Fix crash in url preview when html tag has no text

Signed-off-by: Marcin Bachry <[email protected]>
c-rhodes · Dec 14, 2016 · 24c16fc · 24c16fc
1 parent c3208e4
commit 24c16fc
Show file tree

Hide file tree

Showing 2 changed files with 54 additions and 1 deletion.
diff --git a/synapse/rest/media/v1/preview_url_resource.py b/synapse/rest/media/v1/preview_url_resource.py
@@ -381,7 +381,10 @@ def _calc_og(tree, media_uri):
     if 'og:title' not in og:
         # do some basic spidering of the HTML
         title = tree.xpath("(//title)[1] | (//h1)[1] | (//h2)[1] | (//h3)[1]")
-        og['og:title'] = title[0].text.strip() if title else None
+        if title and title[0].text is not None:
+            og['og:title'] = title[0].text.strip()
+        else:
+            og['og:title'] = None
 
     if 'og:image' not in og:
         # TODO: extract a favicon failing all else

diff --git a/tests/test_preview.py b/tests/test_preview.py
@@ -215,3 +215,53 @@ def test_script(self):
             u"og:title": u"Foo",
             u"og:description": u"Some text."
         })
+
+    def test_missing_title(self):
+        html = u"""
+        <html>
+        <body>
+        Some text.
+        </body>
+        </html>
+        """
+
+        og = decode_and_calc_og(html, "http://example.com/test.html")
+
+        self.assertEquals(og, {
+            u"og:title": None,
+            u"og:description": u"Some text."
+        })
+
+    def test_h1_as_title(self):
+        html = u"""
+        <html>
+        <meta property="og:description" content="Some text."/>
+        <body>
+        <h1>Title</h1>
+        </body>
+        </html>
+        """
+
+        og = decode_and_calc_og(html, "http://example.com/test.html")
+
+        self.assertEquals(og, {
+            u"og:title": u"Title",
+            u"og:description": u"Some text."
+        })
+
+    def test_missing_title_and_broken_h1(self):
+        html = u"""
+        <html>
+        <body>
+        <h1><a href="foo"/></h1>
+        Some text.
+        </body>
+        </html>
+        """
+
+        og = decode_and_calc_og(html, "http://example.com/test.html")
+
+        self.assertEquals(og, {
+            u"og:title": None,
+            u"og:description": u"Some text."
+        })