1*4882a593Smuzhiyun# -*- coding: utf-8 -*- 2*4882a593Smuzhiyun"""Tests for Beautiful Soup's tree traversal methods. 3*4882a593Smuzhiyun 4*4882a593SmuzhiyunThe tree traversal methods are the main advantage of using Beautiful 5*4882a593SmuzhiyunSoup over just using a parser. 6*4882a593Smuzhiyun 7*4882a593SmuzhiyunDifferent parsers will build different Beautiful Soup trees given the 8*4882a593Smuzhiyunsame markup, but all Beautiful Soup trees can be traversed with the 9*4882a593Smuzhiyunmethods tested here. 10*4882a593Smuzhiyun""" 11*4882a593Smuzhiyun 12*4882a593Smuzhiyunimport copy 13*4882a593Smuzhiyunimport pickle 14*4882a593Smuzhiyunimport re 15*4882a593Smuzhiyunimport warnings 16*4882a593Smuzhiyunfrom bs4 import BeautifulSoup 17*4882a593Smuzhiyunfrom bs4.builder import builder_registry 18*4882a593Smuzhiyunfrom bs4.element import ( 19*4882a593Smuzhiyun PY3K, 20*4882a593Smuzhiyun CData, 21*4882a593Smuzhiyun Comment, 22*4882a593Smuzhiyun Declaration, 23*4882a593Smuzhiyun Doctype, 24*4882a593Smuzhiyun NavigableString, 25*4882a593Smuzhiyun SoupStrainer, 26*4882a593Smuzhiyun Tag, 27*4882a593Smuzhiyun) 28*4882a593Smuzhiyunfrom bs4.testing import SoupTest 29*4882a593Smuzhiyun 30*4882a593SmuzhiyunXML_BUILDER_PRESENT = (builder_registry.lookup("xml") is not None) 31*4882a593SmuzhiyunLXML_PRESENT = (builder_registry.lookup("lxml") is not None) 32*4882a593Smuzhiyun 33*4882a593Smuzhiyunclass TreeTest(SoupTest): 34*4882a593Smuzhiyun 35*4882a593Smuzhiyun def assertSelects(self, tags, should_match): 36*4882a593Smuzhiyun """Make sure that the given tags have the correct text. 37*4882a593Smuzhiyun 38*4882a593Smuzhiyun This is used in tests that define a bunch of tags, each 39*4882a593Smuzhiyun containing a single string, and then select certain strings by 40*4882a593Smuzhiyun some mechanism. 41*4882a593Smuzhiyun """ 42*4882a593Smuzhiyun self.assertEqual([tag.string for tag in tags], should_match) 43*4882a593Smuzhiyun 44*4882a593Smuzhiyun def assertSelectsIDs(self, tags, should_match): 45*4882a593Smuzhiyun """Make sure that the given tags have the correct IDs. 46*4882a593Smuzhiyun 47*4882a593Smuzhiyun This is used in tests that define a bunch of tags, each 48*4882a593Smuzhiyun containing a single string, and then select certain strings by 49*4882a593Smuzhiyun some mechanism. 50*4882a593Smuzhiyun """ 51*4882a593Smuzhiyun self.assertEqual([tag['id'] for tag in tags], should_match) 52*4882a593Smuzhiyun 53*4882a593Smuzhiyun 54*4882a593Smuzhiyunclass TestFind(TreeTest): 55*4882a593Smuzhiyun """Basic tests of the find() method. 56*4882a593Smuzhiyun 57*4882a593Smuzhiyun find() just calls find_all() with limit=1, so it's not tested all 58*4882a593Smuzhiyun that thouroughly here. 59*4882a593Smuzhiyun """ 60*4882a593Smuzhiyun 61*4882a593Smuzhiyun def test_find_tag(self): 62*4882a593Smuzhiyun soup = self.soup("<a>1</a><b>2</b><a>3</a><b>4</b>") 63*4882a593Smuzhiyun self.assertEqual(soup.find("b").string, "2") 64*4882a593Smuzhiyun 65*4882a593Smuzhiyun def test_unicode_text_find(self): 66*4882a593Smuzhiyun soup = self.soup('<h1>Räksmörgås</h1>') 67*4882a593Smuzhiyun self.assertEqual(soup.find(string='Räksmörgås'), 'Räksmörgås') 68*4882a593Smuzhiyun 69*4882a593Smuzhiyun def test_unicode_attribute_find(self): 70*4882a593Smuzhiyun soup = self.soup('<h1 id="Räksmörgås">here it is</h1>') 71*4882a593Smuzhiyun str(soup) 72*4882a593Smuzhiyun self.assertEqual("here it is", soup.find(id='Räksmörgås').text) 73*4882a593Smuzhiyun 74*4882a593Smuzhiyun 75*4882a593Smuzhiyun def test_find_everything(self): 76*4882a593Smuzhiyun """Test an optimization that finds all tags.""" 77*4882a593Smuzhiyun soup = self.soup("<a>foo</a><b>bar</b>") 78*4882a593Smuzhiyun self.assertEqual(2, len(soup.find_all())) 79*4882a593Smuzhiyun 80*4882a593Smuzhiyun def test_find_everything_with_name(self): 81*4882a593Smuzhiyun """Test an optimization that finds all tags with a given name.""" 82*4882a593Smuzhiyun soup = self.soup("<a>foo</a><b>bar</b><a>baz</a>") 83*4882a593Smuzhiyun self.assertEqual(2, len(soup.find_all('a'))) 84*4882a593Smuzhiyun 85*4882a593Smuzhiyunclass TestFindAll(TreeTest): 86*4882a593Smuzhiyun """Basic tests of the find_all() method.""" 87*4882a593Smuzhiyun 88*4882a593Smuzhiyun def test_find_all_text_nodes(self): 89*4882a593Smuzhiyun """You can search the tree for text nodes.""" 90*4882a593Smuzhiyun soup = self.soup("<html>Foo<b>bar</b>\xbb</html>") 91*4882a593Smuzhiyun # Exact match. 92*4882a593Smuzhiyun self.assertEqual(soup.find_all(string="bar"), ["bar"]) 93*4882a593Smuzhiyun self.assertEqual(soup.find_all(text="bar"), ["bar"]) 94*4882a593Smuzhiyun # Match any of a number of strings. 95*4882a593Smuzhiyun self.assertEqual( 96*4882a593Smuzhiyun soup.find_all(text=["Foo", "bar"]), ["Foo", "bar"]) 97*4882a593Smuzhiyun # Match a regular expression. 98*4882a593Smuzhiyun self.assertEqual(soup.find_all(text=re.compile('.*')), 99*4882a593Smuzhiyun ["Foo", "bar", '\xbb']) 100*4882a593Smuzhiyun # Match anything. 101*4882a593Smuzhiyun self.assertEqual(soup.find_all(text=True), 102*4882a593Smuzhiyun ["Foo", "bar", '\xbb']) 103*4882a593Smuzhiyun 104*4882a593Smuzhiyun def test_find_all_limit(self): 105*4882a593Smuzhiyun """You can limit the number of items returned by find_all.""" 106*4882a593Smuzhiyun soup = self.soup("<a>1</a><a>2</a><a>3</a><a>4</a><a>5</a>") 107*4882a593Smuzhiyun self.assertSelects(soup.find_all('a', limit=3), ["1", "2", "3"]) 108*4882a593Smuzhiyun self.assertSelects(soup.find_all('a', limit=1), ["1"]) 109*4882a593Smuzhiyun self.assertSelects( 110*4882a593Smuzhiyun soup.find_all('a', limit=10), ["1", "2", "3", "4", "5"]) 111*4882a593Smuzhiyun 112*4882a593Smuzhiyun # A limit of 0 means no limit. 113*4882a593Smuzhiyun self.assertSelects( 114*4882a593Smuzhiyun soup.find_all('a', limit=0), ["1", "2", "3", "4", "5"]) 115*4882a593Smuzhiyun 116*4882a593Smuzhiyun def test_calling_a_tag_is_calling_findall(self): 117*4882a593Smuzhiyun soup = self.soup("<a>1</a><b>2<a id='foo'>3</a></b>") 118*4882a593Smuzhiyun self.assertSelects(soup('a', limit=1), ["1"]) 119*4882a593Smuzhiyun self.assertSelects(soup.b(id="foo"), ["3"]) 120*4882a593Smuzhiyun 121*4882a593Smuzhiyun def test_find_all_with_self_referential_data_structure_does_not_cause_infinite_recursion(self): 122*4882a593Smuzhiyun soup = self.soup("<a></a>") 123*4882a593Smuzhiyun # Create a self-referential list. 124*4882a593Smuzhiyun l = [] 125*4882a593Smuzhiyun l.append(l) 126*4882a593Smuzhiyun 127*4882a593Smuzhiyun # Without special code in _normalize_search_value, this would cause infinite 128*4882a593Smuzhiyun # recursion. 129*4882a593Smuzhiyun self.assertEqual([], soup.find_all(l)) 130*4882a593Smuzhiyun 131*4882a593Smuzhiyun def test_find_all_resultset(self): 132*4882a593Smuzhiyun """All find_all calls return a ResultSet""" 133*4882a593Smuzhiyun soup = self.soup("<a></a>") 134*4882a593Smuzhiyun result = soup.find_all("a") 135*4882a593Smuzhiyun self.assertTrue(hasattr(result, "source")) 136*4882a593Smuzhiyun 137*4882a593Smuzhiyun result = soup.find_all(True) 138*4882a593Smuzhiyun self.assertTrue(hasattr(result, "source")) 139*4882a593Smuzhiyun 140*4882a593Smuzhiyun result = soup.find_all(text="foo") 141*4882a593Smuzhiyun self.assertTrue(hasattr(result, "source")) 142*4882a593Smuzhiyun 143*4882a593Smuzhiyun 144*4882a593Smuzhiyunclass TestFindAllBasicNamespaces(TreeTest): 145*4882a593Smuzhiyun 146*4882a593Smuzhiyun def test_find_by_namespaced_name(self): 147*4882a593Smuzhiyun soup = self.soup('<mathml:msqrt>4</mathml:msqrt><a svg:fill="red">') 148*4882a593Smuzhiyun self.assertEqual("4", soup.find("mathml:msqrt").string) 149*4882a593Smuzhiyun self.assertEqual("a", soup.find(attrs= { "svg:fill" : "red" }).name) 150*4882a593Smuzhiyun 151*4882a593Smuzhiyun 152*4882a593Smuzhiyunclass TestFindAllByName(TreeTest): 153*4882a593Smuzhiyun """Test ways of finding tags by tag name.""" 154*4882a593Smuzhiyun 155*4882a593Smuzhiyun def setUp(self): 156*4882a593Smuzhiyun super(TreeTest, self).setUp() 157*4882a593Smuzhiyun self.tree = self.soup("""<a>First tag.</a> 158*4882a593Smuzhiyun <b>Second tag.</b> 159*4882a593Smuzhiyun <c>Third <a>Nested tag.</a> tag.</c>""") 160*4882a593Smuzhiyun 161*4882a593Smuzhiyun def test_find_all_by_tag_name(self): 162*4882a593Smuzhiyun # Find all the <a> tags. 163*4882a593Smuzhiyun self.assertSelects( 164*4882a593Smuzhiyun self.tree.find_all('a'), ['First tag.', 'Nested tag.']) 165*4882a593Smuzhiyun 166*4882a593Smuzhiyun def test_find_all_by_name_and_text(self): 167*4882a593Smuzhiyun self.assertSelects( 168*4882a593Smuzhiyun self.tree.find_all('a', text='First tag.'), ['First tag.']) 169*4882a593Smuzhiyun 170*4882a593Smuzhiyun self.assertSelects( 171*4882a593Smuzhiyun self.tree.find_all('a', text=True), ['First tag.', 'Nested tag.']) 172*4882a593Smuzhiyun 173*4882a593Smuzhiyun self.assertSelects( 174*4882a593Smuzhiyun self.tree.find_all('a', text=re.compile("tag")), 175*4882a593Smuzhiyun ['First tag.', 'Nested tag.']) 176*4882a593Smuzhiyun 177*4882a593Smuzhiyun 178*4882a593Smuzhiyun def test_find_all_on_non_root_element(self): 179*4882a593Smuzhiyun # You can call find_all on any node, not just the root. 180*4882a593Smuzhiyun self.assertSelects(self.tree.c.find_all('a'), ['Nested tag.']) 181*4882a593Smuzhiyun 182*4882a593Smuzhiyun def test_calling_element_invokes_find_all(self): 183*4882a593Smuzhiyun self.assertSelects(self.tree('a'), ['First tag.', 'Nested tag.']) 184*4882a593Smuzhiyun 185*4882a593Smuzhiyun def test_find_all_by_tag_strainer(self): 186*4882a593Smuzhiyun self.assertSelects( 187*4882a593Smuzhiyun self.tree.find_all(SoupStrainer('a')), 188*4882a593Smuzhiyun ['First tag.', 'Nested tag.']) 189*4882a593Smuzhiyun 190*4882a593Smuzhiyun def test_find_all_by_tag_names(self): 191*4882a593Smuzhiyun self.assertSelects( 192*4882a593Smuzhiyun self.tree.find_all(['a', 'b']), 193*4882a593Smuzhiyun ['First tag.', 'Second tag.', 'Nested tag.']) 194*4882a593Smuzhiyun 195*4882a593Smuzhiyun def test_find_all_by_tag_dict(self): 196*4882a593Smuzhiyun self.assertSelects( 197*4882a593Smuzhiyun self.tree.find_all({'a' : True, 'b' : True}), 198*4882a593Smuzhiyun ['First tag.', 'Second tag.', 'Nested tag.']) 199*4882a593Smuzhiyun 200*4882a593Smuzhiyun def test_find_all_by_tag_re(self): 201*4882a593Smuzhiyun self.assertSelects( 202*4882a593Smuzhiyun self.tree.find_all(re.compile('^[ab]$')), 203*4882a593Smuzhiyun ['First tag.', 'Second tag.', 'Nested tag.']) 204*4882a593Smuzhiyun 205*4882a593Smuzhiyun def test_find_all_with_tags_matching_method(self): 206*4882a593Smuzhiyun # You can define an oracle method that determines whether 207*4882a593Smuzhiyun # a tag matches the search. 208*4882a593Smuzhiyun def id_matches_name(tag): 209*4882a593Smuzhiyun return tag.name == tag.get('id') 210*4882a593Smuzhiyun 211*4882a593Smuzhiyun tree = self.soup("""<a id="a">Match 1.</a> 212*4882a593Smuzhiyun <a id="1">Does not match.</a> 213*4882a593Smuzhiyun <b id="b">Match 2.</a>""") 214*4882a593Smuzhiyun 215*4882a593Smuzhiyun self.assertSelects( 216*4882a593Smuzhiyun tree.find_all(id_matches_name), ["Match 1.", "Match 2."]) 217*4882a593Smuzhiyun 218*4882a593Smuzhiyun 219*4882a593Smuzhiyunclass TestFindAllByAttribute(TreeTest): 220*4882a593Smuzhiyun 221*4882a593Smuzhiyun def test_find_all_by_attribute_name(self): 222*4882a593Smuzhiyun # You can pass in keyword arguments to find_all to search by 223*4882a593Smuzhiyun # attribute. 224*4882a593Smuzhiyun tree = self.soup(""" 225*4882a593Smuzhiyun <a id="first">Matching a.</a> 226*4882a593Smuzhiyun <a id="second"> 227*4882a593Smuzhiyun Non-matching <b id="first">Matching b.</b>a. 228*4882a593Smuzhiyun </a>""") 229*4882a593Smuzhiyun self.assertSelects(tree.find_all(id='first'), 230*4882a593Smuzhiyun ["Matching a.", "Matching b."]) 231*4882a593Smuzhiyun 232*4882a593Smuzhiyun def test_find_all_by_utf8_attribute_value(self): 233*4882a593Smuzhiyun peace = "םולש".encode("utf8") 234*4882a593Smuzhiyun data = '<a title="םולש"></a>'.encode("utf8") 235*4882a593Smuzhiyun soup = self.soup(data) 236*4882a593Smuzhiyun self.assertEqual([soup.a], soup.find_all(title=peace)) 237*4882a593Smuzhiyun self.assertEqual([soup.a], soup.find_all(title=peace.decode("utf8"))) 238*4882a593Smuzhiyun self.assertEqual([soup.a], soup.find_all(title=[peace, "something else"])) 239*4882a593Smuzhiyun 240*4882a593Smuzhiyun def test_find_all_by_attribute_dict(self): 241*4882a593Smuzhiyun # You can pass in a dictionary as the argument 'attrs'. This 242*4882a593Smuzhiyun # lets you search for attributes like 'name' (a fixed argument 243*4882a593Smuzhiyun # to find_all) and 'class' (a reserved word in Python.) 244*4882a593Smuzhiyun tree = self.soup(""" 245*4882a593Smuzhiyun <a name="name1" class="class1">Name match.</a> 246*4882a593Smuzhiyun <a name="name2" class="class2">Class match.</a> 247*4882a593Smuzhiyun <a name="name3" class="class3">Non-match.</a> 248*4882a593Smuzhiyun <name1>A tag called 'name1'.</name1> 249*4882a593Smuzhiyun """) 250*4882a593Smuzhiyun 251*4882a593Smuzhiyun # This doesn't do what you want. 252*4882a593Smuzhiyun self.assertSelects(tree.find_all(name='name1'), 253*4882a593Smuzhiyun ["A tag called 'name1'."]) 254*4882a593Smuzhiyun # This does what you want. 255*4882a593Smuzhiyun self.assertSelects(tree.find_all(attrs={'name' : 'name1'}), 256*4882a593Smuzhiyun ["Name match."]) 257*4882a593Smuzhiyun 258*4882a593Smuzhiyun self.assertSelects(tree.find_all(attrs={'class' : 'class2'}), 259*4882a593Smuzhiyun ["Class match."]) 260*4882a593Smuzhiyun 261*4882a593Smuzhiyun def test_find_all_by_class(self): 262*4882a593Smuzhiyun tree = self.soup(""" 263*4882a593Smuzhiyun <a class="1">Class 1.