Fixed issue with self-closing tags. Fix buriy#125

tianyeeee · Dec 29, 2019 · 0846955 · 0846955
1 parent de20908
commit 0846955
Show file tree

Hide file tree

Showing 2 changed files with 6 additions and 1 deletion.
diff --git a/readability/readability.py b/readability/readability.py
@@ -171,7 +171,7 @@ def get_clean_html(self):
         An internal method, which can be overridden in subclasses, for example,
         to disable or to improve DOM-to-text conversion in .summary() method
         """
-        return clean_attributes(tounicode(self.html))
+        return clean_attributes(tounicode(self.html, method='html'))
 
     def summary(self, html_partial=False):
         """

diff --git a/tests/test_article_only.py b/tests/test_article_only.py
@@ -106,3 +106,8 @@ def test_many_repeated_spaces(self):
         s = doc.summary()
 
         assert 'foo' in s
+
+    def test_not_self_closing(self):
+        sample = '<h2><a href="#"></a>foobar</h2>'
+        doc = Document(sample)
+        assert '<body id="readabilityBody"><h2><a href="#"></a>foobar</h2></body>' == doc.summary()