"""Tests for HTMLParser.py."""
import HTMLParser
import pprint
import unittest
from test import test_support
class EventCollector(HTMLParser.HTMLParser):
def __init__(self):
self.events = []
self.append = self.events.append
HTMLParser.HTMLParser.__init__(self)
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))
def unknown_decl(self, decl):
self.append(("unknown decl", decl))
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):
def _run_check(self, source, expected_events, collector=EventCollector):
parser = collector()
for s in source:
parser.feed(s)
parser.close()
events = parser.get_events()
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))
def _run_check_extra(self, source, events):
self._run_check(source, events, EventCollectorExtra)
def _parse_error(self, source):
def parse(source=source):
parser = HTMLParser.HTMLParser()
parser.feed(source)
parser.close()
self.assertRaises(HTMLParser.HTMLParseError, parse)
class HTMLParserTestCase(TestCaseBase):
def test_processing_instruction_only(self):
self._run_check("", [
("pi", "processing instruction"),
])
self._run_check("", [
("pi", "processing instruction ?"),
])
def test_simple_html(self):
self._run_check("""
&entity;
sample
text
“
""", [
("data", "\n"),
("decl", "DOCTYPE html PUBLIC 'foo'"),
("data", "\n"),
("starttag", "html", []),
("entityref", "entity"),
("charref", "32"),
("data", "\n"),
("comment", "comment1a\n-><
%paramEntity;
]"""
self._run_check("" % inside, [
("decl", inside),
])
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.
self._run_check("", [
("starttag", "a", []),
("starttag", "b", []),
("endtag", "a"),
("endtag", "b"),
])
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"),
])
def test_attr_syntax(self):
output = [
("starttag", "a", [("b", "v"), ("c", "v"), ("d", "v"), ("e", None)])
]
self._run_check("""""", output)
self._run_check("""""", output)
self._run_check("""""", output)
self._run_check("""""", output)
def test_attr_values(self):
self._run_check("""""",
[("starttag", "a", [("b", "xxx\n\txxx"),
("c", "yyy\t\nyyy"),
("d", "\txyz\n")])
])
self._run_check("""""", [
("starttag", "a", [("b", ""), ("c", "")]),
])
# Regression test for SF patch #669683.
self._run_check("", [
("starttag", "e", [("a", "rgb(1,2,3)")]),
])
# Regression test for SF bug #921657.
self._run_check("", [
("starttag", "a", [("href", "mailto:xyz@example.com")]),
])
def test_attr_nonascii(self):
# see issue 7311
self._run_check(u"
", [
("starttag", "img", [("src", "/foo/bar.png"),
("alt", u"\u4e2d\u6587")]),
])
self._run_check(u"", [
("starttag", "a", [("title", u"\u30c6\u30b9\u30c8"),
("href", u"\u30c6\u30b9\u30c8.html")]),
])
self._run_check(u'', [
("starttag", "a", [("title", u"\u30c6\u30b9\u30c8"),
("href", u"\u30c6\u30b9\u30c8.html")]),
])
def test_attr_entity_replacement(self):
self._run_check("""""", [
("starttag", "a", [("b", "&><\"'")]),
])
def test_attr_funky_names(self):
self._run_check("""""", [
("starttag", "a", [("a.b", "v"), ("c:d", "v"), ("e-f", "v")]),
])
def test_illegal_declarations(self):
self._parse_error('')
def test_starttag_end_boundary(self):
self._run_check("""""", [("starttag", "a", [("b", "<")])])
self._run_check("""""", [("starttag", "a", [("b", ">")])])
def test_buffer_artefacts(self):
output = [("starttag", "a", [("b", "<")])]
self._run_check([""], output)
self._run_check([""], output)
self._run_check([""], output)
self._run_check([""], output)
self._run_check([""], output)
self._run_check([""], output)
output = [("starttag", "a", [("b", ">")])]
self._run_check([""], output)
self._run_check([""], output)
self._run_check([""], output)
self._run_check(["'>"], output)
self._run_check([""], output)
self._run_check([""], output)
output = [("comment", "abc")]
self._run_check(["", ""], output)
self._run_check(["<", "!--abc-->"], output)
self._run_check([""], output)
self._run_check([""], output)
self._run_check([""], output)
self._run_check([""], output)
self._run_check([""], output)
self._run_check([""], output)
self._run_check(["", ""], output)
def test_starttag_junk_chars(self):
self._parse_error("")
self._parse_error("")
self._parse_error("")
self._parse_error("")
self._parse_error("")
self._parse_error("'")
self._parse_error("", [
("starttag", "p", []),
("startendtag", "img", [("src", "foo")]),
("endtag", "p"),
])
def test_get_starttag_text(self):
s = """"""
self._run_check_extra(s, [
("starttag", "foo:bar", [("one", "1"), ("two", "2")]),
("starttag_text", s)])
def test_cdata_content(self):
s = """"""
self._run_check(s, [
("starttag", "script", []),
("data", " ¬-an-entity-ref; "),
("endtag", "script"),
])
s = """"""
self._run_check(s, [
("starttag", "script", []),
("data", " "),
("endtag", "script"),
])
def test_entityrefs_in_attributes(self):
self._run_check("", [
("starttag", "html", [("foo", u"\u20AC&aa&unsupported;")])
])
def test_malformatted_charref(self):
self._run_check("&#bad;
", [
("starttag", "p", []),
("data", "&#bad;"),
("endtag", "p"),
])
def test_unescape_function(self):
parser = HTMLParser.HTMLParser()
self.assertEqual(parser.unescape('&#bad;'),'&#bad;')
self.assertEqual(parser.unescape('&'),'&')
def test_main():
test_support.run_unittest(HTMLParserTestCase)
if __name__ == "__main__":
test_main()