</a> 264*4882a593Smuzhiyun <a class="2">Class 2.</a> 265*4882a593Smuzhiyun <b class="1">Class 1.</b> 266*4882a593Smuzhiyun <c class="3 4">Class 3 and 4.</c> 267*4882a593Smuzhiyun """) 268*4882a593Smuzhiyun 269*4882a593Smuzhiyun # Passing in the class_ keyword argument will search against 270*4882a593Smuzhiyun # the 'class' attribute. 271*4882a593Smuzhiyun self.assertSelects(tree.find_all('a', class_='1'), ['Class 1.']) 272*4882a593Smuzhiyun self.assertSelects(tree.find_all('c', class_='3'), ['Class 3 and 4.']) 273*4882a593Smuzhiyun self.assertSelects(tree.find_all('c', class_='4'), ['Class 3 and 4.']) 274*4882a593Smuzhiyun 275*4882a593Smuzhiyun # Passing in a string to 'attrs' will also search the CSS class. 276*4882a593Smuzhiyun self.assertSelects(tree.find_all('a', '1'), ['Class 1.']) 277*4882a593Smuzhiyun self.assertSelects(tree.find_all(attrs='1'), ['Class 1.', 'Class 1.']) 278*4882a593Smuzhiyun self.assertSelects(tree.find_all('c', '3'), ['Class 3 and 4.']) 279*4882a593Smuzhiyun self.assertSelects(tree.find_all('c', '4'), ['Class 3 and 4.']) 280*4882a593Smuzhiyun 281*4882a593Smuzhiyun def test_find_by_class_when_multiple_classes_present(self): 282*4882a593Smuzhiyun tree = self.soup("<gar class='foo bar'>Found it</gar>") 283*4882a593Smuzhiyun 284*4882a593Smuzhiyun f = tree.find_all("gar", class_=re.compile("o")) 285*4882a593Smuzhiyun self.assertSelects(f, ["Found it"]) 286*4882a593Smuzhiyun 287*4882a593Smuzhiyun f = tree.find_all("gar", class_=re.compile("a")) 288*4882a593Smuzhiyun self.assertSelects(f, ["Found it"]) 289*4882a593Smuzhiyun 290*4882a593Smuzhiyun # Since the class is not the string "foo bar", but the two 291*4882a593Smuzhiyun # strings "foo" and "bar", this will not find anything. 292*4882a593Smuzhiyun f = tree.find_all("gar", class_=re.compile("o b")) 293*4882a593Smuzhiyun self.assertSelects(f, []) 294*4882a593Smuzhiyun 295*4882a593Smuzhiyun def test_find_all_with_non_dictionary_for_attrs_finds_by_class(self): 296*4882a593Smuzhiyun soup = self.soup("<a class='bar'>Found it</a>") 297*4882a593Smuzhiyun 298*4882a593Smuzhiyun self.assertSelects(soup.find_all("a", re.compile("ba")), ["Found it"]) 299*4882a593Smuzhiyun 300*4882a593Smuzhiyun def big_attribute_value(value): 301*4882a593Smuzhiyun return len(value) > 3 302*4882a593Smuzhiyun 303*4882a593Smuzhiyun self.assertSelects(soup.find_all("a", big_attribute_value), []) 304*4882a593Smuzhiyun 305*4882a593Smuzhiyun def small_attribute_value(value): 306*4882a593Smuzhiyun return len(value) <= 3 307*4882a593Smuzhiyun 308*4882a593Smuzhiyun self.assertSelects( 309*4882a593Smuzhiyun soup.find_all("a", small_attribute_value), ["Found it"]) 310*4882a593Smuzhiyun 311*4882a593Smuzhiyun def test_find_all_with_string_for_attrs_finds_multiple_classes(self): 312*4882a593Smuzhiyun soup = self.soup('<a class="foo bar"></a><a class="foo"></a>') 313*4882a593Smuzhiyun a, a2 = soup.find_all("a") 314*4882a593Smuzhiyun self.assertEqual([a, a2], soup.find_all("a", "foo")) 315*4882a593Smuzhiyun self.assertEqual([a], soup.find_all("a", "bar")) 316*4882a593Smuzhiyun 317*4882a593Smuzhiyun # If you specify the class as a string that contains a 318*4882a593Smuzhiyun # space, only that specific value will be found. 319*4882a593Smuzhiyun self.assertEqual([a], soup.find_all("a", class_="foo bar")) 320*4882a593Smuzhiyun self.assertEqual([a], soup.find_all("a", "foo bar")) 321*4882a593Smuzhiyun self.assertEqual([], soup.find_all("a", "bar foo")) 322*4882a593Smuzhiyun 323*4882a593Smuzhiyun def test_find_all_by_attribute_soupstrainer(self): 324*4882a593Smuzhiyun tree = self.soup(""" 325*4882a593Smuzhiyun <a id="first">Match.</a> 326*4882a593Smuzhiyun <a id="second">Non-match.</a>""") 327*4882a593Smuzhiyun 328*4882a593Smuzhiyun strainer = SoupStrainer(attrs={'id' : 'first'}) 329*4882a593Smuzhiyun self.assertSelects(tree.find_all(strainer), ['Match.']) 330*4882a593Smuzhiyun 331*4882a593Smuzhiyun def test_find_all_with_missing_atribute(self): 332*4882a593Smuzhiyun # You can pass in None as the value of an attribute to find_all. 333*4882a593Smuzhiyun # This will match tags that do not have that attribute set. 334*4882a593Smuzhiyun tree = self.soup("""<a id="1">ID present.</a> 335*4882a593Smuzhiyun <a>No ID present.</a> 336*4882a593Smuzhiyun <a id="">ID is empty.</a>""") 337*4882a593Smuzhiyun self.assertSelects(tree.find_all('a', id=None), ["No ID present."]) 338*4882a593Smuzhiyun 339*4882a593Smuzhiyun def test_find_all_with_defined_attribute(self): 340*4882a593Smuzhiyun # You can pass in None as the value of an attribute to find_all. 341*4882a593Smuzhiyun # This will match tags that have that attribute set to any value. 342*4882a593Smuzhiyun tree = self.soup("""<a id="1">ID present.</a> 343*4882a593Smuzhiyun <a>No ID present.</a> 344*4882a593Smuzhiyun <a id="">ID is empty.</a>""") 345*4882a593Smuzhiyun self.assertSelects( 346*4882a593Smuzhiyun tree.find_all(id=True), ["ID present.", "ID is empty."]) 347*4882a593Smuzhiyun 348*4882a593Smuzhiyun def test_find_all_with_numeric_attribute(self): 349*4882a593Smuzhiyun # If you search for a number, it's treated as a string. 350*4882a593Smuzhiyun tree = self.soup("""<a id=1>Unquoted attribute.</a> 351*4882a593Smuzhiyun <a id="1">Quoted attribute.</a>""") 352*4882a593Smuzhiyun 353*4882a593Smuzhiyun expected = ["Unquoted attribute.", "Quoted attribute."] 354*4882a593Smuzhiyun self.assertSelects(tree.find_all(id=1), expected) 355*4882a593Smuzhiyun self.assertSelects(tree.find_all(id="1"), expected) 356*4882a593Smuzhiyun 357*4882a593Smuzhiyun def test_find_all_with_list_attribute_values(self): 358*4882a593Smuzhiyun # You can pass a list of attribute values instead of just one, 359*4882a593Smuzhiyun # and you'll get tags that match any of the values. 360*4882a593Smuzhiyun tree = self.soup("""<a id="1">1</a> 361*4882a593Smuzhiyun <a id="2">2</a> 362*4882a593Smuzhiyun <a id="3">3</a> 363*4882a593Smuzhiyun <a>No ID.</a>""") 364*4882a593Smuzhiyun self.assertSelects(tree.find_all(id=["1", "3", "4"]), 365*4882a593Smuzhiyun ["1", "3"]) 366*4882a593Smuzhiyun 367*4882a593Smuzhiyun def test_find_all_with_regular_expression_attribute_value(self): 368*4882a593Smuzhiyun # You can pass a regular expression as an attribute value, and 369*4882a593Smuzhiyun # you'll get tags whose values for that attribute match the 370*4882a593Smuzhiyun # regular expression. 371*4882a593Smuzhiyun tree = self.soup("""<a id="a">One a.</a> 372*4882a593Smuzhiyun <a id="aa">Two as.</a> 373*4882a593Smuzhiyun <a id="ab">Mixed as and bs.</a> 374*4882a593Smuzhiyun <a id="b">One b.</a> 375*4882a593Smuzhiyun <a>No ID.</a>""") 376*4882a593Smuzhiyun 377*4882a593Smuzhiyun self.assertSelects(tree.find_all(id=re.compile("^a+$")), 378*4882a593Smuzhiyun ["One a.", "Two as."]) 379*4882a593Smuzhiyun 380*4882a593Smuzhiyun def test_find_by_name_and_containing_string(self): 381*4882a593Smuzhiyun soup = self.soup("<b>foo</b><b>bar</b><a>foo</a>") 382*4882a593Smuzhiyun a = soup.a 383*4882a593Smuzhiyun 384*4882a593Smuzhiyun self.assertEqual([a], soup.find_all("a", text="foo")) 385*4882a593Smuzhiyun self.assertEqual([], soup.find_all("a", text="bar")) 386*4882a593Smuzhiyun self.assertEqual([], soup.find_all("a", text="bar")) 387*4882a593Smuzhiyun 388*4882a593Smuzhiyun def test_find_by_name_and_containing_string_when_string_is_buried(self): 389*4882a593Smuzhiyun soup = self.soup("<a>foo</a><a><b><c>foo</c></b></a>") 390*4882a593Smuzhiyun self.assertEqual(soup.find_all("a"), soup.find_all("a", text="foo")) 391*4882a593Smuzhiyun 392*4882a593Smuzhiyun def test_find_by_attribute_and_containing_string(self): 393*4882a593Smuzhiyun soup = self.soup('<b id="1">foo</b><a id="2">foo</a>') 394*4882a593Smuzhiyun a = soup.a 395*4882a593Smuzhiyun 396*4882a593Smuzhiyun self.assertEqual([a], soup.find_all(id=2, text="foo")) 397*4882a593Smuzhiyun self.assertEqual([], soup.find_all(id=1, text="bar")) 398*4882a593Smuzhiyun 399*4882a593Smuzhiyun 400*4882a593Smuzhiyun 401*4882a593Smuzhiyun 402*4882a593Smuzhiyunclass TestIndex(TreeTest): 403*4882a593Smuzhiyun """Test Tag.index""" 404*4882a593Smuzhiyun def test_index(self): 405*4882a593Smuzhiyun tree = self.soup("""<div> 406*4882a593Smuzhiyun <a>Identical</a> 407*4882a593Smuzhiyun <b>Not identical</b> 408*4882a593Smuzhiyun <a>Identical</a> 409*4882a593Smuzhiyun 410*4882a593Smuzhiyun <c><d>Identical with child</d></c> 411*4882a593Smuzhiyun <b>Also not identical</b> 412*4882a593Smuzhiyun <c><d>Identical with child</d></c> 413*4882a593Smuzhiyun </div>""") 414*4882a593Smuzhiyun div = tree.div 415*4882a593Smuzhiyun for i, element in enumerate(div.contents): 416*4882a593Smuzhiyun self.assertEqual(i, div.index(element)) 417*4882a593Smuzhiyun self.assertRaises(ValueError, tree.index, 1) 418*4882a593Smuzhiyun 419*4882a593Smuzhiyun 420*4882a593Smuzhiyunclass TestParentOperations(TreeTest): 421*4882a593Smuzhiyun """Test navigation and searching through an element's parents.""" 422*4882a593Smuzhiyun 423*4882a593Smuzhiyun def setUp(self): 424*4882a593Smuzhiyun super(TestParentOperations, self).setUp() 425*4882a593Smuzhiyun self.tree = self.soup('''<ul id="empty"></ul> 426*4882a593Smuzhiyun <ul id="top"> 427*4882a593Smuzhiyun <ul id="middle"> 428*4882a593Smuzhiyun <ul id="bottom"> 429*4882a593Smuzhiyun <b>Start here</b> 430*4882a593Smuzhiyun </ul> 431*4882a593Smuzhiyun </ul>''') 432*4882a593Smuzhiyun self.start = self.tree.b 433*4882a593Smuzhiyun 434*4882a593Smuzhiyun 435*4882a593Smuzhiyun def test_parent(self): 436*4882a593Smuzhiyun self.assertEqual(self.start.parent['id'], 'bottom') 437*4882a593Smuzhiyun self.assertEqual(self.start.parent.parent['id'], 'middle') 438*4882a593Smuzhiyun self.assertEqual(self.start.parent.parent.parent['id'], 'top') 439*4882a593Smuzhiyun 440*4882a593Smuzhiyun def test_parent_of_top_tag_is_soup_object(self): 441*4882a593Smuzhiyun top_tag = self.tree.contents[0] 442*4882a593Smuzhiyun self.assertEqual(top_tag.parent, self.tree) 443*4882a593Smuzhiyun 444*4882a593Smuzhiyun def test_soup_object_has_no_parent(self): 445*4882a593Smuzhiyun self.assertEqual(None, self.tree.parent) 446*4882a593Smuzhiyun 447*4882a593Smuzhiyun def test_find_parents(self): 448*4882a593Smuzhiyun self.assertSelectsIDs( 449*4882a593Smuzhiyun self.start.find_parents('ul'), ['bottom', 'middle', 'top']) 450*4882a593Smuzhiyun self.assertSelectsIDs( 451*4882a593Smuzhiyun self.start.find_parents('ul', id="middle"), ['middle']) 452*4882a593Smuzhiyun 453*4882a593Smuzhiyun def test_find_parent(self): 454*4882a593Smuzhiyun self.assertEqual(self.start.find_parent('ul')['id'], 'bottom') 455*4882a593Smuzhiyun self.assertEqual(self.start.find_parent('ul', id='top')['id'], 'top') 456*4882a593Smuzhiyun 457*4882a593Smuzhiyun def test_parent_of_text_element(self): 458*4882a593Smuzhiyun text = self.tree.find(text="Start here") 459*4882a593Smuzhiyun self.assertEqual(text.parent.name, 'b') 460*4882a593Smuzhiyun 461*4882a593Smuzhiyun def test_text_element_find_parent(self): 462*4882a593Smuzhiyun text = self.tree.find(text="Start here") 463*4882a593Smuzhiyun self.assertEqual(text.find_parent('ul')['id'], 'bottom') 464*4882a593Smuzhiyun 465*4882a593Smuzhiyun def test_parent_generator(self): 466*4882a593Smuzhiyun parents = [parent['id'] for parent in self.start.parents 467*4882a593Smuzhiyun if parent is not None and 'id' in parent.attrs] 468*4882a593Smuzhiyun self.assertEqual(parents, ['bottom', 'middle', 'top']) 469*4882a593Smuzhiyun 470*4882a593Smuzhiyun 471*4882a593Smuzhiyunclass ProximityTest(TreeTest): 472*4882a593Smuzhiyun 473*4882a593Smuzhiyun def setUp(self): 474*4882a593Smuzhiyun super(TreeTest, self).setUp() 475*4882a593Smuzhiyun self.tree = self.soup( 476*4882a593Smuzhiyun '<html id="start"><head></head><body><b id="1">One</b><b id="2">Two</b><b id="3">Three</b></body></html>') 477*4882a593Smuzhiyun 478*4882a593Smuzhiyun 479*4882a593Smuzhiyunclass TestNextOperations(ProximityTest): 480*4882a593Smuzhiyun 481*4882a593Smuzhiyun def setUp(self): 482*4882a593Smuzhiyun super(TestNextOperations, self).setUp() 483*4882a593Smuzhiyun self.start = self.tree.b 484*4882a593Smuzhiyun 485*4882a593Smuzhiyun def test_next(self): 486*4882a593Smuzhiyun self.assertEqual(self.start.next_element, "One") 487*4882a593Smuzhiyun self.assertEqual(self.start.next_element.next_element['id'], "2") 488*4882a593Smuzhiyun 489*4882a593Smuzhiyun def test_next_of_last_item_is_none(self): 490*4882a593Smuzhiyun last = self.tree.find(text="Three") 491*4882a593Smuzhiyun self.assertEqual(last.next_element, None) 492*4882a593Smuzhiyun 493*4882a593Smuzhiyun def test_next_of_root_is_none(self): 494*4882a593Smuzhiyun # The document root is outside the next/previous chain. 495*4882a593Smuzhiyun self.assertEqual(self.tree.next_element, None) 496*4882a593Smuzhiyun 497*4882a593Smuzhiyun def test_find_all_next(self): 498*4882a593Smuzhiyun self.assertSelects(self.start.find_all_next('b'), ["Two", "Three"]) 499*4882a593Smuzhiyun self.start.find_all_next(id=3) 500*4882a593Smuzhiyun self.assertSelects(self.start.find_all_next(id=3), ["Three"]) 501*4882a593Smuzhiyun 502*4882a593Smuzhiyun def test_find_next(self): 503*4882a593Smuzhiyun self.assertEqual(self.start.find_next('b')['id'], '2') 504*4882a593Smuzhiyun self.assertEqual(self.start.find_next(text="Three"), "Three") 505*4882a593Smuzhiyun 506*4882a593Smuzhiyun def test_find_next_for_text_element(self): 507*4882a593Smuzhiyun text = self.tree.find(text="One") 508*4882a593Smuzhiyun self.assertEqual(text.find_next("b").string, "Two") 509*4882a593Smuzhiyun self.assertSelects(text.find_all_next("b"), ["Two", "Three"]) 510*4882a593Smuzhiyun 511*4882a593Smuzhiyun def test_next_generator(self): 512*4882a593Smuzhiyun start = self.tree.find(text="Two") 513*4882a593Smuzhiyun successors = [node for node in start.next_elements] 514*4882a593Smuzhiyun # There are two successors: the final <b> tag and its text contents. 