protect html_to_text as well

talon/quotations.py

@@ -386,9 +386,6 @@ def _extract_from_html(msg_body):
     then checking deleted checkpoints,
     then deleting necessary tags.
     """
-    if _html_too_big(msg_body):
-        return msg_body
-
     if msg_body.strip() == b'':
         return msg_body
 
@@ -483,12 +480,3 @@ def register_xpath_extensions():
     ns.prefix = 'mg'
     ns['text_content'] = text_content
     ns['tail'] = tail
-
-
-def _html_too_big(msg_body):
-    return msg_body.count('<') > _MAX_TAGS_COUNT
-
-
-# an extensive research shows that exceeding this limit
-# might lead to excessive processing time
-_MAX_TAGS_COUNT = 419

talon/utils.py

@@ -178,6 +178,9 @@ def html_fromstring(s):
     """Parse html tree from string. Return None if the string can't be parsed.
     """
     try:
+        if html_too_big(s):
+            return None
+
         return html5parser.fromstring(s, parser=_html5lib_parser())
     except Exception:
         pass
@@ -187,6 +190,9 @@ def html_document_fromstring(s):
     """Parse html tree from string. Return None if the string can't be parsed.
     """
     try:
+        if html_too_big(s):
+            return None
+
         return html5parser.document_fromstring(s, parser=_html5lib_parser())
     except Exception:
         pass
@@ -196,6 +202,10 @@ def cssselect(expr, tree):
     return CSSSelector(expr)(tree)
 
 
+def html_too_big(s):
+    return s.count('<') > _MAX_TAGS_COUNT
+
+
 def _contains_charset_spec(s):
     """Return True if the first 4KB contain charset spec
     """
@@ -243,3 +253,7 @@ _BLOCKTAGS  = ['div', 'p', 'ul', 'li', 'h1', 'h2', 'h3']
 _HARDBREAKS = ['br', 'hr', 'tr']
 
 _RE_EXCESSIVE_NEWLINES = re.compile("\n{2,10}")
+
+# an extensive research shows that exceeding this limit
+# might lead to excessive processing time
+_MAX_TAGS_COUNT = 419