%s</div>" % output_unicode.encode("utf8") - self.assertEqual(without_element, expect) + assert without_element == expect with_element = div.encode(formatter="html") expect = b"<div>%s</div>" % output_element - self.assertEqual(with_element, expect) + assert with_element == expect diff --git a/bs4/tests/test_htmlparser.py b/bs4/tests/test_htmlparser.py index 0d8161e..640cb1f 100644 --- a/bs4/tests/test_htmlparser.py +++ b/bs4/tests/test_htmlparser.py @@ -8,7 +8,7 @@ from bs4.testing import SoupTest, HTMLTreeBuilderSmokeTest from bs4.builder import HTMLParserTreeBuilder from bs4.builder._htmlparser import BeautifulSoupHTMLParser -class HTMLParserTreeBuilderSmokeTest(SoupTest, HTMLTreeBuilderSmokeTest): +class TestHTMLParserTreeBuilder(SoupTest, HTMLTreeBuilderSmokeTest): default_builder = HTMLParserTreeBuilder @@ -27,30 +27,30 @@ class HTMLParserTreeBuilderSmokeTest(SoupTest, HTMLTreeBuilderSmokeTest): tree = self.soup("<a><b>foo</a>") dumped = pickle.dumps(tree, 2) loaded = pickle.loads(dumped) - self.assertTrue(isinstance(loaded.builder, type(tree.builder))) + assert isinstance(loaded.builder, type(tree.builder)) def test_redundant_empty_element_closing_tags(self): - self.assertSoupEquals('<br></br><br></br><br></br>', "<br/><br/><br/>") - self.assertSoupEquals('</br></br></br>', "") + self.assert_soup('<br></br><br></br><br></br>', "<br/><br/><br/>") + self.assert_soup('</br></br></br>', "") def test_empty_element(self): # This verifies that any buffered data present when the parser # finishes working is handled. - self.assertSoupEquals("foo &# bar", "foo &# bar") + self.assert_soup("foo &# bar", "foo &# bar") def test_tracking_line_numbers(self): # The html.parser TreeBuilder keeps track of line number and # position of each element. markup = "\n <p>\n\n<sourceline>\n<b>text</b></sourceline><sourcepos></p>" soup = self.soup(markup) - self.assertEqual(2, soup.p.sourceline) - self.assertEqual(3, soup.p.sourcepos) - self.assertEqual("sourceline", soup.p.find('sourceline').name) + assert 2 == soup.p.sourceline + assert 3 == soup.p.sourcepos + assert "sourceline" == soup.p.find('sourceline').name # You can deactivate this behavior. soup = self.soup(markup, store_line_numbers=False) - self.assertEqual("sourceline", soup.p.sourceline.name) - self.assertEqual("sourcepos", soup.p.sourcepos.name) + assert "sourceline" == soup.p.sourceline.name + assert "sourcepos" == soup.p.sourcepos.name def test_on_duplicate_attribute(self): # The html.parser tree builder has a variety of ways of @@ -61,20 +61,20 @@ class HTMLParserTreeBuilderSmokeTest(SoupTest, HTMLTreeBuilderSmokeTest): # If you don't provide any particular value for # on_duplicate_attribute, later values replace earlier values. soup = self.soup(markup) - self.assertEqual("url3", soup.a['href']) - self.assertEqual(["cls"], soup.a['class']) - self.assertEqual("id", soup.a['id']) + assert "url3" == soup.a['href'] + assert ["cls"] == soup.a['class'] + assert "id" == soup.a['id'] # You can also get this behavior explicitly. def assert_attribute(on_duplicate_attribute, expected): soup = self.soup( markup, on_duplicate_attribute=on_duplicate_attribute ) - self.assertEqual(expected, soup.a['href']) + assert expected == soup.a['href'] # Verify that non-duplicate attributes are treated normally. - self.assertEqual(["cls"], soup.a['class']) - self.assertEqual("id", soup.a['id']) + assert ["cls"] == soup.a['class'] + assert "id" == soup.a['id'] assert_attribute(None, "url3") assert_attribute(BeautifulSoupHTMLParser.REPLACE, "url3") @@ -114,11 +114,11 @@ class HTMLParserTreeBuilderSmokeTest(SoupTest, HTMLTreeBuilderSmokeTest): div = self.soup(markup).div without_element = div.encode() expect = b"<div>%s</div>" % output_unicode.encode("utf8") - self.assertEqual(without_element, expect) + assert without_element == expect with_element = div.encode(formatter="html") expect = b"<div>%s</div>" % output_element - self.assertEqual(with_element, expect) + assert with_element == expect class TestHTMLParserSubclass(SoupTest): diff --git a/bs4/tests/test_lxml.py b/bs4/tests/test_lxml.py index 71931ff..0e505d9 100644 --- a/bs4/tests/test_lxml.py +++ b/bs4/tests/test_lxml.py @@ -31,7 +31,7 @@ from bs4.testing import ( @skipIf( not LXML_PRESENT, "lxml seems not to be present, not testing its tree builder.") -class LXMLTreeBuilderSmokeTest(SoupTest, HTMLTreeBuilderSmokeTest): +class TestLXMLTreeBuilder(SoupTest, HTMLTreeBuilderSmokeTest): """See ``HTMLTreeBuilderSmokeTest``.""" @property @@ -39,11 +39,11 @@ class LXMLTreeBuilderSmokeTest(SoupTest, HTMLTreeBuilderSmokeTest): return LXMLTreeBuilder def test_out_of_range_entity(self): - self.assertSoupEquals( + self.assert_soup( "<p>foo�bar</p>", "<p>foobar</p>") - self.assertSoupEquals( + self.assert_soup( "<p>foo�bar</p>", "<p>foobar</p>") - self.assertSoupEquals( + self.assert_soup( "<p>foo�bar</p>", "<p>foobar</p>") def test_entities_in_foreign_document_encoding(self): @@ -61,15 +61,15 @@ class LXMLTreeBuilderSmokeTest(SoupTest, HTMLTreeBuilderSmokeTest): def test_empty_doctype(self): soup = self.soup("<!DOCTYPE>") doctype = soup.contents[0] - self.assertEqual("", doctype.strip()) + assert "" == 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("<b />") - self.assertEqual("<b/>", str(soup.b)) - self.assertTrue("BeautifulStoneSoup class is deprecated" in str(w[0].message)) + assert "<b/>" == str(soup.b) + assert "BeautifulStoneSoup class is deprecated" in str(w[0].message) def test_tracking_line_numbers(self): # The lxml TreeBuilder cannot keep track of line numbers from @@ -83,8 +83,8 @@ class LXMLTreeBuilderSmokeTest(SoupTest, HTMLTreeBuilderSmokeTest): "\n <p>\n\n<sourceline>\n<b>text</b></sourceline><sourcepos></p>", store_line_numbers=True ) - self.assertEqual("sourceline", soup.p.sourceline.name) - self.assertEqual("sourcepos", soup.p.sourcepos.name) + assert "sourceline" == soup.p.sourceline.name + assert "sourcepos" == soup.p.sourcepos.name @skipIf( not LXML_PRESENT, @@ -109,7 +109,6 @@ class LXMLXMLTreeBuilderSmokeTest(SoupTest, XMLTreeBuilderSmokeTest): '<prefix:tag xmlns:prefix="http://prefixed-namespace.com">content</tag>' '</root>' ) - self.assertEqual( - soup._namespaces, + assert soup._namespaces == ( {'xml': 'http://www.w3.org/XML/1998/namespace', 'prefix': 'http://prefixed-namespace.com'} ) diff --git a/bs4/tests/test_navigablestring.py b/bs4/tests/test_navigablestring.py index 89e92b7..6cd57f7 100644 --- a/bs4/tests/test_navigablestring.py +++ b/bs4/tests/test_navigablestring.py @@ -15,34 +15,33 @@ class TestNavigableString(SoupTest): def test_text_acquisition_methods(self): # These methods are intended for use against Tag, but they # work on NavigableString as well, - eq_ = self.assertEqual s = NavigableString("fee ") cdata = CData("fie ") comment = Comment("foe ") - eq_("fee ", s.get_text()) - eq_("fee", s.get_text(strip=True)) - eq_(["fee "], list(s.strings)) - eq_(["fee"], list(s.stripped_strings)) - eq_(["fee "], list(s._all_strings())) - - eq_("fie ", cdata.get_text()) - eq_("fie", cdata.get_text(strip=True)) - eq_(["fie "], list(cdata.strings)) - eq_(["fie"], list(cdata.stripped_strings)) - eq_(["fie "], list(cdata._all_strings())) + assert "fee " == s.get_text() + assert "fee" == s.get_text(strip=True) + assert ["fee "] == list(s.strings) + assert ["fee"] == list(s.stripped_strings) + assert ["fee "] == list(s._all_strings()) + + assert "fie " == cdata.get_text() + assert "fie" == cdata.get_text(strip=True) + assert ["fie "] == list(cdata.strings) + assert ["fie"] == list(cdata.stripped_strings) + assert ["fie "] == list(cdata._all_strings()) # Since a Comment isn't normally considered 'text', # these methods generally do nothing. - eq_("", comment.get_text()) - eq_([], list(comment.strings)) - eq_([], list(comment.stripped_strings)) - eq_([], list(comment._all_strings())) + assert "" == comment.get_text() + assert [] == list(comment.strings) + assert [] == list(comment.stripped_strings) + assert [] == list(comment._all_strings()) # Unless you specifically say that comments are okay. - eq_("foe", comment.get_text(strip=True, types=Comment)) - eq_("foe ", comment.get_text(types=(Comment, NavigableString))) + assert "foe" == comment.get_text(strip=True, types=Comment) + assert "foe " == comment.get_text(types=(Comment, NavigableString)) class TestNavigableStringSubclasses(SoupTest): @@ -52,9 +51,9 @@ class TestNavigableStringSubclasses(SoupTest): soup = self.soup("") cdata = CData("foo") soup.insert(1, cdata) - self.assertEqual(str(soup), "<![CDATA[foo]]>") - self.assertEqual(soup.find(text="foo"), "foo") - self.assertEqual(soup.contents[0], "foo") + assert str(soup) == "<![CDATA[foo]]>" + assert soup.find(text="foo") == "foo" + assert soup.contents[0] == "foo" def test_cdata_is_never_formatted(self): """Text inside a CData object is passed into the formatter. @@ -70,9 +69,8 @@ class TestNavigableStringSubclasses(SoupTest): soup = self.soup("") cdata = CData("<><><>") soup.insert(1, cdata) - self.assertEqual( - b"<![CDATA[<><><>]]>", soup.encode(formatter=increment)) - self.assertEqual(1, self.count) + assert b"<![CDATA[<><><>]]>" == soup.encode(formatter=increment) + assert 1 == self.count def test_doctype_ends_in_newline(self): # Unlike other NavigableString subclasses, a DOCTYPE always ends @@ -80,11 +78,11 @@ class TestNavigableStringSubclasses(SoupTest): doctype = Doctype("foo") soup = self.soup("") soup.insert(1, doctype) - self.assertEqual(soup.encode(), b"<!DOCTYPE foo>\n") + assert soup.encode() == b"<!DOCTYPE foo>\n" def test_declaration(self): d = Declaration("foo") - self.assertEqual("<?foo?>", d.output_ready()) + assert "<?foo?>" == d.output_ready() def test_default_string_containers(self): # In some cases, we use different NavigableString subclasses for @@ -92,10 +90,9 @@ class TestNavigableStringSubclasses(SoupTest): soup = self.soup( "<div>text</div><script>text</script><style>text</style>" ) - self.assertEqual( - [NavigableString, Script, Stylesheet], - [x.__class__ for x in soup.find_all(text=True)] - ) + assert [NavigableString, Script, Stylesheet] == [ + x.__class__ for x in soup.find_all(text=True) + ] # The TemplateString is a little unusual because it's generally found # _inside_ children of a <template> element, not a direct child of the @@ -119,5 +116,5 @@ class TestNavigableStringSubclasses(SoupTest): # Comment. markup = b"<template>Some text<p>In a tag</p><!--with a comment--></template>" soup = self.soup(markup) - self.assertEqual(markup, soup.template.encode("utf8")) + assert markup == soup.template.encode("utf8") diff --git a/bs4/tests/test_soup.py b/bs4/tests/test_soup.py index 4d00845..f73086a 100644 --- a/bs4/tests/test_soup.py +++ b/bs4/tests/test_soup.py @@ -4,7 +4,7 @@ from pdb import set_trace import logging import os -import unittest +import pytest import sys import tempfile @@ -53,17 +53,17 @@ class TestConstructor(SoupTest): def test_short_unicode_input(self): data = "<h1>éé</h1>" soup = self.soup(data) - self.assertEqual("éé", soup.h1.string) + assert "éé" == soup.h1.string def test_embedded_null(self): data = "<h1>foo\0bar</h1>" soup = self.soup(data) - self.assertEqual("foo\0bar", soup.h1.string) + assert "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) + assert "windows-1252" == soup.original_encoding def test_custom_builder_class(self): # Verify that you can pass in a custom Builder class and @@ -97,8 +97,8 @@ class TestConstructor(SoupTest): with warnings.catch_warnings(record=True): soup = BeautifulSoup('', builder=Mock, **kwargs) assert isinstance(soup.builder, Mock) - self.assertEqual(dict(var="value"), soup.builder.called_with) - self.assertEqual("prepared markup", soup.builder.fed) + assert dict(var="value") == soup.builder.called_with + assert "prepared markup" == soup.builder.fed # You can also instantiate the TreeBuilder yourself. In this # case, that specific object is used and any keyword arguments @@ -110,8 +110,8 @@ class TestConstructor(SoupTest): ) msg = str(w[0].message) assert msg.startswith("Keyword arguments to the BeautifulSoup constructor will be ignored.") - self.assertEqual(builder, soup.builder) - self.assertEqual(kwargs, builder.called_with) + assert builder == soup.builder + assert kwargs == builder.called_with def test_parser_markup_rejection(self): # If markup is completely rejected by the parser, an @@ -126,12 +126,11 @@ class TestConstructor(SoupTest): yield markup, None, None, False yield markup, None, None, False + import re - self.assertRaisesRegex( - ParserRejectedMarkup, - "The markup you provided was rejected by the parser. Trying a different parser or a different encoding may help.", - BeautifulSoup, '', builder=Mock, - ) + with pytest.raises(ParserRejectedMarkup) as exc_info: + BeautifulSoup('', builder=Mock) + assert "The markup you provided was rejected by the parser. Trying a different parser or a different encoding may help." in str(exc_info.value) def test_cdata_list_attributes(self): # Most attribute values are represented as scalars, but the @@ -142,14 +141,14 @@ class TestConstructor(SoupTest): # Note that the spaces are stripped for 'class' but not for 'id'. a = soup.a - self.assertEqual(" an id ", a['id']) - self.assertEqual(["a", "class"], a['class']) + assert " an id " == a['id'] + assert ["a", "class"] == a['class'] # TreeBuilder takes an argument called 'mutli_valued_attributes' which lets # you customize or disable this. As always, you can customize the TreeBuilder # by passing in a keyword argument to the BeautifulSoup constructor. soup = self.soup(markup, builder=default_builder, multi_valued_attributes=None) - self.assertEqual(" a class ", soup.a['class']) + assert " a class " == soup.a['class'] # Here are two ways of saying that `id` is a multi-valued # attribute in this context, but 'class' is not. @@ -159,8 +158,8 @@ class TestConstructor(SoupTest): # specifying a parser, but we'll ignore it. soup = self.soup(markup, builder=None, multi_valued_attributes=switcheroo) a = soup.a - self.assertEqual(["an", "id"], a['id']) - self.assertEqual(" a class ", a['class']) + assert ["an", "id"] == a['id'] + assert " a class " == a['class'] def test_replacement_classes(self): # Test the ability to pass in replacements for element classes @@ -221,7 +220,7 @@ class TestConstructor(SoupTest): # Now that parsing was complete, the string_container_stack # (where this information was kept) has been cleared out. - self.assertEqual([], soup.string_container_stack) + assert [] == soup.string_container_stack class TestWarnings(SoupTest): @@ -235,9 +234,7 @@ class TestWarnings(SoupTest): def _assert_no_parser_specified(self, w): warning = self._assert_warning(w, GuessedAtParserWarning) message = str(warning.message) - self.assertTrue( - message.startswith(BeautifulSoup.NO_PARSER_SPECIFIED_WARNING[:60]) - ) + assert message.startswith(BeautifulSoup.NO_PARSER_SPECIFIED_WARNING[:60]) def test_warning_if_no_parser_specified(self): with warnings.catch_warnings(record=True) as w: @@ -252,28 +249,28 @@ class TestWarnings(SoupTest): def test_no_warning_if_explicit_parser_specified(self): with warnings.catch_warnings(record=True) as w: soup = BeautifulSoup("<a><b></b></a>", "html.parser") - self.assertEqual([], w) + assert [] == w def test_parseOnlyThese_renamed_to_parse_only(self): with warnings.catch_warnings(record=True) as w: soup = self.soup("<a><b></b></a>", parseOnlyThese=SoupStrainer("b")) msg = str(w[0].message) - self.assertTrue("parseOnlyThese" in msg) - self.assertTrue("parse_only" in msg) - self.assertEqual(b"<b></b>", soup.encode()) + assert "parseOnlyThese" in msg + assert "parse_only" in msg + assert b"<b></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) + assert "fromEncoding" in msg + assert "from_encoding" in msg + assert "utf8" == soup.original_encoding def test_unrecognized_keyword_argument(self): - self.assertRaises( - TypeError, self.soup, "<a>", no_such_argument=True) + with pytest.raises(TypeError): + self.soup("<a>", no_such_argument=True) def test_disk_file_warning(self): filehandle = tempfile.NamedTemporaryFile() @@ -282,14 +279,14 @@ class TestWarnings(SoupTest): with warnings.catch_warnings(record=True) as w: soup = self.soup(filename) warning = self._assert_warning(w, MarkupResemblesLocatorWarning) - self.assertTrue("looks like a filename" in str(warning.message)) + assert "looks like a filename" in str(warning.message) 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([], w) + assert [] == w def test_directory_warning(self): try: @@ -297,14 +294,14 @@ class TestWarnings(SoupTest): with warnings.catch_warnings(record=True) as w: soup = self.soup(filename) warning = self._assert_warning(w, MarkupResemblesLocatorWarning) - self.assertTrue("looks like a directory" in str(warning.message)) + assert "looks like a directory" in str(warning.message) finally: os.rmdir(filename) # The directory 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([], w) + assert [] == w def test_url_warning_with_bytes_url(self): with warnings.catch_warnings(record=True) as warning_list: @@ -312,7 +309,7 @@ class TestWarnings(SoupTest): warning = self._assert_warning( warning_list, MarkupResemblesLocatorWarning ) - self.assertTrue("looks like a URL" in str(warning.message)) + assert "looks like a URL" in str(warning.message) def test_url_warning_with_unicode_url(self): with warnings.catch_warnings(record=True) as warning_list: @@ -322,21 +319,21 @@ class TestWarnings(SoupTest): warning = self._assert_warning( warning_list, MarkupResemblesLocatorWarning ) - self.assertTrue("looks like a URL" in str(warning.message)) + assert "looks like a URL" in str(warning.message) def test_url_warning_with_bytes_and_space(self): # Here the markup contains something besides a URL, so no warning # is issued. with warnings.catch_warnings(record=True) as warning_list: soup = self.soup(b"http://www.crummybytes.com/ is great") - self.assertFalse(any("looks like a URL" in str(w.message) - for w in warning_list)) + assert not any("looks like a URL" in str(w.message) + for w in warning_list) def test_url_warning_with_unicode_and_space(self): with warnings.catch_warnings(record=True) as warning_list: - soup = self.soup("http://www.crummyuncode.com/ is great") - self.assertFalse(any("looks like a URL" in str(w.message) - for w in warning_list)) + soup = self.soup("http://www.crummyunicode.com/ is great") + assert not any("looks like a URL" in str(w.message) + for w in warning_list) class TestSelectiveParsing(SoupTest): @@ -345,28 +342,26 @@ class TestSelectiveParsing(SoupTest): markup = "No<b>Yes</b><a>No<b>Yes <c>Yes</c></b>" strainer = SoupStrainer("b") soup = self.soup(markup, parse_only=strainer) - self.assertEqual(soup.encode(), b"<b>Yes</b><b>Yes <c>Yes</c></b>") + assert soup.encode() == b"<b>Yes</b><b>Yes <c>Yes</c></b>" -class TestEntitySubstitution(unittest.TestCase): +class TestEntitySubstitution(object): """Standalone tests of the EntitySubstitution class.""" - def setUp(self): + def setup_method(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") + assert 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“”") + assert self.sub.substitute_html(dammit.markup) == "‘’foo“”" def test_html5_entity(self): # Some HTML5 entities correspond to single- or multi-character @@ -399,7 +394,7 @@ class TestEntitySubstitution(unittest.TestCase): template = '3 %s 4' raw = template % u with_entities = template % entity - self.assertEqual(self.sub.substitute_html(raw), with_entities) + assert self.sub.substitute_html(raw) == with_entities def test_html5_entity_with_variation_selector(self): # Some HTML5 entities correspond either to a single-character @@ -407,73 +402,59 @@ class TestEntitySubstitution(unittest.TestCase): # VARIATION SELECTOR 1. We can handle this. data = "fjords \u2294 penguins" markup = "fjords ⊔ penguins" - self.assertEqual(self.sub.substitute_html(data), markup) + assert self.sub.substitute_html(data) == markup data = "fjords \u2294\ufe00 penguins" markup = "fjords ⊔︀ penguins" - self.assertEqual(self.sub.substitute_html(data), markup) + assert self.sub.substitute_html(data) == markup 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) + assert 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"') + assert self.sub.substitute_xml("Welcome", True) == '"Welcome"' + assert 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\"'") + assert 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""') + assert 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) + assert self.sub.substitute_xml(quoted) == quoted def test_xml_quoting_handles_angle_brackets(self): - self.assertEqual( - self.sub.substitute_xml("foo<bar>"), - "foo<bar>") + assert self.sub.substitute_xml("foo<bar>") == "foo<bar>" def test_xml_quoting_handles_ampersands(self): - self.assertEqual(self.sub.substitute_xml("AT&T"), "AT&T") + assert 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") + assert 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") + assert 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) + assert 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() + def setup_method(self): self.unicode_data = '<html><head><meta charset="utf-8"/></head><body><foo>Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!</foo></body></html>' self.utf8_data = self.unicode_data.encode("utf-8") # Just so you know what it looks like. - self.assertEqual( - self.utf8_data, - b'<html><head><meta charset="utf-8"/></head><body><foo>Sacr\xc3\xa9 bleu!</foo></body></html>') + assert self.utf8_data == b'<html><head><meta charset="utf-8"/></head><body><foo>Sacr\xc3\xa9 bleu!</foo></body></html>' def test_ascii_in_unicode_out(self): # ASCII input is converted to Unicode. The original_encoding @@ -488,9 +469,9 @@ class TestEncodingConversion(SoupTest): ascii = b"<foo>a</foo>" 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") + assert isinstance(unicode_output, str) + assert unicode_output == self.document_for(ascii.decode()) + assert soup_from_ascii.original_encoding.lower() == "utf-8" finally: logging.disable(logging.NOTSET) bs4.dammit.chardet_dammit = chardet @@ -499,81 +480,81 @@ class TestEncodingConversion(SoupTest): # 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) + assert soup_from_unicode.decode() == self.unicode_data + assert soup_from_unicode.foo.string == 'Sacr\xe9 bleu!' + assert 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!') + assert soup_from_utf8.decode() == self.unicode_data + assert 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) + assert soup_from_unicode.encode('utf-8') == self.utf8_data @skipIf( 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 = '<div><a \N{SNOWMAN}="snowman"></a></div>' - self.assertEqual(self.soup(markup).div.encode("utf8"), markup.encode("utf8")) + assert self.soup(markup).div.encode("utf8") == markup.encode("utf8") class TestNamedspacedAttribute(SoupTest): def test_name_may_be_none_or_missing(self): a = NamespacedAttribute("xmlns", None) - self.assertEqual(a, "xmlns") + assert a == "xmlns" a = NamespacedAttribute("xmlns", "") - self.assertEqual(a, "xmlns") + assert a == "xmlns" a = NamespacedAttribute("xmlns") - self.assertEqual(a, "xmlns") + assert a == "xmlns" def test_namespace_may_be_none_or_missing(self): a = NamespacedAttribute(None, "tag") - self.assertEqual(a, "tag") + assert a == "tag" a = NamespacedAttribute("", "tag") - self.assertEqual(a, "tag") + assert a == "tag" def test_attribute_is_equivalent_to_colon_separated_string(self): a = NamespacedAttribute("a", "b") - self.assertEqual("a:b", a) + assert "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) + assert a == b # The actual namespace is not considered. c = NamespacedAttribute("a", "b", None) - self.assertEqual(a, c) + assert a == c # But name and prefix are important. d = NamespacedAttribute("a", "z", "c") - self.assertNotEqual(a, d) + assert a != d e = NamespacedAttribute("z", "b", "c") - self.assertNotEqual(a, e) + assert a != e -class TestAttributeValueWithCharsetSubstitution(unittest.TestCase): +class TestAttributeValueWithCharsetSubstitution(object): 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")) + assert "euc-jp" == value + assert "euc-jp" == value.original_value + assert "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")) + assert "text/html; charset=euc-jp" == value + assert "text/html; charset=euc-jp" == value.original_value + assert "text/html; charset=utf8" == value.encode("utf8") diff --git a/bs4/tests/test_tree.py b/bs4/tests/test_tree.py index 59b51d0..aab773f 100644 --- a/bs4/tests/test_tree.py +++ b/bs4/tests/test_tree.py @@ -12,6 +12,7 @@ methods tested here. from pdb import set_trace import copy import pickle +import pytest import re import warnings from bs4 import BeautifulSoup @@ -44,23 +45,23 @@ LXML_PRESENT = (builder_registry.lookup("lxml") is not None) class TreeTest(SoupTest): - def assertSelects(self, tags, should_match): + def assert_selects(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) + assert [tag.string for tag in tags] == should_match - def assertSelectsIDs(self, tags, should_match): + def assert_selects_ids(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) + assert [tag['id'] for tag in tags] == should_match class TestFind(TreeTest): @@ -72,27 +73,27 @@ class TestFind(TreeTest): def test_find_tag(self): soup = self.soup("<a>1</a><b>2</b><a>3</a><b>4</b>") - self.assertEqual(soup.find("b").string, "2") + assert soup.find("b").string == "2" def test_unicode_text_find(self): soup = self.soup('<h1>Räksmörgås</h1>') - self.assertEqual(soup.find(string='Räksmörgås'), 'Räksmörgås') + assert soup.find(string='Räksmörgås') == 'Räksmörgås' def test_unicode_attribute_find(self): soup = self.soup('<h1 id="Räksmörgås">here it is</h1>') str(soup) - self.assertEqual("here it is", soup.find(id='Räksmörgås').text) + assert "here it is" == soup.find(id='Räksmörgås').text def test_find_everything(self): """Test an optimization that finds all tags.""" soup = self.soup("<a>foo</a><b>bar</b>") - self.assertEqual(2, len(soup.find_all())) + assert 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("<a>foo</a><b>bar</b><a>baz</a>") - self.assertEqual(2, len(soup.find_all('a'))) + assert 2 == len(soup.find_all('a')) class TestFindAll(TreeTest): """Basic tests of the find_all() method.""" @@ -101,34 +102,31 @@ class TestFindAll(TreeTest): """You can search the tree for text nodes.""" soup = self.soup("<html>Foo<b>bar</b>\xbb</html>") # Exact match. - self.assertEqual(soup.find_all(string="bar"), ["bar"]) - self.assertEqual(soup.find_all(text="bar"), ["bar"]) + assert soup.find_all(string="bar") == ["bar"] + assert soup.find_all(text="bar") == ["bar"] # Match any of a number of strings. - self.assertEqual( - soup.find_all(text=["Foo", "bar"]), ["Foo", "bar"]) + assert soup.find_all(text=["Foo", "bar"]) == ["Foo", "bar"] # Match a regular expression. - self.assertEqual(soup.find_all(text=re.compile('.