refactor mediawiki modules into unified library

master
mutantmonkey 2012-06-13 21:58:31 -07:00
parent 63c6adb316
commit ec32741826
5 changed files with 88 additions and 317 deletions

View File

@ -11,74 +11,34 @@ author: mutantmonkey <mutantmonkey@mutantmonkey.in>
""" """
import re, urllib.request, urllib.parse, urllib.error import re, urllib.request, urllib.parse, urllib.error
import web import wiki
import json
wikiapi = 'https://wiki.archlinux.org/api.php?action=query&list=search&srsearch=%s&limit=1&prop=snippet&format=json' wikiapi = 'https://wiki.archlinux.org/api.php?action=query&list=search&srsearch={0}&limit=1&prop=snippet&format=json'
wikiuri = 'https://wiki.archlinux.org/index.php/%s' wikiuri = 'https://wiki.archlinux.org/index.php/{0}'
wikisearch = 'https://wiki.archlinux.org/index.php/Special:Search?' \ wikisearch = 'https://wiki.archlinux.org/index.php/Special:Search?' \
+ 'search=%s&fulltext=Search' + 'search={0}&fulltext=Search'
r_tr = re.compile(r'(?ims)<tr[^>]*>.*?</tr>')
r_content = re.compile(r'(?ims)</p>\n</div>.*?<!-- end content -->')
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', 'transl', 'sess', 'fl', 'Op'] \
+ 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 archwiki(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 awik(phenny, input): def awik(phenny, input):
origterm = input.groups()[1] origterm = input.groups()[1]
if not origterm: if not origterm:
return phenny.say('Perhaps you meant ".awik dwm"?') return phenny.say('Perhaps you meant ".awik dwm"?')
origterm = origterm
term = urllib.parse.unquote(origterm) term = urllib.parse.unquote(origterm)
term = term[0].upper() + term[1:] term = term[0].upper() + term[1:]
term = term.replace(' ', '_') term = term.replace(' ', '_')
try: result = archwiki(term) w = wiki.Wiki(wikiapi, wikiuri, wikisearch)
try:
result = w.search(term)
except IOError: except IOError:
error = "Can't connect to wiki.archlinux.org (%s)" % (wikiuri % term) error = "Can't connect to wiki.archlinux.org ({0})".format(wikiuri.format(term))
return phenny.say(error) return phenny.say(error)
if result is not None: if result is not None:
phenny.say(result) phenny.say(result)
else: phenny.say('Can\'t find anything in the ArchWiki for "%s".' % origterm) else:
phenny.say('Can\'t find anything in the ArchWiki for "{0}".'.format(origterm))
awik.commands = ['awik'] awik.commands = ['awik']
awik.priority = 'high' awik.priority = 'high'

View File

@ -1,165 +0,0 @@
#!/usr/bin/env python
"""
uncyclopedia.py - Phenny Uncyclopedia Module
Copyright 2008-9, Sean B. Palmer, inamidst.com
Licensed under the Eiffel Forum License 2.
http://inamidst.com/phenny/
modified from Wikipedia module
author: mutantmonkey <mutantmonkey@mutantmonkey.in>
"""
import re, urllib.request, urllib.parse, urllib.error
import web
wikiuri = 'http://uncyclopedia.wikia.com/wiki/%s'
wikisearch = 'http://uncyclopedia.wikia.com/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', 'transl', 'sess', 'fl', 'Op'] \
+ 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: from . import search
except ImportError as e:
print(e)
return term
if not isinstance(term, str):
term = term.decode('utf-8')
term = term.replace('_', ' ')
try: uri = search.result('site:uncyclopedia.wikia.com %s' % term)
except IndexError: return term
if uri:
return uri[len('http://uncyclopedia.wikia.com/wiki/'):]
else: return term
def uncyclopedia(term, last=False):
global wikiuri
if not '%' in term:
if isinstance(term, str):
t = term
else: t = term
q = urllib.parse.quote(t)
u = wikiuri % q
bytes = web.get(u)
else: bytes = web.get(wikiuri % term)
bytes = r_tr.sub('', bytes)
if not last:
r = r_redirect.search(bytes[:4096])
if r:
term = urllib.parse.unquote(r.group(1))
return uncyclopedia(term, last=True)
paragraphs = r_paragraph.findall(bytes)
if not paragraphs:
if not last:
term = search(term)
return uncyclopedia(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
and not 'class="thumb' in para
and not 'There is currently no text in this page.' in para]
# and not 'style="display:none"' 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 uncyclopedia(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 (('using the Article Wizard if you wish' in sentence)
or ('or add a request for it' in sentence)):
if not last:
term = search(term)
return uncyclopedia(term, last=True)
return None
sentence = '"' + sentence.replace('"', "'") + '"'
return sentence + ' - ' + (wikiuri % term)
def uncyc(phenny, input):
origterm = input.groups()[1]
if not origterm:
return phenny.say('Perhaps you meant ".uncyc Zen"?')
origterm = origterm
term = urllib.parse.unquote(origterm)
term = term[0].upper() + term[1:]
term = term.replace(' ', '_')
try: result = uncyclopedia(term)
except IOError:
error = "Can't connect to uncyclopedia.wikia.com (%s)" % (wikiuri % term)
return phenny.say(error)
if result is not None:
phenny.say(result)
else: phenny.say('Can\'t find anything in Uncyclopedia for "%s".' % origterm)
uncyc.commands = ['uncyc']
uncyc.priority = 'high'
if __name__ == '__main__':
print(__doc__.strip())

