From 7931fab14599b739c18c8f1ebcc24b75688dbc09 Mon Sep 17 00:00:00 2001
From: "Sean B. Palmer" <http://inamidst.com/sbp/>
Date: Thu, 21 Feb 2008 12:06:33 +0000
Subject: Phenny2, now being tested on Freenode as the main phenny.

---
 modules/wikipedia.py | 146 +++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 146 insertions(+)
 create mode 100644 modules/wikipedia.py

(limited to 'modules/wikipedia.py')

diff --git a/modules/wikipedia.py b/modules/wikipedia.py
new file mode 100644
index 0000000..893ecab
--- /dev/null
+++ b/modules/wikipedia.py
@@ -0,0 +1,146 @@
+#!/usr/bin/env python
+"""
+wikipedia.py - Phenny Wikipedia Module
+Copyright 2008, Sean B. Palmer, inamidst.com
+Licensed under the Eiffel Forum License 2.
+
+http://inamidst.com/phenny/
+"""
+
+import re, urllib
+import web
+
+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(
+   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'] \
+   + list('ABCDEFGHIJKLMNOPQRSTUVWXYZ') \
+   + list('abcdefghijklmnopqrstuvwxyz')
+t_sentence = r'^.{5,}?(?<!\b%s)(?:\.(?= [A-Z0-9]|\Z)|\Z)'
+r_sentence = re.compile(t_sentence % r')(?<!\b'.join(abbrs))
+
+def unescape(s): 
+   s = s.replace('&gt;', '>')
+   s = s.replace('&lt;', '<')
+   s = s.replace('&amp;', '&')
+   s = s.replace('&#160;', ' ')
+   return s
+
+def text(html): 
+   html = r_tag.sub('', html)
+   html = r_whitespace.sub(' ', html)
+   return unescape(html).strip()
+
+def search(term): 
+   try: import google
+   except ImportError, e: 
+      print e
+      return term
+
+   term = term.replace('_', ' ')
+   uri = google.google('site:en.wikipedia.org %s' % term)
+   if uri: 
+      return uri[len('http://en.wikipedia.org/wiki/'):]
+   else: return term
+
+def wikipedia(term, last=False): 
+   bytes = web.get(wikiuri % urllib.quote(term))
+   bytes = r_tr.sub('', bytes)
+
+   if not last: 
+      r = r_redirect.search(bytes[:4096])
+      if r: 
+         term = urllib.unquote(r.group(1))
+         return wikipedia(term, last=True)
+
+   paragraphs = r_paragraph.findall(bytes)
+
+   if not paragraphs: 
+      if not last: 
+         term = search(term)
+         return wikipedia(term, last=True)
+      return None
+
+   # Pre-process
+   paragraphs = [para for para in paragraphs 
+                 if (para and 'technical limitations' not in para 
+                          and 'window.showTocToggle' not in para 
+                          and 'Deletion_policy' not in para 
+                          and 'Template:AfD_footer' not in para 
+                          and not (para.startswith('<p><i>') and 
+                                   para.endswith('</i></p>'))
+                          and not 'disambiguation)"' in para) 
+                          and not '(images and media)' in para
+                          and not 'This article contains a' in para 
+                          and not 'id="coordinates"' in para]
+
+   for i, para in enumerate(paragraphs): 
+      para = para.replace('<sup>', '|')
+      para = para.replace('</sup>', '|')
+      paragraphs[i] = text(para).strip()
+
+   # Post-process
+   paragraphs = [para for para in paragraphs if 
+                 (para and not (para.endswith(':') and len(para) < 150))]
+
+   para = text(paragraphs[0])
+   m = r_sentence.match(para)
+
+   if not m: 
+      if not last: 
+         term = search(term)
+         return wikipedia(term, last=True)
+      return None
+   sentence = m.group(0)
+
+   maxlength = 275
+   if len(sentence) > maxlength: 
+      sentence = sentence[:maxlength]
+      words = sentence[:-5].split(' ')
+      words.pop()
+      sentence = ' '.join(words) + ' [...]'
+
+   if ((sentence == 'Wikipedia does not have an article with this exact name.')
+    or (sentence == 'Wikipedia does not have a page with this exact name.')): 
+      if not last: 
+         term = search(term)
+         return wikipedia(term, last=True)
+      return None
+
+   sentence = '"' + sentence.replace('"', "'") + '"'
+   return sentence + ' - ' + (wikiuri % term)
+
+def wik(phenny, input): 
+   origterm = input.groups()[1]
+   term = urllib.unquote(origterm)
+   if not term: 
+      return phenny.say(origin.sender, 'Maybe you meant ".wik Zen"?')
+
+   term = term[0].upper() + term[1:]
+   term = term.replace(' ', '_')
+
+   try: result = wikipedia(term)
+   except IOError: 
+      error = "Can't connect to en.wikipedia.org (%s)" % (wikiuri % term)
+      return phenny.say(error)
+
+   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__': 
+   print __doc__.strip()
-- 
cgit v1.2.3-1-g7c22