515*4882a593Smuzhiyun tag, contents = successors 516*4882a593Smuzhiyun self.assertEqual(tag['id'], '3') 517*4882a593Smuzhiyun self.assertEqual(contents, "Three") 518*4882a593Smuzhiyun 519*4882a593Smuzhiyunclass TestPreviousOperations(ProximityTest): 520*4882a593Smuzhiyun 521*4882a593Smuzhiyun def setUp(self): 522*4882a593Smuzhiyun super(TestPreviousOperations, self).setUp() 523*4882a593Smuzhiyun self.end = self.tree.find(text="Three") 524*4882a593Smuzhiyun 525*4882a593Smuzhiyun def test_previous(self): 526*4882a593Smuzhiyun self.assertEqual(self.end.previous_element['id'], "3") 527*4882a593Smuzhiyun self.assertEqual(self.end.previous_element.previous_element, "Two") 528*4882a593Smuzhiyun 529*4882a593Smuzhiyun def test_previous_of_first_item_is_none(self): 530*4882a593Smuzhiyun first = self.tree.find('html') 531*4882a593Smuzhiyun self.assertEqual(first.previous_element, None) 532*4882a593Smuzhiyun 533*4882a593Smuzhiyun def test_previous_of_root_is_none(self): 534*4882a593Smuzhiyun # The document root is outside the next/previous chain. 535*4882a593Smuzhiyun # XXX This is broken! 536*4882a593Smuzhiyun #self.assertEqual(self.tree.previous_element, None) 537*4882a593Smuzhiyun pass 538*4882a593Smuzhiyun 539*4882a593Smuzhiyun def test_find_all_previous(self): 540*4882a593Smuzhiyun # The <b> tag containing the "Three" node is the predecessor 541*4882a593Smuzhiyun # of the "Three" node itself, which is why "Three" shows up 542*4882a593Smuzhiyun # here. 543*4882a593Smuzhiyun self.assertSelects( 544*4882a593Smuzhiyun self.end.find_all_previous('b'), ["Three", "Two", "One"]) 545*4882a593Smuzhiyun self.assertSelects(self.end.find_all_previous(id=1), ["One"]) 546*4882a593Smuzhiyun 547*4882a593Smuzhiyun def test_find_previous(self): 548*4882a593Smuzhiyun self.assertEqual(self.end.find_previous('b')['id'], '3') 549*4882a593Smuzhiyun self.assertEqual(self.end.find_previous(text="One"), "One") 550*4882a593Smuzhiyun 551*4882a593Smuzhiyun def test_find_previous_for_text_element(self): 552*4882a593Smuzhiyun text = self.tree.find(text="Three") 553*4882a593Smuzhiyun self.assertEqual(text.find_previous("b").string, "Three") 554*4882a593Smuzhiyun self.assertSelects( 555*4882a593Smuzhiyun text.find_all_previous("b"), ["Three", "Two", "One"]) 556*4882a593Smuzhiyun 557*4882a593Smuzhiyun def test_previous_generator(self): 558*4882a593Smuzhiyun start = self.tree.find(text="One") 559*4882a593Smuzhiyun predecessors = [node for node in start.previous_elements] 560*4882a593Smuzhiyun 561*4882a593Smuzhiyun # There are four predecessors: the <b> tag containing "One" 562*4882a593Smuzhiyun # the <body> tag, the <head> tag, and the <html> tag. 563*4882a593Smuzhiyun b, body, head, html = predecessors 564*4882a593Smuzhiyun self.assertEqual(b['id'], '1') 565*4882a593Smuzhiyun self.assertEqual(body.name, "body") 566*4882a593Smuzhiyun self.assertEqual(head.name, "head") 567*4882a593Smuzhiyun self.assertEqual(html.name, "html") 568*4882a593Smuzhiyun 569*4882a593Smuzhiyun 570*4882a593Smuzhiyunclass SiblingTest(TreeTest): 571*4882a593Smuzhiyun 572*4882a593Smuzhiyun def setUp(self): 573*4882a593Smuzhiyun super(SiblingTest, self).setUp() 574*4882a593Smuzhiyun markup = '''<html> 575*4882a593Smuzhiyun <span id="1"> 576*4882a593Smuzhiyun <span id="1.1"></span> 577*4882a593Smuzhiyun </span> 578*4882a593Smuzhiyun <span id="2"> 579*4882a593Smuzhiyun <span id="2.1"></span> 580*4882a593Smuzhiyun </span> 581*4882a593Smuzhiyun <span id="3"> 582*4882a593Smuzhiyun <span id="3.1"></span> 583*4882a593Smuzhiyun </span> 584*4882a593Smuzhiyun <span id="4"></span> 585*4882a593Smuzhiyun </html>''' 586*4882a593Smuzhiyun # All that whitespace looks good but makes the tests more 587*4882a593Smuzhiyun # difficult. Get rid of it. 588*4882a593Smuzhiyun markup = re.compile("\n\s*").sub("", markup) 589*4882a593Smuzhiyun self.tree = self.soup(markup) 590*4882a593Smuzhiyun 591*4882a593Smuzhiyun 592*4882a593Smuzhiyunclass TestNextSibling(SiblingTest): 593*4882a593Smuzhiyun 594*4882a593Smuzhiyun def setUp(self): 595*4882a593Smuzhiyun super(TestNextSibling, self).setUp() 596*4882a593Smuzhiyun self.start = self.tree.find(id="1") 597*4882a593Smuzhiyun 598*4882a593Smuzhiyun def test_next_sibling_of_root_is_none(self): 599*4882a593Smuzhiyun self.assertEqual(self.tree.next_sibling, None) 600*4882a593Smuzhiyun 601*4882a593Smuzhiyun def test_next_sibling(self): 602*4882a593Smuzhiyun self.assertEqual(self.start.next_sibling['id'], '2') 603*4882a593Smuzhiyun self.assertEqual(self.start.next_sibling.next_sibling['id'], '3') 604*4882a593Smuzhiyun 605*4882a593Smuzhiyun # Note the difference between next_sibling and next_element. 606*4882a593Smuzhiyun self.assertEqual(self.start.next_element['id'], '1.1') 607*4882a593Smuzhiyun 608*4882a593Smuzhiyun def test_next_sibling_may_not_exist(self): 609*4882a593Smuzhiyun self.assertEqual(self.tree.html.next_sibling, None) 610*4882a593Smuzhiyun 611*4882a593Smuzhiyun nested_span = self.tree.find(id="1.1") 612*4882a593Smuzhiyun self.assertEqual(nested_span.next_sibling, None) 613*4882a593Smuzhiyun 614*4882a593Smuzhiyun last_span = self.tree.find(id="4") 615*4882a593Smuzhiyun self.assertEqual(last_span.next_sibling, None) 616*4882a593Smuzhiyun 617*4882a593Smuzhiyun def test_find_next_sibling(self): 618*4882a593Smuzhiyun self.assertEqual(self.start.find_next_sibling('span')['id'], '2') 619*4882a593Smuzhiyun 620*4882a593Smuzhiyun def test_next_siblings(self): 621*4882a593Smuzhiyun self.assertSelectsIDs(self.start.find_next_siblings("span"), 622*4882a593Smuzhiyun ['2', '3', '4']) 623*4882a593Smuzhiyun 624*4882a593Smuzhiyun self.assertSelectsIDs(self.start.find_next_siblings(id='3'), ['3']) 625*4882a593Smuzhiyun 626*4882a593Smuzhiyun def test_next_sibling_for_text_element(self): 627*4882a593Smuzhiyun soup = self.soup("Foo<b>bar</b>baz") 628*4882a593Smuzhiyun start = soup.find(text="Foo") 629*4882a593Smuzhiyun self.assertEqual(start.next_sibling.name, 'b') 630*4882a593Smuzhiyun self.assertEqual(start.next_sibling.next_sibling, 'baz') 631*4882a593Smuzhiyun 632*4882a593Smuzhiyun self.assertSelects(start.find_next_siblings('b'), ['bar']) 633*4882a593Smuzhiyun self.assertEqual(start.find_next_sibling(text="baz"), "baz") 634*4882a593Smuzhiyun self.assertEqual(start.find_next_sibling(text="nonesuch"), None) 635*4882a593Smuzhiyun 636*4882a593Smuzhiyun 637*4882a593Smuzhiyunclass TestPreviousSibling(SiblingTest): 638*4882a593Smuzhiyun 639*4882a593Smuzhiyun def setUp(self): 640*4882a593Smuzhiyun super(TestPreviousSibling, self).setUp() 641*4882a593Smuzhiyun self.end = self.tree.find(id="4") 642*4882a593Smuzhiyun 643*4882a593Smuzhiyun def test_previous_sibling_of_root_is_none(self): 644*4882a593Smuzhiyun self.assertEqual(self.tree.previous_sibling, None) 645*4882a593Smuzhiyun 646*4882a593Smuzhiyun def test_previous_sibling(self): 647*4882a593Smuzhiyun self.assertEqual(self.end.previous_sibling['id'], '3') 648*4882a593Smuzhiyun self.assertEqual(self.end.previous_sibling.previous_sibling['id'], '2') 649*4882a593Smuzhiyun 650*4882a593Smuzhiyun # Note the difference between previous_sibling and previous_element. 651*4882a593Smuzhiyun self.assertEqual(self.end.previous_element['id'], '3.1') 652*4882a593Smuzhiyun 653*4882a593Smuzhiyun def test_previous_sibling_may_not_exist(self): 654*4882a593Smuzhiyun self.assertEqual(self.tree.html.previous_sibling, None) 655*4882a593Smuzhiyun 656*4882a593Smuzhiyun nested_span = self.tree.find(id="1.1") 657*4882a593Smuzhiyun self.assertEqual(nested_span.previous_sibling, None) 658*4882a593Smuzhiyun 659*4882a593Smuzhiyun first_span = self.tree.find(id="1") 660*4882a593Smuzhiyun self.assertEqual(first_span.previous_sibling, None) 661*4882a593Smuzhiyun 662*4882a593Smuzhiyun def test_find_previous_sibling(self): 663*4882a593Smuzhiyun self.assertEqual(self.end.find_previous_sibling('span')['id'], '3') 664*4882a593Smuzhiyun 665*4882a593Smuzhiyun def test_previous_siblings(self): 666*4882a593Smuzhiyun self.assertSelectsIDs(self.end.find_previous_siblings("span"), 667*4882a593Smuzhiyun ['3', '2', '1']) 668*4882a593Smuzhiyun 669*4882a593Smuzhiyun self.assertSelectsIDs(self.end.find_previous_siblings(id='1'), ['1']) 670*4882a593Smuzhiyun 671*4882a593Smuzhiyun def test_previous_sibling_for_text_element(self): 672*4882a593Smuzhiyun soup = self.soup("Foo<b>bar</b>baz") 673*4882a593Smuzhiyun start = soup.find(text="baz") 674*4882a593Smuzhiyun self.assertEqual(start.previous_sibling.name, 'b') 675*4882a593Smuzhiyun self.assertEqual(start.previous_sibling.previous_sibling, 'Foo') 676*4882a593Smuzhiyun 677*4882a593Smuzhiyun self.assertSelects(start.find_previous_siblings('b'), ['bar']) 678*4882a593Smuzhiyun self.assertEqual(start.find_previous_sibling(text="Foo"), "Foo") 679*4882a593Smuzhiyun self.assertEqual(start.find_previous_sibling(text="nonesuch"), None) 680*4882a593Smuzhiyun 681*4882a593Smuzhiyun 682*4882a593Smuzhiyunclass TestTagCreation(SoupTest): 683*4882a593Smuzhiyun """Test the ability to create new tags.""" 684*4882a593Smuzhiyun def test_new_tag(self): 685*4882a593Smuzhiyun soup = self.soup("") 686*4882a593Smuzhiyun new_tag = soup.new_tag("foo", bar="baz") 687*4882a593Smuzhiyun self.assertTrue(isinstance(new_tag, Tag)) 688*4882a593Smuzhiyun self.assertEqual("foo", new_tag.name) 689*4882a593Smuzhiyun self.assertEqual(dict(bar="baz"), new_tag.attrs) 690*4882a593Smuzhiyun self.assertEqual(None, new_tag.parent) 691*4882a593Smuzhiyun 692*4882a593Smuzhiyun def test_tag_inherits_self_closing_rules_from_builder(self): 693*4882a593Smuzhiyun if XML_BUILDER_PRESENT: 694*4882a593Smuzhiyun xml_soup = BeautifulSoup("", "lxml-xml") 695*4882a593Smuzhiyun xml_br = xml_soup.new_tag("br") 696*4882a593Smuzhiyun xml_p = xml_soup.new_tag("p") 697*4882a593Smuzhiyun 698*4882a593Smuzhiyun # Both the <br> and <p> tag are empty-element, just because 699*4882a593Smuzhiyun # they have no contents. 700*4882a593Smuzhiyun self.assertEqual(b"<br/>", xml_br.encode()) 701*4882a593Smuzhiyun self.assertEqual(b"<p/>", xml_p.encode()) 702*4882a593Smuzhiyun 703*4882a593Smuzhiyun html_soup = BeautifulSoup("", "html.parser") 704*4882a593Smuzhiyun html_br = html_soup.new_tag("br") 705*4882a593Smuzhiyun html_p = html_soup.new_tag("p") 706*4882a593Smuzhiyun 707*4882a593Smuzhiyun # The HTML builder users HTML's rules about which tags are 708*4882a593Smuzhiyun # empty-element tags, and the new tags reflect these rules. 709*4882a593Smuzhiyun self.assertEqual(b"<br/>", html_br.encode()) 710*4882a593Smuzhiyun self.assertEqual(b"<p></p>", html_p.encode()) 711*4882a593Smuzhiyun 712*4882a593Smuzhiyun def test_new_string_creates_navigablestring(self): 713*4882a593Smuzhiyun soup = self.soup("") 714*4882a593Smuzhiyun s = soup.new_string("foo") 715*4882a593Smuzhiyun self.assertEqual("foo", s) 716*4882a593Smuzhiyun self.assertTrue(isinstance(s, NavigableString)) 717*4882a593Smuzhiyun 718*4882a593Smuzhiyun def test_new_string_can_create_navigablestring_subclass(self): 719*4882a593Smuzhiyun soup = self.soup("") 720*4882a593Smuzhiyun s = soup.new_string("foo", Comment) 721*4882a593Smuzhiyun self.assertEqual("foo", s) 722*4882a593Smuzhiyun self.assertTrue(isinstance(s, Comment)) 723*4882a593Smuzhiyun 724*4882a593Smuzhiyunclass TestTreeModification(SoupTest): 725*4882a593Smuzhiyun 726*4882a593Smuzhiyun def test_attribute_modification(self): 727*4882a593Smuzhiyun soup = self.soup('<a id="1"></a>') 728*4882a593Smuzhiyun soup.a['id'] = 2 729*4882a593Smuzhiyun self.assertEqual(soup.decode(), self.document_for('<a id="2"></a>')) 730*4882a593Smuzhiyun del(soup.a['id']) 731*4882a593Smuzhiyun self.assertEqual(soup.decode(), self.document_for('<a></a>')) 732*4882a593Smuzhiyun soup.a['id2'] = 'foo' 733*4882a593Smuzhiyun self.assertEqual(soup.decode(), self.document_for('<a id2="foo"></a>')) 734*4882a593Smuzhiyun 735*4882a593Smuzhiyun def test_new_tag_creation(self): 736*4882a593Smuzhiyun builder = builder_registry.lookup('html')() 737*4882a593Smuzhiyun soup = self.soup("<body></body>", builder=builder) 738*4882a593Smuzhiyun a = Tag(soup, builder, 'a') 739*4882a593Smuzhiyun ol = Tag(soup, builder, 'ol') 740*4882a593Smuzhiyun a['href'] = 'http://foo.com/' 741*4882a593Smuzhiyun soup.body.insert(0, a) 742*4882a593Smuzhiyun soup.body.insert(1, ol) 743*4882a593Smuzhiyun self.assertEqual( 744*4882a593Smuzhiyun soup.body.encode(), 745*4882a593Smuzhiyun b'<body><a href="http://foo.com/"></a><ol></ol></body>') 746*4882a593Smuzhiyun 747*4882a593Smuzhiyun def test_append_to_contents_moves_tag(self): 748*4882a593Smuzhiyun doc = """<p id="1">Don't leave me <b>here</b>.</p> 749*4882a593Smuzhiyun <p id="2">Don\'t leave!</p>""" 750*4882a593Smuzhiyun soup = self.soup(doc) 751*4882a593Smuzhiyun second_para = soup.find(id='2') 752*4882a593Smuzhiyun bold = soup.b 753*4882a593Smuzhiyun 754*4882a593Smuzhiyun # Move the <b> tag to the end of the second paragraph. 755*4882a593Smuzhiyun soup.find(id='2').append(soup.b) 756*4882a593Smuzhiyun 757*4882a593Smuzhiyun # The <b> tag is now a child of the second paragraph. 758*4882a593Smuzhiyun self.assertEqual(bold.parent, second_para) 759*4882a593Smuzhiyun 760*4882a593Smuzhiyun self.assertEqual( 761*4882a593Smuzhiyun soup.decode(), self.document_for( 762*4882a593Smuzhiyun '<p id="1">Don\'t leave me .