User:WindBOT/Source: Difference between revisions

From the Portal Wiki
Jump to navigation Jump to search
No edit summary
 
(4 intermediate revisions by one other user not shown)
Line 6: Line 6:
* I am a [http://python.org/ Python] bot. That means you need [http://www.python.org/download/releases/2.6.6/ Python 2.6]. On Ubuntu:
* I am a [http://python.org/ Python] bot. That means you need [http://www.python.org/download/releases/2.6.6/ Python 2.6]. On Ubuntu:
<pre>$ sudo apt-get install python2.6</pre>
<pre>$ sudo apt-get install python2.6</pre>
* I also need [http://upload.gaiatools.com/files/wikitools_mod.tar_0.gz a modified version of wikitools] to use the MediaWiki API. Comes with a separate hand-crafted module called wikiUpload to allow file uploading without MediaWiki 1.16. On Ubuntu:
* I also need [http://upload.gaiatools.com/files/wikitools_mod_5.7z a modified version of wikitools] to use the MediaWiki API. Comes with a separate hand-crafted module called wikiUpload to allow file uploading without MediaWiki 1.16. On Ubuntu:
<pre>$ wget 'http://upload.gaiatools.com/files/wikitools_mod.tar_0.gz'
<pre>$ wget 'http://upload.gaiatools.com/files/wikitools_mod_5.7z'
$ tar xfv wikitools*.tar.gz && rm wikitools*.tar.gz
$ p7zip -d wikitools*.7z && wikitools*.7z
$ mv wikiUpload.py MultipartPostHandler.py `dirname "/path/to/bot.py"`
$ mv wikiUpload.py MultipartPostHandler.py `dirname "/path/to/bot.py"`
$ cd wikitools*
$ cd wikitools*
Line 14: Line 14:
* (''Optional'') [http://cpulimit.sourceforge.net/ cpulimit] can be used to limit the bot's often-high CPU usage when applying filters to a lengthy page, to ensure sufficient resources to other programs running concurrently on the server. This does slow the bot down but makes it act nicer to others, which is welcome on a shared host. On Ubuntu:
* (''Optional'') [http://cpulimit.sourceforge.net/ cpulimit] can be used to limit the bot's often-high CPU usage when applying filters to a lengthy page, to ensure sufficient resources to other programs running concurrently on the server. This does slow the bot down but makes it act nicer to others, which is welcome on a shared host. On Ubuntu:
<pre>$ sudo apt-get install cpulimit</pre>
<pre>$ sudo apt-get install cpulimit</pre>
* (''Optional'') [http://git.optf2.com/steamodd.git steamodd] can be used to communicate with the TF2 [[WebAPI|Web API]]. On Ubuntu:
<pre>$ sudo apt-get install git
$ git clone http://git.optf2.com/steamodd.git
$ cd steamodd
$ sudo python2.6 setup.py install</pre>
* (''Optional'') [http://pmt.sourceforge.net/pngcrush/ pngcrush] can be used to crush PNG images in file filters. Without it, images will not be considered for crushing. On Ubuntu:
* (''Optional'') [http://pmt.sourceforge.net/pngcrush/ pngcrush] can be used to crush PNG images in file filters. Without it, images will not be considered for crushing. On Ubuntu:
<pre>$ sudo apt-get install pngcrush</pre>
<pre>$ sudo apt-get install pngcrush</pre>

Latest revision as of 05:49, 25 March 2013

I am open-source, licensed under the GPL. That means you're free to see my code, run it, modify it, redistribute it, and even sell paid copies of me, as long as you keep the source open and report back all the modifications you've made, if any.

Installation

Dependencies

$ sudo apt-get install python2.6
  • I also need a modified version of wikitools to use the MediaWiki API. Comes with a separate hand-crafted module called wikiUpload to allow file uploading without MediaWiki 1.16. On Ubuntu:
$ wget 'http://upload.gaiatools.com/files/wikitools_mod_5.7z'
$ p7zip -d wikitools*.7z && wikitools*.7z
$ mv wikiUpload.py MultipartPostHandler.py `dirname "/path/to/bot.py"`
$ cd wikitools*
$ sudo python2.6 setup.py install
  • (Optional) cpulimit can be used to limit the bot's often-high CPU usage when applying filters to a lengthy page, to ensure sufficient resources to other programs running concurrently on the server. This does slow the bot down but makes it act nicer to others, which is welcome on a shared host. On Ubuntu:
$ sudo apt-get install cpulimit
  • (Optional) pngcrush can be used to crush PNG images in file filters. Without it, images will not be considered for crushing. On Ubuntu:
$ sudo apt-get install pngcrush

That's it!