*')), - ["Foo", "bar", '\xbb']) + assert soup.find_all(text=re.compile('.*')) == ["Foo", "bar", '\xbb'] # Match anything. - self.assertEqual(soup.find_all(text=True), - ["Foo", "bar", '\xbb']) + assert 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("<a>1</a><a>2</a><a>3</a><a>4</a><a>5</a>") - self.assertSelects(soup.find_all('a', limit=3), ["1", "2", "3"]) - self.assertSelects(soup.find_all('a', limit=1), ["1"]) - self.assertSelects( + self.assert_selects(soup.find_all('a', limit=3), ["1", "2", "3"]) + self.assert_selects(soup.find_all('a', limit=1), ["1"]) + self.assert_selects( soup.find_all('a', limit=10), ["1", "2", "3", "4", "5"]) # A limit of 0 means no limit. - self.assertSelects( + self.assert_selects( soup.find_all('a', limit=0), ["1", "2", "3", "4", "5"]) def test_calling_a_tag_is_calling_findall(self): soup = self.soup("<a>1</a><b>2<a id='foo'>3</a></b>") - self.assertSelects(soup('a', limit=1), ["1"]) - self.assertSelects(soup.b(id="foo"), ["3"]) + self.assert_selects(soup('a', limit=1), ["1"]) + self.assert_selects(soup.b(id="foo"), ["3"]) def test_find_all_with_self_referential_data_structure_does_not_cause_infinite_recursion(self): soup = self.soup("<a></a>") @@ -138,79 +136,78 @@ class TestFindAll(TreeTest): # Without special code in _normalize_search_value, this would cause infinite # recursion. - self.assertEqual([], soup.find_all(l)) + assert [] == soup.find_all(l) def test_find_all_resultset(self): """All find_all calls return a ResultSet""" soup = self.soup("<a></a>") result = soup.find_all("a") - self.assertTrue(hasattr(result, "source")) + assert hasattr(result, "source") result = soup.find_all(True) - self.assertTrue(hasattr(result, "source")) + assert hasattr(result, "source") result = soup.find_all(text="foo") - self.assertTrue(hasattr(result, "source")) + assert hasattr(result, "source") class TestFindAllBasicNamespaces(TreeTest): def test_find_by_namespaced_name(self): soup = self.soup('<mathml:msqrt>4</mathml:msqrt><a svg:fill="red">') - self.assertEqual("4", soup.find("mathml:msqrt").string) - self.assertEqual("a", soup.find(attrs= { "svg:fill" : "red" }).name) + assert "4" == soup.find("mathml:msqrt").string + assert "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() + def setup_method(self): self.tree = self.soup("""<a>First tag.</a> <b>Second tag.</b> <c>Third <a>Nested tag.</a> tag.</c>""") def test_find_all_by_tag_name(self): # Find all the <a> tags. - self.assertSelects( + self.assert_selects( self.tree.find_all('a'), ['First tag.', 'Nested tag.']) def test_find_all_by_name_and_text(self): - self.assertSelects( + self.assert_selects( self.tree.find_all('a', text='First tag.'), ['First tag.']) - self.assertSelects( + self.assert_selects( self.tree.find_all('a', text=True), ['First tag.', 'Nested tag.']) - self.assertSelects( + self.assert_selects( 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.']) + self.assert_selects(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.']) + self.assert_selects(self.tree('a'), ['First tag.', 'Nested tag.']) def test_find_all_by_tag_strainer(self): - self.assertSelects( + self.assert_selects( self.tree.find_all(SoupStrainer('a')), ['First tag.', 'Nested tag.']) def test_find_all_by_tag_names(self): - self.assertSelects( + self.assert_selects( self.tree.find_all(['a', 'b']), ['First tag.', 'Second tag.', 'Nested tag.']) def test_find_all_by_tag_dict(self): - self.assertSelects( + self.assert_selects( 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.assert_selects( self.tree.find_all(re.compile('^[ab]$')), ['First tag.', 'Second tag.', 'Nested tag.']) @@ -224,7 +221,7 @@ class TestFindAllByName(TreeTest): <a id="1">Does not match.</a> <b id="b">Match 2.</a>""") - self.assertSelects( + self.assert_selects( tree.find_all(id_matches_name), ["Match 1.", "Match 2."]) def test_find_with_multi_valued_attribute(self): @@ -234,10 +231,10 @@ class TestFindAllByName(TreeTest): r1 = soup.find('div', 'a d'); r2 = soup.find('div', re.compile(r'a d')); r3, r4 = soup.find_all('div', ['a b', 'a d']); - self.assertEqual('3', r1.string) - self.assertEqual('3', r2.string) - self.assertEqual('1', r3.string) - self.assertEqual('3', r4.string) + assert '3' == r1.string + assert '3' == r2.string + assert '1' == r3.string + assert '3' == r4.string class TestFindAllByAttribute(TreeTest): @@ -250,16 +247,16 @@ class TestFindAllByAttribute(TreeTest): <a id="second"> Non-matching <b id="first">Matching b.</b>a. </a>""") - self.assertSelects(tree.find_all(id='first'), + self.assert_selects(tree.find_all(id='first'), ["Matching a.", "Matching b."]) def test_find_all_by_utf8_attribute_value(self): peace = "םולש".encode("utf8") data = '<a title="םולש"></a>'.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"])) + assert [soup.a] == soup.find_all(title=peace) + assert [soup.a] == soup.find_all(title=peace.decode("utf8")) + assert [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 @@ -273,13 +270,13 @@ class TestFindAllByAttribute(TreeTest): """) # This doesn't do what you want. - self.assertSelects(tree.find_all(name='name1'), + self.assert_selects(tree.find_all(name='name1'), ["A tag called 'name1'."]) # This does what you want. - self.assertSelects(tree.find_all(attrs={'name' : 'name1'}), + self.assert_selects(tree.find_all(attrs={'name' : 'name1'}), ["Name match."]) - self.assertSelects(tree.find_all(attrs={'class' : 'class2'}), + self.assert_selects(tree.find_all(attrs={'class' : 'class2'}), ["Class match."]) def test_find_all_by_class(self): @@ -292,57 +289,57 @@ class TestFindAllByAttribute(TreeTest): # 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.']) + self.assert_selects(tree.find_all('a', class_='1'), ['Class 1.']) + self.assert_selects(tree.find_all('c', class_='3'), ['Class 3 and 4.']) + self.assert_selects(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.']) + self.assert_selects(tree.find_all('a', '1'), ['Class 1.']) + self.assert_selects(tree.find_all(attrs='1'), ['Class 1.', 'Class 1.']) + self.assert_selects(tree.find_all('c', '3'), ['Class 3 and 4.']) + self.assert_selects(tree.find_all('c', '4'), ['Class 3 and 4.']) def test_find_by_class_when_multiple_classes_present(self): tree = self.soup("<gar class='foo bar'>Found it</gar>") f = tree.find_all("gar", class_=re.compile("o")) - self.assertSelects(f, ["Found it"]) + self.assert_selects(f, ["Found it"]) f = tree.find_all("gar", class_=re.compile("a")) - self.assertSelects(f, ["Found it"]) + self.assert_selects(f, ["Found it"]) # If the search fails to match the individual strings "foo" and "bar", # it will be tried against the combined string "foo bar". f = tree.find_all("gar", class_=re.compile("o b")) - self.assertSelects(f, ["Found it"]) + self.assert_selects(f, ["Found it"]) def test_find_all_with_non_dictionary_for_attrs_finds_by_class(self): soup = self.soup("<a class='bar'>Found it</a>") - self.assertSelects(soup.find_all("a", re.compile("ba")), ["Found it"]) + self.assert_selects(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), []) + self.assert_selects(soup.find_all("a", big_attribute_value), []) def small_attribute_value(value): return len(value) <= 3 - self.assertSelects( + self.assert_selects( 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 class="foo bar"></a><a class="foo"></a>') a, a2 = soup.find_all("a") - self.assertEqual([a, a2], soup.find_all("a", "foo")) - self.assertEqual([a], soup.find_all("a", "bar")) + assert [a, a2], soup.find_all("a", "foo") + assert [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")) + assert [a] == soup.find_all("a", class_="foo bar") + assert [a] == soup.find_all("a", "foo bar") + assert [] == soup.find_all("a", "bar foo") def test_find_all_by_attribute_soupstrainer(self): tree = self.soup(""" @@ -350,7 +347,7 @@ class TestFindAllByAttribute(TreeTest): <a id="second">Non-match.</a>""") strainer = SoupStrainer(attrs={'id' : 'first'}) - self.assertSelects(tree.find_all(strainer), ['Match.']) + self.assert_selects(tree.find_all(strainer), ['Match.']) def test_find_all_with_missing_attribute(self): # You can pass in None as the value of an attribute to find_all. @@ -358,7 +355,7 @@ class TestFindAllByAttribute(TreeTest): tree = self.soup("""<a id="1">ID present.</a> <a>No ID present.</a> <a id="">ID is empty.</a>""") - self.assertSelects(tree.find_all('a', id=None), ["No ID present."]) + self.assert_selects(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. @@ -366,7 +363,7 @@ class TestFindAllByAttribute(TreeTest): tree = self.soup("""<a id="1">ID present.</a> <a>No ID present.</a> <a id="">ID is empty.</a>""") - self.assertSelects( + self.assert_selects( tree.find_all(id=True), ["ID present.", "ID is empty."]) def test_find_all_with_numeric_attribute(self): @@ -375,8 +372,8 @@ class TestFindAllByAttribute(TreeTest): <a id="1">Quoted attribute.</a>""") expected = ["Unquoted attribute.", "Quoted attribute."] - self.assertSelects(tree.find_all(id=1), expected) - self.assertSelects(tree.find_all(id="1"), expected) + self.assert_selects(tree.find_all(id=1), expected) + self.assert_selects(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, @@ -385,7 +382,7 @@ class TestFindAllByAttribute(TreeTest): <a id="2">2</a> <a id="3">3</a> <a>No ID.</a>""") - self.assertSelects(tree.find_all(id=["1", "3", "4"]), + self.assert_selects(tree.find_all(id=["1", "3", "4"]), ["1", "3"]) def test_find_all_with_regular_expression_attribute_value(self): @@ -398,27 +395,26 @@ class TestFindAllByAttribute(TreeTest): <a id="b">One b.</a> <a>No ID.