2001-05-18 12:32:59 -03:00
|
|
|
"""Tests for HTMLParser.py."""
|
|
|
|
|
|
2008-05-20 04:58:42 -03:00
|
|
|
import HTMLParser
|
2001-08-20 18:24:19 -03:00
|
|
|
import pprint
|
2001-05-18 12:32:59 -03:00
|
|
|
import unittest
|
2002-07-23 16:04:11 -03:00
|
|
|
from test import test_support
|
2001-05-18 12:32:59 -03:00
|
|
|
|
|
|
|
|
|
2008-05-20 04:58:42 -03:00
|
|
|
class EventCollector(HTMLParser.HTMLParser):
|
2001-05-18 12:32:59 -03:00
|
|
|
|
|
|
|
|
def __init__(self):
|
|
|
|
|
self.events = []
|
|
|
|
|
self.append = self.events.append
|
2008-05-20 04:58:42 -03:00
|
|
|
HTMLParser.HTMLParser.__init__(self)
|
2001-05-18 12:32:59 -03:00
|
|
|
|
|
|
|
|
def get_events(self):
|
|
|
|
|
# Normalize the list of events so that buffer artefacts don't
|
|
|
|
|
# separate runs of contiguous characters.
|
|
|
|
|
L = []
|
|
|
|
|
prevtype = None
|
|
|
|
|
for event in self.events:
|
|
|
|
|
type = event[0]
|
|
|
|
|
if type == prevtype == "data":
|
|
|
|
|
L[-1] = ("data", L[-1][1] + event[1])
|
|
|
|
|
else:
|
|
|
|
|
L.append(event)
|
|
|
|
|
prevtype = type
|
|
|
|
|
self.events = L
|
|
|
|
|
return L
|
|
|
|
|
|
|
|
|
|
# structure markup
|
|
|
|
|
|
|
|
|
|
def handle_starttag(self, tag, attrs):
|
|
|
|
|
self.append(("starttag", tag, attrs))
|
|
|
|
|
|
|
|
|
|
def handle_startendtag(self, tag, attrs):
|
|
|
|
|
self.append(("startendtag", tag, attrs))
|
|
|
|
|
|
|
|
|
|
def handle_endtag(self, tag):
|
|
|
|
|
self.append(("endtag", tag))
|
|
|
|
|
|
|
|
|
|
# all other markup
|
|
|
|
|
|
|
|
|
|
def handle_comment(self, data):
|
|
|
|
|
self.append(("comment", data))
|
|
|
|
|
|
|
|
|
|
def handle_charref(self, data):
|
|
|
|
|
self.append(("charref", data))
|
|
|
|
|
|
|
|
|
|
def handle_data(self, data):
|
|
|
|
|
self.append(("data", data))
|
|
|
|
|
|
|
|
|
|
def handle_decl(self, data):
|
|
|
|
|
self.append(("decl", data))
|
|
|
|
|
|
|
|
|
|
def handle_entityref(self, data):
|
|
|
|
|
self.append(("entityref", data))
|
|
|
|
|
|
|
|
|
|
def handle_pi(self, data):
|
|
|
|
|
self.append(("pi", data))
|
|
|
|
|
|
2001-09-04 12:13:04 -03:00
|
|
|
def unknown_decl(self, decl):
|
|
|
|
|
self.append(("unknown decl", decl))
|
|
|
|
|
|
2001-05-18 12:32:59 -03:00
|
|
|
|
|
|
|
|
class EventCollectorExtra(EventCollector):
|
|
|
|
|
|
|
|
|
|
def handle_starttag(self, tag, attrs):
|
|
|
|
|
EventCollector.handle_starttag(self, tag, attrs)
|
|
|
|
|
self.append(("starttag_text", self.get_starttag_text()))
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
class TestCaseBase(unittest.TestCase):
|
|
|
|
|
|
2001-09-04 12:13:04 -03:00
|
|
|
def _run_check(self, source, expected_events, collector=EventCollector):
|
