Fix issue #366 - pip throws IndexError when it calls `scraped_rel_links`

docs/news.txt

@@ -17,6 +17,9 @@ develop (unreleased)
   restart-in-venv mechanism that's broken, and neither one is useful since
   every virtualenv now has pip inside it.
 
+* Fixed issue #366 - pip throws IndexError when it calls `scraped_rel_links`
+
+
 1.0.2 (2011-07-16)
 ------------------
 

pip/index.py

@@ -549,7 +549,7 @@ class HTMLPage(object):
             href_match = self._href_re.search(self.content, pos=match.end())
             if not href_match:
                 continue
-            url = match.group(1) or match.group(2) or match.group(3)
+            url = href_match.group(1) or href_match.group(2) or href_match.group(3)
             if not url:
                 continue
             url = self.clean_link(urlparse.urljoin(self.base_url, url))

tests/test_index.py

-from pip.index import package_to_requirement
+from pip.index import package_to_requirement, HTMLPage
 
 
 def test_package_name_should_be_converted_to_requirement():
@@ -8,3 +8,21 @@ def test_package_name_should_be_converted_to_requirement():
     assert package_to_requirement('Foo-1.2') == 'Foo==1.2'
     assert package_to_requirement('Foo-dev') == 'Foo==dev'
     assert package_to_requirement('Foo') == 'Foo'
+
+
+def test_html_page_should_be_able_to_scrap_rel_links():
+    """
+    Test scraping page looking for url in href
+    """
+    page = HTMLPage("""
+        <!-- The <th> elements below are a terrible terrible hack for setuptools -->
+        <li>
+        <strong>Home Page:</strong>
+        <!-- <th>Home Page -->
+        <a href="http://supervisord.org/">http://supervisord.org/</a>
+        </li>""", "supervisor")
+
+    links = list(page.scraped_rel_links())
+    assert len(links) == 1
+    assert links[0].url == 'http://supervisord.org/'
+