</a>""") - self.assertSelects(tree.find_all(id=re.compile("^a+$")), + self.assert_selects(tree.find_all(id=re.compile("^a+$")), ["One a.", "Two as."]) def test_find_by_name_and_containing_string(self): soup = self.soup("<b>foo</b><b>bar</b><a>foo</a>") 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")) + assert [a] == soup.find_all("a", text="foo") + assert [] == soup.find_all("a", text="bar") def test_find_by_name_and_containing_string_when_string_is_buried(self): soup = self.soup("<a>foo</a><a><b><c>foo</c></b></a>") - self.assertEqual(soup.find_all("a"), soup.find_all("a", text="foo")) + assert soup.find_all("a") == soup.find_all("a", text="foo") def test_find_by_attribute_and_containing_string(self): soup = self.soup('<b id="1">foo</b><a id="2">foo</a>') a = soup.a - self.assertEqual([a], soup.find_all(id=2, text="foo")) - self.assertEqual([], soup.find_all(id=1, text="bar")) + assert [a] == soup.find_all(id=2, text="foo") + assert [] == soup.find_all(id=1, text="bar") class TestSmooth(TreeTest): @@ -444,25 +440,25 @@ class TestSmooth(TreeTest): # output. # Since the <span> tag has two children, its .string is None. - self.assertEqual(None, div.span.string) + assert None == div.span.string - self.assertEqual(7, len(div.contents)) + assert 7 == len(div.contents) div.smooth() - self.assertEqual(5, len(div.contents)) + assert 5 == len(div.contents) # The three strings at the beginning of div.contents have been # merged into on string. # - self.assertEqual('abc', div.contents[0]) + assert 'abc' == div.contents[0] # The call is recursive -- the <span> tag was also smoothed. - self.assertEqual('12', div.span.string) + assert '12' == div.span.string # The two comments have _not_ been merged, even though # comments are strings. Merging comments would change the # meaning of the HTML. - self.assertEqual('Comment 1', div.contents[1]) - self.assertEqual('Comment 2', div.contents[2]) + assert 'Comment 1' == div.contents[1] + assert 'Comment 2' == div.contents[2] class TestIndex(TreeTest): @@ -479,15 +475,15 @@ class TestIndex(TreeTest): </div>""") div = tree.div for i, element in enumerate(div.contents): - self.assertEqual(i, div.index(element)) - self.assertRaises(ValueError, tree.index, 1) + assert i == div.index(element) + with pytest.raises(ValueError): + tree.index(1) class TestParentOperations(TreeTest): """Test navigation and searching through an element's parents.""" - def setUp(self): - super(TestParentOperations, self).setUp() + def setup_method(self): self.tree = self.soup('''<ul id="empty"></ul> <ul id="top"> <ul id="middle"> @@ -499,125 +495,124 @@ class TestParentOperations(TreeTest): 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') + assert self.start.parent['id'] == 'bottom' + assert self.start.parent.parent['id'] == 'middle' + assert 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) + assert top_tag.parent == self.tree def test_soup_object_has_no_parent(self): - self.assertEqual(None, self.tree.parent) + assert None == self.tree.parent def test_find_parents(self): - self.assertSelectsIDs( + self.assert_selects_ids( self.start.find_parents('ul'), ['bottom', 'middle', 'top']) - self.assertSelectsIDs( + self.assert_selects_ids( 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') + assert self.start.find_parent('ul')['id'] == 'bottom' + assert 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') + assert 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') + assert 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']) + assert parents, ['bottom', 'middle' == 'top'] class ProximityTest(TreeTest): - def setUp(self): - super(TreeTest, self).setUp() + def setup_method(self): self.tree = self.soup( '<html id="start"><head></head><body><b id="1">One</b><b id="2">Two</b><b id="3">Three</b></body></html>') class TestNextOperations(ProximityTest): - def setUp(self): - super(TestNextOperations, self).setUp() + def setup_method(self): + super(TestNextOperations, self).setup_method() 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") + assert self.start.next_element == "One" + assert 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) + assert 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) + assert self.tree.next_element == None def test_find_all_next(self): - self.assertSelects(self.start.find_all_next('b'), ["Two", "Three"]) + self.assert_selects(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"]) + self.assert_selects(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") + assert self.start.find_next('b')['id'] == '2' + assert 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"]) + assert text.find_next("b").string == "Two" + self.assert_selects(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 <b> tag and its text contents. tag, contents = successors - self.assertEqual(tag['id'], '3') - self.assertEqual(contents, "Three") + assert tag['id'] == '3' + assert contents == "Three" class TestPreviousOperations(ProximityTest): - def setUp(self): - super(TestPreviousOperations, self).setUp() + def setup_method(self): + super(TestPreviousOperations, self).setup_method() 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") + assert self.end.previous_element['id'] == "3" + assert 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) + assert 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) + #assert self.tree.previous_element == None pass def test_find_all_previous(self): # The <b> tag containing the "Three" node is the predecessor # of the "Three" node itself, which is why "Three" shows up # here. - self.assertSelects( + self.assert_selects( self.end.find_all_previous('b'), ["Three", "Two", "One"]) - self.assertSelects(self.end.find_all_previous(id=1), ["One"]) + self.assert_selects(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") + assert self.end.find_previous('b')['id'] == '3' + assert 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( + assert text.find_previous("b").string == "Three" + self.assert_selects( text.find_all_previous("b"), ["Three", "Two", "One"]) def test_previous_generator(self): @@ -627,16 +622,15 @@ class TestPreviousOperations(ProximityTest): # There are four predecessors: the <b> tag containing "One" # the <body> tag, the <head> tag, and the <html> 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") + assert b['id'] == '1' + assert body.name == "body" + assert head.name == "head" + assert html.name == "html" class SiblingTest(TreeTest): - def setUp(self): - super(SiblingTest, self).setUp() + def setup_method(self): markup = '''<html> <span id="1"> <span id="1.1"></span> @@ -657,92 +651,92 @@ class SiblingTest(TreeTest): class TestNextSibling(SiblingTest): - def setUp(self): - super(TestNextSibling, self).setUp() + def setup_method(self): + super(TestNextSibling, self).setup_method() self.start = self.tree.find(id="1") def test_next_sibling_of_root_is_none(self): - self.assertEqual(self.tree.next_sibling, None) + assert 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') + assert self.start.next_sibling['id'] == '2' + assert 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') + assert self.start.next_element['id'] == '1.1' def test_next_sibling_may_not_exist(self): - self.assertEqual(self.tree.html.next_sibling, None) + assert self.tree.html.next_sibling == None nested_span = self.tree.find(id="1.1") - self.assertEqual(nested_span.next_sibling, None) + assert nested_span.next_sibling == None last_span = self.tree.find(id="4") - self.assertEqual(last_span.next_sibling, None) + assert last_span.next_sibling == None def test_find_next_sibling(self): - self.assertEqual(self.start.find_next_sibling('span')['id'], '2') + assert self.start.find_next_sibling('span')['id'] == '2' def test_next_siblings(self): - self.assertSelectsIDs(self.start.find_next_siblings("span"), + self.assert_selects_ids(self.start.find_next_siblings("span"), ['2', '3', '4']) - self.assertSelectsIDs(self.start.find_next_siblings(id='3'), ['3']) + self.assert_selects_ids(self.start.find_next_siblings(id='3'), ['3']) def test_next_sibling_for_text_element(self): soup = self.soup("Foo<b>bar</b>baz") start = soup.find(text="Foo") - self.assertEqual(start.next_sibling.name, 'b') - self.assertEqual(start.next_sibling.next_sibling, 'baz') + assert start.next_sibling.name == 'b' + assert 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) + self.assert_selects(start.find_next_siblings('b'), ['bar']) + assert start.find_next_sibling(text="baz") == "baz" + assert start.find_next_sibling(text="nonesuch") == None class TestPreviousSibling(SiblingTest): - def setUp(self): - super(TestPreviousSibling, self).setUp() + def setup_method(self): + super(TestPreviousSibling, self).setup_method() self.end = self.tree.find(id="4") def test_previous_sibling_of_root_is_none(self): - self.assertEqual(self.tree.previous_sibling, None) + assert 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') + assert self.end.previous_sibling['id'] == '3' + assert 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') + assert self.end.previous_element['id'] == '3.1' def test_previous_sibling_may_not_exist(self): - self.assertEqual(self.tree.html.previous_sibling, None) + assert self.tree.html.previous_sibling == None nested_span = self.tree.find(id="1.1") - self.assertEqual(nested_span.previous_sibling, None) + assert nested_span.previous_sibling == None first_span = self.tree.find(id="1") - self.assertEqual(first_span.previous_sibling, None) + assert first_span.previous_sibling == None def test_find_previous_sibling(self): - self.assertEqual(self.end.find_previous_sibling('span')['id'], '3') + assert self.end.find_previous_sibling('span')['id'] == '3' def test_previous_siblings(self): - self.assertSelectsIDs(self.end.find_previous_siblings("span"), + self.assert_selects_ids(self.end.find_previous_siblings("span"), ['3', '2', '1']) - self.assertSelectsIDs(self.end.find_previous_siblings(id='1'), ['1']) + self.assert_selects_ids(self.end.find_previous_siblings(id='1'), ['1']) def test_previous_sibling_for_text_element(self): soup = self.soup("Foo<b>bar</b>baz") start = soup.find(text="baz") - self.assertEqual(start.previous_sibling.name, 'b') - self.assertEqual(start.previous_sibling.previous_sibling, 'Foo') + assert start.previous_sibling.name == 'b' + assert 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) + self.assert_selects(start.find_previous_siblings('b'), ['bar']) + assert start.find_previous_sibling(text="Foo") == "Foo" + assert start.find_previous_sibling(text="nonesuch") == None class TestTag(SoupTest): @@ -755,18 +749,18 @@ class TestTag(SoupTest): # No list of whitespace-preserving tags -> pretty-print tag._preserve_whitespace_tags = None - self.assertEqual(True, tag._should_pretty_print(0)) + assert True == tag._should_pretty_print(0) # List exists but tag is not on the list -> pretty-print tag.preserve_whitespace_tags = ["some_other_tag"] - self.assertEqual(True, tag._should_pretty_print(1)) + assert True == tag._should_pretty_print(1) # Indent level is None -> don't pretty-print - self.assertEqual(False, tag._should_pretty_print(None)) + assert False == tag._should_pretty_print(None) # Tag is on the whitespace-preserving list -> don't pretty-print tag.preserve_whitespace_tags = ["some_other_tag", "a_tag"] - self.assertEqual(False, tag._should_pretty_print(1)) + assert False == tag._should_pretty_print(1) class TestTagCreation(SoupTest): @@ -774,10 +768,10 @@ class TestTagCreation(SoupTest): def test_new_tag(self): soup = self.soup("") new_tag = soup.new_tag("foo", bar="baz", attrs={"name": "a name"}) - self.assertTrue(isinstance(new_tag, Tag)) - self.assertEqual("foo", new_tag.name) - self.assertEqual(dict(bar="baz", name="a name"), new_tag.attrs) - self.assertEqual(None, new_tag.parent) + assert isinstance(new_tag, Tag) + assert "foo" == new_tag.name + assert dict(bar="baz", name="a name") == new_tag.attrs + assert None == new_tag.parent def test_tag_inherits_self_closing_rules_from_builder(self): if XML_BUILDER_PRESENT: @@ -787,8 +781,8 @@ class TestTagCreation(SoupTest): # Both the <br> and <p> tag are empty-element, just because # they have no contents. - self.assertEqual(b"<br/>", xml_br.encode()) - self.assertEqual(b"<p/>", xml_p.encode()) + assert b"<br/>" == xml_br.encode() + assert b"<p/>" == xml_p.encode() html_soup = BeautifulSoup("", "html.parser") html_br = html_soup.new_tag("br") @@ -796,31 +790,31 @@ class TestTagCreation(SoupTest): # The HTML builder users HTML's rules about which tags are # empty-element tags, and the new tags reflect these rules. - self.assertEqual(b"<br/>", html_br.encode()) - self.assertEqual(b"<p></p>", html_p.encode()) + assert b"<br/>" == html_br.encode() + assert b"<p></p>" == 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)) + assert "foo" == s + assert 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)) + assert "foo" == s + assert isinstance(s, Comment) class TestTreeModification(SoupTest): def test_attribute_modification(self): soup = self.soup('<a id="1"></a>') soup.a['id'] = 2 - self.assertEqual(soup.decode(), self.document_for('<a id="2"></a>')) + assert soup.decode() == self.document_for('<a id="2"></a>') del(soup.a['id']) - self.assertEqual(soup.decode(), self.document_for('<a></a>')) + assert soup.decode() == self.document_for('<a></a>') soup.a['id2'] = 'foo' - self.assertEqual(soup.decode(), self.document_for('<a id2="foo"></a>')) + assert soup.decode() == self.document_for('<a id2="foo"></a>') def test_new_tag_creation(self): builder = builder_registry.lookup('html')() @@ -830,9 +824,7 @@ class TestTreeModification(SoupTest): a['href'] = 'http://foo.com/' soup.body.insert(0, a) soup.body.insert(1, ol) - self.assertEqual( - soup.body.encode(), - b'<body><a href="http://foo.com/"></a><ol></ol></body>') + assert soup.body.encode() == b'<body><a href="http://foo.com/"></a><ol></ol></body>' def test_append_to_contents_moves_tag(self): doc = """<p id="1">Don't leave me <b>here</b>.</p> @@ -845,51 +837,55 @@ class TestTreeModification(SoupTest): soup.find(id='2').append(soup.b) # The <b> tag is now a child of the second paragraph. - self.assertEqual(bold.parent, second_para) + assert bold.parent == second_para - self.assertEqual( - soup.decode(), self.document_for( + assert soup.decode() == self.document_for( '<p id="1">Don\'t leave me .</p>\n' - '<p id="2">Don\'t leave!<b>here</b></p>')) + '<p id="2">Don\'t leave!