Add checks to links to remove really bad links from the scripting site

src/breadability/readable.py

@@ -251,6 +251,16 @@ def clean_conditionally(node):
 
     LNODE.log(node, 2, 'Cleaning conditionally node.')
 
+    # Clean out links with really large href anchor values.
+    if node.tag == 'a':
+        name = node.get('name')
+        href = node.get('href')
+        if name and not href:
+            return True
+
+        if href and href.startswith('#') and len(href) > 50:
+            return True
+
     if node.tag not in target_tags:
         # this is not the tag you're looking for
         LNODE.log(node, 2, 'Node cleared.')

src/breadability/tests/test_articles/test_scripting-com/test.py

@@ -29,11 +29,13 @@ class TestArticle(TestCase):
         doc = Article(self.article)
         self.assertTrue('Amazon and Google' in doc.readable)
         self.assertFalse('Linkblog updated' in doc.readable)
+        self.assertFalse(
+            '#anExampleGoogleDoesntIntendToShareBlogAndItWill' in doc.readable)
 
     def test_candidates(self):
         """Verify we have candidates."""
         doc = Article(self.article)
-        from lxml.etree import tounicode
+        # from lxml.etree import tounicode
         found = False
         wanted_hash = '04e46055'
         # from breadability.logconfig import LNODE