"""Tests for HTMLParser.py."""
import html.parser
import pprint
import unittest
class EventCollector(html.parser.HTMLParser):
def __init__(self, *args, **kw):
self.events = []
self.append = self.events.append
html.parser.HTMLParser.__init__(self, *args, **kw)
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 EventCollectorCharrefs(EventCollector):
def handle_charref(self, data):
self.fail('This should never be called with convert_charrefs=True')
def handle_entityref(self, data):
self.fail('This should never be called with convert_charrefs=True')
class TestCaseBase(unittest.TestCase):
def get_collector(self):
return EventCollector(convert_charrefs=False)
def _run_check(self, source, expected_events, collector=None):
if collector is None:
collector = self.get_collector()
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" +
"\nSource:\n" + repr(source) +
"\nExpected:\n" + pprint.pformat(expected_events) +
"\nReceived:\n" + pprint.pformat(events))
def _run_check_extra(self, source, events):
self._run_check(source, events,
EventCollectorExtra(convert_charrefs=False))
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->
', 'foo = "";', 'foo = "";', 'foo = <\n/script> ', '', ('\n//<\\/s\'+\'cript>\');\n//]]>'), '\n\n', 'foo = "";', '', # 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 = '<{element}>{content}{element}>'.format(element=element, content=content) self._run_check(s, [("starttag", element_lower, []), ("data", content), ("endtag", element_lower)]) def test_cdata_with_closing_tags(self): # see issue #13358 # make sure that HTMLParser calls handle_data only once for each CDATA. # The normal event collector normalizes the events in get_events, # so we override it to return the original list of events. class Collector(EventCollector): def get_events(self): return self.events content = """ ¬-an-entity-ref;
''"""
for element in [' script', 'script ', ' script ',
'\nscript', 'script\n', '\nscript\n']:
element_lower = element.lower().strip()
s = '{1}'
'{1}'.format(text, charref),
expected, collector=collector())
# check truncated charrefs at the end of the file
html = '&quo '
for x in range(1, len(html)):
self._run_check(html[:x], [('data', html[:x])],
collector=collector())
# check a string with no charrefs
self._run_check('no charrefs here', [('data', 'no charrefs here')],
collector=collector())
# the remaining tests were for the "tolerant" parser (which is now
# the default), and check various kind of broken markup
def test_tolerant_parsing(self):
self._run_check('te>>xt&a<
" "- software-and-i" "- library |