<b>here</b></p>' + ) def test_replace_with_returns_thing_that_was_replaced(self): text = "<a></a><b><c></c></b>" soup = self.soup(text) a = soup.a new_a = a.replace_with(soup.c) - self.assertEqual(a, new_a) + assert a == new_a def test_unwrap_returns_thing_that_was_replaced(self): text = "<a><b></b><c></c></a>" soup = self.soup(text) a = soup.a new_a = a.unwrap() - self.assertEqual(a, new_a) + assert a == new_a def test_replace_with_and_unwrap_give_useful_exception_when_tag_has_no_parent(self): soup = self.soup("<a><b>Foo</b></a><c>Bar</c>") a = soup.a a.extract() - self.assertEqual(None, a.parent) - self.assertRaises(ValueError, a.unwrap) - self.assertRaises(ValueError, a.replace_with, soup.c) + assert None == a.parent + with pytest.raises(ValueError): + a.unwrap() + with pytest.raises(ValueError): + a.replace_with(soup.c) def test_replace_tag_with_itself(self): text = "<a><b></b><c>Foo<d></d></c></a><a><e></e></a>" soup = self.soup(text) c = soup.c soup.c.replace_with(c) - self.assertEqual(soup.decode(), self.document_for(text)) + assert soup.decode() == self.document_for(text) def test_replace_tag_with_its_parent_raises_exception(self): text = "<a><b></b></a>" soup = self.soup(text) - self.assertRaises(ValueError, soup.b.replace_with, soup.a) + with pytest.raises(ValueError): + soup.b.replace_with(soup.a) def test_insert_tag_into_itself_raises_exception(self): text = "<a><b></b></a>" soup = self.soup(text) - self.assertRaises(ValueError, soup.a.insert, 0, soup.a) + with pytest.raises(ValueError): + soup.a.insert(0, soup.a) def test_insert_beautifulsoup_object_inserts_children(self): """Inserting one BeautifulSoup object into another actually inserts all @@ -905,10 +901,10 @@ class TestTreeModification(SoupTest): assert not isinstance(i, BeautifulSoup) p1, p2, p3, p4 = list(soup.children) - self.assertEqual("And now, a word:", p1.string) - self.assertEqual("p2", p2.string) - self.assertEqual("p3", p3.string) - self.assertEqual("And we're back.", p4.string) + assert "And now, a word:" == p1.string + assert "p2" == p2.string + assert "p3" == p3.string + assert "And we're back." == p4.string def test_replace_with_maintains_next_element_throughout(self): @@ -924,17 +920,17 @@ class TestTreeModification(SoupTest): right.replaceWith('') # The <b> tag is still connected to the tree. - self.assertEqual("three", soup.b.string) + assert "three" == soup.b.string def test_replace_final_node(self): soup = self.soup("<b>Argh!</b>") 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) + assert new_text.previous_element == b + assert new_text.parent == b + assert new_text.previous_element.next_element == new_text + assert new_text.next_element == None def test_consecutive_text_nodes(self): # A builder should never create two consecutive text nodes, @@ -943,28 +939,28 @@ class TestTreeModification(SoupTest): soup = self.soup("<a><b>Argh!</b><c></c></a>") soup.b.insert(1, "Hooray!") - self.assertEqual( - soup.decode(), self.document_for( - "<a><b>Argh!Hooray!</b><c></c></a>")) + assert soup.decode() == self.document_for( + "<a><b>Argh!Hooray!</b><c></c></a>" + ) new_text = soup.find(text="Hooray!") - self.assertEqual(new_text.previous_element, "Argh!") - self.assertEqual(new_text.previous_element.next_element, new_text) + assert new_text.previous_element == "Argh!" + assert 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) + assert new_text.previous_sibling == "Argh!" + assert new_text.previous_sibling.next_sibling == new_text - self.assertEqual(new_text.next_sibling, None) - self.assertEqual(new_text.next_element, soup.c) + assert new_text.next_sibling == None + assert new_text.next_element == soup.c def test_insert_string(self): soup = self.soup("<a></a>") soup.a.insert(0, "bar") soup.a.insert(0, "foo") # The string were added to the tag. - self.assertEqual(["foo", "bar"], soup.a.contents) + assert ["foo", "bar"] == soup.a.contents # And they were converted to NavigableStrings. - self.assertEqual(soup.a.contents[0].next_element, "bar") + assert soup.a.contents[0].next_element == "bar" def test_insert_tag(self): builder = self.default_builder() @@ -974,40 +970,40 @@ class TestTreeModification(SoupTest): magic_tag.insert(0, "the") soup.a.insert(1, magic_tag) - self.assertEqual( - soup.decode(), self.document_for( - "<a><b>Find</b><magictag>the</magictag><c>lady!</c><d></d></a>")) + assert soup.decode() == self.document_for( + "<a><b>Find</b><magictag>the</magictag><c>lady!</c><d></d></a>" + ) # 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) + assert b_tag.next_sibling == magic_tag + assert 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) + assert find.next_element == magic_tag + assert 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) + assert magic_tag.next_sibling == c_tag + assert 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) + assert the.parent == magic_tag + assert the.next_element == c_tag + assert c_tag.previous_element == the def test_append_child_thats_already_at_the_end(self): data = "<a><b></b></a>" soup = self.soup(data) soup.a.append(soup.b) - self.assertEqual(data, soup.decode()) + assert data == soup.decode() def test_extend(self): data = "<a><b><c><d><e><f><g></g></f></e></d></c></b></a>" soup = self.soup(data) l = [soup.g, soup.f, soup.e, soup.d, soup.c, soup.b] soup.a.extend(l) - self.assertEqual("<a><g></g><f></f><e></e><d></d><c></c><b></b></a>", soup.decode()) + assert "<a><g></g><f></f><e></e><d></d><c></c><b></b></a>" == soup.decode() def test_extend_with_another_tags_contents(self): data = '<body><div id="d1"><a>1</a><a>2</a><a>3</a><a>4</a></div><div id="d2"></div></body>' @@ -1015,14 +1011,14 @@ class TestTreeModification(SoupTest): d1 = soup.find('div', id='d1') d2 = soup.find('div', id='d2') d2.extend(d1) - self.assertEqual('<div id="d1"></div>', d1.decode()) - self.assertEqual('<div id="d2"><a>1</a><a>2</a><a>3</a><a>4</a></div>', d2.decode()) + assert '<div id="d1"></div>' == d1.decode() + assert '<div id="d2"><a>1</a><a>2</a><a>3</a><a>4</a></div>' == d2.decode() def test_move_tag_to_beginning_of_parent(self): data = "<a><b></b><c></c><d></d></a>" soup = self.soup(data) soup.a.insert(0, soup.d) - self.assertEqual("<a><d></d><b></b><c></c></a>", soup.decode()) + assert "<a><d></d><b></b><c></c></a>" == soup.decode() def test_insert_works_on_empty_element_tag(self): # This is a little strange, since most HTML parsers don't allow @@ -1031,117 +1027,136 @@ class TestTreeModification(SoupTest): # I'm letting this succeed for now. soup = self.soup("<br/>") soup.br.insert(1, "Contents") - self.assertEqual(str(soup.br), "<br>Contents</br>") + assert str(soup.br) == "<br>Contents</br>" def test_insert_before(self): soup = self.soup("<a>foo</a><b>bar</b>") soup.b.insert_before("BAZ") soup.a.insert_before("QUUX") - self.assertEqual( - soup.decode(), self.document_for("QUUX<a>foo</a>BAZ<b>bar</b>")) + assert soup.decode() == self.document_for( + "QUUX<a>foo</a>BAZ<b>bar</b>" + ) soup.a.insert_before(soup.b) - self.assertEqual( - soup.decode(), self.document_for("QUUX<b>bar</b><a>foo</a>BAZ")) + assert soup.decode() == self.document_for("QUUX<b>bar</b><a>foo</a>BAZ") # Can't insert an element before itself. b = soup.b - self.assertRaises(ValueError, b.insert_before, b) + with pytest.raises(ValueError): + b.insert_before(b) # Can't insert before if an element has no parent. b.extract() - self.assertRaises(ValueError, b.insert_before, "nope") + with pytest.raises(ValueError): + b.insert_before("nope") # Can insert an identical element soup = self.soup("<a>") soup.a.insert_before(soup.new_tag("a")) + + # TODO: OK but what happens? def test_insert_multiple_before(self): soup = self.soup("<a>foo</a><b>bar</b>") soup.b.insert_before("BAZ", " ", "QUUX") soup.a.insert_before("QUUX", " ", "BAZ") - self.assertEqual( - soup.decode(), self.document_for("QUUX BAZ<a>foo</a>BAZ QUUX<b>bar</b>")) + assert soup.decode() == self.document_for( + "QUUX BAZ<a>foo</a>BAZ QUUX<b>bar</b>" + ) soup.a.insert_before(soup.b, "FOO") - self.assertEqual( - soup.decode(), self.document_for("QUUX BAZ<b>bar</b>FOO<a>foo</a>BAZ QUUX")) + assert soup.decode() == self.document_for( + "QUUX BAZ<b>bar</b>FOO<a>foo</a>BAZ QUUX" + ) def test_insert_after(self): soup = self.soup("<a>foo</a><b>bar</b>") soup.b.insert_after("BAZ") soup.a.insert_after("QUUX") - self.assertEqual( - soup.decode(), self.document_for("<a>foo</a>QUUX<b>bar</b>BAZ")) + assert soup.decode() == self.document_for( + "<a>foo</a>QUUX<b>bar</b>BAZ" + ) soup.b.insert_after(soup.a) - self.assertEqual( - soup.decode(), self.document_for("QUUX<b>bar</b><a>foo</a>BAZ")) + assert soup.decode() == self.document_for("QUUX<b>bar</b><a>foo</a>BAZ") # Can't insert an element after itself. b = soup.b - self.assertRaises(ValueError, b.insert_after, b) + with pytest.raises(ValueError): + b.insert_after(b) # Can't insert after if an element has no parent. b.extract() - self.assertRaises(ValueError, b.insert_after, "nope") + with pytest.raises(ValueError): + b.insert_after("nope") # Can insert an identical element soup = self.soup("<a>") soup.a.insert_before(soup.new_tag("a")) + + # TODO: OK but what does it look like? def test_insert_multiple_after(self): soup = self.soup("<a>foo</a><b>bar</b>") soup.b.insert_after("BAZ", " ", "QUUX") soup.a.insert_after("QUUX", " ", "BAZ") - self.assertEqual( - soup.decode(), self.document_for("<a>foo</a>QUUX BAZ<b>bar</b>BAZ QUUX")) + assert soup.decode() == self.document_for( + "<a>foo</a>QUUX BAZ<b>bar</b>BAZ QUUX" + ) soup.b.insert_after(soup.a, "FOO ") - self.assertEqual( - soup.decode(), self.document_for("QUUX BAZ<b>bar</b><a>foo</a>FOO BAZ QUUX")) + assert soup.decode() == self.document_for( + "QUUX BAZ<b>bar</b><a>foo</a>FOO BAZ QUUX" + ) 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) + with pytest.raises(ValueError): + string.insert_after(tag) + with pytest.raises(NotImplementedError): + soup.insert_after(tag) + with pytest.raises(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) + with pytest.raises(ValueError): + string.insert_before(tag) + with pytest.raises(NotImplementedError): + soup.insert_before(tag) + with pytest.raises(ValueError): + tag.insert_before(tag) def test_replace_with(self): soup = self.soup( "<p>There's <b>no</b> business like <b>show</b> business</p>") no, show = soup.find_all('b') show.replace_with(no) - self.assertEqual( - soup.decode(), - self.document_for( - "<p>There's business like <b>no</b> business</p>")) + assert soup.decode() == self.document_for( + "<p>There's business like <b>no</b> business</p>" + ) - self.assertEqual(show.parent, None) - self.assertEqual(no.parent, soup.p) - self.assertEqual(no.next_element, "no") - self.assertEqual(no.next_sibling, " business") + assert show.parent == None + assert no.parent == soup.p + assert no.next_element == "no" + assert no.next_sibling == " business" def test_replace_with_errors(self): # Can't replace a tag that's not part of a tree. a_tag = Tag(name="a") - self.assertRaises(ValueError, a_tag.replace_with, "won't work") + with pytest.raises(ValueError): + a_tag.replace_with("won't work") # Can't replace a tag with its parent. a_tag = self.soup("<a><b></b></a>").a - self.assertRaises(ValueError, a_tag.b.replace_with, a_tag) + with pytest.raises(ValueError): + a_tag.b.replace_with(a_tag) # Or with a list that includes its parent. - self.assertRaises(ValueError, a_tag.b.replace_with, - "string1", a_tag, "string2") + with pytest.raises(ValueError): + a_tag.b.replace_with("string1", a_tag, "string2") def test_replace_with_multiple(self): data = "<a><b></b><c></c></a>" @@ -1152,10 +1167,7 @@ class TestTreeModification(SoupTest): f_tag = soup.new_tag("f") a_string = "Random Text" soup.c.replace_with(d_tag, e_tag, a_string, f_tag) - self.assertEqual( - "<a><b></b><d>Text In D Tag</d><e></e>Random Text<f></f></a>", - soup.decode() - ) + assert soup.decode() == "<a><b></b><d>Text In D Tag</d><e></e>Random Text<f></f></a>" assert soup.b.next_element == d_tag assert d_tag.string.next_element==e_tag assert e_tag.next_element.string == a_string @@ -1165,13 +1177,13 @@ class TestTreeModification(SoupTest): data = "<a><b></b><c></c></a>" soup = self.soup(data) soup.b.replace_with(soup.c) - self.assertEqual("<a><c></c></a>", soup.decode()) + assert "<a><c></c></a>" == soup.decode() def test_replace_last_child(self): data = "<a><b></b><c></c></a>" soup = self.soup(data) soup.c.replace_with(soup.b) - self.assertEqual("<a><b></b></a>", soup.decode()) + assert "<a><b></b></a>" == soup.decode() def test_nested_tag_replace_with(self): soup = self.soup( @@ -1183,85 +1195,82 @@ class TestTreeModification(SoupTest): move_tag = soup.f remove_tag.replace_with(move_tag) - self.assertEqual( - soup.decode(), self.document_for( - "<a>We<f>refuse</f></a><e>to<g>service</g></e>")) + assert soup.decode() == self.document_for( + "<a>We<f>refuse</f></a><e>to<g>service</g></e>" + ) # The <b> 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) + assert remove_tag.parent == None + assert remove_tag.find(text="right").next_element == None + assert remove_tag.previous_element == None + assert remove_tag.next_sibling == None + assert remove_tag.previous_sibling == None # The <f> tag is now connected to the <a> 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) + assert move_tag.parent == soup.a + assert move_tag.previous_element == "We" + assert move_tag.next_element.next_element == soup.e + assert move_tag.next_sibling == None # The gap where the <f> tag used to be has been mended, and # the word "to" is now connected to the <g> 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) + assert to_text.next_element == g_tag + assert to_text.next_sibling == g_tag + assert g_tag.previous_element == to_text + assert g_tag.previous_sibling == to_text def test_unwrap(self): tree = self.soup(""" <p>Unneeded <em>formatting</em> is unneeded</p> """) tree.em.unwrap() - self.assertEqual(tree.em, None) - self.assertEqual(tree.p.text, "Unneeded formatting is unneeded") + assert tree.em == None + assert 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(), "<b>I wish I was bold.