2001-05-18 12:32:59 -03:00
|
|
|
parser = collector()
|
|
|
|
|
for s in source:
|
|
|
|
|
parser.feed(s)
|
|
|
|
|
parser.close()
|
2001-08-20 18:24:19 -03:00
|
|
|
events = parser.get_events()
|
2001-09-04 12:13:04 -03:00
|
|
|
if events != expected_events:
|
|
|
|
|
self.fail("received events did not match expected events\n"
|
|
|
|
|
"Expected:\n" + pprint.pformat(expected_events) +
|
|
|
|
|
"\nReceived:\n" + pprint.pformat(events))
|
2001-05-18 12:32:59 -03:00
|
|
|
|
|
|
|
|
def _run_check_extra(self, source, events):
|
|
|
|
|
self._run_check(source, events, EventCollectorExtra)
|
|
|
|
|
|
|
|
|
|
def _parse_error(self, source):
|
|
|
|
|
def parse(source=source):
|
2008-05-20 04:58:42 -03:00
|
|
|
parser = HTMLParser.HTMLParser()
|
2001-05-18 12:32:59 -03:00
|
|
|
parser.feed(source)
|
|
|
|
|
parser.close()
|
2008-05-20 04:58:42 -03:00
|
|
|
self.assertRaises(HTMLParser.HTMLParseError, parse)
|
2001-05-18 12:32:59 -03:00
|
|
|
|
|
|
|
|
|
|
|
|
|
class HTMLParserTestCase(TestCaseBase):
|
|
|
|
|
|
2001-08-03 16:53:01 -03:00
|
|
|
def test_processing_instruction_only(self):
|
2001-05-18 12:32:59 -03:00
|
|
|
self._run_check("<?processing instruction>", [
|
|
|
|
|
("pi", "processing instruction"),
|
|
|
|
|
])
|
2003-04-17 19:19:26 -03:00
|
|
|
self._run_check("<?processing instruction ?>", [
|
|
|
|
|
("pi", "processing instruction ?"),
|
|
|
|
|
])
|
2001-05-18 12:32:59 -03:00
|
|
|
|
2001-08-03 16:53:01 -03:00
|
|
|
def test_simple_html(self):
|
2001-05-18 12:32:59 -03:00
|
|
|
self._run_check("""
|
|
|
|
|
<!DOCTYPE html PUBLIC 'foo'>
|
|
|
|
|
<HTML>&entity; 
|
|
|
|
|
<!--comment1a
|
|
|
|
|
-></foo><bar><<?pi?></foo<bar
|
|
|
|
|
comment1b-->
|
|
|
|
|
<Img sRc='Bar' isMAP>sample
|
|
|
|
|
text
|
2001-08-03 16:53:01 -03:00
|
|
|
“
|
2006-03-09 09:27:14 -04:00
|
|
|
<!--comment2a-- --comment2b--><!>
|
2001-05-18 12:32:59 -03:00
|
|
|
</Html>
|
|
|
|
|
""", [
|
|
|
|
|
("data", "\n"),
|
|
|
|
|
("decl", "DOCTYPE html PUBLIC 'foo'"),
|
|
|
|
|
("data", "\n"),
|
|
|
|
|
("starttag", "html", []),
|
|
|
|
|
("entityref", "entity"),
|
|
|
|
|
("charref", "32"),
|
|
|
|
|
("data", "\n"),
|
|
|
|
|
("comment", "comment1a\n-></foo><bar><<?pi?></foo<bar\ncomment1b"),
|
|
|
|
|
("data", "\n"),
|
|
|
|
|
("starttag", "img", [("src", "Bar"), ("ismap", None)]),
|
|
|
|
|
("data", "sample\ntext\n"),
|
2001-08-03 16:53:01 -03:00
|
|
|
("charref", "x201C"),
|
|
|
|
|
("data", "\n"),
|
2001-05-18 12:32:59 -03:00
|
|
|
("comment", "comment2a-- --comment2b"),
|
|
|
|
|
("data", "\n"),
|
|
|
|
|
("endtag", "html"),
|
|
|
|
|
("data", "\n"),
|
|
|
|
|
])
|
|
|
|
|
|
2001-12-03 12:44:09 -04:00
|
|
|
def test_unclosed_entityref(self):
|