File structure

  • botConfig.py: The bot's configuration. A sample configuration file is given below.
  • bot.py: The bot's core. Full source is given below. Make sure to make it executable:
chmod +x bot.py

Both files have to be in the same directory. To run the bot, assuming Python is in your path and that you are in the bot's source directory:

./bot.py

Files

botConfig.py (sample)

# -*- coding: utf-8 -*-
config = {
	'api': 'http://theportalwiki.com/w/api.php', # API URL
	'steamAPI': 'PootAPIKeyHeer', # Steam API key
	'username': 'MyCoolBot', # Username
	'password': 'LeetPassword0MG', # Password
	'maxrequests': 16, # Max PageRequests to process per run
	'rcidrate': 50, # Edit RCID every n edits
	'freshnessThreshold': 300, # In seconds
	'pagePasses': 8, # Maximum number of parsing/filtering passes
	'filterPasses': 64, # Maximum number of times to run a fitler on a filtering pass
	'tempPrefix': 'pwiki', # Prefix for temporary filenames
	'pages': {
		'filters': 'User:WindBOT/Filters', # Filters page
		'blacklist': 'User:WindBOT/Blacklist', # Blacklist
		'pagerequests': 'User:WindBOT/PageRequests', # PageRequests
		'pagerequestsforce': 'User:WindBOT/PageRequestsForce', # PageRequests bypassing blacklist
		'rcid': 'User:WindBOT/RCID', # RCID page
		'editcount': 'User:WindBOT/EditCount' # Edit count page
	}
}

bot.py

#!/usr/bin/python -OO
# -*- coding: utf-8 -*-
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program.  If not, see <http://www.gnu.org/licenses/>.

import sys
import os
import time, datetime
import re
import hashlib
import urllib2
import tempfile
import traceback
import threading
import random
import subprocess
import cStringIO as StringIO
import shutil
import wikitools
import wikiUpload
import feedparser
try:
	import steam
except:
	steam = None
from wikiUpload import wikiUploader

from botConfig import config
if steam is not None and 'steamAPI' in config:
	steam.set_api_key(config['steamAPI'])
config['runtime'] = {
	'rcid': -1,
	'onlinercid': -1,
	'wiki': None,
	'edits': 0,
	'regexes': {},
	'pages': {},
	'uploader': wikiUploader(config['username'], config['password'], config['api'])
}

def u(s):
	if type(s) is type(u''):
		return s
	if type(s) is type(''):
		try:
			return unicode(s)
		except:
			try:
				return unicode(s.decode('utf8'))
			except:
				try:
					return unicode(s.decode('windows-1252'))
				except:
					return unicode(s, errors='ignore')
	try:
		return unicode(s)
	except:
		try:
			return u(str(s))
		except:
			return s
class curry:
	def __init__(self, func, *args, **kwargs):
		self.func = func
		self.pending = args[:]
		self.kwargs = kwargs
	def __str__(self):
		if u(self.func) == u(regSub):
			return 'Regex' + u(self.pending)
		return u'<Curry of ' + u(self.func) + u'; args = ' + u(self.pending) + u'; kwargs = ' + u(self.kwargs) + u'>'
	def __repr__(self):
		return self.__str__()
	def __call__(self, *args, **kwargs):
		if kwargs and self.kwargs:
			kw = self.kwargs.copy()
			kw.update(kwargs)
		else:
			kw = kwargs or self.kwargs
		return self.func(*(self.pending + args), **kw)
class BatchScheduler:
	def __init__(self, concurrency=16):
		self.concurrency = 16
		self.tasks = []
	def schedule(self, target, *args, **kwargs):
		self.tasks.append((target, args, kwargs))
	def execute(self):
		while len(self.tasks):
			pool = []
			numThreads = min(self.concurrency, len(self.tasks))
			for task in range(numThreads):
				task = self.tasks[task]
				t = threading.Thread(target=task[0], args=task[1], kwargs=task[2])
				t.start()
				pool.append(t)
			for t in pool:
				t.join()
			self.tasks = self.tasks[numThreads:]
def getTempFilename(extension=None):
	global config
	if extension is None:
		f = tempfile.mkstemp(prefix=config['tempPrefix'])
	else:
		f = tempfile.mkstemp(suffix=u'.' + u(extension), prefix=config['tempPrefix'])
	os.close(f[0]) # Damn you Python I just want a filename
	return u(f[1])

