diff --git a/Lib/html/parser.py b/Lib/html/parser.py
--- a/Lib/html/parser.py
+++ b/Lib/html/parser.py
@@ -139,8 +139,16 @@
if self.convert_charrefs and not self.cdata_elem:
j = rawdata.find('<', i)
if j < 0:
- if not end:
- break # wait till we get all the text
+ # if we can't find the next <, either we are at the end
+ # or there's more text incoming. If the latter is True,
+ # we can't pass the text to handle_data in case we have
+ # a charref cut in half at end. Try to determine if
+ # this is the case before proceding by looking for an
+ # & near the end and see if it's followed by a space or ;.
+ amppos = rawdata.rfind('&', max(i, n-34))
+ if (amppos >= 0 and
+ not re.compile(r'[\s;]').search(rawdata, amppos)):
+ break # wait till we get all the text
j = n
else:
match = self.interesting.search(rawdata, i) # < or &
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
@@ -72,9 +72,6 @@
class EventCollectorCharrefs(EventCollector):
- def get_events(self):
- return self.events
-
def handle_charref(self, data):
self.fail('This should never be called with convert_charrefs=True')
@@ -633,6 +630,18 @@
]
self._run_check(html, expected)
+ def test_convert_charrefs_dropped_text(self):
+ # #23144: make sure that all the events are triggered when
+ # convert_charrefs is True, even if we don't call .close()
+ parser = EventCollector(convert_charrefs=True)
+ # before the fix, bar & baz was missing
+ parser.feed("foo link bar & baz")
+ self.assertEqual(
+ parser.get_events(),
+ [('data', 'foo '), ('starttag', 'a', []), ('data', 'link'),
+ ('endtag', 'a'), ('data', ' bar & baz')]
+ )
+
class AttributesTestCase(TestCaseBase):