</p>\n' 763*4882a593Smuzhiyun '<p id="2">Don\'t leave!<b>here</b></p>')) 764*4882a593Smuzhiyun 765*4882a593Smuzhiyun def test_replace_with_returns_thing_that_was_replaced(self): 766*4882a593Smuzhiyun text = "<a></a><b><c></c></b>" 767*4882a593Smuzhiyun soup = self.soup(text) 768*4882a593Smuzhiyun a = soup.a 769*4882a593Smuzhiyun new_a = a.replace_with(soup.c) 770*4882a593Smuzhiyun self.assertEqual(a, new_a) 771*4882a593Smuzhiyun 772*4882a593Smuzhiyun def test_unwrap_returns_thing_that_was_replaced(self): 773*4882a593Smuzhiyun text = "<a><b></b><c></c></a>" 774*4882a593Smuzhiyun soup = self.soup(text) 775*4882a593Smuzhiyun a = soup.a 776*4882a593Smuzhiyun new_a = a.unwrap() 777*4882a593Smuzhiyun self.assertEqual(a, new_a) 778*4882a593Smuzhiyun 779*4882a593Smuzhiyun def test_replace_with_and_unwrap_give_useful_exception_when_tag_has_no_parent(self): 780*4882a593Smuzhiyun soup = self.soup("<a><b>Foo</b></a><c>Bar</c>") 781*4882a593Smuzhiyun a = soup.a 782*4882a593Smuzhiyun a.extract() 783*4882a593Smuzhiyun self.assertEqual(None, a.parent) 784*4882a593Smuzhiyun self.assertRaises(ValueError, a.unwrap) 785*4882a593Smuzhiyun self.assertRaises(ValueError, a.replace_with, soup.c) 786*4882a593Smuzhiyun 787*4882a593Smuzhiyun def test_replace_tag_with_itself(self): 788*4882a593Smuzhiyun text = "<a><b></b><c>Foo<d></d></c></a><a><e></e></a>" 789*4882a593Smuzhiyun soup = self.soup(text) 790*4882a593Smuzhiyun c = soup.c 791*4882a593Smuzhiyun soup.c.replace_with(c) 792*4882a593Smuzhiyun self.assertEqual(soup.decode(), self.document_for(text)) 793*4882a593Smuzhiyun 794*4882a593Smuzhiyun def test_replace_tag_with_its_parent_raises_exception(self): 795*4882a593Smuzhiyun text = "<a><b></b></a>" 796*4882a593Smuzhiyun soup = self.soup(text) 797*4882a593Smuzhiyun self.assertRaises(ValueError, soup.b.replace_with, soup.a) 798*4882a593Smuzhiyun 799*4882a593Smuzhiyun def test_insert_tag_into_itself_raises_exception(self): 800*4882a593Smuzhiyun text = "<a><b></b></a>" 801*4882a593Smuzhiyun soup = self.soup(text) 802*4882a593Smuzhiyun self.assertRaises(ValueError, soup.a.insert, 0, soup.a) 803*4882a593Smuzhiyun 804*4882a593Smuzhiyun def test_replace_with_maintains_next_element_throughout(self): 805*4882a593Smuzhiyun soup = self.soup('<p><a>one</a><b>three</b></p>') 806*4882a593Smuzhiyun a = soup.a 807*4882a593Smuzhiyun b = a.contents[0] 808*4882a593Smuzhiyun # Make it so the <a> tag has two text children. 809*4882a593Smuzhiyun a.insert(1, "two") 810*4882a593Smuzhiyun 811*4882a593Smuzhiyun # Now replace each one with the empty string. 812*4882a593Smuzhiyun left, right = a.contents 813*4882a593Smuzhiyun left.replaceWith('') 814*4882a593Smuzhiyun right.replaceWith('') 815*4882a593Smuzhiyun 816*4882a593Smuzhiyun # The <b> tag is still connected to the tree. 817*4882a593Smuzhiyun self.assertEqual("three", soup.b.string) 818*4882a593Smuzhiyun 819*4882a593Smuzhiyun def test_replace_final_node(self): 820*4882a593Smuzhiyun soup = self.soup("<b>Argh!</b>") 821*4882a593Smuzhiyun soup.find(text="Argh!").replace_with("Hooray!") 822*4882a593Smuzhiyun new_text = soup.find(text="Hooray!") 823*4882a593Smuzhiyun b = soup.b 824*4882a593Smuzhiyun self.assertEqual(new_text.previous_element, b) 825*4882a593Smuzhiyun self.assertEqual(new_text.parent, b) 826*4882a593Smuzhiyun self.assertEqual(new_text.previous_element.next_element, new_text) 827*4882a593Smuzhiyun self.assertEqual(new_text.next_element, None) 828*4882a593Smuzhiyun 829*4882a593Smuzhiyun def test_consecutive_text_nodes(self): 830*4882a593Smuzhiyun # A builder should never create two consecutive text nodes, 831*4882a593Smuzhiyun # but if you insert one next to another, Beautiful Soup will 832*4882a593Smuzhiyun # handle it correctly. 833*4882a593Smuzhiyun soup = self.soup("<a><b>Argh!</b><c></c></a>") 834*4882a593Smuzhiyun soup.b.insert(1, "Hooray!") 835*4882a593Smuzhiyun 836*4882a593Smuzhiyun self.assertEqual( 837*4882a593Smuzhiyun soup.decode(), self.document_for( 838*4882a593Smuzhiyun "<a><b>Argh!Hooray!</b><c></c></a>")) 839*4882a593Smuzhiyun 840*4882a593Smuzhiyun new_text = soup.find(text="Hooray!") 841*4882a593Smuzhiyun self.assertEqual(new_text.previous_element, "Argh!") 842*4882a593Smuzhiyun self.assertEqual(new_text.previous_element.next_element, new_text) 843*4882a593Smuzhiyun 844*4882a593Smuzhiyun self.assertEqual(new_text.previous_sibling, "Argh!") 845*4882a593Smuzhiyun self.assertEqual(new_text.previous_sibling.next_sibling, new_text) 846*4882a593Smuzhiyun 847*4882a593Smuzhiyun self.assertEqual(new_text.next_sibling, None) 848*4882a593Smuzhiyun self.assertEqual(new_text.next_element, soup.c) 849*4882a593Smuzhiyun 850*4882a593Smuzhiyun def test_insert_string(self): 851*4882a593Smuzhiyun soup = self.soup("<a></a>") 852*4882a593Smuzhiyun soup.a.insert(0, "bar") 853*4882a593Smuzhiyun soup.a.insert(0, "foo") 854*4882a593Smuzhiyun # The string were added to the tag. 855*4882a593Smuzhiyun self.assertEqual(["foo", "bar"], soup.a.contents) 856*4882a593Smuzhiyun # And they were converted to NavigableStrings. 857*4882a593Smuzhiyun self.assertEqual(soup.a.contents[0].next_element, "bar") 858*4882a593Smuzhiyun 859*4882a593Smuzhiyun def test_insert_tag(self): 860*4882a593Smuzhiyun builder = self.default_builder 861*4882a593Smuzhiyun soup = self.soup( 862*4882a593Smuzhiyun "<a><b>Find</b><c>lady!</c><d></d></a>", builder=builder) 863*4882a593Smuzhiyun magic_tag = Tag(soup, builder, 'magictag') 864*4882a593Smuzhiyun magic_tag.insert(0, "the") 865*4882a593Smuzhiyun soup.a.insert(1, magic_tag) 866*4882a593Smuzhiyun 867*4882a593Smuzhiyun self.assertEqual( 868*4882a593Smuzhiyun soup.decode(), self.document_for( 869*4882a593Smuzhiyun "<a><b>Find</b><magictag>the</magictag><c>lady!</c><d></d></a>")) 870*4882a593Smuzhiyun 871*4882a593Smuzhiyun # Make sure all the relationships are hooked up correctly. 872*4882a593Smuzhiyun b_tag = soup.b 873*4882a593Smuzhiyun self.assertEqual(b_tag.next_sibling, magic_tag) 874*4882a593Smuzhiyun self.assertEqual(magic_tag.previous_sibling, b_tag) 875*4882a593Smuzhiyun 876*4882a593Smuzhiyun find = b_tag.find(text="Find") 877*4882a593Smuzhiyun self.assertEqual(find.next_element, magic_tag) 878*4882a593Smuzhiyun self.assertEqual(magic_tag.previous_element, find) 879*4882a593Smuzhiyun 880*4882a593Smuzhiyun c_tag = soup.c 881*4882a593Smuzhiyun self.assertEqual(magic_tag.next_sibling, c_tag) 882*4882a593Smuzhiyun self.assertEqual(c_tag.previous_sibling, magic_tag) 883*4882a593Smuzhiyun 884*4882a593Smuzhiyun the = magic_tag.find(text="the") 885*4882a593Smuzhiyun self.assertEqual(the.parent, magic_tag) 886*4882a593Smuzhiyun self.assertEqual(the.next_element, c_tag) 887*4882a593Smuzhiyun self.assertEqual(c_tag.previous_element, the) 888*4882a593Smuzhiyun 889*4882a593Smuzhiyun def test_append_child_thats_already_at_the_end(self): 890*4882a593Smuzhiyun data = "<a><b></b></a>" 891*4882a593Smuzhiyun soup = self.soup(data) 892*4882a593Smuzhiyun soup.a.append(soup.b) 893*4882a593Smuzhiyun self.assertEqual(data, soup.decode()) 894*4882a593Smuzhiyun 895*4882a593Smuzhiyun def test_move_tag_to_beginning_of_parent(self): 896*4882a593Smuzhiyun data = "<a><b></b><c></c><d></d></a>" 897*4882a593Smuzhiyun soup = self.soup(data) 898*4882a593Smuzhiyun soup.a.insert(0, soup.d) 899*4882a593Smuzhiyun self.assertEqual("<a><d></d><b></b><c></c></a>", soup.decode()) 900*4882a593Smuzhiyun 901*4882a593Smuzhiyun def test_insert_works_on_empty_element_tag(self): 902*4882a593Smuzhiyun # This is a little strange, since most HTML parsers don't allow 903*4882a593Smuzhiyun # markup like this to come through. But in general, we don't 904*4882a593Smuzhiyun # know what the parser would or wouldn't have allowed, so 905*4882a593Smuzhiyun # I'm letting this succeed for now. 906*4882a593Smuzhiyun soup = self.soup("<br/>") 907*4882a593Smuzhiyun soup.br.insert(1, "Contents") 908*4882a593Smuzhiyun self.assertEqual(str(soup.br), "<br>Contents</br>") 909*4882a593Smuzhiyun 910*4882a593Smuzhiyun def test_insert_before(self): 911*4882a593Smuzhiyun soup = self.soup("<a>foo</a><b>bar</b>") 912*4882a593Smuzhiyun soup.b.insert_before("BAZ") 913*4882a593Smuzhiyun soup.a.insert_before("QUUX") 914*4882a593Smuzhiyun self.assertEqual( 915*4882a593Smuzhiyun soup.decode(), self.document_for("QUUX<a>foo</a>BAZ<b>bar</b>")) 916*4882a593Smuzhiyun 917*4882a593Smuzhiyun soup.a.insert_before(soup.b) 918*4882a593Smuzhiyun self.assertEqual( 919*4882a593Smuzhiyun soup.decode(), self.document_for("QUUX<b>bar</b><a>foo</a>BAZ")) 920*4882a593Smuzhiyun 921*4882a593Smuzhiyun def test_insert_after(self): 922*4882a593Smuzhiyun soup = self.soup("<a>foo</a><b>bar</b>") 923*4882a593Smuzhiyun soup.b.insert_after("BAZ") 924*4882a593Smuzhiyun soup.a.insert_after("QUUX") 925*4882a593Smuzhiyun self.assertEqual( 926*4882a593Smuzhiyun soup.decode(), self.document_for("<a>foo</a>QUUX<b>bar</b>BAZ")) 927*4882a593Smuzhiyun soup.b.insert_after(soup.a) 928*4882a593Smuzhiyun self.assertEqual( 929*4882a593Smuzhiyun soup.decode(), self.document_for("QUUX<b>bar</b><a>foo</a>BAZ")) 930*4882a593Smuzhiyun 931*4882a593Smuzhiyun def test_insert_after_raises_exception_if_after_has_no_meaning(self): 932*4882a593Smuzhiyun soup = self.soup("") 933*4882a593Smuzhiyun tag = soup.new_tag("a") 934*4882a593Smuzhiyun string = soup.new_string("") 935*4882a593Smuzhiyun self.assertRaises(ValueError, string.insert_after, tag) 936*4882a593Smuzhiyun self.assertRaises(NotImplementedError, soup.insert_after, tag) 937*4882a593Smuzhiyun self.assertRaises(ValueError, tag.insert_after, tag) 938*4882a593Smuzhiyun 939*4882a593Smuzhiyun def test_insert_before_raises_notimplementederror_if_before_has_no_meaning(self): 940*4882a593Smuzhiyun soup = self.soup("") 941*4882a593Smuzhiyun tag = soup.new_tag("a") 942*4882a593Smuzhiyun string = soup.new_string("") 943*4882a593Smuzhiyun self.assertRaises(ValueError, string.insert_before, tag) 944*4882a593Smuzhiyun self.assertRaises(NotImplementedError, soup.insert_before, tag) 945*4882a593Smuzhiyun self.assertRaises(ValueError, tag.insert_before, tag) 946*4882a593Smuzhiyun 947*4882a593Smuzhiyun def test_replace_with(self): 948*4882a593Smuzhiyun soup = self.soup( 949*4882a593Smuzhiyun "<p>There's <b>no</b> business like <b>show</b> business</p>") 950*4882a593Smuzhiyun no, show = soup.find_all('b') 951*4882a593Smuzhiyun show.replace_with(no) 952*4882a593Smuzhiyun self.assertEqual( 953*4882a593Smuzhiyun soup.decode(), 954*4882a593Smuzhiyun self.document_for( 955*4882a593Smuzhiyun "<p>There's business like <b>no</b> business</p>")) 956*4882a593Smuzhiyun 957*4882a593Smuzhiyun self.assertEqual(show.parent, None) 958*4882a593Smuzhiyun self.assertEqual(no.parent, soup.p) 959*4882a593Smuzhiyun self.assertEqual(no.next_element, "no") 960*4882a593Smuzhiyun self.assertEqual(no.next_sibling, " business") 961*4882a593Smuzhiyun 962*4882a593Smuzhiyun def test_replace_first_child(self): 963*4882a593Smuzhiyun data = "<a><b></b><c></c></a>" 964*4882a593Smuzhiyun soup = self.soup(data) 965*4882a593Smuzhiyun soup.b.replace_with(soup.c) 966*4882a593Smuzhiyun self.assertEqual("<a><c></c></a>", soup.decode()) 967*4882a593Smuzhiyun 968*4882a593Smuzhiyun def test_replace_last_child(self): 969*4882a593Smuzhiyun data = "<a><b></b><c></c></a>" 970*4882a593Smuzhiyun soup = self.soup(data) 971*4882a593Smuzhiyun soup.c.replace_with(soup.b) 972*4882a593Smuzhiyun self.assertEqual("<a><b></b></a>", soup.decode()) 973*4882a593Smuzhiyun 974*4882a593Smuzhiyun def test_nested_tag_replace_with(self): 975*4882a593Smuzhiyun soup = self.soup( 976*4882a593Smuzhiyun """<a>We<b>reserve<c>the</c><d>right</d></b></a><e>to<f>refuse</f><g>service</g></e>""") 977*4882a593Smuzhiyun 978*4882a593Smuzhiyun # Replace the entire <b> tag and its contents ("reserve the 979*4882a593Smuzhiyun # right") with the <f> tag ("refuse"). 980*4882a593Smuzhiyun remove_tag = soup.b 981*4882a593Smuzhiyun move_tag = soup.f 982*4882a593Smuzhiyun remove_tag.replace_with(move_tag) 983*4882a593Smuzhiyun 984*4882a593Smuzhiyun self.assertEqual( 985*4882a593Smuzhiyun soup.decode(), self.document_for( 986*4882a593Smuzhiyun "<a>We<f>refuse</f></a><e>to<g>service</g></e>")) 987*4882a593Smuzhiyun 988*4882a593Smuzhiyun # The <b> tag is now an orphan. 989*4882a593Smuzhiyun self.assertEqual(remove_tag.parent, None) 990*4882a593Smuzhiyun self.assertEqual(remove_tag.find(text="right").next_element, None) 991*4882a593Smuzhiyun self.assertEqual(remove_tag.previous_element, None) 992*4882a593Smuzhiyun self.assertEqual(remove_tag.next_sibling, None) 993*4882a593Smuzhiyun self.assertEqual(remove_tag.previous_sibling, None) 994*4882a593Smuzhiyun 995*4882a593Smuzhiyun # The <f> tag is now connected to the <a> tag. 996*4882a593Smuzhiyun self.assertEqual(move_tag.parent, soup.a) 997*4882a593Smuzhiyun self.assertEqual(move_tag.previous_element, "We") 998*4882a593Smuzhiyun self.assertEqual(move_tag.next_element.next_element, soup.e) 999*4882a593Smuzhiyun self.assertEqual(move_tag.next_sibling, None) 1000*4882a593Smuzhiyun 1001*4882a593Smuzhiyun # The gap where the <f> tag used to be has been mended, and 1002*4882a593Smuzhiyun # the word "to" is now connected to the <g> tag. 1003*4882a593Smuzhiyun to_text = soup.find(text="to") 1004*4882a593Smuzhiyun g_tag = soup.g 1005*4882a593Smuzhiyun self.assertEqual(to_text.next_element, g_tag) 1006*4882a593Smuzhiyun self.assertEqual(to_text.next_sibling, g_tag) 1007*4882a593Smuzhiyun self.assertEqual(g_tag.previous_element, to_text) 1008*4882a593Smuzhiyun self.assertEqual(g_tag.previous_sibling, to_text) 1009*4882a593Smuzhiyun 1010*4882a593Smuzhiyun def test_unwrap(self): 1011*4882a593Smuzhiyun tree = self.soup(""" 1012*4882a593Smuzhiyun <p>Unneeded <em>formatting</em> is unneeded</p> 1013*4882a593Smuzhiyun """) 1014*4882a593Smuzhiyun tree.em.unwrap() 1015*4882a593Smuzhiyun self.assertEqual(tree.em, None) 1016*4882a593Smuzhiyun self.assertEqual(tree.p.text, "Unneeded formatting is unneeded") 1017*4882a593Smuzhiyun 1018*4882a593Smuzhiyun def test_wrap(self): 1019*4882a593Smuzhiyun soup = self.soup("I wish I was bold.") 1020*4882a593Smuzhiyun value = soup.string.wrap(soup.new_tag("b")) 1021*4882a593Smuzhiyun self.assertEqual(value.decode(), "<b>I wish I was bold.