def wiki():
	global config
	if config['runtime']['wiki'] is None:
		config['runtime']['wiki'] = wikitools.wiki.Wiki(config['api'])
		print 'Logging in as', config['username'], '...'
		config['runtime']['wiki'].login(config['username'], config['password'])
		try:
			config['runtime']['onlinercid'] = int(u(wikitools.page.Page(wiki(), config['pages']['rcid']).getWikiText()).strip())
			config['runtime']['rcid'] = config['runtime']['onlinercid']
		except:
			error('Couldn\'t read RCID.')
		print 'Logged in.'
	return config['runtime']['wiki']
def page(p):
	global config
	if type(p) in (type(''), type(u'')):
		p = u(p)
		if p not in config['runtime']['pages']:
			config['runtime']['pages'][p] = wikitools.page.Page(wiki(), p, followRedir=False)
		return config['runtime']['pages'][p]
	# Else, it is a page object
	title = u(p.title)
	if title not in config['runtime']['pages']:
		config['runtime']['pages'][title] = p
	return config['runtime']['pages'][title]
def getSummary(summary):
	summary = u(summary)
	while len(summary) > 250:
		if summary.find(u' ') == -1:
			summary = summary[:summary.rfind(u' ')] + u'...'
		else:
			summary = summary[:247] + u'...'
	return summary
def editPage(p, content, summary=u'', minor=True, bot=True, nocreate=True):
	global config
	summary = getSummary(summary)
	p = page(p)
	try:
		print 'Editing', p.title, 'with summary', summary
	except:
		pass
	try:
		if nocreate:
			if minor:
				result = p.edit(u(content), summary=summary, minor=True, bot=bot, nocreate=nocreate)
			else:
				result = p.edit(u(content), summary=summary, notminor=True, bot=bot, nocreate=nocreate)
		else:
			if minor:
				result = p.edit(u(content), summary=summary, minor=True, bot=bot)
			else:
				result = p.edit(u(content), summary=summary, notminor=True, bot=bot)
	except:
		warning('Couldn\'t edit', p.title)
		return None
	try:
		if result['edit']['result']:
			config['runtime']['edits'] += 1
	except:
		warning('Couldn\'t edit', p.title)
	return result
def deletePage(p, summary=False):
	if summary:
		summary = getSummary(summary)
	return page(p).delete(summary)
def uploadFile(filename, destfile, pagecontent='', license='', overwrite=False, reupload=False):
	global config
	return config['runtime']['uploader'].upload(filename, destfile, pagecontent, license, overwrite=overwrite, reupload=reupload)
def updateRCID():
	if abs(config['runtime']['rcid'] - config['runtime']['onlinercid']) >= config['rcidrate']:
		print 'Updating last RCID...'
		try:
			editPage(config['pages']['rcid'], config['runtime']['rcid'], summary=u'Updated Recent Changes log position to ' + u(config['runtime']['rcid']))
			config['runtime']['onlinercid'] = config['runtime']['rcid']
		except:
			warning('Couldn\'t update RCID.')
def updateEditCount(force=False):
	global config
	if not config['runtime']['edits']:
		return
	if not force and random.randint(0, 40) != 7:
		return
	try:
		editPage(config['pages']['editcount'], int(wikitools.api.APIRequest(wiki(), {
			'action': 'query',
			'list': 'users',
			'usprop': 'editcount',
			'ususers': config['username']
		}).query(querycontinue=False)['query']['users'][0]['editcount']) + 1, summary=u'Updated edit count.')
		config['runtime']['edits'] = 0
	except:
		warning('Couldn\'t update edit count.')

# Because SOME LIBRARY will not use singletons, this has to be done at the bot level
# rather than the individual filter level to avoid loading the damn thing twice.
steamGameSchemas = {}
def steamGetGameSchema(game):
	global steamGameSchemas
	if steam is None:
		return None
	if game not in steamGameSchemas:
		steamGameSchemas[game] = game.item_schema()
	return steamGameSchemas[game]
steamGameAssets = {}
def steamGetGameAssets(game):
	global steamGameAssets
	if steam is None:
		return None
	if game not in steamGameAssets:
		steamGameAssets[game] = game.assets()
	return steamGameAssets[game]

def compileRegex(regex, flags=re.IGNORECASE):
	global config
	regex = u(regex)
	if regex in config['runtime']['regexes']:
		return config['runtime']['regexes'][regex]
	config['runtime']['regexes'][regex] = re.compile(regex, flags)
	return config['runtime']['regexes'][regex]

def warning(*info):
	s = []
	print info
	import traceback
	traceback.print_exc()
def error(*info):
	warning(*info)
	sys.exit(1)