|
|
|
|
self._run_check("&entityref foo", [
|
|
|
|
|
("entityref", "entityref"),
|
|
|
|
|
("data", " foo"),
|
|
|
|
|
])
|
|
|
|
|
|
2001-08-20 18:24:19 -03:00
|
|
|
def test_doctype_decl(self):
|
|
|
|
|
inside = """\
|
|
|
|
|
DOCTYPE html [
|
|
|
|
|
<!ELEMENT html - O EMPTY>
|
|
|
|
|
<!ATTLIST html
|
2001-09-04 12:13:04 -03:00
|
|
|
version CDATA #IMPLIED
|
|
|
|
|
profile CDATA 'DublinCore'>
|
|
|
|
|
<!NOTATION datatype SYSTEM 'http://xml.python.org/notations/python-module'>
|
|
|
|
|
<!ENTITY myEntity 'internal parsed entity'>
|
|
|
|
|
<!ENTITY anEntity SYSTEM 'http://xml.python.org/entities/something.xml'>
|
|
|
|
|
<!ENTITY % paramEntity 'name|name|name'>
|
|
|
|
|
%paramEntity;
|
2001-08-20 18:24:19 -03:00
|
|
|
<!-- comment -->
|
|
|
|
|
]"""
|
|
|
|
|
self._run_check("<!%s>" % inside, [
|
|
|
|
|
("decl", inside),
|
|
|
|
|
])
|
|
|
|
|
|
2001-08-03 16:53:01 -03:00
|
|
|
def test_bad_nesting(self):
|
|
|
|
|
# Strangely, this *is* supposed to test that overlapping
|
|
|
|
|
# elements are allowed. HTMLParser is more geared toward
|
|
|
|
|
# lexing the input that parsing the structure.
|
2001-05-18 12:32:59 -03:00
|
|
|
self._run_check("<a><b></a></b>", [
|
|
|
|
|
("starttag", "a", []),
|
|
|
|
|
("starttag", "b", []),
|
|
|
|
|
("endtag", "a"),
|
|
|
|
|
("endtag", "b"),
|
|
|
|
|
])
|
|
|
|
|
|
2001-08-20 18:24:19 -03:00
|
|
|
def test_bare_ampersands(self):
|
|
|
|
|
self._run_check("this text & contains & ampersands &", [
|
|
|
|
|
("data", "this text & contains & ampersands &"),
|
|
|
|
|
])
|
|
|
|
|
|
|
|
|
|
def test_bare_pointy_brackets(self):
|
|
|
|
|
self._run_check("this < text > contains < bare>pointy< brackets", [
|
|
|
|
|
("data", "this < text > contains < bare>pointy< brackets"),
|
|
|
|
|
])
|
|
|
|
|
|
2001-08-03 16:53:01 -03:00
|
|
|
def test_attr_syntax(self):
|
2001-05-18 12:32:59 -03:00
|
|
|
output = [
|
|
|
|
|
("starttag", "a", [("b", "v"), ("c", "v"), ("d", "v"), ("e", None)])
|
|
|
|
|
]
|
|
|
|
|
self._run_check("""<a b='v' c="v" d=v e>""", output)
|
|
|
|
|
self._run_check("""<a b = 'v' c = "v" d = v e>""", output)
|
|
|
|
|
self._run_check("""<a\nb\n=\n'v'\nc\n=\n"v"\nd\n=\nv\ne>""", output)
|
|
|
|
|
self._run_check("""<a\tb\t=\t'v'\tc\t=\t"v"\td\t=\tv\te>""", output)
|
|
|
|
|
|
2001-08-03 16:53:01 -03:00
|
|
|
def test_attr_values(self):
|
2001-05-18 12:32:59 -03:00
|
|
|
self._run_check("""<a b='xxx\n\txxx' c="yyy\t\nyyy" d='\txyz\n'>""",
|
|
|
|
|
[("starttag", "a", [("b", "xxx\n\txxx"),
|
|
|
|
|
("c", "yyy\t\nyyy"),
|
|
|
|
|
("d", "\txyz\n")])
|
|
|
|
|
])
|
|
|
|
|
self._run_check("""<a b='' c="">""", [
|
|
|
|
|
("starttag", "a", [("b", ""), ("c", "")]),
|
|
|
|
|
])
