Extended URL regexp. Enhanced wikipedia quoting.

git-svn-id: file:///home/phd/archive/SVN/phdru.name/scripts@81 7bb0bf08-9e0d-0410-b083-99cee3bf18b8

diff --git a/phd_pp.py b/phd_pp.py
index 45b17aa766def132e9f07b3e32c0c26a7ab601b4..80331b05c91025397253bc5cb3e8467592931deb 100644 (file)
--- a/phd_pp.py
+++ b/phd_pp.py
@@ -1,19 +1,12 @@
 import os, re, time, urllib
 from Cheetah.Template import Template
 
-
-# Copied from ZWiki and Gajim
-
-urlchars         = r'[A-Za-z0-9/:@_%~#=&\.\-\?\+\$,]+'
-urlendchar       = r'[A-Za-z0-9/]'
-url              = r'["=]?((about|file|ftp|gopher|http|https|mailto|news|telnet|ed2k|irc|sip|magnet):%s)' % \
-                   (urlchars+urlendchar)
+url_re = r"(((https?|ftp|gopher|telnet)://|(mailto|file|news|about|ed2k|irc|sip|magnet):)[^' \t<>\"]+|(www|web|w3)[A-Za-z0-9_-]*\.[A-Za-z0-9._-]+\.[^' \t<>\"]+)[A-Za-z0-9/]"
 
 def _url2href(match):
    url = match.group(0)
    return '<a href="%s">%s</a>' % (url, url)
 
-
 full_dirs = len(os.getcwd().split('/')) + 1
 
 class phd_pp(Template):
@@ -50,7 +43,7 @@ class phd_pp(Template):
          return self.rst2html()
 
    def text2html(self):
-      body = re.sub(url, _url2href, self.body_text())
+      body = re.sub(url_re, _url2href, self.body_text())
 
       paragraphs = body.split("\n\n")
 
@@ -102,11 +95,10 @@ class phd_pp(Template):
       return "http://www.yandex.ru/yandsearch?text=%s&rpt=rad" % quote_string(query, "cp1251")
 
    def wikipedia(self, query):
-      return "http://en.wikipedia.org/wiki/%s" % quote_string(query.replace(' ', '_'))
+      return "http://en.wikipedia.org/wiki/%s" % quote_string(query.replace(' ', '_'), ext_safe=',')
 
    def wikipedia_ru(self, query):
-      return "http://ru.wikipedia.org/wiki/%s" % quote_string(query.replace(' ', '_'))
-
+      return "http://ru.wikipedia.org/wiki/%s" % quote_string(query.replace(' ', '_'), ext_safe=',')
 
-def quote_string(s, to_encoding="utf-8"):
-   return urllib.quote(unicode(s, "koi8-r").encode(to_encoding))
+def quote_string(s, to_encoding="utf-8", ext_safe=''):
+   return urllib.quote(unicode(s, "koi8-r").encode(to_encoding), '/' + ext_safe)