def setFilterName(f, name):
	name = u(name)
	f.__unicode__ = lambda: name
	f.__str__ = lambda: name.encode('utf8')
	f.filterName = name
	return f
class link:
	def __init__(self, content):
		content = u(content)
		self.joined = False
		self.setBody(content)
		self.setType(u'unknown')
		self.setLabel(None)
		self.setLink(u'')
		self.anchor = None
		self.joined = False
		if len(content) > 2:
			if content[:2] == u'[[' and content[-2:] == u']]':
				split = content[2:-2].split(u'|')
				if len(split) in (1, 2):
					self.setType(u'internal')
					lnk = split[0]
					if lnk.find(u':') == -1:
						lnk = lnk.replace(u'_', u' ')
					anchor = None
					if lnk.find(u'#') != -1:
						lnk, anchor = lnk.split(u'#', 1)
						self.setAnchor(anchor)
					self.setLink(lnk)
					if len(split) == 2:
						self.setLabel(split[1])
					else:
						self.setLabel(split[0])
						self.joined = anchor is None
			elif content[0] == u'[' and content[-1] == u']':
				split = content[1:-1].split(u' ', 1)
				self.setType(u'external')
				self.setLink(split[0])
				if len(split) == 2:
					self.setLabel(split[1])
				else:
					self.setLabel(None)
	def getType(self):
		return u(self.kind)
	def getBody(self):
		return u(self.body)
	def getLink(self, withAnchor=False):
		if withAnchor and self.getAnchor() is not None:
			return u(self.link) + u'#' + self.getAnchor()
		return u(self.link)
	def getAnchor(self):
		return self.anchor
	def getLabel(self):
		if self.label is None:
			return None
		if self.joined:
			return self.getLink()
		return u(self.label)
	def setType(self, kind):
		self.kind = u(kind)
	def setBody(self, body):
		self.body = u(body)
	def setLink(self, link):
		link = u(link)
		if self.getType() == u'internal' and link.find(u'#') != -1:
			link, anchor = link.split(u'#', 1)
			self.setAnchor(anchor)
		self.link = link
		if self.joined:
			self.label = u(link)
	replaceDots = compileRegex(r'(?:\.[a-f\d][a-f\d])+')
	def _replaceDots(self, g):
		s = ''
		g = g.group(0)
		for i in xrange(0, len(g), 3):
			s += chr(int(g[i + 1:i + 3], 16))
		return s.decode('utf8')
	def setAnchor(self, anchor):
		if self.getType() == u'internal':
			u(anchor).replace(u'_', u' ')
			try:
				anchor = link.replaceDots.sub(self._replaceDots, anchor)
			except:
				pass
			self.anchor = anchor
	def setLabel(self, label):
		if label is None:
			self.label = None
		else:
			self.label = u(label)
		if self.joined:
			self.link = u(label)
	def __str__(self):
		return self.__unicode__()
	def __repr__(self):
		return u'<Link-' + self.getType() + u': ' + self.__unicode__() + u'>'
	def __unicode__(self):
		label = self.getLabel()
		tmpLink = self.getLink(withAnchor=True)
		if self.getType() == u'internal':
			tmpLink2 = tmpLink.replace(u'_', u' ')
			if label in (tmpLink2, tmpLink) or (label and tmpLink and (label[0].lower() == tmpLink[0].lower() and tmpLink[1:] == label[1:]) or (label[0].lower() == tmpLink2[0].lower() and tmpLink2[1:] == label[1:])):
				return u'[[' + label + u']]'
			elif tmpLink and label and len(label) > len(tmpLink) and (label.lower().find(tmpLink2.lower()) == 0 or label.lower().find(tmpLink.lower()) == 0):
				index = max(label.lower().find(tmpLink2.lower()), label.lower().find(tmpLink.lower()))
				badchars = (u' ', u'_')
				nobadchars = True
				for c in badchars:
					if label[:index].find(c) != -1 or label[index+len(tmpLink):].find(c) != -1:
						nobadchars = False
				if nobadchars:
					return label[:index] + u(link(u'[[' + tmpLink + u'|' + label[index:index+len(tmpLink)] + u']]')) + label[index+len(tmpLink):]
			return u'[[' + tmpLink + u'|' + label + u']]'
		if self.getType() == u'external':
			if label is None:
				return u'[' + tmpLink + u']'
			return u'[' + tmpLink + u' ' + label + u']'
		return self.getBody()
class template:
	maxInlineParams = 1
	def __init__(self, content):
		content = u(content)
		self.changed = False
		self.content = content
		self.name = None
		self.order = None
		self.links = []
		self.params = []
		self.paramNum = 0
		self.indentation = {}
		self.defaultIndent = 0
		if len(content) > 4 and content[:2] == '{{' and content[-2:] == '}}':
			innerRegex = compileRegex(r'\s*\|\s*')
			itemRegex = compileRegex(r'^(\S[^=]*?)\s*=\s*([\s\S]*?)$')
			content = content[2:-2]
			content, self.links = linkExtract(content)
			innerStuff = innerRegex.split(content)
			if innerStuff[0][:9].lower() == 'template:':
				innerStuff[0] = innerStuff[0][9:]
			self.name = u(innerStuff[0][0].upper() + innerStuff[0][1:]).replace(u'_', u' ').strip()
			innerStuff = innerStuff[1:]
			for i in innerStuff:
				i = linkRestore(i.strip(), self.links, restore=True)
				itemRes = itemRegex.search(i)
				if itemRes:
					self.params.append((u(itemRes.group(1)).lower(), u(itemRes.group(2))))
				else:
					self.paramNum += 1
					self.params.append((u(self.paramNum), i))
		self.originalContent = self.content
		self.originalParams = self.params[:]
		self.originalName = self.name
		self.forceindent = False
	def indentationMatters(self, doesitmatter=True):
		self.forceindent = doesitmatter
	def getName(self):
