diff --git a/Lib/html/parser.py b/Lib/html/parser.py
--- a/Lib/html/parser.py
+++ b/Lib/html/parser.py
@@ -184,7 +184,17 @@
elif startswith("", i):
k = self.parse_pi(i)
elif startswith(" or
+ # . When strict is True an
+ # error is raised, when it's False they will be considered
+ # as bogus comments and parsed (see parse_bogus_comment).
+ if self.strict:
+ k = self.parse_declaration(i)
+ else:
+ try:
+ k = self.parse_declaration(i)
+ except HTMLParseError:
+ k = self.parse_bogus_comment(i)
elif (i + 1) < n:
self.handle_data("<")
k = i + 1
@@ -256,6 +266,19 @@
i = self.updatepos(i, n)
self.rawdata = rawdata[i:]
+ # Internal -- parse bogus comment, return length or -1 if not terminated
+ # see http://www.w3.org/TR/html5/tokenization.html#bogus-comment-state
+ def parse_bogus_comment(self, i, report=1):
+ rawdata = self.rawdata
+ if rawdata[i:i+2] != '', i+2)
+ if pos == -1:
+ return -1
+ if report:
+ self.handle_comment(rawdata[i+2: pos])
+ return pos + 1
+
# Internal -- parse processing instr, return end or -1 if not terminated
def parse_pi(self, i):
rawdata = self.rawdata
diff --git a/Lib/test/test_htmlparser.py b/Lib/test/test_htmlparser.py
--- a/Lib/test/test_htmlparser.py
+++ b/Lib/test/test_htmlparser.py
@@ -426,6 +426,19 @@
# see #12888
self.assertEqual(p.unescape('{ ' * 1050), '{ ' * 1050)
+ def test_broken_comments(self):
+ html = (''
+ ''
+ ''
+ '')
+ expected = [
+ ('comment', ' not really a comment '),
+ ('comment', ' not a comment either --'),
+ ('comment', ' -- close enough --'),
+ ('comment', ' <-- this one is supposed to be an empty comment')
+ ]
+ self._run_check(html, expected)
+
def test_broken_condcoms(self):
# these condcoms are missing the '--' after ''
html = ('broken condcom'