</b>") 1022*4882a593Smuzhiyun self.assertEqual( 1023*4882a593Smuzhiyun soup.decode(), self.document_for("<b>I wish I was bold.</b>")) 1024*4882a593Smuzhiyun 1025*4882a593Smuzhiyun def test_wrap_extracts_tag_from_elsewhere(self): 1026*4882a593Smuzhiyun soup = self.soup("<b></b>I wish I was bold.") 1027*4882a593Smuzhiyun soup.b.next_sibling.wrap(soup.b) 1028*4882a593Smuzhiyun self.assertEqual( 1029*4882a593Smuzhiyun soup.decode(), self.document_for("<b>I wish I was bold.</b>")) 1030*4882a593Smuzhiyun 1031*4882a593Smuzhiyun def test_wrap_puts_new_contents_at_the_end(self): 1032*4882a593Smuzhiyun soup = self.soup("<b>I like being bold.</b>I wish I was bold.") 1033*4882a593Smuzhiyun soup.b.next_sibling.wrap(soup.b) 1034*4882a593Smuzhiyun self.assertEqual(2, len(soup.b.contents)) 1035*4882a593Smuzhiyun self.assertEqual( 1036*4882a593Smuzhiyun soup.decode(), self.document_for( 1037*4882a593Smuzhiyun "<b>I like being bold.I wish I was bold.</b>")) 1038*4882a593Smuzhiyun 1039*4882a593Smuzhiyun def test_extract(self): 1040*4882a593Smuzhiyun soup = self.soup( 1041*4882a593Smuzhiyun '<html><body>Some content. <div id="nav">Nav crap</div> More content.</body></html>') 1042*4882a593Smuzhiyun 1043*4882a593Smuzhiyun self.assertEqual(len(soup.body.contents), 3) 1044*4882a593Smuzhiyun extracted = soup.find(id="nav").extract() 1045*4882a593Smuzhiyun 1046*4882a593Smuzhiyun self.assertEqual( 1047*4882a593Smuzhiyun soup.decode(), "<html><body>Some content. More content.</body></html>") 1048*4882a593Smuzhiyun self.assertEqual(extracted.decode(), '<div id="nav">Nav crap</div>') 1049*4882a593Smuzhiyun 1050*4882a593Smuzhiyun # The extracted tag is now an orphan. 1051*4882a593Smuzhiyun self.assertEqual(len(soup.body.contents), 2) 1052*4882a593Smuzhiyun self.assertEqual(extracted.parent, None) 1053*4882a593Smuzhiyun self.assertEqual(extracted.previous_element, None) 1054*4882a593Smuzhiyun self.assertEqual(extracted.next_element.next_element, None) 1055*4882a593Smuzhiyun 1056*4882a593Smuzhiyun # The gap where the extracted tag used to be has been mended. 1057*4882a593Smuzhiyun content_1 = soup.find(text="Some content. ") 1058*4882a593Smuzhiyun content_2 = soup.find(text=" More content.") 1059*4882a593Smuzhiyun self.assertEqual(content_1.next_element, content_2) 1060*4882a593Smuzhiyun self.assertEqual(content_1.next_sibling, content_2) 1061*4882a593Smuzhiyun self.assertEqual(content_2.previous_element, content_1) 1062*4882a593Smuzhiyun self.assertEqual(content_2.previous_sibling, content_1) 1063*4882a593Smuzhiyun 1064*4882a593Smuzhiyun def test_extract_distinguishes_between_identical_strings(self): 1065*4882a593Smuzhiyun soup = self.soup("<a>foo</a><b>bar</b>") 1066*4882a593Smuzhiyun foo_1 = soup.a.string 1067*4882a593Smuzhiyun bar_1 = soup.b.string 1068*4882a593Smuzhiyun foo_2 = soup.new_string("foo") 1069*4882a593Smuzhiyun bar_2 = soup.new_string("bar") 1070*4882a593Smuzhiyun soup.a.append(foo_2) 1071*4882a593Smuzhiyun soup.b.append(bar_2) 1072*4882a593Smuzhiyun 1073*4882a593Smuzhiyun # Now there are two identical strings in the <a> tag, and two 1074*4882a593Smuzhiyun # in the <b> tag. Let's remove the first "foo" and the second 1075*4882a593Smuzhiyun # "bar". 1076*4882a593Smuzhiyun foo_1.extract() 1077*4882a593Smuzhiyun bar_2.extract() 1078*4882a593Smuzhiyun self.assertEqual(foo_2, soup.a.string) 1079*4882a593Smuzhiyun self.assertEqual(bar_2, soup.b.string) 1080*4882a593Smuzhiyun 1081*4882a593Smuzhiyun def test_extract_multiples_of_same_tag(self): 1082*4882a593Smuzhiyun soup = self.soup(""" 1083*4882a593Smuzhiyun<html> 1084*4882a593Smuzhiyun<head> 1085*4882a593Smuzhiyun<script>foo</script> 1086*4882a593Smuzhiyun</head> 1087*4882a593Smuzhiyun<body> 1088*4882a593Smuzhiyun <script>bar</script> 1089*4882a593Smuzhiyun <a></a> 1090*4882a593Smuzhiyun</body> 1091*4882a593Smuzhiyun<script>baz</script> 1092*4882a593Smuzhiyun</html>""") 1093*4882a593Smuzhiyun [soup.script.extract() for i in soup.find_all("script")] 1094*4882a593Smuzhiyun self.assertEqual("<body>\n\n<a></a>\n</body>", str(soup.body)) 1095*4882a593Smuzhiyun 1096*4882a593Smuzhiyun 1097*4882a593Smuzhiyun def test_extract_works_when_element_is_surrounded_by_identical_strings(self): 1098*4882a593Smuzhiyun soup = self.soup( 1099*4882a593Smuzhiyun '<html>\n' 1100*4882a593Smuzhiyun '<body>hi</body>\n' 1101*4882a593Smuzhiyun '</html>') 1102*4882a593Smuzhiyun soup.find('body').extract() 1103*4882a593Smuzhiyun self.assertEqual(None, soup.find('body')) 1104*4882a593Smuzhiyun 1105*4882a593Smuzhiyun 1106*4882a593Smuzhiyun def test_clear(self): 1107*4882a593Smuzhiyun """Tag.clear()""" 1108*4882a593Smuzhiyun soup = self.soup("<p><a>String <em>Italicized</em></a> and another</p>") 1109*4882a593Smuzhiyun # clear using extract() 1110*4882a593Smuzhiyun a = soup.a 1111*4882a593Smuzhiyun soup.p.clear() 1112*4882a593Smuzhiyun self.assertEqual(len(soup.p.contents), 0) 1113*4882a593Smuzhiyun self.assertTrue(hasattr(a, "contents")) 1114*4882a593Smuzhiyun 1115*4882a593Smuzhiyun # clear using decompose() 1116*4882a593Smuzhiyun em = a.em 1117*4882a593Smuzhiyun a.clear(decompose=True) 1118*4882a593Smuzhiyun self.assertEqual(0, len(em.contents)) 1119*4882a593Smuzhiyun 1120*4882a593Smuzhiyun def test_string_set(self): 1121*4882a593Smuzhiyun """Tag.string = 'string'""" 1122*4882a593Smuzhiyun soup = self.soup("<a></a> <b><c></c></b>") 1123*4882a593Smuzhiyun soup.a.string = "foo" 1124*4882a593Smuzhiyun self.assertEqual(soup.a.contents, ["foo"]) 1125*4882a593Smuzhiyun soup.b.string = "bar" 1126*4882a593Smuzhiyun self.assertEqual(soup.b.contents, ["bar"]) 1127*4882a593Smuzhiyun 1128*4882a593Smuzhiyun def test_string_set_does_not_affect_original_string(self): 1129*4882a593Smuzhiyun soup = self.soup("<a><b>foo</b><c>bar</c>") 1130*4882a593Smuzhiyun soup.b.string = soup.c.string 1131*4882a593Smuzhiyun self.assertEqual(soup.a.encode(), b"<a><b>bar</b><c>bar</c></a>") 1132*4882a593Smuzhiyun 1133*4882a593Smuzhiyun def test_set_string_preserves_class_of_string(self): 1134*4882a593Smuzhiyun soup = self.soup("<a></a>") 1135*4882a593Smuzhiyun cdata = CData("foo") 1136*4882a593Smuzhiyun soup.a.string = cdata 1137*4882a593Smuzhiyun self.assertTrue(isinstance(soup.a.string, CData)) 1138*4882a593Smuzhiyun 1139*4882a593Smuzhiyunclass TestElementObjects(SoupTest): 1140*4882a593Smuzhiyun """Test various features of element objects.""" 1141*4882a593Smuzhiyun 1142*4882a593Smuzhiyun def test_len(self): 1143*4882a593Smuzhiyun """The length of an element is its number of children.""" 1144*4882a593Smuzhiyun soup = self.soup("<top>1<b>2</b>3</top>") 1145*4882a593Smuzhiyun 1146*4882a593Smuzhiyun # The BeautifulSoup object itself contains one element: the 1147*4882a593Smuzhiyun # <top> tag. 1148*4882a593Smuzhiyun self.assertEqual(len(soup.contents), 1) 1149*4882a593Smuzhiyun self.assertEqual(len(soup), 1) 1150*4882a593Smuzhiyun 1151*4882a593Smuzhiyun # The <top> tag contains three elements: the text node "1", the 1152*4882a593Smuzhiyun # <b> tag, and the text node "3". 1153*4882a593Smuzhiyun self.assertEqual(len(soup.top), 3) 1154*4882a593Smuzhiyun self.assertEqual(len(soup.top.contents), 3) 1155*4882a593Smuzhiyun 1156*4882a593Smuzhiyun def test_member_access_invokes_find(self): 1157*4882a593Smuzhiyun """Accessing a Python member .foo invokes find('foo')""" 1158*4882a593Smuzhiyun soup = self.soup('<b><i></i></b>') 1159*4882a593Smuzhiyun self.assertEqual(soup.b, soup.find('b')) 1160*4882a593Smuzhiyun self.assertEqual(soup.b.i, soup.find('b').find('i')) 1161*4882a593Smuzhiyun self.assertEqual(soup.a, None) 1162*4882a593Smuzhiyun 1163*4882a593Smuzhiyun def test_deprecated_member_access(self): 1164*4882a593Smuzhiyun soup = self.soup('<b><i></i></b>') 1165*4882a593Smuzhiyun with warnings.catch_warnings(record=True) as w: 1166*4882a593Smuzhiyun tag = soup.bTag 1167*4882a593Smuzhiyun self.assertEqual(soup.b, tag) 1168*4882a593Smuzhiyun self.assertEqual( 1169*4882a593Smuzhiyun '.bTag is deprecated, use .find("b") instead.', 1170*4882a593Smuzhiyun str(w[0].message)) 1171*4882a593Smuzhiyun 1172*4882a593Smuzhiyun def test_has_attr(self): 1173*4882a593Smuzhiyun """has_attr() checks for the presence of an attribute. 1174*4882a593Smuzhiyun 1175*4882a593Smuzhiyun Please note note: has_attr() is different from 1176*4882a593Smuzhiyun __in__. has_attr() checks the tag's attributes and __in__ 1177*4882a593Smuzhiyun checks the tag's chidlren. 1178*4882a593Smuzhiyun """ 1179*4882a593Smuzhiyun soup = self.soup("<foo attr='bar'>") 1180*4882a593Smuzhiyun self.assertTrue(soup.foo.has_attr('attr')) 1181*4882a593Smuzhiyun self.assertFalse(soup.foo.has_attr('attr2')) 1182*4882a593Smuzhiyun 1183*4882a593Smuzhiyun 1184*4882a593Smuzhiyun def test_attributes_come_out_in_alphabetical_order(self): 1185*4882a593Smuzhiyun markup = '<b a="1" z="5" m="3" f="2" y="4"></b>' 1186*4882a593Smuzhiyun self.assertSoupEquals(markup, '<b a="1" f="2" m="3" y="4" z="5"></b>') 1187*4882a593Smuzhiyun 1188*4882a593Smuzhiyun def test_string(self): 1189*4882a593Smuzhiyun # A tag that contains only a text node makes that node 1190*4882a593Smuzhiyun # available as .string. 1191*4882a593Smuzhiyun soup = self.soup("<b>foo</b>") 1192*4882a593Smuzhiyun self.assertEqual(soup.b.string, 'foo') 1193*4882a593Smuzhiyun 1194*4882a593Smuzhiyun def test_empty_tag_has_no_string(self): 1195*4882a593Smuzhiyun # A tag with no children has no .stirng. 1196*4882a593Smuzhiyun soup = self.soup("<b></b>") 1197*4882a593Smuzhiyun self.assertEqual(soup.b.string, None) 1198*4882a593Smuzhiyun 1199*4882a593Smuzhiyun def test_tag_with_multiple_children_has_no_string(self): 1200*4882a593Smuzhiyun # A tag with no children has no .string. 1201*4882a593Smuzhiyun soup = self.soup("<a>foo<b></b><b></b></b>") 1202*4882a593Smuzhiyun self.assertEqual(soup.b.string, None) 1203*4882a593Smuzhiyun 1204*4882a593Smuzhiyun soup = self.soup("<a>foo<b></b>bar</b>") 1205*4882a593Smuzhiyun self.assertEqual(soup.b.string, None) 1206*4882a593Smuzhiyun 1207*4882a593Smuzhiyun # Even if all the children are strings, due to trickery, 1208*4882a593Smuzhiyun # it won't work--but this would be a good optimization. 1209*4882a593Smuzhiyun soup = self.soup("<a>foo</b>") 1210*4882a593Smuzhiyun soup.a.insert(1, "bar") 1211*4882a593Smuzhiyun self.assertEqual(soup.a.string, None) 1212*4882a593Smuzhiyun 1213*4882a593Smuzhiyun def test_tag_with_recursive_string_has_string(self): 1214*4882a593Smuzhiyun # A tag with a single child which has a .string inherits that 1215*4882a593Smuzhiyun # .string. 1216*4882a593Smuzhiyun soup = self.soup("<a><b>foo</b></a>") 1217*4882a593Smuzhiyun self.assertEqual(soup.a.string, "foo") 1218*4882a593Smuzhiyun self.assertEqual(soup.string, "foo") 1219*4882a593Smuzhiyun 1220*4882a593Smuzhiyun def test_lack_of_string(self): 1221*4882a593Smuzhiyun """Only a tag containing a single text node has a .string.""" 1222*4882a593Smuzhiyun soup = self.soup("<b>f<i>e</i>o</b>") 1223*4882a593Smuzhiyun self.assertFalse(soup.b.string) 1224*4882a593Smuzhiyun 1225*4882a593Smuzhiyun soup = self.soup("<b></b>") 1226*4882a593Smuzhiyun self.assertFalse(soup.b.string) 1227*4882a593Smuzhiyun 1228*4882a593Smuzhiyun def test_all_text(self): 1229*4882a593Smuzhiyun """Tag.text and Tag.get_text(sep=u"") -> all child text, concatenated""" 1230*4882a593Smuzhiyun soup = self.soup("<a>a<b>r</b> <r> t </r></a>") 1231*4882a593Smuzhiyun self.assertEqual(soup.a.text, "ar t ") 1232*4882a593Smuzhiyun self.assertEqual(soup.a.get_text(strip=True), "art") 1233*4882a593Smuzhiyun self.assertEqual(soup.a.get_text(","), "a,r, , t ") 1234*4882a593Smuzhiyun self.assertEqual(soup.a.get_text(",", strip=True), "a,r,t") 1235*4882a593Smuzhiyun 1236*4882a593Smuzhiyun def test_get_text_ignores_comments(self): 1237*4882a593Smuzhiyun soup = self.soup("foo<!--IGNORE-->bar") 1238*4882a593Smuzhiyun self.assertEqual(soup.get_text(), "foobar") 1239*4882a593Smuzhiyun 1240*4882a593Smuzhiyun self.assertEqual( 1241*4882a593Smuzhiyun soup.get_text(types=(NavigableString, Comment)), "fooIGNOREbar") 1242*4882a593Smuzhiyun self.assertEqual( 1243*4882a593Smuzhiyun soup.get_text(types=None), "fooIGNOREbar") 1244*4882a593Smuzhiyun 1245*4882a593Smuzhiyun def test_all_strings_ignores_comments(self): 1246*4882a593Smuzhiyun soup = self.soup("foo<!--IGNORE-->bar") 1247*4882a593Smuzhiyun self.assertEqual(['foo', 'bar'], list(soup.strings)) 1248*4882a593Smuzhiyun 1249*4882a593Smuzhiyunclass TestCDAtaListAttributes(SoupTest): 1250*4882a593Smuzhiyun 1251*4882a593Smuzhiyun """Testing cdata-list attributes like 'class'. 