		return self.name
	def setName(self, name):
		self.name = u(name).replace(u'_', u' ')
		self.changed = self.changed or self.name != self.originalName
	def getParam(self, key):
		key = u(key).lower()
		for k, v in self.params:
			if k == key:
				return v
		return None
	def delParam(self, *indexes):
		for index in indexes:
			index = u(index)
			isNumber = self.isInt(index)
			for p in range(len(self.params)):
				k, v = self.params[p]
				if k == index:
					self.changed = True
					self.params = self.params[:p] + self.params[p+1:]
					if isNumber:
						if int(index) == self.paramNum:
							self.paramNum -= 1
					break
	def setParam(self, index=None, value=u''):
		if value is None:
			return self.delParam(index)
		if index is None:
			index = u(self.paramNum)
		else:
			index = u(index).lower()
		isNumber = self.isInt(index)
		value = u(value)
		hasChanged = False
		for p in range(len(self.params)):
			k, v = self.params[p]
			if k == index:
				self.changed = self.changed or v != value
				self.params[p] = (k, value)
				hasChanged = True
				break
		if not hasChanged:
			if isNumber:
				while self.numParam < int(index) - 1:
					self.appendParam(u'')
			self.params.append((index, value))
			self.changed = True
	def appendParam(self, value=u''):
		self.paramNum += 1
		self.params.append((u(self.paramNum), value))
	def setPreferedIndentation(self, index, indent=0):
		self.indentation[u(index)] = indent
		self.changed = self.changed or self.forceindent
	def setDefaultIndentation(self, indent=0):
		self.defaultIndent = indent
		self.changed = self.changed or self.forceindent
	def setPreferedOrder(self, order=None):
		order2 = []
		for o in order:
			order2.append(u(o))
		self.order = order2
		oldParams = self.params[:]
		self.changed = self.changed or self.fixOrder() == oldParams
	def renameParam(self, oldkey, newkey):
		oldkey, newkey = u(oldkey).lower(), u(newkey).lower()
		if oldkey == newkey:
			return
		for p in range(len(self.params)):
			k, v = self.params[p]
			if k == oldkey:
				self.params[p] = (newkey, v)
				self.changed = True
				break
	def fixOrder(self):
		if self.order is None:
			return self.params
		newParams = []
		doneParams = []
		for k in self.order:
			k = u(k)
			if self.getParam(k) is not None:
				newParams.append((k, self.getParam(k)))
				doneParams.append(k)
		for k, v in self.params:
			if k not in doneParams:
				doneParams.append(k)
				newParams.append((k, v))
		self.params = newParams
		return self.params
	def defined(self):
		return self.name is not None
	def isInt(self, i):
		try:
			return u(int(i)) == u(i) and int(i) > 0
		except:
			return False
	def __str__(self):
		return self.__unicode__()
	def __repr__(self):
		return u'<Template-' + self.getName() + u': ' + self.__unicode__() + u'>'
	def __unicode__(self):
		if not self.defined():
			return u''
		if not self.changed:
			return self.originalContent
		self.fixOrder()
		params = [self.name]
		indentMode = len(self.params) > template.maxInlineParams or self.forceindent
		maxIndent = 0
		if indentMode:
			for k, v in self.params:
				l = len(k) + self.defaultIndent
				if k in self.indentation:
					l = len(k) + self.indentation[k]
				if not self.isInt(k) and l > maxIndent:
					maxIndent = l
		numParam = 1
		for k, v in self.params:
			indent = self.defaultIndent
			if k in self.indentation and indentMode:
				indent = self.indentation[k]
			try:
				isNumber = u(int(index)) == u(index) and int(index) > 0
			except:
				isNumber = False
			if indentMode:
				key = u' ' * indent + u'| '
				addKey = True
				if self.isInt(k):
					if int(k) == numParam:
						addKey = False
					numParam += 1
				if addKey:
					key += k + (u' ' * max(0, maxIndent - len(k) - indent)) + u' = '
			else:
				key = u''
				addKey = True
				if self.isInt(k):
					if int(k) == numParam:
						addKey = False
					numParam += 1
				if addKey:
					key += k + u' = '
			params.append(key + v)
		if indentMode:
			params = u'\n'.join(params) + u'\n'
		else:
			params = u' | '.join(params)
		return u'{{' + params + u'}}'
def linkExtract(content):
	content = u(content)
	links1 = compileRegex(r'\[\[([^\[\]]+)\]\]')
	links2 = compileRegex(r'\[([^\[\]]+)\](?!\])')
	linkcount = 0
	linklist = []
	res = links1.search(content)
	while res:
		linklist.append(link(res.group()))
		content = content[:res.start()] + u'~!~!~!~OMGLINK-' + u(linkcount) + u'~!~!~!~' + content[res.end():]
		linkcount += 1
		res = links1.search(content)
	res = links2.search(content)
	while res:
		linklist.append(link(res.group()))
		content = content[:res.start()] + u'~!~!~!~OMGLINK-' + u(linkcount) + u'~!~!~!~' + content[res.end():]
		linkcount += 1
		res = links2.search(content)
	return content, linklist
def templateExtract(content):
	content = u(content)
	templatesR = compileRegex(r'\{\{([^\{\}]+)\}\}')
	templatecount = 0
	templatelist = []
	res = templatesR.search(content)
	while res:
		templatelist.append(template(res.group()))
		content = content[:res.start()] + u'~!~!~!~OMGTEMPLATE-' + u(templatecount) + u'~!~!~!~' + content[res.end():]
