PsychoticNinja/JLU5-SupyPlugins
1
0
Fork 0
mirror of https://github.com/jlu5/SupyPlugins.git synced 2025-04-30 07:21:12 -05:00
Code Issues Projects Releases Wiki Activity

Wikifetch: modularize wiki fetcher, use new-style wrap, simplify URL display

Browse Source
This commit is contained in:
James Lu 2015-10-25 17:20:40 -07:00
parent f1bd086098
commit 08a2942d99
1 changed files with 22 additions and 21 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

39
Wikifetch/plugin.py
View File

@ -59,28 +59,25 @@ class Wikifetch(callbacks.Plugin):
"""Grabs data from Wikipedia and other MediaWiki-powered sites."""
threaded = True
@internationalizeDocstring
def wiki(self, irc, msg, args, optlist, search):
"""[--site <site>] <search term>
Returns the first paragraph of a Wikipedia article."""
def _wiki(self, irc, msg, search, baseurl):
"""Fetches and replies content from a MediaWiki-powered website."""
reply = ''
optlist = dict(optlist)
# Different instances of MediaWiki use different URLs... This tries
# to make the parser work for most sites, but still use resonable defaults
# such as filling in http:// and appending /wiki to links...
# Try using '--site lyrics.wikia.com' or '--site wiki.archlinux.org'.
baseurl = optlist.get('site') or self.registryValue('url', msg.args[0])
# Special cases: Wikia, Wikipedia, Arch Linux Wiki
if 'wikia.com' in baseurl or 'wikipedia.org' in baseurl:
baseurl += '/wiki'
elif 'wiki.archlinux.org' in baseurl:
baseurl += '/index.php'
if not baseurl.lower().startswith(('http://', 'https://')):
baseurl = 'http://' + baseurl
# first, we get the page
addr = '%s/Special:Search?search=%s' % \
(baseurl, quote_plus(search))
self.log.debug('Wikifetch: using URL %s', addr)
article = utils.web.getUrl(addr)
if sys.version_info[0] >= 3:
article = article.decode()
@ -140,16 +137,9 @@ class Wikifetch(callbacks.Plugin):
reply += '"%s" (Redirected from "%s"): ' % (title, redirect)
except IndexError:
pass
# extract the address we got it from
# We only care about formatting this if we're actually on Wikipedia
# (i.e. not using --site <site>)
# extract the address we got it from - most sites have the perm link
# inside the page itself
try:
if 'site' not in optlist:
addr = tree.find(".//link[@rel='canonical']")
addr = addr.attrib['href']
# force serving HTTPS links
addr = 'https://' + addr.split("//")[1]
else:
addr = tree.find(".//div[@class='printfooter']/a").attrib['href']
addr = re.sub('([&?]|(amp;)?)oldid=\d+$', '', addr)
except:
@ -211,7 +201,18 @@ class Wikifetch(callbacks.Plugin):
# Remove inline citations (text[1][2][3], etc.)
reply = re.sub('\[\d+\]', '', reply)
irc.reply(reply)
wiki = wrap(wiki, [getopts({'site': 'somethingWithoutSpaces'}), 'text'])
@internationalizeDocstring
@wrap([getopts({'site': 'somethingWithoutSpaces'}), 'text'])
def wiki(self, irc, msg, args, optlist, search):
"""[--site <site>] <search term>
Returns the first paragraph of a Wikipedia article. Optionally, a site
argument can be given to override the default (usually Wikipedia) -
try using '--site lyrics.wikia.com' or '--site wiki.archlinux.org'."""
optlist = dict(optlist)
baseurl = optlist.get('site') or self.registryValue('url', msg.args[0])
self._wiki(irc, msg, search, baseurl)
Class = Wikifetch