|
2003-03-14 12:21:57 -04:00
|
|
|
# Regression test for SF patch #669683.
|
|
|
|
|
self._run_check("<e a=rgb(1,2,3)>", [
|
|
|
|
|
("starttag", "e", [("a", "rgb(1,2,3)")]),
|
|
|
|
|
])
|
2004-07-08 01:22:35 -03:00
|
|
|
# Regression test for SF bug #921657.
|
2004-06-05 12:31:45 -03:00
|
|
|
self._run_check("<a href=mailto:xyz@example.com>", [
|
|
|
|
|
("starttag", "a", [("href", "mailto:xyz@example.com")]),
|
|
|
|
|
])
|
2001-05-18 12:32:59 -03:00
|
|
|
|
2011-04-05 14:40:52 -03:00
|
|
|
def test_attr_nonascii(self):
|
|
|
|
|
# see issue 7311
|
|
|
|
|
self._run_check(u"<img src=/foo/bar.png alt=\u4e2d\u6587>", [
|
|
|
|
|
("starttag", "img", [("src", "/foo/bar.png"),
|
|
|
|
|
("alt", u"\u4e2d\u6587")]),
|
|
|
|
|
])
|
|
|
|
|
self._run_check(u"<a title='\u30c6\u30b9\u30c8' "
|
|
|
|
|
u"href='\u30c6\u30b9\u30c8.html'>", [
|
|
|
|
|
("starttag", "a", [("title", u"\u30c6\u30b9\u30c8"),
|
|
|
|
|
("href", u"\u30c6\u30b9\u30c8.html")]),
|
|
|
|
|
])
|
|
|
|
|
self._run_check(u'<a title="\u30c6\u30b9\u30c8" '
|
|
|
|
|
u'href="\u30c6\u30b9\u30c8.html">', [
|
|
|
|
|
("starttag", "a", [("title", u"\u30c6\u30b9\u30c8"),
|
|
|
|
|
("href", u"\u30c6\u30b9\u30c8.html")]),
|
|
|
|
|
])
|
|
|
|
|
|
2001-08-03 16:53:01 -03:00
|
|
|
def test_attr_entity_replacement(self):
|
2001-05-18 12:32:59 -03:00
|
|
|
self._run_check("""<a b='&><"''>""", [
|
|
|
|
|
("starttag", "a", [("b", "&><\"'")]),
|
|
|
|
|
])
|
|
|
|
|
|
2001-08-03 16:53:01 -03:00
|
|
|
def test_attr_funky_names(self):
|
2001-05-18 12:32:59 -03:00
|
|
|
self._run_check("""<a a.b='v' c:d=v e-f=v>""", [
|
|
|
|
|
("starttag", "a", [("a.b", "v"), ("c:d", "v"), ("e-f", "v")]),
|
|
|
|
|
])
|
|
|
|
|
|
2001-09-04 12:13:04 -03:00
|
|
|
def test_illegal_declarations(self):
|
2001-09-04 13:26:03 -03:00
|
|
|
self._parse_error('<!spacer type="block" height="25">')
|
2001-09-04 12:13:04 -03:00
|
|
|
|
2001-08-03 16:53:01 -03:00
|
|
|
def test_starttag_end_boundary(self):
|
2001-05-18 12:32:59 -03:00
|
|
|
self._run_check("""<a b='<'>""", [("starttag", "a", [("b", "<")])])
|
|
|
|
|
self._run_check("""<a b='>'>""", [("starttag", "a", [("b", ">")])])
|
|
|
|
|
|
2001-08-03 16:53:01 -03:00
|
|
|
def test_buffer_artefacts(self):
|
2001-05-18 12:32:59 -03:00
|
|
|
output = [("starttag", "a", [("b", "<")])]
|
|
|
|
|
self._run_check(["<a b='<'>"], output)
|
|
|
|
|
self._run_check(["<a ", "b='<'>"], output)
|
|
|
|
|
self._run_check(["<a b", "='<'>"], output)
|
|
|
|
|
self._run_check(["<a b=", "'<'>"], output)
|
|
|
|
|