		templatecount += 1
		res = templatesR.search(content)
	return content, templatelist
def blankAround(content, search, repl=u''):
	content = u(content)
	search = u(search)
	repl = u(repl)
	blank = compileRegex(u'(\\s*)' + u(re.escape(search)) + u'(\\s*)')
	res = blank.search(content)
	if not res:
		return content.replace(search, repl)
	if u(res.group(0)) == content:
		return repl
	if len(res.group(1)) < len(res.group(2)):
		return content[:res.end(1)] + content[res.end(2):]
	else:
		return content[:res.start()] + content[res.start(2):]
def linkRestore(content, links=[], restore=False):
	linklist=links[:]
	linkcount = len(linklist)
	i = 0
	linklist.reverse()
	for l in linklist:
		i += 1
		if l is None:
			content = blankAround(content, u'~!~!~!~OMGLINK-' + u(linkcount - i) + u'~!~!~!~', u'')
		else:
			if restore:
				l = l.getBody()
			content = content.replace(u'~!~!~!~OMGLINK-' + u(linkcount - i) + u'~!~!~!~', u(l))
	return content
def templateRestore(content, templatelist=[]):
	templatecount = len(templatelist)
	i = 0
	templatelist.reverse()
	for t in templatelist:
		i += 1
		if t is None:
			content = blankAround(content, u'~!~!~!~OMGTEMPLATE-' + u(templatecount - i) + u'~!~!~!~', u'')
		else:
			content = content.replace(u'~!~!~!~OMGTEMPLATE-' + u(templatecount - i) + u'~!~!~!~', u(t))
	return content
def safeContent(content):
	safelist = []
	tags = compileRegex(r'<(?:ref|gallery|pre|code)[^<>]*>[\S\s]*?</(?:ref|gallery|pre|code)>', re.IGNORECASE | re.MULTILINE)
	comments = compileRegex(r'<!--[\S\s]*?-->')
	tagcount = 0
	res = tags.search(content)
	if not res:
		res = comments.search(content)
	while res:
		safelist.append(('~!~!~!~OMGTAG-' + u(tagcount) + u'~!~!~!~', u(res.group())))
		content = content[:res.start()] + u'~!~!~!~OMGTAG-' + u(tagcount) + u'~!~!~!~' + content[res.end():]
		tagcount += 1
		res = tags.search(content)
		if not res:
			res = comments.search(content)
	return content, safelist
def safeContentRestore(content, safelist=[]):
	safelist.reverse()
	for s in safelist:
		content = content.replace(s[0], s[1])
	return content
def regReplaceCallBack(sub, match):
	groupcount = 1
	for g in match.groups():
		if g is not None:
			sub = sub.replace(u'$' + u(groupcount), g)
		else:
			sub = sub.replace(u'$' + u(groupcount), u'')
		groupcount += 1
	return sub
def regSub(regexes, content, **kwargs):
	content = u(content)
	for regex in regexes.keys():
		if type(regex) in (type(()), type([])):
			compiled = compileRegex(u(regex[0]), regex[1])
		else:
			compiled = compileRegex(u(regex), re.IGNORECASE | re.DOTALL | re.MULTILINE)
		callback = curry(regReplaceCallBack, u(regexes[regex]))
		oldcontent = u''
		while content != oldcontent:
			oldcontent = content
			content = compiled.sub(callback, content)
	return u(content)
def dumbReplacement(strings, content, **kwargs):
	content = u(content)
	for s in strings.keys():
		content = content.replace(u(s), u(strings[s]))
	return content
def filterEnabled(f, **kwargs):
	if type(f) is not type(()):
		return True
	if len(f) < 2:
		return True
	if type(f[1]) is not type({}):
		return True
	article = None
	if 'article' in kwargs.keys():
		article = kwargs['article']
		if article is None:
			return True
		if type(article) not in (type(u''), type('')):
			article = article.title
	if article is None:
		return True
	if 'languageBlacklist' in f[1].keys():
		for i in f[1]['languageBlacklist']:
			if compileRegex(u'/' + u(i) + u'$').search(u(article)):
				return False
		return True
	if 'languageWhitelist' in f[1].keys():
		for i in f[1]['languageWhitelist']:
			if compileRegex(u'/' + u(i) + u'$').search(u(article)):
				return True
		return False
	if 'language' in f[1].keys():
		return compileRegex(u'/' + u(f[1]['language']) + u'$').search(u(article))
	return True
scheduledTasks = []
def scheduleTask(task, oneinevery):
	global scheduledTasks
	result = random.randint(0, oneinevery-1)
	print 'Task:', task, '; result:', result
	if not result:
		scheduledTasks.append(task)
def runScheduledTasks():
	global scheduledTasks
	if not len(scheduledTasks):
		print 'No tasks scheduled.'
		return
	print 'Running scheduled tasks...'
	for t in scheduledTasks:
		print 'Running task:', t
		try:
			t()
			print 'End of task:', t
		except:
			print 'Error while executing task:', t
def sFilter(filters, content, returnActive=False, **kwargs):
	content = u(content)
	lenfilters = len(filters)
	if not lenfilters:
		if returnActive:
			return content, []
		return content
	filtercount = 0
	activeFilters = []
	for f in filters:
		if not filterEnabled(f, **kwargs):
			continue
		if type(f) is type(()):
			f, params = f
		filtercount += 1
		#print 'Filter', f, '(', filtercount, '/', lenfilters, ')'
		loopTimes = 0
		beforeFilter = u''
		while not loopTimes or beforeFilter != content:
			loopTimes += 1
			if loopTimes >= config['filterPasses']:
				print 'Warning: More than', config['filterPasses'], 'loops with filter', u(f)
				break
			beforeFilter = content
			content = u(f(content, **kwargs))
			if content != beforeFilter and f not in activeFilters:
				activeFilters.append(f)
