1from __future__ import absolute_import, division, unicode_literals 2 3from six import PY2, text_type, unichr 4 5import io 6 7from . import support # noqa 8 9from html5lib.constants import namespaces, tokenTypes 10from html5lib import parse, parseFragment, HTMLParser 11 12 13# tests that aren't autogenerated from text files 14def test_assertDoctypeCloneable(): 15 doc = parse('<!DOCTYPE HTML>', treebuilder="dom") 16 assert doc.cloneNode(True) is not None 17 18 19def test_line_counter(): 20 # http://groups.google.com/group/html5lib-discuss/browse_frm/thread/f4f00e4a2f26d5c0 21 assert parse("<pre>\nx\n>\n</pre>") is not None 22 23 24def test_namespace_html_elements_0_dom(): 25 doc = parse("<html></html>", 26 treebuilder="dom", 27 namespaceHTMLElements=True) 28 assert doc.childNodes[0].namespaceURI == namespaces["html"] 29 30 31def test_namespace_html_elements_1_dom(): 32 doc = parse("<html></html>", 33 treebuilder="dom", 34 namespaceHTMLElements=False) 35 assert doc.childNodes[0].namespaceURI is None 36 37 38def test_namespace_html_elements_0_etree(): 39 doc = parse("<html></html>", 40 treebuilder="etree", 41 namespaceHTMLElements=True) 42 assert doc.tag == "{%s}html" % (namespaces["html"],) 43 44 45def test_namespace_html_elements_1_etree(): 46 doc = parse("<html></html>", 47 treebuilder="etree", 48 namespaceHTMLElements=False) 49 assert doc.tag == "html" 50 51 52def test_unicode_file(): 53 assert parse(io.StringIO("a")) is not None 54 55 56def test_maintain_attribute_order(): 57 # This is here because we impl it in parser and not tokenizer 58 p = HTMLParser() 59 # generate loads to maximize the chance a hash-based mutation will occur 60 attrs = [(unichr(x), i) for i, x in enumerate(range(ord('a'), ord('z')))] 61 token = {'name': 'html', 62 'selfClosing': False, 63 'selfClosingAcknowledged': False, 64 'type': tokenTypes["StartTag"], 65 'data': attrs} 66 out = p.normalizeToken(token) 67 attr_order = list(out["data"].keys()) 68 assert attr_order == [x for x, i in attrs] 69 70 71def test_duplicate_attribute(): 72 # This is here because we impl it in parser and not tokenizer 73 doc = parse('<p class=a class=b>') 74 el = doc[1][0] 75 assert el.get("class") == "a" 76 77 78def test_maintain_duplicate_attribute_order(): 79 # This is here because we impl it in parser and not tokenizer 80 p = HTMLParser() 81 attrs = [(unichr(x), i) for i, x in enumerate(range(ord('a'), ord('z')))] 82 token = {'name': 'html', 83 'selfClosing': False, 84 'selfClosingAcknowledged': False, 85 'type': tokenTypes["StartTag"], 86 'data': attrs + [('a', len(attrs))]} 87 out = p.normalizeToken(token) 88 attr_order = list(out["data"].keys()) 89 assert attr_order == [x for x, i in attrs] 90 91 92def test_debug_log(): 93 parser = HTMLParser(debug=True) 94 parser.parse("<!doctype html><title>a</title><p>b<script>c</script>d</p>e") 95 96 expected = [('dataState', 'InitialPhase', 'InitialPhase', 'processDoctype', {'type': 'Doctype'}), 97 ('dataState', 'BeforeHtmlPhase', 'BeforeHtmlPhase', 'processStartTag', {'name': 'title', 'type': 'StartTag'}), 98 ('dataState', 'BeforeHeadPhase', 'BeforeHeadPhase', 'processStartTag', {'name': 'title', 'type': 'StartTag'}), 99 ('dataState', 'InHeadPhase', 'InHeadPhase', 'processStartTag', {'name': 'title', 'type': 'StartTag'}), 100 ('rcdataState', 'TextPhase', 'TextPhase', 'processCharacters', {'type': 'Characters'}), 101 ('dataState', 'TextPhase', 'TextPhase', 'processEndTag', {'name': 'title', 'type': 'EndTag'}), 102 ('dataState', 'InHeadPhase', 'InHeadPhase', 'processStartTag', {'name': 'p', 'type': 'StartTag'}), 103 ('dataState', 'AfterHeadPhase', 'AfterHeadPhase', 'processStartTag', {'name': 'p', 'type': 'StartTag'}), 104 ('dataState', 'InBodyPhase', 'InBodyPhase', 'processStartTag', {'name': 'p', 'type': 'StartTag'}), 105 ('dataState', 'InBodyPhase', 'InBodyPhase', 'processCharacters', {'type': 'Characters'}), 106 ('dataState', 'InBodyPhase', 'InBodyPhase', 'processStartTag', {'name': 'script', 'type': 'StartTag'}), 107 ('dataState', 'InBodyPhase', 'InHeadPhase', 'processStartTag', {'name': 'script', 'type': 'StartTag'}), 108 ('scriptDataState', 'TextPhase', 'TextPhase', 'processCharacters', {'type': 'Characters'}), 109 ('dataState', 'TextPhase', 'TextPhase', 'processEndTag', {'name': 'script', 'type': 'EndTag'}), 110 ('dataState', 'InBodyPhase', 'InBodyPhase', 'processCharacters', {'type': 'Characters'}), 111 ('dataState', 'InBodyPhase', 'InBodyPhase', 'processEndTag', {'name': 'p', 'type': 'EndTag'}), 112 ('dataState', 'InBodyPhase', 'InBodyPhase', 'processCharacters', {'type': 'Characters'})] 113 114 if PY2: 115 for i, log in enumerate(expected): 116 log = [x.encode("ascii") if isinstance(x, text_type) else x for x in log] 117 expected[i] = tuple(log) 118 119 assert parser.log == expected 120 121 122def test_no_duplicate_clone(): 123 frag = parseFragment("<b><em><foo><foob><fooc><aside></b></em>") 124 assert len(frag) == 2 125 126 127def test_self_closing_col(): 128 parser = HTMLParser() 129 parser.parseFragment('<table><colgroup><col /></colgroup></table>') 130 assert not parser.errors 131