Update to catch back up to craig's image helper

readability/scoring.py

@@ -90,7 +90,15 @@ def get_link_density(node, node_text=None):
     if text_length == 0:
         return 0.0
 
-    links_length = sum(map(_get_normalized_text_length, node.findall(".//a")))
+
+    link_length = sum([len(a.text_content()) or 0
+        for a in node.findall(".//a")])
+
+    # For each img, give 50 bonus chars worth of length.
+    # Tweaking this 50 down a notch should help if we hit false positives.
+    links_length = max(link_length -
+                      sum([50 for img in node.findall(".//img")]), 0)
+
     return links_length / text_length
 
 

readability/scripts/client.py

@@ -37,7 +37,10 @@ from ..readable import Article
 
 
 HEADERS = {
-    "User-Agent": "Readability (Readable content parser) Version/%s" % __version__,
+    "User-Agent": 'breadability/{version} ({url})'.format(
+        url="https://github.com/bookieio/breadability",
+        version=__version__
+    )
 }
 
 

tests/test_articles/test_businessinsider-com/test.py

@@ -5,7 +5,7 @@ try:
 except ImportError:
     import unittest
 
-from breadability.readable import Article
+from readability.readable import Article
 
 
 class TestBusinessInsiderArticle(unittest.TestCase):

tests/test_articles/test_sweetshark/test.py

@@ -5,7 +5,7 @@ try:
 except ImportError:
     import unittest
 
-from breadability.readable import Article
+from readability.readable import Article
 
 
 class TestSweetsharkBlog(unittest.TestCase):