self._run_check(["<a b='<", "'>"], output)
|
|
|
|
|
self._run_check(["<a b='<'", ">"], output)
|
|
|
|
|
|
|
|
|
|
output = [("starttag", "a", [("b", ">")])]
|
|
|
|
|
self._run_check(["<a b='>'>"], output)
|
|
|
|
|
self._run_check(["<a ", "b='>'>"], output)
|
|
|
|
|
self._run_check(["<a b", "='>'>"], output)
|
|
|
|
|
self._run_check(["<a b=", "'>'>"], output)
|
|
|
|
|
self._run_check(["<a b='>", "'>"], output)
|
|
|
|
|
self._run_check(["<a b='>'", ">"], output)
|
|
|
|
|
|
2004-09-08 19:57:01 -03:00
|
|
|
output = [("comment", "abc")]
|
|
|
|
|
self._run_check(["", "<!--abc-->"], output)
|
|
|
|
|
self._run_check(["<", "!--abc-->"], output)
|
|
|
|
|
self._run_check(["<!", "--abc-->"], output)
|
|
|
|
|
self._run_check(["<!-", "-abc-->"], output)
|
|
|
|
|
self._run_check(["<!--", "abc-->"], output)
|
|
|
|
|
self._run_check(["<!--a", "bc-->"], output)
|
|
|
|
|
self._run_check(["<!--ab", "c-->"], output)
|
|
|
|
|
self._run_check(["<!--abc", "-->"], output)
|
|
|
|
|
self._run_check(["<!--abc-", "->"], output)
|
|
|
|
|
self._run_check(["<!--abc--", ">"], output)
|
|
|
|
|
self._run_check(["<!--abc-->", ""], output)
|
|
|
|
|
|
2001-08-03 16:53:01 -03:00
|
|
|
def test_starttag_junk_chars(self):
|
2001-05-18 12:32:59 -03:00
|
|
|
self._parse_error("</>")
|
|
|
|
|
self._parse_error("</$>")
|
|
|
|
|
self._parse_error("</")
|
|
|
|
|
self._parse_error("</a")
|
|
|
|
|
self._parse_error("<a<a>")
|
|
|
|
|
self._parse_error("</a<a>")
|
|
|
|
|
self._parse_error("<!")
|
|
|
|
|
self._parse_error("<a $>")
|
|
|
|
|
self._parse_error("<a")
|
|
|
|
|
self._parse_error("<a foo='bar'")
|
|
|
|
|
self._parse_error("<a foo='bar")
|
|
|
|
|
self._parse_error("<a foo='>'")
|
|
|
|
|
self._parse_error("<a foo='>")
|
|
|
|
|
self._parse_error("<a foo=>")
|
|
|
|
|
|
2001-08-03 16:53:01 -03:00
|
|
|
def test_declaration_junk_chars(self):
|
2001-05-18 12:32:59 -03:00
|
|
|
self._parse_error("<!DOCTYPE foo $ >")
|
|
|
|
|
|
2001-08-03 16:53:01 -03:00
|
|
|
def test_startendtag(self):
|
2001-05-18 12:32:59 -03:00
|
|
|
self._run_check("<p/>", [
|
|
|
|
|
("startendtag", "p", []),
|
|
|
|
|
])
|
|
|
|
|
self._run_check("<p></p>", [
|
|
|
|
|
("starttag", "p", []),
|
|
|
|
|
("endtag", "p"),
|
|
|
|
|
])
|
|
|
|
|
self._run_check("<p><img src='foo' /></p>", [
|
|
|
|
|
("starttag", "p", []),
|
|
|
|
|
("startendtag", "img", [("src", "foo")]),
|
|
|
|
|
("endtag", "p"),
|
|
|
|
|
])
|
|
|
|
|
|
2001-08-03 16:53:01 -03:00
|
|
|
def test_get_starttag_text(self):
|
2001-05-18 12:32:59 -03:00