</b>") - self.assertEqual( - soup.decode(), self.document_for("<b>I wish I was bold.</b>")) + assert value.decode() == "<b>I wish I was bold.</b>" + assert soup.decode() == self.document_for("<b>I wish I was bold.</b>") def test_wrap_extracts_tag_from_elsewhere(self): soup = self.soup("<b></b>I wish I was bold.") soup.b.next_sibling.wrap(soup.b) - self.assertEqual( - soup.decode(), self.document_for("<b>I wish I was bold.</b>")) + assert soup.decode() == self.document_for("<b>I wish I was bold.</b>") def test_wrap_puts_new_contents_at_the_end(self): soup = self.soup("<b>I like being bold.</b>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( - "<b>I like being bold.I wish I was bold.</b>")) + assert 2 == len(soup.b.contents) + assert soup.decode() == self.document_for( + "<b>I like being bold.I wish I was bold.</b>" + ) def test_extract(self): soup = self.soup( '<html><body>Some content. <div id="nav">Nav crap</div> More content.</body></html>') - self.assertEqual(len(soup.body.contents), 3) + assert len(soup.body.contents) == 3 extracted = soup.find(id="nav").extract() - self.assertEqual( - soup.decode(), "<html><body>Some content. More content.</body></html>") - self.assertEqual(extracted.decode(), '<div id="nav">Nav crap</div>') + assert soup.decode() == "<html><body>Some content. More content.</body></html>" + assert extracted.decode() == '<div id="nav">Nav crap</div>' # 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) + assert len(soup.body.contents) == 2 + assert extracted.parent == None + assert extracted.previous_element == None + assert 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) + assert content_1.next_element == content_2 + assert content_1.next_sibling == content_2 + assert content_2.previous_element == content_1 + assert content_2.previous_sibling == content_1 def test_extract_distinguishes_between_identical_strings(self): soup = self.soup("<a>foo</a><b>bar</b>") @@ -1277,8 +1286,8 @@ class TestTreeModification(SoupTest): # "bar". foo_1.extract() bar_2.extract() - self.assertEqual(foo_2, soup.a.string) - self.assertEqual(bar_2, soup.b.string) + assert foo_2 == soup.a.string + assert bar_2 == soup.b.string def test_extract_multiples_of_same_tag(self): soup = self.soup(""" @@ -1293,7 +1302,7 @@ class TestTreeModification(SoupTest): <script>baz</script> </html>""") [soup.script.extract() for i in soup.find_all("script")] - self.assertEqual("<body>\n\n<a></a>\n</body>", str(soup.body)) + assert "<body>\n\n<a></a>\n</body>" == str(soup.body) def test_extract_works_when_element_is_surrounded_by_identical_strings(self): @@ -1302,7 +1311,7 @@ class TestTreeModification(SoupTest): '<body>hi</body>\n' '</html>') soup.find('body').extract() - self.assertEqual(None, soup.find('body')) + assert None == soup.find('body') def test_clear(self): @@ -1311,13 +1320,13 @@ class TestTreeModification(SoupTest): # clear using extract() a = soup.a soup.p.clear() - self.assertEqual(len(soup.p.contents), 0) - self.assertTrue(hasattr(a, "contents")) + assert len(soup.p.contents) == 0 + assert hasattr(a, "contents") # clear using decompose() em = a.em a.clear(decompose=True) - self.assertEqual(0, len(em.contents)) + assert 0 == len(em.contents) def test_decompose(self): @@ -1327,33 +1336,33 @@ class TestTreeModification(SoupTest): a = p1.a text = p1.em.string for i in [p1, p2, a, text]: - self.assertEqual(False, i.decomposed) + assert False == i.decomposed # This sets p1 and everything beneath it to decomposed. p1.decompose() for i in [p1, a, text]: - self.assertEqual(True, i.decomposed) + assert True == i.decomposed # p2 is unaffected. - self.assertEqual(False, p2.decomposed) + assert False == p2.decomposed def test_string_set(self): """Tag.string = 'string'""" soup = self.soup("<a></a> <b><c></c></b>") soup.a.string = "foo" - self.assertEqual(soup.a.contents, ["foo"]) + assert soup.a.contents == ["foo"] soup.b.string = "bar" - self.assertEqual(soup.b.contents, ["bar"]) + assert soup.b.contents == ["bar"] def test_string_set_does_not_affect_original_string(self): soup = self.soup("<a><b>foo</b><c>bar</c>") soup.b.string = soup.c.string - self.assertEqual(soup.a.encode(), b"<a><b>bar</b><c>bar</c></a>") + assert soup.a.encode() == b"<a><b>bar</b><c>bar</c></a>" def test_set_string_preserves_class_of_string(self): soup = self.soup("<a></a>") cdata = CData("foo") soup.a.string = cdata - self.assertTrue(isinstance(soup.a.string, CData)) + assert isinstance(soup.a.string, CData) class TestElementObjects(SoupTest): """Test various features of element objects.""" @@ -1364,29 +1373,27 @@ class TestElementObjects(SoupTest): # The BeautifulSoup object itself contains one element: the # <top> tag. - self.assertEqual(len(soup.contents), 1) - self.assertEqual(len(soup), 1) + assert len(soup.contents) == 1 + assert len(soup) == 1 # The <top> tag contains three elements: the text node "1", the # <b> tag, and the text node "3". - self.assertEqual(len(soup.top), 3) - self.assertEqual(len(soup.top.contents), 3) + assert len(soup.top) == 3 + assert len(soup.top.contents) == 3 def test_member_access_invokes_find(self): """Accessing a Python member .foo invokes find('foo')""" soup = self.soup('<b><i></i></b>') - self.assertEqual(soup.b, soup.find('b')) - self.assertEqual(soup.b.i, soup.find('b').find('i')) - self.assertEqual(soup.a, None) + assert soup.b == soup.find('b') + assert soup.b.i == soup.find('b').find('i') + assert soup.a == None def test_deprecated_member_access(self): soup = self.soup('<b><i></i></b>') 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. If you really were looking for a tag called bTag, use .find("bTag")', - str(w[0].message)) + assert soup.b == tag + assert '.bTag is deprecated, use .find("b") instead. If you really were looking for a tag called bTag, use .find("bTag")' == str(w[0].message) def test_has_attr(self): """has_attr() checks for the presence of an attribute. @@ -1396,8 +1403,8 @@ class TestElementObjects(SoupTest): checks the tag's chidlren. """ soup = self.soup("<foo attr='bar'>") - self.assertTrue(soup.foo.has_attr('attr')) - self.assertFalse(soup.foo.has_attr('attr2')) + assert soup.foo.has_attr('attr') + assert not soup.foo.has_attr('attr2') def test_attributes_come_out_in_alphabetical_order(self): @@ -1408,68 +1415,66 @@ class TestElementObjects(SoupTest): # A tag that contains only a text node makes that node # available as .string. soup = self.soup("<b>foo</b>") - self.assertEqual(soup.b.string, 'foo') + assert soup.b.string == 'foo' def test_empty_tag_has_no_string(self): # A tag with no children has no .stirng. soup = self.soup("<b></b>") - self.assertEqual(soup.b.string, None) + assert 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("<a>foo<b></b><b></b></b>") - self.assertEqual(soup.b.string, None) + assert soup.b.string == None soup = self.soup("<a>foo<b></b>bar</b>") - self.assertEqual(soup.b.string, None) + assert 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("<a>foo</b>") soup.a.insert(1, "bar") - self.assertEqual(soup.a.string, None) + assert 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("<a><b>foo</b></a>") - self.assertEqual(soup.a.string, "foo") - self.assertEqual(soup.string, "foo") + assert soup.a.string == "foo" + assert soup.string == "foo" def test_lack_of_string(self): """Only a tag containing a single text node has a .string.""" soup = self.soup("<b>f<i>e</i>o</b>") - self.assertFalse(soup.b.string) + assert soup.b.string is None soup = self.soup("<b></b>") - self.assertFalse(soup.b.string) + assert soup.b.string is None def test_all_text(self): """Tag.text and Tag.get_text(sep=u"") -> all child text, concatenated""" soup = self.soup("<a>a<b>r</b> <r> t </r></a>") - 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") + assert soup.a.text == "ar t " + assert soup.a.get_text(strip=True) == "art" + assert soup.a.get_text(",") == "a,r, , t " + assert soup.a.get_text(",", strip=True) == "a,r,t" def test_get_text_ignores_special_string_containers(self): soup = self.soup("foo<!--IGNORE-->bar") - self.assertEqual(soup.get_text(), "foobar") + assert soup.get_text() == "foobar" - self.assertEqual( - soup.get_text(types=(NavigableString, Comment)), "fooIGNOREbar") - self.assertEqual( - soup.get_text(types=None), "fooIGNOREbar") + assert soup.get_text(types=(NavigableString, Comment)) == "fooIGNOREbar" + assert soup.get_text(types=None) == "fooIGNOREbar" soup = self.soup("foo<style>CSS</style><script>Javascript</script>bar") - self.assertEqual(soup.get_text(), "foobar") + assert soup.get_text() == "foobar" def test_all_strings_ignores_special_string_containers(self): soup = self.soup("foo<!--IGNORE-->bar") - self.assertEqual(['foo', 'bar'], list(soup.strings)) + assert ['foo', 'bar'] == list(soup.strings) soup = self.soup("foo<style>CSS</style><script>Javascript</script>bar") - self.assertEqual(['foo', 'bar'], list(soup.strings)) + assert ['foo', 'bar'] == list(soup.strings) def test_string_methods_inside_special_string_container_tags(self): # Strings inside tags like <script> are generally ignored by @@ -1482,29 +1487,25 @@ class TestElementObjects(SoupTest): template = self.soup("<div>a<template><p>Templated <b>text</b>.</p><!--With a comment.--></template></div>") script = self.soup("<div>a<script><!--a comment-->Some text</script></div>") - self.assertEqual(style.div.get_text(), "a") - self.assertEqual(list(style.div.strings), ["a"]) - self.assertEqual(style.div.style.get_text(), "Some CSS") - self.assertEqual(list(style.div.style.strings), - ['Some CSS']) + assert style.div.get_text() == "a" + assert list(style.div.strings) == ["a"] + assert style.div.style.get_text() == "Some CSS" + assert list(style.div.style.strings) == ['Some CSS'] # The comment is not picked up here. That's because it was # parsed into a Comment object, which is not considered # interesting by template.strings. - self.assertEqual(template.div.get_text(), "a") - self.assertEqual(list(template.div.strings), ["a"]) - self.assertEqual(template.div.template.get_text(), "Templated text.") - self.assertEqual(list(template.div.template.strings), - ["Templated ", "text", "."]) + assert template.div.get_text() == "a" + assert list(template.div.strings) == ["a"] + assert template.div.template.get_text() == "Templated text." + assert list(template.div.template.strings) == ["Templated ", "text", "."] # The comment is included here, because it didn't get parsed # into a Comment object--it's part of the Script string. - self.assertEqual(script.div.get_text(), "a") - self.assertEqual(list(script.div.strings), ["a"]) - self.assertEqual(script.div.script.get_text(), - "<!--a comment-->Some text") - self.assertEqual(list(script.div.script.strings), - ['<!--a comment-->Some text']) + assert script.div.get_text() == "a" + assert list(script.div.strings) == ["a"] + assert script.div.script.get_text() == "<!--a comment-->Some text" + assert list(script.div.script.strings) == ['<!