1252*4882a593Smuzhiyun """ 1253*4882a593Smuzhiyun def test_single_value_becomes_list(self): 1254*4882a593Smuzhiyun soup = self.soup("<a class='foo'>") 1255*4882a593Smuzhiyun self.assertEqual(["foo"],soup.a['class']) 1256*4882a593Smuzhiyun 1257*4882a593Smuzhiyun def test_multiple_values_becomes_list(self): 1258*4882a593Smuzhiyun soup = self.soup("<a class='foo bar'>") 1259*4882a593Smuzhiyun self.assertEqual(["foo", "bar"], soup.a['class']) 1260*4882a593Smuzhiyun 1261*4882a593Smuzhiyun def test_multiple_values_separated_by_weird_whitespace(self): 1262*4882a593Smuzhiyun soup = self.soup("<a class='foo\tbar\nbaz'>") 1263*4882a593Smuzhiyun self.assertEqual(["foo", "bar", "baz"],soup.a['class']) 1264*4882a593Smuzhiyun 1265*4882a593Smuzhiyun def test_attributes_joined_into_string_on_output(self): 1266*4882a593Smuzhiyun soup = self.soup("<a class='foo\tbar'>") 1267*4882a593Smuzhiyun self.assertEqual(b'<a class="foo bar"></a>', soup.a.encode()) 1268*4882a593Smuzhiyun 1269*4882a593Smuzhiyun def test_accept_charset(self): 1270*4882a593Smuzhiyun soup = self.soup('<form accept-charset="ISO-8859-1 UTF-8">') 1271*4882a593Smuzhiyun self.assertEqual(['ISO-8859-1', 'UTF-8'], soup.form['accept-charset']) 1272*4882a593Smuzhiyun 1273*4882a593Smuzhiyun def test_cdata_attribute_applying_only_to_one_tag(self): 1274*4882a593Smuzhiyun data = '<a accept-charset="ISO-8859-1 UTF-8"></a>' 1275*4882a593Smuzhiyun soup = self.soup(data) 1276*4882a593Smuzhiyun # We saw in another test that accept-charset is a cdata-list 1277*4882a593Smuzhiyun # attribute for the <form> tag. But it's not a cdata-list 1278*4882a593Smuzhiyun # attribute for any other tag. 1279*4882a593Smuzhiyun self.assertEqual('ISO-8859-1 UTF-8', soup.a['accept-charset']) 1280*4882a593Smuzhiyun 1281*4882a593Smuzhiyun def test_string_has_immutable_name_property(self): 1282*4882a593Smuzhiyun string = self.soup("s").string 1283*4882a593Smuzhiyun self.assertEqual(None, string.name) 1284*4882a593Smuzhiyun def t(): 1285*4882a593Smuzhiyun string.name = 'foo' 1286*4882a593Smuzhiyun self.assertRaises(AttributeError, t) 1287*4882a593Smuzhiyun 1288*4882a593Smuzhiyunclass TestPersistence(SoupTest): 1289*4882a593Smuzhiyun "Testing features like pickle and deepcopy." 1290*4882a593Smuzhiyun 1291*4882a593Smuzhiyun def setUp(self): 1292*4882a593Smuzhiyun super(TestPersistence, self).setUp() 1293*4882a593Smuzhiyun self.page = """<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" 1294*4882a593Smuzhiyun"http://www.w3.org/TR/REC-html40/transitional.dtd"> 1295*4882a593Smuzhiyun<html> 1296*4882a593Smuzhiyun<head> 1297*4882a593Smuzhiyun<meta http-equiv="Content-Type" content="text/html; charset=utf-8"> 1298*4882a593Smuzhiyun<title>Beautiful Soup: We called him Tortoise because he taught us.</title> 1299*4882a593Smuzhiyun<link rev="made" href="mailto:leonardr@segfault.org"> 1300*4882a593Smuzhiyun<meta name="Description" content="Beautiful Soup: an HTML parser optimized for screen-scraping."> 1301*4882a593Smuzhiyun<meta name="generator" content="Markov Approximation 1.4 (module: leonardr)"> 1302*4882a593Smuzhiyun<meta name="author" content="Leonard Richardson"> 1303*4882a593Smuzhiyun</head> 1304*4882a593Smuzhiyun<body> 1305*4882a593Smuzhiyun<a href="foo">foo</a> 1306*4882a593Smuzhiyun<a href="foo"><b>bar</b></a> 1307*4882a593Smuzhiyun</body> 1308*4882a593Smuzhiyun</html>""" 1309*4882a593Smuzhiyun self.tree = self.soup(self.page) 1310*4882a593Smuzhiyun 1311*4882a593Smuzhiyun def test_pickle_and_unpickle_identity(self): 1312*4882a593Smuzhiyun # Pickling a tree, then unpickling it, yields a tree identical 1313*4882a593Smuzhiyun # to the original. 1314*4882a593Smuzhiyun dumped = pickle.dumps(self.tree, 2) 1315*4882a593Smuzhiyun loaded = pickle.loads(dumped) 1316*4882a593Smuzhiyun self.assertEqual(loaded.__class__, BeautifulSoup) 1317*4882a593Smuzhiyun self.assertEqual(loaded.decode(), self.tree.decode()) 1318*4882a593Smuzhiyun 1319*4882a593Smuzhiyun def test_deepcopy_identity(self): 1320*4882a593Smuzhiyun # Making a deepcopy of a tree yields an identical tree. 1321*4882a593Smuzhiyun copied = copy.deepcopy(self.tree) 1322*4882a593Smuzhiyun self.assertEqual(copied.decode(), self.tree.decode()) 1323*4882a593Smuzhiyun 1324*4882a593Smuzhiyun def test_unicode_pickle(self): 1325*4882a593Smuzhiyun # A tree containing Unicode characters can be pickled. 1326*4882a593Smuzhiyun html = "<b>\N{SNOWMAN}</b>" 1327*4882a593Smuzhiyun soup = self.soup(html) 1328*4882a593Smuzhiyun dumped = pickle.dumps(soup, pickle.HIGHEST_PROTOCOL) 1329*4882a593Smuzhiyun loaded = pickle.loads(dumped) 1330*4882a593Smuzhiyun self.assertEqual(loaded.decode(), soup.decode()) 1331*4882a593Smuzhiyun 1332*4882a593Smuzhiyun def test_copy_navigablestring_is_not_attached_to_tree(self): 1333*4882a593Smuzhiyun html = "<b>Foo<a></a></b><b>Bar</b>" 1334*4882a593Smuzhiyun soup = self.soup(html) 1335*4882a593Smuzhiyun s1 = soup.find(string="Foo") 1336*4882a593Smuzhiyun s2 = copy.copy(s1) 1337*4882a593Smuzhiyun self.assertEqual(s1, s2) 1338*4882a593Smuzhiyun self.assertEqual(None, s2.parent) 1339*4882a593Smuzhiyun self.assertEqual(None, s2.next_element) 1340*4882a593Smuzhiyun self.assertNotEqual(None, s1.next_sibling) 1341*4882a593Smuzhiyun self.assertEqual(None, s2.next_sibling) 1342*4882a593Smuzhiyun self.assertEqual(None, s2.previous_element) 1343*4882a593Smuzhiyun 1344*4882a593Smuzhiyun def test_copy_navigablestring_subclass_has_same_type(self): 1345*4882a593Smuzhiyun html = "<b><!--Foo--></b>" 1346*4882a593Smuzhiyun soup = self.soup(html) 1347*4882a593Smuzhiyun s1 = soup.string 1348*4882a593Smuzhiyun s2 = copy.copy(s1) 1349*4882a593Smuzhiyun self.assertEqual(s1, s2) 1350*4882a593Smuzhiyun self.assertTrue(isinstance(s2, Comment)) 1351*4882a593Smuzhiyun 1352*4882a593Smuzhiyun def test_copy_entire_soup(self): 1353*4882a593Smuzhiyun html = "<div><b>Foo<a></a></b><b>Bar</b></div>end" 1354*4882a593Smuzhiyun soup = self.soup(html) 1355*4882a593Smuzhiyun soup_copy = copy.copy(soup) 1356*4882a593Smuzhiyun self.assertEqual(soup, soup_copy) 1357*4882a593Smuzhiyun 1358*4882a593Smuzhiyun def test_copy_tag_copies_contents(self): 1359*4882a593Smuzhiyun html = "<div><b>Foo<a></a></b><b>Bar</b></div>end" 1360*4882a593Smuzhiyun soup = self.soup(html) 1361*4882a593Smuzhiyun div = soup.div 1362*4882a593Smuzhiyun div_copy = copy.copy(div) 1363*4882a593Smuzhiyun 1364*4882a593Smuzhiyun # The two tags look the same, and evaluate to equal. 1365*4882a593Smuzhiyun self.assertEqual(str(div), str(div_copy)) 1366*4882a593Smuzhiyun self.assertEqual(div, div_copy) 1367*4882a593Smuzhiyun 1368*4882a593Smuzhiyun # But they're not the same object. 1369*4882a593Smuzhiyun self.assertFalse(div is div_copy) 1370*4882a593Smuzhiyun 1371*4882a593Smuzhiyun # And they don't have the same relation to the parse tree. The 1372*4882a593Smuzhiyun # copy is not associated with a parse tree at all. 1373*4882a593Smuzhiyun self.assertEqual(None, div_copy.parent) 1374*4882a593Smuzhiyun self.assertEqual(None, div_copy.previous_element) 1375*4882a593Smuzhiyun self.assertEqual(None, div_copy.find(string='Bar').next_element) 1376*4882a593Smuzhiyun self.assertNotEqual(None, div.find(string='Bar').next_element) 1377*4882a593Smuzhiyun 1378*4882a593Smuzhiyunclass TestSubstitutions(SoupTest): 1379*4882a593Smuzhiyun 1380*4882a593Smuzhiyun def test_default_formatter_is_minimal(self): 1381*4882a593Smuzhiyun markup = "<b><<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>></b>" 1382*4882a593Smuzhiyun soup = self.soup(markup) 1383*4882a593Smuzhiyun decoded = soup.decode(formatter="minimal") 1384*4882a593Smuzhiyun # The < is converted back into < but the e-with-acute is left alone. 1385*4882a593Smuzhiyun self.assertEqual( 1386*4882a593Smuzhiyun decoded, 1387*4882a593Smuzhiyun self.document_for( 1388*4882a593Smuzhiyun "<b><<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>></b>")) 1389*4882a593Smuzhiyun 1390*4882a593Smuzhiyun def test_formatter_html(self): 1391*4882a593Smuzhiyun markup = "<b><<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>></b>" 1392*4882a593Smuzhiyun soup = self.soup(markup) 1393*4882a593Smuzhiyun decoded = soup.decode(formatter="html") 1394*4882a593Smuzhiyun self.assertEqual( 1395*4882a593Smuzhiyun decoded, 1396*4882a593Smuzhiyun self.document_for("<b><<Sacré bleu!>></b>")) 1397*4882a593Smuzhiyun 1398*4882a593Smuzhiyun def test_formatter_minimal(self): 1399*4882a593Smuzhiyun markup = "<b><<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>></b>" 1400*4882a593Smuzhiyun soup = self.soup(markup) 1401*4882a593Smuzhiyun decoded = soup.decode(formatter="minimal") 1402*4882a593Smuzhiyun # The < is converted back into < but the e-with-acute is left alone. 1403*4882a593Smuzhiyun self.assertEqual( 1404*4882a593Smuzhiyun decoded, 1405*4882a593Smuzhiyun self.document_for( 1406*4882a593Smuzhiyun "<b><<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>></b>")) 1407*4882a593Smuzhiyun 1408*4882a593Smuzhiyun def test_formatter_null(self): 1409*4882a593Smuzhiyun markup = "<b><<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>></b>" 1410*4882a593Smuzhiyun soup = self.soup(markup) 1411*4882a593Smuzhiyun decoded = soup.decode(formatter=None) 1412*4882a593Smuzhiyun # Neither the angle brackets nor the e-with-acute are converted. 1413*4882a593Smuzhiyun # This is not valid HTML, but it's what the user wanted. 1414*4882a593Smuzhiyun self.assertEqual(decoded, 1415*4882a593Smuzhiyun self.document_for("<b><<Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>></b>")) 1416*4882a593Smuzhiyun 1417*4882a593Smuzhiyun def test_formatter_custom(self): 1418*4882a593Smuzhiyun markup = "<b><foo></b><b>bar</b>" 1419*4882a593Smuzhiyun soup = self.soup(markup) 1420*4882a593Smuzhiyun decoded = soup.decode(formatter = lambda x: x.upper()) 1421*4882a593Smuzhiyun # Instead of normal entity conversion code, the custom 1422*4882a593Smuzhiyun # callable is called on every string. 1423*4882a593Smuzhiyun self.assertEqual( 1424*4882a593Smuzhiyun decoded, 1425*4882a593Smuzhiyun self.document_for("<b><FOO></b><b>BAR</b>")) 1426*4882a593Smuzhiyun 1427*4882a593Smuzhiyun def test_formatter_is_run_on_attribute_values(self): 1428*4882a593Smuzhiyun markup = '<a href="http://a.com?a=b&c=é">e</a>' 1429*4882a593Smuzhiyun soup = self.soup(markup) 1430*4882a593Smuzhiyun a = soup.a 1431*4882a593Smuzhiyun 1432*4882a593Smuzhiyun expect_minimal = '<a href="http://a.com?a=b&c=é">e</a>' 1433*4882a593Smuzhiyun 1434*4882a593Smuzhiyun self.assertEqual(expect_minimal, a.decode()) 1435*4882a593Smuzhiyun self.assertEqual(expect_minimal, a.decode(formatter="minimal")) 1436*4882a593Smuzhiyun 1437*4882a593Smuzhiyun expect_html = '<a href="http://a.com?a=b&c=é">e</a>' 1438*4882a593Smuzhiyun self.assertEqual(expect_html, a.decode(formatter="html")) 1439*4882a593Smuzhiyun 1440*4882a593Smuzhiyun self.assertEqual(markup, a.decode(formatter=None)) 1441*4882a593Smuzhiyun expect_upper = '<a href="HTTP://A.COM?A=B&C=É">E</a>' 1442*4882a593Smuzhiyun self.assertEqual(expect_upper, a.decode(formatter=lambda x: x.upper())) 1443*4882a593Smuzhiyun 1444*4882a593Smuzhiyun def test_formatter_skips_script_tag_for_html_documents(self): 1445*4882a593Smuzhiyun doc = """ 1446*4882a593Smuzhiyun <script type="text/javascript"> 1447*4882a593Smuzhiyun console.log("< < hey > > "); 1448*4882a593Smuzhiyun </script> 1449*4882a593Smuzhiyun""" 1450*4882a593Smuzhiyun encoded = BeautifulSoup(doc, 'html.parser').encode() 1451*4882a593Smuzhiyun self.assertTrue(b"< < hey > >" in encoded) 1452*4882a593Smuzhiyun 1453*4882a593Smuzhiyun def test_formatter_skips_style_tag_for_html_documents(self): 1454*4882a593Smuzhiyun doc = """ 1455*4882a593Smuzhiyun <style type="text/css"> 1456*4882a593Smuzhiyun console.log("< < hey > > "); 1457*4882a593Smuzhiyun </style> 1458*4882a593Smuzhiyun""" 1459*4882a593Smuzhiyun encoded = BeautifulSoup(doc, 'html.parser').encode() 1460*4882a593Smuzhiyun self.assertTrue(b"< < hey > >" in encoded) 1461*4882a593Smuzhiyun 1462*4882a593Smuzhiyun def test_prettify_leaves_preformatted_text_alone(self): 1463*4882a593Smuzhiyun soup = self.soup("<div> foo <pre> \tbar\n \n </pre> baz ") 1464*4882a593Smuzhiyun # Everything outside the <pre> tag is reformatted, but everything 1465*4882a593Smuzhiyun # inside is left alone. 1466*4882a593Smuzhiyun self.assertEqual( 1467*4882a593Smuzhiyun '<div>\n foo\n <pre> \tbar\n \n </pre>\n baz\n</div>', 1468*4882a593Smuzhiyun soup.div.prettify()) 1469*4882a593Smuzhiyun 1470*4882a593Smuzhiyun def test_prettify_accepts_formatter(self): 1471*4882a593Smuzhiyun soup = BeautifulSoup("<html><body>foo</body></html>", 'html.parser') 1472*4882a593Smuzhiyun pretty = soup.prettify(formatter = lambda x: x.upper()) 1473*4882a593Smuzhiyun self.assertTrue("FOO" in pretty) 1474*4882a593Smuzhiyun 1475*4882a593Smuzhiyun def test_prettify_outputs_unicode_by_default(self): 1476*4882a593Smuzhiyun soup = self.soup("<a></a>") 1477*4882a593Smuzhiyun self.assertEqual(str, type(soup.prettify())) 1478*4882a593Smuzhiyun 1479*4882a593Smuzhiyun def test_prettify_can_encode_data(self): 1480*4882a593Smuzhiyun soup = self.soup("<a></a>") 1481*4882a593Smuzhiyun self.assertEqual(bytes, type(soup.prettify("utf-8"))) 1482*4882a593Smuzhiyun 1483*4882a593Smuzhiyun def test_html_entity_substitution_off_by_default(self): 1484*4882a593Smuzhiyun markup = "<b>Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!</b>" 1485*4882a593Smuzhiyun soup = self.soup(markup) 1486*4882a593Smuzhiyun encoded = soup.b.encode("utf-8") 1487*4882a593Smuzhiyun self.assertEqual(encoded, markup.encode('utf-8')) 1488*4882a593Smuzhiyun 1489*4882a593Smuzhiyun def test_encoding_substitution(self): 1490*4882a593Smuzhiyun # Here's the <meta> tag saying that a document is 1491*4882a593Smuzhiyun # encoded in Shift-JIS. 1492*4882a593Smuzhiyun meta_tag = ('<meta content="text/html; charset=x-sjis" ' 1493*4882a593Smuzhiyun 'http-equiv="Content-type"/>') 1494*4882a593Smuzhiyun soup = self.soup(meta_tag) 1495*4882a593Smuzhiyun 1496*4882a593Smuzhiyun # Parse the document, and the charset apprears unchanged. 