|
|
|
s = """<foo:bar \n one="1"\ttwo=2 >"""
|
|
|
|
|
self._run_check_extra(s, [
|
|
|
|
|
("starttag", "foo:bar", [("one", "1"), ("two", "2")]),
|
|
|
|
|
("starttag_text", s)])
|
|
|
|
|
|
2001-08-03 16:53:01 -03:00
|
|
|
def test_cdata_content(self):
|
2011-11-01 09:09:56 -03:00
|
|
|
contents = [
|
|
|
|
|
'<!-- not a comment --> ¬-an-entity-ref;',
|
|
|
|
|
"<not a='start tag'>",
|
|
|
|
|
'<a href="" /> <p> <span></span>',
|
|
|
|
|
'foo = "</scr" + "ipt>";',
|
|
|
|
|
'foo = "</SCRIPT" + ">";',
|
|
|
|
|
'foo = <\n/script> ',
|
|
|
|
|
'<!-- document.write("</scr" + "ipt>"); -->',
|
|
|
|
|
('\n//<![CDATA[\n'
|
|
|
|
|
'document.write(\'<s\'+\'cript type="text/javascript" '
|
|
|
|
|
'src="http://www.example.org/r=\'+new '
|
|
|
|
|
'Date().getTime()+\'"><\\/s\'+\'cript>\');\n//]]>'),
|
|
|
|
|
'\n<!-- //\nvar foo = 3.14;\n// -->\n',
|
|
|
|
|
'foo = "</sty" + "le>";',
|
|
|
|
|
u'<!-- \u2603 -->',
|
|
|
|
|
# these two should be invalid according to the HTML 5 spec,
|
|
|
|
|
# section 8.1.2.2
|
|
|
|
|
#'foo = </\nscript>',
|
|
|
|
|
#'foo = </ script>',
|
|
|
|
|
]
|
|
|
|
|
elements = ['script', 'style', 'SCRIPT', 'STYLE', 'Script', 'Style']
|
|
|
|
|
for content in contents:
|
|
|
|
|
for element in elements:
|
|
|
|
|
element_lower = element.lower()
|
|
|
|
|
s = u'<{element}>{content}</{element}>'.format(element=element,
|
|
|
|
|
content=content)
|
|
|
|
|
self._run_check(s, [("starttag", element_lower, []),
|
|
|
|
|
("data", content),
|
|
|
|
|
("endtag", element_lower)])
|
|
|
|
|
|
2001-05-18 12:32:59 -03:00
|
|
|
|
2007-03-06 10:43:00 -04:00
|
|
|
def test_entityrefs_in_attributes(self):
|
|
|
|
|
self._run_check("<html foo='€&aa&unsupported;'>", [
|
|
|
|
|
("starttag", "html", [("foo", u"\u20AC&aa&unsupported;")])
|
|
|
|
|
])
|
|
|
|
|
|
2010-05-24 18:33:24 -03:00
|
|
|
def test_malformatted_charref(self):
|
|
|
|
|
self._run_check("<p>&#bad;</p>", [
|
|
|
|
|
("starttag", "p", []),
|
|
|
|
|
("data", "&#bad;"),
|
|
|
|
|
("endtag", "p"),
|
|
|
|
|
])
|
|
|
|
|
|
2010-12-28 12:05:07 -04:00
|
|
|
def test_unescape_function(self):
|
|
|
|
|
parser = HTMLParser.HTMLParser()
|
|
|
|
|
self.assertEqual(parser.unescape('&#bad;'),'&#bad;')
|
|
|
|
|
self.assertEqual(parser.unescape('&'),'&')
|
|
|
|
|
|
2001-05-18 12:32:59 -03:00
|
|
|
|
2001-09-24 17:19:08 -03:00
|
|
|
def test_main():
|
|
|
|
|
test_support.run_unittest(HTMLParserTestCase)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
if __name__ == "__main__":
|
|
|
|
|
test_main()
|
