phenny-1/modules/wikipedia.py

84 lines
2.5 KiB
Python
Raw Normal View History

#!/usr/bin/env python
"""
wikipedia.py - Phenny Wikipedia Module
2009-06-07 05:08:49 -04:00
Copyright 2008-9, Sean B. Palmer, inamidst.com
Licensed under the Eiffel Forum License 2.
http://inamidst.com/phenny/
"""
2011-09-22 14:17:27 -04:00
import re, urllib.request, urllib.parse, urllib.error, gzip, io
import web
2011-07-15 13:43:27 -04:00
import json
2011-07-15 13:43:27 -04:00
wikiapi = 'http://en.wikipedia.org/w/api.php?action=query&list=search&srsearch=%s&limit=1&prop=snippet&format=json'
wikiuri = 'http://en.wikipedia.org/wiki/%s'
wikisearch = 'http://en.wikipedia.org/wiki/Special:Search?' \
+ 'search=%s&fulltext=Search'
r_tr = re.compile(r'(?ims)<tr[^>]*>.*?</tr>')
r_paragraph = re.compile(r'(?ims)<p[^>]*>.*?</p>|<li(?!n)[^>]*>.*?</li>')
r_tag = re.compile(r'<(?!!)[^>]+>')
r_whitespace = re.compile(r'[\t\r\n ]+')
r_redirect = re.compile(
2011-07-15 13:43:27 -04:00
r'(?ims)class=.redirectText.>\s*<a\s*href=./wiki/([^"/]+)'
)
abbrs = ['etc', 'ca', 'cf', 'Co', 'Ltd', 'Inc', 'Mt', 'Mr', 'Mrs',
'Dr', 'Ms', 'Rev', 'Fr', 'St', 'Sgt', 'pron', 'approx', 'lit',
'syn', 'transl', 'sess', 'fl', 'Op', 'Dec', 'Brig', 'Gen'] \
+ list('ABCDEFGHIJKLMNOPQRSTUVWXYZ') \
+ list('abcdefghijklmnopqrstuvwxyz')
2009-01-19 11:47:05 -05:00
t_sentence = r'^.{5,}?(?<!\b%s)(?:\.(?=[\[ ][A-Z0-9]|\Z)|\Z)'
r_sentence = re.compile(t_sentence % r')(?<!\b'.join(abbrs))
def unescape(s):
2011-07-15 13:43:27 -04:00
s = s.replace('&gt;', '>')
s = s.replace('&lt;', '<')
s = s.replace('&amp;', '&')
s = s.replace('&#160;', ' ')
return s
def text(html):
2011-07-15 13:43:27 -04:00
html = r_tag.sub('', html)
html = r_whitespace.sub(' ', html)
return unescape(html).strip()
def wikipedia(term, last=False):
global wikiapi, wikiuri
url = wikiapi % term
bytes = web.get(url)
result = json.loads(bytes)
result = result['query']['search']
if len(result) <= 0:
return None
term = result[0]['title']
term = term.replace(' ', '_')
snippet = text(result[0]['snippet'])
return "%s - %s" % (snippet, wikiuri % term)
def wik(phenny, input):
2011-07-15 13:43:27 -04:00
origterm = input.groups()[1]
if not origterm:
return phenny.say('Perhaps you meant ".wik Zen"?')
2011-09-22 14:17:27 -04:00
origterm = origterm
2011-09-22 14:17:27 -04:00
term = urllib.parse.unquote(origterm)
2011-07-15 13:43:27 -04:00
term = term[0].upper() + term[1:]
term = term.replace(' ', '_')
2011-07-15 13:43:27 -04:00
try: result = wikipedia(term)
except IOError:
error = "Can't connect to en.wikipedia.org (%s)" % (wikiuri % term)
return phenny.say(error)
2011-07-15 13:43:27 -04:00
if result is not None:
phenny.say(result)
else: phenny.say('Can\'t find anything in Wikipedia for "%s".' % origterm)
wik.commands = ['wik']
wik.priority = 'high'
if __name__ == '__main__':
2011-09-22 14:17:27 -04:00
print(__doc__.strip())