From 865d57780930ba36f25008bc4c702b764a2d1d5e Mon Sep 17 00:00:00 2001
From: =?utf8?q?Stefan=20K=C3=B6gl?= <stefan@skoegl.net>
Date: Mon, 10 Jan 2011 09:55:44 +0200
Subject: [PATCH] include URLs of a feed in the response

---
 feedservice/feeddownloader.py |  2 ++
 feedservice/urls.py           | 32 ++++++++++++++++++++++++++++++++
 2 files changed, 34 insertions(+)
 create mode 100644 feedservice/urls.py

diff --git a/feedservice/feeddownloader.py b/feedservice/feeddownloader.py
index 2bf3880..e767ad4 100644
--- a/feedservice/feeddownloader.py
+++ b/feedservice/feeddownloader.py
@@ -35,6 +35,7 @@ import feedcore
 from utils import parse_time
 import youtube
 from mimetype import get_mimetype, check_mimetype, get_podcast_types
+from urls import get_redirects
 
 #socket.setdefaulttimeout(10)
 fetcher = feedcore.Fetcher(USER_AGENT)
@@ -155,6 +156,7 @@ def parse_feed(feed_url, inline_logo, scale_to):
         podcast['description'] = feed.feed.get('subtitle', '')
         podcast['author'] = feed.feed.get('author', feed.feed.get('itunes_author', ''))
         podcast['language'] = feed.feed.get('language', '')
+        podcast['urls'] = get_redirects(feed_url)
 
         logo_url = get_podcast_logo(feed)
         podcast['logo'] = logo_url
diff --git a/feedservice/urls.py b/feedservice/urls.py
new file mode 100644
index 0000000..211ceb5
--- /dev/null
+++ b/feedservice/urls.py
@@ -0,0 +1,32 @@
+import urllib2
+import urlparse
+
+class RedirectCollector(urllib2.HTTPRedirectHandler):
+    """Collects all seen (intermediate) redirects for a HTTP request"""
+
+    def __init__(self, *args, **kwargs):
+        self.urls = []
+
+    def redirect_request(self, req, fp, code, msg, hdrs, newurl):
+        self.urls.append(newurl)
+        return urllib2.HTTPRedirectHandler.redirect_request(self, req, fp, code, msg, hdrs, newurl)
+
+
+def get_redirects(url):
+    """ Returns the complete redirect chain, starting from url """
+    collector = RedirectCollector()
+    collector.urls.append(url)
+    opener = urllib2.build_opener(collector)
+    opener.open(url)
+    return map(basic_sanitizing, collector.urls)
+
+
+def basic_sanitizing(url):
+    """
+    does basic sanitizing through urlparse and additionally converts the netloc to lowercase
+    """
+    r = urlparse.urlsplit(url)
+    netloc = r.netloc.lower()
+    r2 = urlparse.SplitResult(r.scheme, netloc, r.path or '/', r.query, r.fragment)
+    return r2.geturl()
+
-- 
2.11.4.GIT