1497*4882a593Smuzhiyun self.assertEqual(soup.meta['content'], 'text/html; charset=x-sjis') 1498*4882a593Smuzhiyun 1499*4882a593Smuzhiyun # Encode the document into some encoding, and the encoding is 1500*4882a593Smuzhiyun # substituted into the meta tag. 1501*4882a593Smuzhiyun utf_8 = soup.encode("utf-8") 1502*4882a593Smuzhiyun self.assertTrue(b"charset=utf-8" in utf_8) 1503*4882a593Smuzhiyun 1504*4882a593Smuzhiyun euc_jp = soup.encode("euc_jp") 1505*4882a593Smuzhiyun self.assertTrue(b"charset=euc_jp" in euc_jp) 1506*4882a593Smuzhiyun 1507*4882a593Smuzhiyun shift_jis = soup.encode("shift-jis") 1508*4882a593Smuzhiyun self.assertTrue(b"charset=shift-jis" in shift_jis) 1509*4882a593Smuzhiyun 1510*4882a593Smuzhiyun utf_16_u = soup.encode("utf-16").decode("utf-16") 1511*4882a593Smuzhiyun self.assertTrue("charset=utf-16" in utf_16_u) 1512*4882a593Smuzhiyun 1513*4882a593Smuzhiyun def test_encoding_substitution_doesnt_happen_if_tag_is_strained(self): 1514*4882a593Smuzhiyun markup = ('<head><meta content="text/html; charset=x-sjis" ' 1515*4882a593Smuzhiyun 'http-equiv="Content-type"/></head><pre>foo</pre>') 1516*4882a593Smuzhiyun 1517*4882a593Smuzhiyun # Beautiful Soup used to try to rewrite the meta tag even if the 1518*4882a593Smuzhiyun # meta tag got filtered out by the strainer. This test makes 1519*4882a593Smuzhiyun # sure that doesn't happen. 1520*4882a593Smuzhiyun strainer = SoupStrainer('pre') 1521*4882a593Smuzhiyun soup = self.soup(markup, parse_only=strainer) 1522*4882a593Smuzhiyun self.assertEqual(soup.contents[0].name, 'pre') 1523*4882a593Smuzhiyun 1524*4882a593Smuzhiyunclass TestEncoding(SoupTest): 1525*4882a593Smuzhiyun """Test the ability to encode objects into strings.""" 1526*4882a593Smuzhiyun 1527*4882a593Smuzhiyun def test_unicode_string_can_be_encoded(self): 1528*4882a593Smuzhiyun html = "<b>\N{SNOWMAN}</b>" 1529*4882a593Smuzhiyun soup = self.soup(html) 1530*4882a593Smuzhiyun self.assertEqual(soup.b.string.encode("utf-8"), 1531*4882a593Smuzhiyun "\N{SNOWMAN}".encode("utf-8")) 1532*4882a593Smuzhiyun 1533*4882a593Smuzhiyun def test_tag_containing_unicode_string_can_be_encoded(self): 1534*4882a593Smuzhiyun html = "<b>\N{SNOWMAN}</b>" 1535*4882a593Smuzhiyun soup = self.soup(html) 1536*4882a593Smuzhiyun self.assertEqual( 1537*4882a593Smuzhiyun soup.b.encode("utf-8"), html.encode("utf-8")) 1538*4882a593Smuzhiyun 1539*4882a593Smuzhiyun def test_encoding_substitutes_unrecognized_characters_by_default(self): 1540*4882a593Smuzhiyun html = "<b>\N{SNOWMAN}</b>" 1541*4882a593Smuzhiyun soup = self.soup(html) 1542*4882a593Smuzhiyun self.assertEqual(soup.b.encode("ascii"), b"<b>☃</b>") 1543*4882a593Smuzhiyun 1544*4882a593Smuzhiyun def test_encoding_can_be_made_strict(self): 1545*4882a593Smuzhiyun html = "<b>\N{SNOWMAN}</b>" 1546*4882a593Smuzhiyun soup = self.soup(html) 1547*4882a593Smuzhiyun self.assertRaises( 1548*4882a593Smuzhiyun UnicodeEncodeError, soup.encode, "ascii", errors="strict") 1549*4882a593Smuzhiyun 1550*4882a593Smuzhiyun def test_decode_contents(self): 1551*4882a593Smuzhiyun html = "<b>\N{SNOWMAN}</b>" 1552*4882a593Smuzhiyun soup = self.soup(html) 1553*4882a593Smuzhiyun self.assertEqual("\N{SNOWMAN}", soup.b.decode_contents()) 1554*4882a593Smuzhiyun 1555*4882a593Smuzhiyun def test_encode_contents(self): 1556*4882a593Smuzhiyun html = "<b>\N{SNOWMAN}</b>" 1557*4882a593Smuzhiyun soup = self.soup(html) 1558*4882a593Smuzhiyun self.assertEqual( 1559*4882a593Smuzhiyun "\N{SNOWMAN}".encode("utf8"), soup.b.encode_contents( 1560*4882a593Smuzhiyun encoding="utf8")) 1561*4882a593Smuzhiyun 1562*4882a593Smuzhiyun def test_deprecated_renderContents(self): 1563*4882a593Smuzhiyun html = "<b>\N{SNOWMAN}</b>" 1564*4882a593Smuzhiyun soup = self.soup(html) 1565*4882a593Smuzhiyun self.assertEqual( 1566*4882a593Smuzhiyun "\N{SNOWMAN}".encode("utf8"), soup.b.renderContents()) 1567*4882a593Smuzhiyun 1568*4882a593Smuzhiyun def test_repr(self): 1569*4882a593Smuzhiyun html = "<b>\N{SNOWMAN}</b>" 1570*4882a593Smuzhiyun soup = self.soup(html) 1571*4882a593Smuzhiyun if PY3K: 1572*4882a593Smuzhiyun self.assertEqual(html, repr(soup)) 1573*4882a593Smuzhiyun else: 1574*4882a593Smuzhiyun self.assertEqual(b'<b>\\u2603</b>', repr(soup)) 1575*4882a593Smuzhiyun 1576*4882a593Smuzhiyunclass TestNavigableStringSubclasses(SoupTest): 1577*4882a593Smuzhiyun 1578*4882a593Smuzhiyun def test_cdata(self): 1579*4882a593Smuzhiyun # None of the current builders turn CDATA sections into CData 1580*4882a593Smuzhiyun # objects, but you can create them manually. 1581*4882a593Smuzhiyun soup = self.soup("") 1582*4882a593Smuzhiyun cdata = CData("foo") 1583*4882a593Smuzhiyun soup.insert(1, cdata) 1584*4882a593Smuzhiyun self.assertEqual(str(soup), "<![CDATA[foo]]>") 1585*4882a593Smuzhiyun self.assertEqual(soup.find(text="foo"), "foo") 1586*4882a593Smuzhiyun self.assertEqual(soup.contents[0], "foo") 1587*4882a593Smuzhiyun 1588*4882a593Smuzhiyun def test_cdata_is_never_formatted(self): 1589*4882a593Smuzhiyun """Text inside a CData object is passed into the formatter. 1590*4882a593Smuzhiyun 1591*4882a593Smuzhiyun But the return value is ignored. 1592*4882a593Smuzhiyun """ 1593*4882a593Smuzhiyun 1594*4882a593Smuzhiyun self.count = 0 1595*4882a593Smuzhiyun def increment(*args): 1596*4882a593Smuzhiyun self.count += 1 1597*4882a593Smuzhiyun return "BITTER FAILURE" 1598*4882a593Smuzhiyun 1599*4882a593Smuzhiyun soup = self.soup("") 1600*4882a593Smuzhiyun cdata = CData("<><><>") 1601*4882a593Smuzhiyun soup.insert(1, cdata) 1602*4882a593Smuzhiyun self.assertEqual( 1603*4882a593Smuzhiyun b"<![CDATA[<><><>]]>", soup.encode(formatter=increment)) 1604*4882a593Smuzhiyun self.assertEqual(1, self.count) 1605*4882a593Smuzhiyun 1606*4882a593Smuzhiyun def test_doctype_ends_in_newline(self): 1607*4882a593Smuzhiyun # Unlike other NavigableString subclasses, a DOCTYPE always ends 1608*4882a593Smuzhiyun # in a newline. 1609*4882a593Smuzhiyun doctype = Doctype("foo") 1610*4882a593Smuzhiyun soup = self.soup("") 1611*4882a593Smuzhiyun soup.insert(1, doctype) 1612*4882a593Smuzhiyun self.assertEqual(soup.encode(), b"<!DOCTYPE foo>\n") 1613*4882a593Smuzhiyun 1614*4882a593Smuzhiyun def test_declaration(self): 1615*4882a593Smuzhiyun d = Declaration("foo") 1616*4882a593Smuzhiyun self.assertEqual("<?foo?>", d.output_ready()) 1617*4882a593Smuzhiyun 1618*4882a593Smuzhiyunclass TestSoupSelector(TreeTest): 1619*4882a593Smuzhiyun 1620*4882a593Smuzhiyun HTML = """ 1621*4882a593Smuzhiyun<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" 1622*4882a593Smuzhiyun"http://www.w3.org/TR/html4/strict.dtd"> 1623*4882a593Smuzhiyun<html> 1624*4882a593Smuzhiyun<head> 1625*4882a593Smuzhiyun<title>The title</title> 1626*4882a593Smuzhiyun<link rel="stylesheet" href="blah.css" type="text/css" id="l1"> 1627*4882a593Smuzhiyun</head> 1628*4882a593Smuzhiyun<body> 1629*4882a593Smuzhiyun<custom-dashed-tag class="dashed" id="dash1">Hello there.</custom-dashed-tag> 1630*4882a593Smuzhiyun<div id="main" class="fancy"> 1631*4882a593Smuzhiyun<div id="inner"> 1632*4882a593Smuzhiyun<h1 id="header1">An H1</h1> 1633*4882a593Smuzhiyun<p>Some text</p> 1634*4882a593Smuzhiyun<p class="onep" id="p1">Some more text</p> 1635*4882a593Smuzhiyun<h2 id="header2">An H2</h2> 1636*4882a593Smuzhiyun<p class="class1 class2 class3" id="pmulti">Another</p> 1637*4882a593Smuzhiyun<a href="http://bob.example.org/" rel="friend met" id="bob">Bob</a> 1638*4882a593Smuzhiyun<h2 id="header3">Another H2</h2> 1639*4882a593Smuzhiyun<a id="me" href="http://simonwillison.net/" rel="me">me</a> 1640*4882a593Smuzhiyun<span class="s1"> 1641*4882a593Smuzhiyun<a href="#" id="s1a1">span1a1</a> 1642*4882a593Smuzhiyun<a href="#" id="s1a2">span1a2 <span id="s1a2s1">test</span></a> 1643*4882a593Smuzhiyun<span class="span2"> 1644*4882a593Smuzhiyun<a href="#" id="s2a1">span2a1</a> 1645*4882a593Smuzhiyun</span> 1646*4882a593Smuzhiyun<span class="span3"></span> 1647*4882a593Smuzhiyun<custom-dashed-tag class="dashed" id="dash2"/> 1648*4882a593Smuzhiyun<div data-tag="dashedvalue" id="data1"/> 1649*4882a593Smuzhiyun</span> 1650*4882a593Smuzhiyun</div> 1651*4882a593Smuzhiyun<x id="xid"> 1652*4882a593Smuzhiyun<z id="zida"/> 1653*4882a593Smuzhiyun<z id="zidab"/> 1654*4882a593Smuzhiyun<z id="zidac"/> 1655*4882a593Smuzhiyun</x> 1656*4882a593Smuzhiyun<y id="yid"> 1657*4882a593Smuzhiyun<z id="zidb"/> 1658*4882a593Smuzhiyun</y> 1659*4882a593Smuzhiyun<p lang="en" id="lang-en">English</p> 1660*4882a593Smuzhiyun<p lang="en-gb" id="lang-en-gb">English UK</p> 1661*4882a593Smuzhiyun<p lang="en-us" id="lang-en-us">English US</p> 1662*4882a593Smuzhiyun<p lang="fr" id="lang-fr">French</p> 1663*4882a593Smuzhiyun</div> 1664*4882a593Smuzhiyun 1665*4882a593Smuzhiyun<div id="footer"> 1666*4882a593Smuzhiyun</div> 1667*4882a593Smuzhiyun""" 1668*4882a593Smuzhiyun 1669*4882a593Smuzhiyun def setUp(self): 1670*4882a593Smuzhiyun self.soup = BeautifulSoup(self.HTML, 'html.parser') 1671*4882a593Smuzhiyun 1672*4882a593Smuzhiyun def assertSelects(self, selector, expected_ids): 1673*4882a593Smuzhiyun el_ids = [el['id'] for el in self.soup.select(selector)] 1674*4882a593Smuzhiyun el_ids.sort() 1675*4882a593Smuzhiyun expected_ids.sort() 1676*4882a593Smuzhiyun self.assertEqual(expected_ids, el_ids, 1677*4882a593Smuzhiyun "Selector %s, expected [%s], got [%s]" % ( 1678*4882a593Smuzhiyun selector, ', '.join(expected_ids), ', '.join(el_ids) 1679*4882a593Smuzhiyun ) 1680*4882a593Smuzhiyun ) 1681*4882a593Smuzhiyun 1682*4882a593Smuzhiyun assertSelect = assertSelects 1683*4882a593Smuzhiyun 1684*4882a593Smuzhiyun def assertSelectMultiple(self, *tests): 1685*4882a593Smuzhiyun for selector, expected_ids in tests: 1686*4882a593Smuzhiyun self.assertSelect(selector, expected_ids) 1687*4882a593Smuzhiyun 1688*4882a593Smuzhiyun def test_one_tag_one(self): 1689*4882a593Smuzhiyun els = self.soup.select('title') 1690*4882a593Smuzhiyun self.assertEqual(len(els), 1) 1691*4882a593Smuzhiyun self.assertEqual(els[0].name, 'title') 1692*4882a593Smuzhiyun self.assertEqual(els[0].contents, ['The title']) 1693*4882a593Smuzhiyun 1694*4882a593Smuzhiyun def test_one_tag_many(self): 1695*4882a593Smuzhiyun els = self.soup.select('div') 1696*4882a593Smuzhiyun self.assertEqual(len(els), 4) 1697*4882a593Smuzhiyun for div in els: 1698*4882a593Smuzhiyun self.assertEqual(div.name, 'div') 1699*4882a593Smuzhiyun 1700*4882a593Smuzhiyun el = self.soup.select_one('div') 1701*4882a593Smuzhiyun self.assertEqual('main', el['id']) 1702*4882a593Smuzhiyun 1703*4882a593Smuzhiyun def test_select_one_returns_none_if_no_match(self): 1704*4882a593Smuzhiyun match = self.soup.select_one('nonexistenttag') 1705*4882a593Smuzhiyun self.assertEqual(None, match) 1706*4882a593Smuzhiyun 1707*4882a593Smuzhiyun 1708*4882a593Smuzhiyun def test_tag_in_tag_one(self): 1709*4882a593Smuzhiyun els = self.soup.select('div div') 1710*4882a593Smuzhiyun self.assertSelects('div div', ['inner', 'data1']) 1711*4882a593Smuzhiyun 1712*4882a593Smuzhiyun def test_tag_in_tag_many(self): 1713*4882a593Smuzhiyun for selector in ('html div', 'html body div', 'body div'): 1714*4882a593Smuzhiyun self.assertSelects(selector, ['data1', 'main', 'inner', 'footer']) 1715*4882a593Smuzhiyun 1716*4882a593Smuzhiyun def test_tag_no_match(self): 1717*4882a593Smuzhiyun self.assertEqual(len(self.soup.select('del')), 0) 1718*4882a593Smuzhiyun 1719*4882a593Smuzhiyun def test_invalid_tag(self): 1720*4882a593Smuzhiyun self.assertRaises(ValueError, self.soup.select, 'tag%t') 1721*4882a593Smuzhiyun 1722*4882a593Smuzhiyun def test_select_dashed_tag_ids(self): 1723*4882a593Smuzhiyun self.assertSelects('custom-dashed-tag', ['dash1', 'dash2']) 1724*4882a593Smuzhiyun 1725*4882a593Smuzhiyun def test_select_dashed_by_id(self): 1726*4882a593Smuzhiyun dashed = self.soup.select('custom-dashed-tag[id=\"dash2\"]') 1727*4882a593Smuzhiyun self.assertEqual(dashed[0].name, 'custom-dashed-tag') 1728*4882a593Smuzhiyun self.assertEqual(dashed[0]['id'], 'dash2') 1729*4882a593Smuzhiyun 1730*4882a593Smuzhiyun def test_dashed_tag_text(self): 1731*4882a593Smuzhiyun self.assertEqual(self.soup.select('body > custom-dashed-tag')[0].text, 'Hello there.') 