make url titles be sent back on one line

[erebus.git] / modules / urls.py

diff --git a/modules/urls.py b/modules/urls.py
index bcfb0e300dd8d8407ff4a8d4256e4023b39886ee..72bdb99c8eeb88a2cdb79194dd6f37b772910f18 100644 (file)
--- a/modules/urls.py
+++ b/modules/urls.py
@@ -1,4 +1,5 @@
 # Erebus IRC bot - Author: Erebus Team
+# vim: fileencoding=utf-8
 # URL Checker
 # This file is released into the public domain; see http://unlicense.org/
 
@@ -20,16 +21,29 @@ modstart = lib.modstart
 modstop = lib.modstop
 
 # module code
-import re, urllib2, urlparse, json, HTMLParser
-from BeautifulSoup import BeautifulSoup
+import sys
+if sys.version_info.major < 3:
+       import urllib2
+       import urlparse
+       import HTMLParser
+       from BeautifulSoup import BeautifulSoup
+       import re
+else:
+       import urllib.request as urllib2
+       import urllib.parse as urlparse
+       import html.parser as HTMLParser
+       from bs4 import BeautifulSoup
+       import re
+
+import re, json
 
 html_parser = HTMLParser.HTMLParser()
 
 hostmask_regex = re.compile(r'^(.*)!(.*)@(.*)$')
-url_regex = re.compile(r'((?:https?://|spotify:)[^\s]+)')
+url_regex = re.compile(r'https?://[^/\s]+\.[^/\s]+(?:/\S+)?')
 spotify_regex = (
        re.compile(r'spotify:(?P<type>\w+):(?P<track_id>\w{22})'),
-       re.compile(r'https?://open.spotify.com/(?P<type>\w+)/(?P<track_id>\w{22})')
+       re.compile(r'https?://open.spotify.com/(?P<type>\w+)/(?P<track_id>\w+)')
 )
 youtube_regex = (
        re.compile(r'https?://(?:www\.)?youtube\.com/watch\?[a-zA-Z0-9=&_\-]+'),
@@ -85,28 +99,15 @@ def privmsg_hook(bot, textline):
        except IndexError:
                line = ''
 
+       responses = []
        for match in url_regex.findall(line):
                if match:
-                       if 'open.spotify.com' in match or 'spotify:' in match:
-                               for r in spotify_regex:
-                                       for sptype, track in r.findall(match):
-                                               bot.msg(chan, gotspotify(sptype, track))
-
-                       elif 'youtube.com' in match or 'youtu.be' in match:
-                               for r in youtube_regex:
-                                       for url in r.findall(match):
-                                               bot.msg(chan, gotyoutube(url))
-
-                       elif 'twitch.tv' in match:
-                               for r in twitch_regex:
-                                       for uri in r.findall(match):
-                                               bot.msg(chan, gottwitch(uri))
-
-                       else:
-                               bot.msg(chan, goturl(match))
+                       responses.append(goturl(match))
+       if len(responses) > 0:
+               bot.msg(chan, ' | '.join(responses), True)
 
 def unescape(line):
-       return html_parser.unescape(line)
+       return re.sub('\s+', ' ', html_parser.unescape(line))
 
 def gotspotify(type, track):
        url = 'http://ws.spotify.com/lookup/1/?uri=spotify:%s:%s' % (type, track)
@@ -165,7 +166,9 @@ def goturl(url):
        request = urllib2.Request(url)
        opener = urllib2.build_opener(SmartRedirectHandler())
        try:
-               soup = BeautifulSoup(opener.open(request, timeout=2))
+               soup = BeautifulSoup(opener.open(request, timeout=0.5))
                return unescape('Title: %s' % (soup.title.string))
-       except:
-               return 'Invalid URL/Timeout'
+       except urllib2.HTTPError as e:
+               return 'Error: %s %s' % (e.code, e.reason)
+       except Exception as e:
+               return 'Error: %r' % (e.message)