Merge lp:~roignac/gwibber/Bug324809 into lp:gwibber

[Vadim Rutkovsky]

Fixes lp:324809 - urls like www.google.com are recognized correctly, http:// is added to the href

[Omer Akram (om26er)]

works

review: Approve

[Ken VanDine (ken-vandine)]

Thanks!

review: Approve

=== modified file 'gwibber/microblog/util/__init__.py'
--- gwibber/microblog/util/__init__.py	2010-08-22 18:30:36 +0000
+++ gwibber/microblog/util/__init__.py	2010-08-28 12:33:42 +0000
@@ -17,8 +17,8 @@
 URL_SCHEMES = ('http', 'https', 'ftp', 'mailto', 'news', 'gopher',
                'nntp', 'telnet', 'wais', 'prospero', 'aim', 'webcal')
 
-URL_FORMAT = (r'(?<!\w)((?:%s):' # protocol + :
-    '/*(?!/)(?:' # get any starting /'s
+URL_FORMAT = (r'(?<!\w)(((?:%s)://|(?:www\.))' # protocol + :// or www.
+    '(?!/)(?:' # get any starting /'s
     '[\w$\+\*@&=\-/]' # reserved | unreserved
     '|%%[a-fA-F0-9]{2}' # escape
     '|[\?\.:\(\),;!\'\~](?!(?:\s|$))' # punctuation
@@ -37,7 +37,9 @@
 def linkify(text, subs=[], escape=True):
   if escape: text = cgi.escape(text)
   for f, r in subs: text = f.sub(r, text)
-  return PARSE_LINK.sub('<a href="\\1">\\1</a>', text)
+  link = PARSE_LINK.sub('<a href="\\1">\\1</a>', text)
+  # if link has www only, add http:// to the href
+  return re.compile(r'"www.', re.U).sub('"http://www.', link)
 
 def imgpreview(text):
   thumbre = {