--a comment-->Some text'] class TestCDAtaListAttributes(SoupTest): @@ -1512,27 +1513,27 @@ class TestCDAtaListAttributes(SoupTest): """ def test_single_value_becomes_list(self): soup = self.soup("<a class='foo'>") - self.assertEqual(["foo"],soup.a['class']) + assert ["foo"] ==soup.a['class'] def test_multiple_values_becomes_list(self): soup = self.soup("<a class='foo bar'>") - self.assertEqual(["foo", "bar"], soup.a['class']) + assert ["foo", "bar"] == soup.a['class'] def test_multiple_values_separated_by_weird_whitespace(self): soup = self.soup("<a class='foo\tbar\nbaz'>") - self.assertEqual(["foo", "bar", "baz"],soup.a['class']) + assert ["foo", "bar", "baz"] ==soup.a['class'] def test_attributes_joined_into_string_on_output(self): soup = self.soup("<a class='foo\tbar'>") - self.assertEqual(b'<a class="foo bar"></a>', soup.a.encode()) + assert b'<a class="foo bar"></a>' == soup.a.encode() def test_get_attribute_list(self): soup = self.soup("<a id='abc def'>") - self.assertEqual(['abc def'], soup.a.get_attribute_list('id')) + assert ['abc def'] == soup.a.get_attribute_list('id') def test_accept_charset(self): soup = self.soup('<form accept-charset="ISO-8859-1 UTF-8">') - self.assertEqual(['ISO-8859-1', 'UTF-8'], soup.form['accept-charset']) + assert ['ISO-8859-1', 'UTF-8'] == soup.form['accept-charset'] def test_cdata_attribute_applying_only_to_one_tag(self): data = '<a accept-charset="ISO-8859-1 UTF-8"></a>' @@ -1540,20 +1541,18 @@ class TestCDAtaListAttributes(SoupTest): # We saw in another test that accept-charset is a cdata-list # attribute for the <form> tag. But it's not a cdata-list # attribute for any other tag. - self.assertEqual('ISO-8859-1 UTF-8', soup.a['accept-charset']) + assert '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(): + assert None == string.name + with pytest.raises(AttributeError): string.name = 'foo' - self.assertRaises(AttributeError, t) class TestPersistence(SoupTest): "Testing features like pickle and deepcopy." - def setUp(self): - super(TestPersistence, self).setUp() + def setup_method(self): self.page = """<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/transitional.dtd"> <html> @@ -1577,20 +1576,20 @@ class TestPersistence(SoupTest): # 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()) + assert loaded.__class__ == BeautifulSoup + assert 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()) + assert copied.decode() == self.tree.decode() def test_copy_preserves_encoding(self): soup = BeautifulSoup(b'<p> </p>', 'html.parser') encoding = soup.original_encoding copy = soup.__copy__() - self.assertEqual("<p> </p>", str(copy)) - self.assertEqual(encoding, copy.original_encoding) + assert "<p> </p>" == str(copy) + assert encoding == copy.original_encoding def test_copy_preserves_builder_information(self): @@ -1604,18 +1603,11 @@ class TestPersistence(SoupTest): # The TreeBuilder object is no longer availble, but information # obtained from it gets copied over to the new Tag object. - self.assertEqual(tag.sourceline, copied.sourceline) - self.assertEqual(tag.sourcepos, copied.sourcepos) - self.assertEqual( - tag.can_be_empty_element, copied.can_be_empty_element - ) - self.assertEqual( - tag.cdata_list_attributes, copied.cdata_list_attributes - ) - self.assertEqual( - tag.preserve_whitespace_tags, copied.preserve_whitespace_tags - ) - + assert tag.sourceline == copied.sourceline + assert tag.sourcepos == copied.sourcepos + assert tag.can_be_empty_element == copied.can_be_empty_element + assert tag.cdata_list_attributes == copied.cdata_list_attributes + assert tag.preserve_whitespace_tags == copied.preserve_whitespace_tags def test_unicode_pickle(self): # A tree containing Unicode characters can be pickled. @@ -1623,33 +1615,33 @@ class TestPersistence(SoupTest): soup = self.soup(html) dumped = pickle.dumps(soup, pickle.HIGHEST_PROTOCOL) loaded = pickle.loads(dumped) - self.assertEqual(loaded.decode(), soup.decode()) + assert loaded.decode() == soup.decode() def test_copy_navigablestring_is_not_attached_to_tree(self): html = "<b>Foo<a></a></b><b>Bar</b>" 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) + assert s1 == s2 + assert None == s2.parent + assert None == s2.next_element + assert None != s1.next_sibling + assert None == s2.next_sibling + assert None == s2.previous_element def test_copy_navigablestring_subclass_has_same_type(self): html = "<b><!--Foo--></b>" soup = self.soup(html) s1 = soup.string s2 = copy.copy(s1) - self.assertEqual(s1, s2) - self.assertTrue(isinstance(s2, Comment)) + assert s1 == s2 + assert isinstance(s2, Comment) def test_copy_entire_soup(self): html = "<div><b>Foo<a></a></b><b>Bar</b></div>end" soup = self.soup(html) soup_copy = copy.copy(soup) - self.assertEqual(soup, soup_copy) + assert soup == soup_copy def test_copy_tag_copies_contents(self): html = "<div><b>Foo<a></a></b><b>Bar</b></div>end" @@ -1658,18 +1650,18 @@ class TestPersistence(SoupTest): 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) + assert str(div) == str(div_copy) + assert div == div_copy # But they're not the same object. - self.assertFalse(div is div_copy) + assert div is not 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) + assert None == div_copy.parent + assert None == div_copy.previous_element + assert None == div_copy.find(string='Bar').next_element + assert None != div.find(string='Bar').next_element class TestSubstitutions(SoupTest): @@ -1678,36 +1670,34 @@ class TestSubstitutions(SoupTest): 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( - "<b><<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>></b>")) + assert decoded == self.document_for( + "<b><<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>></b>" + ) def test_formatter_html(self): markup = "<br><b><<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>></b>" soup = self.soup(markup) decoded = soup.decode(formatter="html") - self.assertEqual( - decoded, - self.document_for("<br/><b><<Sacré bleu!>></b>")) + assert decoded == self.document_for( + "<br/><b><<Sacré bleu!>></b>" + ) def test_formatter_html5(self): markup = "<br><b><<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>></b>" soup = self.soup(markup) decoded = soup.decode(formatter="html5") - self.assertEqual( - decoded, - self.document_for("<br><b><<Sacré bleu!>></b>")) + assert decoded == self.document_for( + "<br><b><<Sacré bleu!>></b>" + ) def test_formatter_minimal(self): markup = "<b><<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>></b>" 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( - "<b><<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>></b>")) + assert decoded == self.document_for( + "<b><<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>></b>" + ) def test_formatter_null(self): markup = "<b><<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>></b>" @@ -1715,8 +1705,9 @@ class TestSubstitutions(SoupTest): 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("<b><<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>></b>")) + assert decoded == self.document_for( + "<b><<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>></b>" + ) def test_formatter_custom(self): markup = "<b><foo></b><b>bar</b><br/>" @@ -1724,9 +1715,7 @@ class TestSubstitutions(SoupTest): 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("<b><FOO></b><b>BAR</b><br/>")) + assert decoded == self.document_for("<b><FOO></b><b>BAR</b><br/>") def test_formatter_is_run_on_attribute_values(self): markup = '<a href="http://a.com?a=b&c=é">e</a>' @@ -1735,15 +1724,15 @@ class TestSubstitutions(SoupTest): expect_minimal = '<a href="http://a.com?a=b&c=é">e</a>' - self.assertEqual(expect_minimal, a.decode()) - self.assertEqual(expect_minimal, a.decode(formatter="minimal")) + assert expect_minimal == a.decode() + assert expect_minimal == a.decode(formatter="minimal") expect_html = '<a href="http://a.com?a=b&c=é">e</a>' - self.assertEqual(expect_html, a.decode(formatter="html")) + assert expect_html == a.decode(formatter="html") - self.assertEqual(markup, a.decode(formatter=None)) + assert markup == a.decode(formatter=None) expect_upper = '<a href="HTTP://A.COM?A=B&C=É">E</a>' - self.assertEqual(expect_upper, a.decode(formatter=lambda x: x.upper())) + assert expect_upper == a.decode(formatter=lambda x: x.upper()) def test_formatter_skips_script_tag_for_html_documents(self): doc = """ @@ -1752,7 +1741,7 @@ class TestSubstitutions(SoupTest): </script> """ encoded = BeautifulSoup(doc, 'html.parser').encode() - self.assertTrue(b"< < hey > >" in encoded) + assert b"< < hey > >" in encoded def test_formatter_skips_style_tag_for_html_documents(self): doc = """ @@ -1761,34 +1750,32 @@ class TestSubstitutions(SoupTest): </style> """ encoded = BeautifulSoup(doc, 'html.parser').encode() - self.assertTrue(b"< < hey > >" in encoded) + assert b"< < hey > >" in encoded def test_prettify_leaves_preformatted_text_alone(self): soup = self.soup("<div> foo <pre> \tbar\n \n </pre> baz <textarea> eee\nfff\t</textarea></div>") # Everything outside the <pre> tag is reformatted, but everything # inside is left alone. - self.assertEqual( - '<div>\n foo\n <pre> \tbar\n \n </pre>\n baz\n <textarea> eee\nfff\t</textarea>\n</div>', - soup.div.prettify()) + assert '<div>\n foo\n <pre> \tbar\n \n </pre>\n baz\n <textarea> eee\nfff\t</textarea>\n</div>' == soup.div.prettify() def test_prettify_accepts_formatter_function(self): soup = BeautifulSoup("<html><body>foo</body></html>", 'html.parser') pretty = soup.prettify(formatter = lambda x: x.upper()) - self.assertTrue("FOO" in pretty) + assert "FOO" in pretty def test_prettify_outputs_unicode_by_default(self): soup = self.soup("<a></a>") - self.assertEqual(str, type(soup.prettify())) + assert str == type(soup.prettify()) def test_prettify_can_encode_data(self): soup = self.soup("<a></a>") - self.assertEqual(bytes, type(soup.prettify("utf-8"))) + assert bytes == type(soup.prettify("utf-8")) def test_html_entity_substitution_off_by_default(self): markup = "<b>Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!</b>" soup = self.soup(markup) encoded = soup.b.encode("utf-8") - self.assertEqual(encoded, markup.encode('utf-8')) + assert encoded == markup.encode('utf-8') def test_encoding_substitution(self): # Here's the <meta> tag saying that a document is @@ -1798,21 +1785,21 @@ class TestSubstitutions(SoupTest): soup = self.soup(meta_tag) # Parse the document, and the charset apprears unchanged. - self.assertEqual(soup.meta['content'], 'text/html; charset=x-sjis') + assert 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) + assert b"charset=utf-8" in utf_8 euc_jp = soup.encode("euc_jp") - self.assertTrue(b"charset=euc_jp" in euc_jp) + assert b"charset=euc_jp" in euc_jp shift_jis = soup.encode("shift-jis") - self.assertTrue(b"charset=shift-jis" in shift_jis) + assert 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) + assert "charset=utf-16" in utf_16_u def test_encoding_substitution_doesnt_happen_if_tag_is_strained(self): markup = ('<head><meta content="text/html; charset=x-sjis" ' @@ -1823,7 +1810,7 @@ class TestSubstitutions(SoupTest): # sure that doesn't happen. strainer = SoupStrainer('pre') soup = self.soup(markup, parse_only=strainer) - self.assertEqual(soup.contents[0].name, 'pre') + assert soup.contents[0].name == 'pre' class TestEncoding(SoupTest): """Test the ability to encode objects into strings.""" @@ -1831,51 +1818,48 @@ class TestEncoding(SoupTest): def test_unicode_string_can_be_encoded(self): html = "<b>\N{SNOWMAN}</b>" soup = self.soup(html) - self.assertEqual(soup.b.string.encode("utf-8"), - "\N{SNOWMAN}".encode("utf-8")) + assert soup.b.string.encode("utf-8") == "\N{SNOWMAN}".encode("utf-8") def test_tag_containing_unicode_string_can_be_encoded(self): html = "<b>\N{SNOWMAN}</b>" soup = self.soup(html) - self.assertEqual( - soup.b.encode("utf-8"), html.encode("utf-8")) + assert soup.b.encode("utf-8") == html.encode("utf-8") def test_encoding_substitutes_unrecognized_characters_by_default(self): html = "<b>\N{SNOWMAN}</b>" soup = self.soup(html) - self.assertEqual(soup.b.encode("ascii"), b"<b>☃</b>") + assert soup.b.encode("ascii") == b"<b>☃</b>" def test_encoding_can_be_made_strict(self): html = "<b>\N{SNOWMAN}</b>" soup = self.soup(html) - self.assertRaises( - UnicodeEncodeError, soup.encode, "ascii", errors="strict") + with pytest.raises(UnicodeEncodeError): + soup.encode("ascii", errors="strict") def test_decode_contents(self): html = "<b>\N{SNOWMAN}</b>" soup = self.soup(html) - self.assertEqual("\N{SNOWMAN}", soup.b.decode_contents()) + assert "\N{SNOWMAN}" == soup.b.decode_contents() def test_encode_contents(self): html = "<b>\N{SNOWMAN}</b>" soup = self.soup(html) - self.assertEqual( - "\N{SNOWMAN}".encode("utf8"), soup.b.encode_contents( - encoding="utf8")) + assert "\N{SNOWMAN}".encode("utf8") == soup.b.encode_contents( + encoding="utf8" + ) def test_deprecated_renderContents(self): html = "<b>\N{SNOWMAN}</b>" soup = self.soup(html) - self.assertEqual( - "\N{SNOWMAN}".encode("utf8"), soup.b.renderContents()) + assert "\N{SNOWMAN}".encode("utf8") == soup.b.renderContents() def test_repr(self): html = "<b>\N{SNOWMAN}</b>" soup = self.soup(html) if PY3K: - self.assertEqual(html, repr(soup)) + assert html == repr(soup) else: - self.assertEqual(b'<b>\\u2603</b>', repr(soup)) + assert b'<b>\\u2603</b>' == repr(soup) class TestSoupSelector(TreeTest): @@ -1929,82 +1913,81 @@ class TestSoupSelector(TreeTest): </div> """ - def setUp(self): + def setup_method(self): self.soup = BeautifulSoup(self.HTML, 'html.parser') - def assertSelects(self, selector, expected_ids, **kwargs): + def assert_selects(self, selector, expected_ids, **kwargs): el_ids = [el['id'] for el in self.soup.select(selector, **kwargs)] el_ids.sort() expected_ids.sort() - self.assertEqual(expected_ids, el_ids, - "Selector %s, expected [%s], got [%s]" % ( + assert expected_ids == el_ids, "Selector %s, expected [%s], got [%s]" % ( selector, ', '.join(expected_ids), ', '.join(el_ids) - ) ) - assertSelect = assertSelects + assertSelect = assert_selects - def assertSelectMultiple(self, *tests): + def assert_select_multiple(self, *tests): for selector, expected_ids in tests: - self.assertSelect(selector, expected_ids) + self.assert_selects(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']) + assert len(els) == 1 + assert els[0].name == 'title' + assert els[0].contents == ['The title'] def test_one_tag_many(self): els = self.soup.select('div') - self.assertEqual(len(els), 4) + assert len(els) == 4 for div in els: - self.assertEqual(div.name, 'div') + assert div.name == 'div' el = self.soup.select_one('div') - self.assertEqual('main', el['id']) + assert 'main' == el['id'] def test_select_one_returns_none_if_no_match(self): match = self.soup.select_one('nonexistenttag') - self.assertEqual(None, match) + assert None == match def test_tag_in_tag_one(self): els = self.soup.select('div div') - self.assertSelects('div div', ['inner', 'data1']) + self.assert_selects('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']) + self.assert_selects(selector, ['data1', 'main', 'inner', 'footer']) def test_limit(self): - self.assertSelects('html div', ['main'], limit=1) - self.assertSelects('html body div', ['inner', 'main'], limit=2) - self.assertSelects('body div', ['data1', 'main', 'inner', 'footer'], + self.assert_selects('html div', ['main'], limit=1) + self.assert_selects('html body div', ['inner', 'main'], limit=2) + self.assert_selects('body div', ['data1', 'main', 'inner', 'footer'], limit=10) def test_tag_no_match(self): - self.assertEqual(len(self.soup.select('del')), 0) + assert len(self.soup.select('del')) == 0 def test_invalid_tag(self): - self.assertRaises(SelectorSyntaxError, self.soup.select, 'tag%t') + with pytest.raises(SelectorSyntaxError): + self.soup.select('tag%t') def test_select_dashed_tag_ids(self): - self.assertSelects('custom-dashed-tag', ['dash1', 'dash2']) + self.assert_selects('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') + assert dashed[0].name == 'custom-dashed-tag' + assert dashed[0]['id'] == 'dash2' def test_dashed_tag_text(self): - self.assertEqual(self.soup.select('body > custom-dashed-tag')[0].text, 'Hello there.') + assert 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')) + assert self.soup.select('custom-dashed-tag') == self.soup.find_all('custom-dashed-tag') def test_header_tags(self): - self.assertSelectMultiple( + self.assert_select_multiple( ('h1', ['header1']), ('h2', ['header2', 'header3']), ) @@ -2012,53 +1995,53 @@ class TestSoupSelector(TreeTest): 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']) + assert len(els) == 1 + assert els[0].name == 'p' + assert els[0]['class'] == ['onep'] def test_class_mismatched_tag(self): els = self.soup.select('div.onep') - self.assertEqual(len(els), 0) + assert len(els) == 0 def test_one_id(self): for selector in ('div#inner', '#inner', 'div div#inner'): - self.assertSelects(selector, ['inner']) + self.assert_selects(selector, ['inner']) def test_bad_id(self): els = self.soup.select('#doesnotexist') - self.assertEqual(len(els), 0) + assert len(els) == 0 def test_items_in_id(self): els = self.soup.select('div#inner p') - self.assertEqual(len(els), 3) + assert 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')) + assert el.name == 'p' + assert els[1]['class'] == ['onep'] + assert not 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) + assert 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']) + self.assert_selects(selector, ['pmulti']) def test_multi_class_selection(self): for selector in ('.class1.class3', '.class3.class2', '.class1.class2.class3'): - self.assertSelects(selector, ['pmulti']) + self.assert_selects(selector, ['pmulti']) def test_child_selector(self): - self.assertSelects('.s1 > a', ['s1a1', 's1a2']) - self.assertSelects('.s1 > a span', ['s1a2s1']) + self.assert_selects('.s1 > a', ['s1a1', 's1a2']) + self.assert_selects('.s1 > a span', ['s1a2s1']) def test_child_selector_id(self): - self.assertSelects('.s1 > a#s1a2 span', ['s1a2s1']) + self.assert_selects('.s1 > a#s1a2 span', ['s1a2s1']) def test_attribute_equals(self): - self.assertSelectMultiple( + self.assert_select_multiple( ('p[class="onep"]', ['p1']), ('p[id="p1"]', ['p1']), ('[class="onep"]', ['p1']), @@ -2076,7 +2059,7 @@ class TestSoupSelector(TreeTest): ) def test_attribute_tilde(self): - self.assertSelectMultiple( + self.assert_select_multiple( ('p[class~="class1"]', ['pmulti']), ('p[class~="class2"]', ['pmulti']), ('p[class~="class3"]', ['pmulti']), @@ -2090,7 +2073,7 @@ class TestSoupSelector(TreeTest): ) def test_attribute_startswith(self): - self.assertSelectMultiple( + self.assert_select_multiple( ('[rel^="style"]', ['l1']), ('link[rel^="style"]', ['l1']), ('notlink[rel^="notstyle"]', []), @@ -2107,7 +2090,7 @@ class TestSoupSelector(TreeTest): ) def test_attribute_endswith(self): - self.assertSelectMultiple( + self.assert_select_multiple( ('[href$=".css"]', ['l1']), ('link[href$=".css"]', ['l1']), ('link[id$="1"]', ['l1']), @@ -2117,7 +2100,7 @@ class TestSoupSelector(TreeTest): ) def test_attribute_contains(self): - self.assertSelectMultiple( + self.assert_select_multiple( # From test_attribute_startswith ('[rel*="style"]', ['l1']), ('link[rel*="style"]', ['l1']), @@ -2146,7 +2129,7 @@ class TestSoupSelector(TreeTest): ) def test_attribute_exact_or_hypen(self): - self.assertSelectMultiple( + self.assert_select_multiple( ('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']), @@ -2154,7 +2137,7 @@ class TestSoupSelector(TreeTest): ) def test_attribute_exists(self): - self.assertSelectMultiple( + self.assert_select_multiple( ('[rel]', ['l1', 'bob', 'me']), ('link[rel]', ['l1']), ('a[rel]', ['bob', 'me']), @@ -2171,41 +2154,41 @@ class TestSoupSelector(TreeTest): """ soup = BeautifulSoup(html, 'html.parser') [chosen] = soup.select('div[style="display: right"]') - self.assertEqual("yes", chosen.string) + assert "yes" == chosen.string def test_unsupported_pseudoclass(self): - self.assertRaises( - NotImplementedError, self.soup.select, "a:no-such-pseudoclass") + with pytest.raises(NotImplementedError): + self.soup.select("a:no-such-pseudoclass") - self.assertRaises( - SelectorSyntaxError, self.soup.select, "a:nth-of-type(a)") + with pytest.raises(SelectorSyntaxError): + 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') + assert len(els) == 1 + assert 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') + assert len(els) == 1 + assert 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) + assert len(els) == 0 # Zero will select no tags. els = self.soup.select('div p:nth-of-type(0)') - self.assertEqual(len(els), 0) + assert len(els) == 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') + assert len(els) == 1 + assert els[0].string == 'Some text' def test_id_child_selector_nth_of_type(self): - self.assertSelects('#inner > p:nth-of-type(2)', ['p1']) + self.assert_selects('#inner > p:nth-of-type(2)', ['p1']) def test_select_on_element(self): # Other tests operate on the tree; this operates on an element @@ -2214,68 +2197,71 @@ class TestSoupSelector(TreeTest): selected = inner.select("div") # The <div id="inner"> tag was selected. The <div id="footer"> # tag was not. - self.assertSelectsIDs(selected, ['inner', 'data1']) + self.assert_selects_ids(selected, ['inner', 'data1']) def test_overspecified_child_id(self): - self.assertSelects(".fancy #inner", ['inner']) - self.assertSelects(".normal #inner", []) + self.assert_selects(".fancy #inner", ['inner']) + self.assert_selects(".normal #inner", []) def test_adjacent_sibling_selector(self): - self.assertSelects('#p1 + h2', ['header2']) - self.assertSelects('#p1 + h2 + p', ['pmulti']) - self.assertSelects('#p1 + #header2 + .class1', ['pmulti']) - self.assertEqual([], self.soup.select('#p1 + p')) + self.assert_selects('#p1 + h2', ['header2']) + self.assert_selects('#p1 + h2 + p', ['pmulti']) + self.assert_selects('#p1 + #header2 + .class1', ['pmulti']) + assert [] == self.soup.select('#p1 + p') def test_general_sibling_selector(self): - self.assertSelects('#p1 ~ h2', ['header2', 'header3']) - self.assertSelects('#p1 ~ #header2', ['header2']) - self.assertSelects('#p1 ~ h2 + a', ['me']) - self.assertSelects('#p1 ~ h2 + [rel="me"]', ['me']) - self.assertEqual([], self.soup.select('#inner ~ h2')) + self.assert_selects('#p1 ~ h2', ['header2', 'header3']) + self.assert_selects('#p1 ~ #header2', ['header2']) + self.assert_selects('#p1 ~ h2 + a', ['me']) + self.assert_selects('#p1 ~ h2 + [rel="me"]', ['me']) + assert [] == self.soup.select('#inner ~ h2') def test_dangling_combinator(self): - self.assertRaises(SelectorSyntaxError, self.soup.select, 'h1 >') + with pytest.raises(SelectorSyntaxError): + self.soup.select('h1 >') def test_sibling_combinator_wont_select_same_tag_twice(self): - self.assertSelects('p[lang] ~ p', ['lang-en-gb', 'lang-en-us', 'lang-fr']) + self.assert_selects('p[lang] ~ p', ['lang-en-gb', 'lang-en-us', 'lang-fr']) # Test the selector grouping operator (the comma) def test_multiple_select(self): - self.assertSelects('x, y', ['xid', 'yid']) + self.assert_selects('x, y', ['xid', 'yid']) def test_multiple_select_with_no_space(self): - self.assertSelects('x,y', ['xid', 'yid']) + self.assert_selects('x,y', ['xid', 'yid']) def test_multiple_select_with_more_space(self): - self.assertSelects('x, y', ['xid', 'yid']) + self.assert_selects('x, y', ['xid', 'yid']) def test_multiple_select_duplicated(self): - self.assertSelects('x, x', ['xid']) + self.assert_selects('x, x', ['xid']) def test_multiple_select_sibling(self): - self.assertSelects('x, y ~ p[lang=fr]', ['xid', 'lang-fr']) + self.assert_selects('x, y ~ p[lang=fr]', ['xid', 'lang-fr']) def test_multiple_select_tag_and_direct_descendant(self): - self.assertSelects('x, y > z', ['xid', 'zidb']) + self.assert_selects('x, y > z', ['xid', 'zidb']) def test_multiple_select_direct_descendant_and_tags(self): - self.assertSelects('div > x, y, z', ['xid', 'yid', 'zida', 'zidb', 'zidab', 'zidac']) + self.assert_selects('div > x, y, z', ['xid', 'yid', 'zida', 'zidb', 'zidab', 'zidac']) def test_multiple_select_indirect_descendant(self): - self.assertSelects('div x,y, z', ['xid', 'yid', 'zida', 'zidb', 'zidab', 'zidac']) + self.assert_selects('div x,y, z', ['xid', 'yid', 'zida', 'zidb', 'zidab', 'zidac']) def test_invalid_multiple_select(self): - self.assertRaises(SelectorSyntaxError, self.soup.select, ',x, y') - self.assertRaises(SelectorSyntaxError, self.soup.select, 'x,,y') + with pytest.raises(SelectorSyntaxError): + self.soup.select(',x, y') + with pytest.raises(SelectorSyntaxError): + self.soup.select('x,,y') def test_multiple_select_attrs(self): - self.assertSelects('p[lang=en], p[lang=en-gb]', ['lang-en', 'lang-en-gb']) + self.assert_selects('p[lang=en], p[lang=en-gb]', ['lang-en', 'lang-en-gb']) def test_multiple_select_ids(self): - self.assertSelects('x, y > z[id=zida], z[id=zidab], z[id=zidb]', ['xid', 'zidb', 'zidab']) + self.assert_selects('x, y > z[id=zida], z[id=zidab], z[id=zidb]', ['xid', 'zidb', 'zidab']) def test_multiple_select_nested(self): - self.assertSelects('body > div > x, y > z', ['xid', 'zidb']) + self.assert_selects('body > div > x, y > z', ['xid', 'zidb']) def test_select_duplicate_elements(self): # When markup contains duplicate elements, a multiple select @@ -2283,7 +2269,7 @@ class TestSoupSelector(TreeTest): markup = '<div class="c1"/><div class="c2"/><div class="c1"/>' soup = BeautifulSoup(markup, 'html.parser') selected = soup.select(".c1, .c2") - self.assertEqual(3, len(selected)) + assert 3 == len(selected) # Verify that find_all finds the same elements, though because # of an implementation detail it finds them in a different @@ -17,10 +17,11 @@ setup( url="http://www.crummy.com/software/BeautifulSoup/bs4/", download_url = "http://www.crummy.com/software/BeautifulSoup/bs4/download/", description="Screen-scraping library", - python_requires='>3.0.0', + python_requires='>=3.6.0', install_requires=[ "soupsieve >1.2", ], + tests_require=['pytest'], long_description=long_description, long_description_content_type="text/markdown", license="MIT", |