View File

@ -11,73 +11,34 @@ author: mutantmonkey <mutantmonkey@mutantmonkey.in>
""" """
import re, urllib.request, urllib.parse, urllib.error import re, urllib.request, urllib.parse, urllib.error
import web import wiki
import json
wikiapi = 'https://vtluug.org/w/api.php?action=query&list=search&srsearch=%s&limit=1&prop=snippet&format=json' wikiapi = 'https://vtluug.org/w/api.php?action=query&list=search&srsearch={0}&limit=1&prop=snippet&format=json'
wikiuri = 'https://vtluug.org/wiki/%s' wikiuri = 'https://vtluug.org/wiki/{0}'
wikisearch = 'https://vtluug.org/wiki/Special:Search?' \ wikisearch = 'https://vtluug.org/wiki/Special:Search?' \
+ 'search=%s&fulltext=Search' + 'search={0}&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', 'transl', 'sess', 'fl', 'Op'] \
+ 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 vtluugwiki(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 vtluug(phenny, input): def vtluug(phenny, input):
origterm = input.groups()[1] origterm = input.groups()[1]
if not origterm: if not origterm:
return phenny.say('Perhaps you meant ".vtluug Zen"?') return phenny.say('Perhaps you meant ".vtluug VT-Wireless"?')
origterm = origterm
term = urllib.parse.unquote(origterm) term = urllib.parse.unquote(origterm)
term = term[0].upper() + term[1:] term = term[0].upper() + term[1:]
term = term.replace(' ', '_') term = term.replace(' ', '_')
try: result = vtluugwiki(term) w = wiki.Wiki(wikiapi, wikiuri, wikisearch)
try:
result = w.search(term)
except IOError: except IOError:
error = "Can't connect to vtluug.org (%s)" % (wikiuri % term) error = "Can't connect to vtluug.org ({0})".format(wikiuri.format(term))
return phenny.say(error) return phenny.say(error)
if result is not None: if result is not None:
phenny.say(result) phenny.say(result)
else: phenny.say('Can\'t find anything in the VTLUUG Wiki for "%s".' % origterm) else:
phenny.say('Can\'t find anything in the VTLUUG Wiki for "{0}".'.format(origterm))
vtluug.commands = ['vtluug'] vtluug.commands = ['vtluug']
vtluug.priority = 'high' vtluug.priority = 'high'

View File

@ -8,73 +8,34 @@ http://inamidst.com/phenny/
""" """
import re, urllib.request, urllib.parse, urllib.error, gzip, io import re, urllib.request, urllib.parse, urllib.error, gzip, io
import web import wiki
import json
wikiapi = 'http://en.wikipedia.org/w/api.php?action=query&list=search&srsearch=%s&limit=1&prop=snippet&format=json' wikiapi = 'http://en.wikipedia.org/w/api.php?action=query&list=search&srsearch={0}&limit=1&prop=snippet&format=json'
wikiuri = 'http://en.wikipedia.org/wiki/%s' wikiuri = 'http://en.wikipedia.org/wiki/{0}'
wikisearch = 'http://en.wikipedia.org/wiki/Special:Search?' \ wikisearch = 'http://en.wikipedia.org/wiki/Special:Search?' \
+ 'search=%s&fulltext=Search' + 'search={0}&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', 'transl', 'sess', 'fl', 'Op', 'Dec', 'Brig', 'Gen'] \
+ 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 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): def wik(phenny, input):
origterm = input.groups()[1] origterm = input.groups()[1]
if not origterm: if not origterm:
return phenny.say('Perhaps you meant ".wik Zen"?') return phenny.say('Perhaps you meant ".wik Zen"?')
origterm = origterm
term = urllib.parse.unquote(origterm) term = urllib.parse.unquote(origterm)
term = term[0].upper() + term[1:] term = term[0].upper() + term[1:]
term = term.replace(' ', '_') term = term.replace(' ', '_')
try: result = wikipedia(term) w = wiki.Wiki(wikiapi, wikiuri, wikisearch)
try:
result = w.search(term)
except IOError: except IOError:
error = "Can't connect to en.wikipedia.org (%s)" % (wikiuri % term) error = "Can't connect to en.wikipedia.org ({0})".format(wikiuri.format(term))
return phenny.say(error) return phenny.say(error)
if result is not None: if result is not None:
phenny.say(result) phenny.say(result)
else: phenny.say('Can\'t find anything in Wikipedia for "%s".' % origterm) else:
phenny.say('Can\'t find anything in Wikipedia for "{0}".'.format(origterm))
wik.commands = ['wik'] wik.commands = ['wik']
wik.priority = 'high' wik.priority = 'high'

54
wiki.py Normal file
View File

@ -0,0 +1,54 @@
import json
import re
import web
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', 'transl', 'sess', 'fl', 'Op', 'Dec', 'Brig', 'Gen'] \
+ 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))
class Wiki(object):
def __init__(self, api, url, searchurl=""):
self.api = api
self.url = url
self.searchurl = searchurl
@staticmethod
def unescape(s):
s = s.replace('&gt;', '>')
s = s.replace('&lt;', '<')
s = s.replace('&amp;', '&')
s = s.replace('&#160;', ' ')
return s
@staticmethod
def text(html):
html = r_tag.sub('', html)
html = r_whitespace.sub(' ', html)
return Wiki.unescape(html).strip()
def search(self, term, last=False):
url = self.api.format(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 = self.text(result[0]['snippet'])
return "{0} - {1}".format(snippet, self.url.format(term))