	if returnActive:
		return content, activeFilters
	return content
def linkFilter(filters, linklist, returnActive=False, **kwargs):
	activeFilters = []
	for f in filters:
		if not filterEnabled(f, **kwargs):
			continue
		if type(f) is type(()):
			f, params = f
		for i in range(len(linklist)):
			if linklist[i] is not None and isinstance(linklist[i], link):
				oldLink = u(linklist[i])
				linklist[i] = f(linklist[i], **kwargs)
				if oldLink != u(linklist[i]) and f not in activeFilters:
					activeFilters.append(f)
	if returnActive:
		return linklist, activeFilters
	return linklist
def templateFilter(filters, templatelist, returnActive=False, **kwargs):
	activeFilters = []
	for f in filters:
		if not filterEnabled(f, **kwargs):
			continue
		if type(f) is type(()):
			f, params = f
		for i in range(len(templatelist)):
			if templatelist[i] is not None and isinstance(templatelist[i], template):
				oldTemplate = u(templatelist[i])
				templatelist[i] = f(templatelist[i], **kwargs)
				if oldTemplate != u(templatelist[i]) and f not in activeFilters:
					activeFilters.append(f)
	if returnActive:
		return templatelist, activeFilters
	return templatelist
def linkTextFilter(subfilters, l, linksafe=False, **kwargs):
	if l.getType() == u'internal' and l.getLink().find(u':') == -1 and pageFilter(l.getLink()):
		if linksafe:
			l.setLink(sFilter(subfilters, l.getLink(), **kwargs))
		if l.getLabel().find(u':') == -1:
			l.setLabel(sFilter(subfilters, l.getLabel(), **kwargs))
	return l
def linkDomainSub(fromDomain, toDomain, link, **kwargs):
	domainR = compileRegex(r'^(https?://(?:[-\w]+\.)*)' + u(re.escape(fromDomain)) + r'(\S+)$')
	toDomain = u(toDomain)
	if link.getType() == 'external':
		linkInfo = domainR.search(link.getLink())
		if linkInfo:
			link.setLink(u(linkInfo.group(1)) + toDomain + u(linkInfo.group(2)))
	return link
def linkDomainFilter(fromDomain, toDomain):
	return curry(linkDomainSub, fromDomain, toDomain)
def regexes(rs):
	return curry(regSub, rs)
def regex(reg, replace):
	return regexes({reg: replace})
def dumbReplaces(rs):
	return setFilterName(curry(dumbReplacement, rs), u'DumbReplacements(' + u(rs) + u')')
def dumbReplace(subject, replacement):
	return setFilterName(dumbReplaces({subject: replacement}), u'DumbReplacement(' + u(subject) + u' \u2192 ' + u(replacement) + u')')
def wordRegex(word, **kwargs):
	flags = None
	if type(word) in (type(()), type([])):
		flags = word[1]
		word = word[0]
	word = u(re.sub(r'[-_ ]+', r'[-_ ]', u(word)))
	word = u(r"(?<![\u00E8-\u00F8\xe8-\xf8\w])(?<!'')(?<!" + r'"' + ")(?:\\b|(?<=[ \\[\\]\\(\\):;.,\"'*\\xab\\xbb])|^)" + word + r"(?:\b(?![\u00E8-\u00F8\xe8-\xf8\w])(?!''|" + r'"' + ")|(?=[ \\[\\]\(\\):;.,\"'*\\xab\\xbb])|$)")
	if flags is None:
		return word
	return (word, flags)
def wordFilter(correct, *badwords, **kwargs):
	correct = u(correct)
	rs = {}
	badwords2 = []
	for i in badwords:
		if type(i) in (type(()), type([])):
			badwords2.extend(map(u, i))
		else:
			badwords2.append(u(i))
	if not len(badwords2):
		badwords2.append(correct)
	for w in badwords2:
		if 'keepcapitalization' in kwargs and kwargs['keepcapitalization']:
			if type(w) not in (type(()), type([])):
				w = (w, 0)
			else:
				w = (w[0], 0)
			rs[wordRegex(w, **kwargs)] = correct
			rs[wordRegex((w[0][0].swapcase() + w[0][1:], w[1]), **kwargs)] = correct[0].swapcase() + correct[1:]
		else:
			rs[wordRegex(w, **kwargs)] = correct
	return setFilterName(regexes(rs), u'WordFilter(' + u'/'.join(badwords2) + u' \u2192 ' + correct + u')')
def enforceCapitalization(*words, **kwargs):
	for w in words:
		addSafeFilter(setFilterName(wordFilter(u(w)), u'EnforceCapitalization(' + u(w) + u')'), **kwargs)
pageFilters = []
pageWhitelist = []
categoryFilters = []
def pageFilter(page):
	global pageFilters, pageWhitelist
	if type(page) in (type(()), type([])):
		pages = []
		for p in page:
			if pageFilter(p):
				pages.append(p)
		return pages
	if type(page) not in (type(u''), type('')):
		page = page.title
	page = u(page)
	if page in pageWhitelist:
		return True
	for f in pageFilters:
		if f.search(page):
			return False
	return True
def categoryFilter(page):
	global categoryFilters
	pageCategories = page.getCategories()
	for c in pageCategories:
		if u(c).replace(u'_', ' ') in categoryFilters:
			return False
	return True
def addPageFilter(*filters):
	global pageFilters
	for f in filters:
		pageFilters.append(compileRegex(f))
def addBlacklistPage(*pages):
	for p in pages:
		addPageFilter(re.escape(u(p)))
def addWhitelistPage(*pages):
	global pageWhitelist
	for p in pages:
		if type(p) in (type([]), type(())):
			addWhitelistPage(*p)
		elif u(p) not in pageWhitelist:
			pageWhitelist.append(u(p))
def addBlacklistCategory(*categories):
	global categoryFilters
	for c in categories:
		categoryFilters.append(u(c).replace(u'_', ' '))
def loadBlacklist():
	global config
	for l in page(config['pages']['blacklist']).getLinks():
		l = u(l)
		if l.find(u':') != -1:
			if l[:l.find(u':')].lower() == 'category':
				addBlacklistCategory(l)
				continue
		addBlacklistPage(l)