1732*4882a593Smuzhiyun 1733*4882a593Smuzhiyun def test_select_dashed_matches_find_all(self): 1734*4882a593Smuzhiyun self.assertEqual(self.soup.select('custom-dashed-tag'), self.soup.find_all('custom-dashed-tag')) 1735*4882a593Smuzhiyun 1736*4882a593Smuzhiyun def test_header_tags(self): 1737*4882a593Smuzhiyun self.assertSelectMultiple( 1738*4882a593Smuzhiyun ('h1', ['header1']), 1739*4882a593Smuzhiyun ('h2', ['header2', 'header3']), 1740*4882a593Smuzhiyun ) 1741*4882a593Smuzhiyun 1742*4882a593Smuzhiyun def test_class_one(self): 1743*4882a593Smuzhiyun for selector in ('.onep', 'p.onep', 'html p.onep'): 1744*4882a593Smuzhiyun els = self.soup.select(selector) 1745*4882a593Smuzhiyun self.assertEqual(len(els), 1) 1746*4882a593Smuzhiyun self.assertEqual(els[0].name, 'p') 1747*4882a593Smuzhiyun self.assertEqual(els[0]['class'], ['onep']) 1748*4882a593Smuzhiyun 1749*4882a593Smuzhiyun def test_class_mismatched_tag(self): 1750*4882a593Smuzhiyun els = self.soup.select('div.onep') 1751*4882a593Smuzhiyun self.assertEqual(len(els), 0) 1752*4882a593Smuzhiyun 1753*4882a593Smuzhiyun def test_one_id(self): 1754*4882a593Smuzhiyun for selector in ('div#inner', '#inner', 'div div#inner'): 1755*4882a593Smuzhiyun self.assertSelects(selector, ['inner']) 1756*4882a593Smuzhiyun 1757*4882a593Smuzhiyun def test_bad_id(self): 1758*4882a593Smuzhiyun els = self.soup.select('#doesnotexist') 1759*4882a593Smuzhiyun self.assertEqual(len(els), 0) 1760*4882a593Smuzhiyun 1761*4882a593Smuzhiyun def test_items_in_id(self): 1762*4882a593Smuzhiyun els = self.soup.select('div#inner p') 1763*4882a593Smuzhiyun self.assertEqual(len(els), 3) 1764*4882a593Smuzhiyun for el in els: 1765*4882a593Smuzhiyun self.assertEqual(el.name, 'p') 1766*4882a593Smuzhiyun self.assertEqual(els[1]['class'], ['onep']) 1767*4882a593Smuzhiyun self.assertFalse(els[0].has_attr('class')) 1768*4882a593Smuzhiyun 1769*4882a593Smuzhiyun def test_a_bunch_of_emptys(self): 1770*4882a593Smuzhiyun for selector in ('div#main del', 'div#main div.oops', 'div div#main'): 1771*4882a593Smuzhiyun self.assertEqual(len(self.soup.select(selector)), 0) 1772*4882a593Smuzhiyun 1773*4882a593Smuzhiyun def test_multi_class_support(self): 1774*4882a593Smuzhiyun for selector in ('.class1', 'p.class1', '.class2', 'p.class2', 1775*4882a593Smuzhiyun '.class3', 'p.class3', 'html p.class2', 'div#inner .class2'): 1776*4882a593Smuzhiyun self.assertSelects(selector, ['pmulti']) 1777*4882a593Smuzhiyun 1778*4882a593Smuzhiyun def test_multi_class_selection(self): 1779*4882a593Smuzhiyun for selector in ('.class1.class3', '.class3.class2', 1780*4882a593Smuzhiyun '.class1.class2.class3'): 1781*4882a593Smuzhiyun self.assertSelects(selector, ['pmulti']) 1782*4882a593Smuzhiyun 1783*4882a593Smuzhiyun def test_child_selector(self): 1784*4882a593Smuzhiyun self.assertSelects('.s1 > a', ['s1a1', 's1a2']) 1785*4882a593Smuzhiyun self.assertSelects('.s1 > a span', ['s1a2s1']) 1786*4882a593Smuzhiyun 1787*4882a593Smuzhiyun def test_child_selector_id(self): 1788*4882a593Smuzhiyun self.assertSelects('.s1 > a#s1a2 span', ['s1a2s1']) 1789*4882a593Smuzhiyun 1790*4882a593Smuzhiyun def test_attribute_equals(self): 1791*4882a593Smuzhiyun self.assertSelectMultiple( 1792*4882a593Smuzhiyun ('p[class="onep"]', ['p1']), 1793*4882a593Smuzhiyun ('p[id="p1"]', ['p1']), 1794*4882a593Smuzhiyun ('[class="onep"]', ['p1']), 1795*4882a593Smuzhiyun ('[id="p1"]', ['p1']), 1796*4882a593Smuzhiyun ('link[rel="stylesheet"]', ['l1']), 1797*4882a593Smuzhiyun ('link[type="text/css"]', ['l1']), 1798*4882a593Smuzhiyun ('link[href="blah.css"]', ['l1']), 1799*4882a593Smuzhiyun ('link[href="no-blah.css"]', []), 1800*4882a593Smuzhiyun ('[rel="stylesheet"]', ['l1']), 1801*4882a593Smuzhiyun ('[type="text/css"]', ['l1']), 1802*4882a593Smuzhiyun ('[href="blah.css"]', ['l1']), 1803*4882a593Smuzhiyun ('[href="no-blah.css"]', []), 1804*4882a593Smuzhiyun ('p[href="no-blah.css"]', []), 1805*4882a593Smuzhiyun ('[href="no-blah.css"]', []), 1806*4882a593Smuzhiyun ) 1807*4882a593Smuzhiyun 1808*4882a593Smuzhiyun def test_attribute_tilde(self): 1809*4882a593Smuzhiyun self.assertSelectMultiple( 1810*4882a593Smuzhiyun ('p[class~="class1"]', ['pmulti']), 1811*4882a593Smuzhiyun ('p[class~="class2"]', ['pmulti']), 1812*4882a593Smuzhiyun ('p[class~="class3"]', ['pmulti']), 1813*4882a593Smuzhiyun ('[class~="class1"]', ['pmulti']), 1814*4882a593Smuzhiyun ('[class~="class2"]', ['pmulti']), 1815*4882a593Smuzhiyun ('[class~="class3"]', ['pmulti']), 1816*4882a593Smuzhiyun ('a[rel~="friend"]', ['bob']), 1817*4882a593Smuzhiyun ('a[rel~="met"]', ['bob']), 1818*4882a593Smuzhiyun ('[rel~="friend"]', ['bob']), 1819*4882a593Smuzhiyun ('[rel~="met"]', ['bob']), 1820*4882a593Smuzhiyun ) 1821*4882a593Smuzhiyun 1822*4882a593Smuzhiyun def test_attribute_startswith(self): 1823*4882a593Smuzhiyun self.assertSelectMultiple( 1824*4882a593Smuzhiyun ('[rel^="style"]', ['l1']), 1825*4882a593Smuzhiyun ('link[rel^="style"]', ['l1']), 1826*4882a593Smuzhiyun ('notlink[rel^="notstyle"]', []), 1827*4882a593Smuzhiyun ('[rel^="notstyle"]', []), 1828*4882a593Smuzhiyun ('link[rel^="notstyle"]', []), 1829*4882a593Smuzhiyun ('link[href^="bla"]', ['l1']), 1830*4882a593Smuzhiyun ('a[href^="http://"]', ['bob', 'me']), 1831*4882a593Smuzhiyun ('[href^="http://"]', ['bob', 'me']), 1832*4882a593Smuzhiyun ('[id^="p"]', ['pmulti', 'p1']), 1833*4882a593Smuzhiyun ('[id^="m"]', ['me', 'main']), 1834*4882a593Smuzhiyun ('div[id^="m"]', ['main']), 1835*4882a593Smuzhiyun ('a[id^="m"]', ['me']), 1836*4882a593Smuzhiyun ('div[data-tag^="dashed"]', ['data1']) 1837*4882a593Smuzhiyun ) 1838*4882a593Smuzhiyun 1839*4882a593Smuzhiyun def test_attribute_endswith(self): 1840*4882a593Smuzhiyun self.assertSelectMultiple( 1841*4882a593Smuzhiyun ('[href$=".css"]', ['l1']), 1842*4882a593Smuzhiyun ('link[href$=".css"]', ['l1']), 1843*4882a593Smuzhiyun ('link[id$="1"]', ['l1']), 1844*4882a593Smuzhiyun ('[id$="1"]', ['data1', 'l1', 'p1', 'header1', 's1a1', 's2a1', 's1a2s1', 'dash1']), 1845*4882a593Smuzhiyun ('div[id$="1"]', ['data1']), 1846*4882a593Smuzhiyun ('[id$="noending"]', []), 1847*4882a593Smuzhiyun ) 1848*4882a593Smuzhiyun 1849*4882a593Smuzhiyun def test_attribute_contains(self): 1850*4882a593Smuzhiyun self.assertSelectMultiple( 1851*4882a593Smuzhiyun # From test_attribute_startswith 1852*4882a593Smuzhiyun ('[rel*="style"]', ['l1']), 1853*4882a593Smuzhiyun ('link[rel*="style"]', ['l1']), 1854*4882a593Smuzhiyun ('notlink[rel*="notstyle"]', []), 1855*4882a593Smuzhiyun ('[rel*="notstyle"]', []), 1856*4882a593Smuzhiyun ('link[rel*="notstyle"]', []), 1857*4882a593Smuzhiyun ('link[href*="bla"]', ['l1']), 1858*4882a593Smuzhiyun ('[href*="http://"]', ['bob', 'me']), 1859*4882a593Smuzhiyun ('[id*="p"]', ['pmulti', 'p1']), 1860*4882a593Smuzhiyun ('div[id*="m"]', ['main']), 1861*4882a593Smuzhiyun ('a[id*="m"]', ['me']), 1862*4882a593Smuzhiyun # From test_attribute_endswith 1863*4882a593Smuzhiyun ('[href*=".css"]', ['l1']), 1864*4882a593Smuzhiyun ('link[href*=".css"]', ['l1']), 1865*4882a593Smuzhiyun ('link[id*="1"]', ['l1']), 1866*4882a593Smuzhiyun ('[id*="1"]', ['data1', 'l1', 'p1', 'header1', 's1a1', 's1a2', 's2a1', 's1a2s1', 'dash1']), 1867*4882a593Smuzhiyun ('div[id*="1"]', ['data1']), 1868*4882a593Smuzhiyun ('[id*="noending"]', []), 1869*4882a593Smuzhiyun # New for this test 1870*4882a593Smuzhiyun ('[href*="."]', ['bob', 'me', 'l1']), 1871*4882a593Smuzhiyun ('a[href*="."]', ['bob', 'me']), 1872*4882a593Smuzhiyun ('link[href*="."]', ['l1']), 1873*4882a593Smuzhiyun ('div[id*="n"]', ['main', 'inner']), 1874*4882a593Smuzhiyun ('div[id*="nn"]', ['inner']), 1875*4882a593Smuzhiyun ('div[data-tag*="edval"]', ['data1']) 1876*4882a593Smuzhiyun ) 1877*4882a593Smuzhiyun 1878*4882a593Smuzhiyun def test_attribute_exact_or_hypen(self): 1879*4882a593Smuzhiyun self.assertSelectMultiple( 1880*4882a593Smuzhiyun ('p[lang|="en"]', ['lang-en', 'lang-en-gb', 'lang-en-us']), 1881*4882a593Smuzhiyun ('[lang|="en"]', ['lang-en', 'lang-en-gb', 'lang-en-us']), 1882*4882a593Smuzhiyun ('p[lang|="fr"]', ['lang-fr']), 1883*4882a593Smuzhiyun ('p[lang|="gb"]', []), 1884*4882a593Smuzhiyun ) 1885*4882a593Smuzhiyun 1886*4882a593Smuzhiyun def test_attribute_exists(self): 1887*4882a593Smuzhiyun self.assertSelectMultiple( 1888*4882a593Smuzhiyun ('[rel]', ['l1', 'bob', 'me']), 1889*4882a593Smuzhiyun ('link[rel]', ['l1']), 1890*4882a593Smuzhiyun ('a[rel]', ['bob', 'me']), 1891*4882a593Smuzhiyun ('[lang]', ['lang-en', 'lang-en-gb', 'lang-en-us', 'lang-fr']), 1892*4882a593Smuzhiyun ('p[class]', ['p1', 'pmulti']), 1893*4882a593Smuzhiyun ('[blah]', []), 1894*4882a593Smuzhiyun ('p[blah]', []), 1895*4882a593Smuzhiyun ('div[data-tag]', ['data1']) 1896*4882a593Smuzhiyun ) 1897*4882a593Smuzhiyun 1898*4882a593Smuzhiyun def test_unsupported_pseudoclass(self): 1899*4882a593Smuzhiyun self.assertRaises( 1900*4882a593Smuzhiyun NotImplementedError, self.soup.select, "a:no-such-pseudoclass") 1901*4882a593Smuzhiyun 1902*4882a593Smuzhiyun self.assertRaises( 1903*4882a593Smuzhiyun NotImplementedError, self.soup.select, "a:nth-of-type(a)") 1904*4882a593Smuzhiyun 1905*4882a593Smuzhiyun 1906*4882a593Smuzhiyun def test_nth_of_type(self): 1907*4882a593Smuzhiyun # Try to select first paragraph 1908*4882a593Smuzhiyun els = self.soup.select('div#inner p:nth-of-type(1)') 1909*4882a593Smuzhiyun self.assertEqual(len(els), 1) 1910*4882a593Smuzhiyun self.assertEqual(els[0].string, 'Some text') 1911*4882a593Smuzhiyun 1912*4882a593Smuzhiyun # Try to select third paragraph 1913*4882a593Smuzhiyun els = self.soup.select('div#inner p:nth-of-type(3)') 1914*4882a593Smuzhiyun self.assertEqual(len(els), 1) 1915*4882a593Smuzhiyun self.assertEqual(els[0].string, 'Another') 1916*4882a593Smuzhiyun 1917*4882a593Smuzhiyun # Try to select (non-existent!) fourth paragraph 1918*4882a593Smuzhiyun els = self.soup.select('div#inner p:nth-of-type(4)') 1919*4882a593Smuzhiyun self.assertEqual(len(els), 0) 1920*4882a593Smuzhiyun 1921*4882a593Smuzhiyun # Pass in an invalid value. 1922*4882a593Smuzhiyun self.assertRaises( 1923*4882a593Smuzhiyun ValueError, self.soup.select, 'div p:nth-of-type(0)') 1924*4882a593Smuzhiyun 1925*4882a593Smuzhiyun def test_nth_of_type_direct_descendant(self): 1926*4882a593Smuzhiyun els = self.soup.select('div#inner > p:nth-of-type(1)') 1927*4882a593Smuzhiyun self.assertEqual(len(els), 1) 1928*4882a593Smuzhiyun self.assertEqual(els[0].string, 'Some text') 1929*4882a593Smuzhiyun 1930*4882a593Smuzhiyun def test_id_child_selector_nth_of_type(self): 1931*4882a593Smuzhiyun self.assertSelects('#inner > p:nth-of-type(2)', ['p1']) 1932*4882a593Smuzhiyun 1933*4882a593Smuzhiyun def test_select_on_element(self): 1934*4882a593Smuzhiyun # Other tests operate on the tree; this operates on an element 1935*4882a593Smuzhiyun # within the tree. 1936*4882a593Smuzhiyun inner = self.soup.find("div", id="main") 1937*4882a593Smuzhiyun selected = inner.select("div") 1938*4882a593Smuzhiyun # The <div id="inner"> tag was selected. The <div id="footer"> 1939*4882a593Smuzhiyun # tag was not. 1940*4882a593Smuzhiyun self.assertSelectsIDs(selected, ['inner', 'data1']) 1941*4882a593Smuzhiyun 1942*4882a593Smuzhiyun def test_overspecified_child_id(self): 1943*4882a593Smuzhiyun self.assertSelects(".fancy #inner", ['inner']) 1944*4882a593Smuzhiyun self.assertSelects(".normal #inner", []) 1945*4882a593Smuzhiyun 1946*4882a593Smuzhiyun def test_adjacent_sibling_selector(self): 1947*4882a593Smuzhiyun self.assertSelects('#p1 + h2', ['header2']) 1948*4882a593Smuzhiyun self.assertSelects('#p1 + h2 + p', ['pmulti']) 1949*4882a593Smuzhiyun self.assertSelects('#p1 + #header2 + .class1', ['pmulti']) 1950*4882a593Smuzhiyun self.assertEqual([], self.soup.select('#p1 + p')) 1951*4882a593Smuzhiyun 1952*4882a593Smuzhiyun def test_general_sibling_selector(self): 1953*4882a593Smuzhiyun self.assertSelects('#p1 ~ h2', ['header2', 'header3']) 1954*4882a593Smuzhiyun self.assertSelects('#p1 ~ #header2', ['header2']) 1955*4882a593Smuzhiyun self.assertSelects('#p1 ~ h2 + a', ['me']) 1956*4882a593Smuzhiyun self.assertSelects('#p1 ~ h2 + [rel="me"]', ['me']) 1957*4882a593Smuzhiyun self.assertEqual([], self.soup.select('#inner ~ h2')) 1958*4882a593Smuzhiyun 1959*4882a593Smuzhiyun def test_dangling_combinator(self): 1960*4882a593Smuzhiyun self.assertRaises(ValueError, self.soup.select, 'h1 >') 1961*4882a593Smuzhiyun 1962*4882a593Smuzhiyun def test_sibling_combinator_wont_select_same_tag_twice(self): 1963*4882a593Smuzhiyun self.assertSelects('p[lang] ~ p', ['lang-en-gb', 'lang-en-us', 'lang-fr']) 1964*4882a593Smuzhiyun 1965*4882a593Smuzhiyun # Test the selector grouping operator (the comma) 1966*4882a593Smuzhiyun def test_multiple_select(self): 1967*4882a593Smuzhiyun self.assertSelects('x, y', ['xid', 'yid']) 1968*4882a593Smuzhiyun 1969*4882a593Smuzhiyun def test_multiple_select_with_no_space(self): 1970*4882a593Smuzhiyun self.assertSelects('x,y', ['xid', 'yid']) 1971*4882a593Smuzhiyun 1972*4882a593Smuzhiyun def test_multiple_select_with_more_space(self): 1973*4882a593Smuzhiyun self.assertSelects('x, y', ['xid', 'yid']) 1974*4882a593Smuzhiyun 1975*4882a593Smuzhiyun def test_multiple_select_duplicated(self): 1976*4882a593Smuzhiyun self.assertSelects('x, x', ['xid']) 1977*4882a593Smuzhiyun 1978*4882a593Smuzhiyun def test_multiple_select_sibling(self): 1979*4882a593Smuzhiyun self.assertSelects('x, y ~ p[lang=fr]', ['xid', 'lang-fr']) 1980*4882a593Smuzhiyun 1981*4882a593Smuzhiyun def test_multiple_select_tag_and_direct_descendant(self): 1982*4882a593Smuzhiyun self.assertSelects('x, y > z', ['xid', 'zidb']) 1983*4882a593Smuzhiyun 1984*4882a593Smuzhiyun def test_multiple_select_direct_descendant_and_tags(self): 1985*4882a593Smuzhiyun self.assertSelects('div > x, y, z', ['xid', 'yid', 'zida', 'zidb', 'zidab', 'zidac']) 1986*4882a593Smuzhiyun 1987*4882a593Smuzhiyun def test_multiple_select_indirect_descendant(self): 1988*4882a593Smuzhiyun self.assertSelects('div x,y, z', ['xid', 'yid', 'zida', 'zidb', 'zidab', 'zidac']) 1989*4882a593Smuzhiyun 1990*4882a593Smuzhiyun def test_invalid_multiple_select(self): 1991*4882a593Smuzhiyun self.assertRaises(ValueError, self.soup.select, ',x, y') 1992*4882a593Smuzhiyun self.assertRaises(ValueError, self.soup.select, 'x,,y') 1993*4882a593Smuzhiyun 1994*4882a593Smuzhiyun def test_multiple_select_attrs(self): 1995*4882a593Smuzhiyun self.assertSelects('p[lang=en], p[lang=en-gb]', ['lang-en', 'lang-en-gb']) 1996*4882a593Smuzhiyun 1997*4882a593Smuzhiyun def test_multiple_select_ids(self): 1998*4882a593Smuzhiyun self.assertSelects('x, y > z[id=zida], z[id=zidab], z[id=zidb]', ['xid', 'zidb', 'zidab']) 1999*4882a593Smuzhiyun 2000*4882a593Smuzhiyun def test_multiple_select_nested(self): 2001*4882a593Smuzhiyun self.assertSelects('body > div > x, y > z', ['xid', 'zidb']) 2002*4882a593Smuzhiyun 2003*4882a593Smuzhiyun 2004*4882a593Smuzhiyun 2005