filters = {
	'regular': [],
	'safe': [],
	'link': [],
	'template': [],
	'file': []
}
def addFilterType(filterType, *fs, **kwargs):
	global filters
	for f in fs:
		f = (f, kwargs)
		if f not in filters[filterType]:
			filters[filterType].append(f)
def delFilterType(filterType, *fs, **kwargs):
	global filters
	for f in fs:
		f = (f, kwargs)
		if f in filters[filterType]:
			filters[filterType].remove(f)
def addFilter(*fs, **kwargs):
	addFilterType('regular', *fs, **kwargs)
def delFilter(*fs, **kwargs):
	delFilterType('regular', *fs, **kwargs)
def addSafeFilter(*fs, **kwargs):
	addFilterType('safe', *fs, **kwargs)
def delSafeFilter(*fs, **kwargs):
	delFilterType('safe', *fs, **kwargs)
def addLinkFilter(*fs, **kwargs):
	addFilterType('link', *fs, **kwargs)
def delLinkFilter(*fs, **kwargs):
	delFilterType('link', *fs, **kwargs)
def addTemplateFilter(*fs, **kwargs):
	addFilterType('template', *fs, **kwargs)
def delTemplateFilter(*fs, **kwargs):
	delFilterType('template', *fs, **kwargs)
def addFileFilter(*fs, **kwargs):
	addFilterType('file', *fs, **kwargs)
def delFileFilter(*fs, **kwargs):
	delFilterType('file', *fs, **kwargs)
def filterRepr(filters):
	s = []
	reprRegex = compileRegex(r'^<function (\S+)')
	for f in filters:
		if type(f) in (type([]), type(())) and not len(f[1]):
			f = f[0] # Omit parameters if there are none
		try:
			name = f.filterName
			s.append(name)
		except:
			res = reprRegex.search(u(f))
			if res:
				filterR = u(res.group(1))
				if filterR not in s:
					s.append(filterR)
			elif u(f) not in s:
				s.append(u(f))
	if not len(s):
		return u'Built-in filters' # Link simplification, template formatting, etc
	return u', '.join(s)
def fixContent(content, article=None, returnActive=False, **kwargs):
	global filters
	content = u(content)
	oldcontent = u''
	loopTimes = 0
	redirect = False
	activeFilters = []
	if len(content) > 9:
		redirect = content[:9] == u'#REDIRECT'
	if article is not None:
		article = page(article)
	while not loopTimes or content != oldcontent:
		loopTimes += 1
		if loopTimes > 2:
			print 'Pass', loopTimes, 'on', article
		if loopTimes >= config['pagePasses']:
			print 'Warning: More than', config['pagePasses'], 'fix passes on article', u(article.title)
			break
		oldcontent = content
		# Apply unsafe filters
		content, activeF = sFilter(filters['regular'], content, returnActive=True, article=article, redirect=redirect)
		activeFilters.extend(activeF)
		# Apply safe filters
		content, safelist = safeContent(content)
		content, templatelist = templateExtract(content)
		content, linklist = linkExtract(content)
		content, activeF = sFilter(filters['safe'], content, returnActive=True, article=article, redirect=redirect)
		activeFilters.extend(activeF)
		extraLinks = setFilterName(curry(linkTextFilter, filters['safe']), u'(Content filters applied to links)')
		addLinkFilter(extraLinks)
		if not redirect:
			linklist, activeF = linkFilter(filters['link'], linklist, returnActive=True, article=article, redirect=redirect)
			activeFilters.extend(activeF)
		content = linkRestore(content, linklist)
		templatelist, activeF = templateFilter(filters['template'], templatelist, returnActive=True, article=article, redirect=redirect)
		activeFilters.extend(activeF)
		content = templateRestore(content, templatelist)
		content = safeContentRestore(content, safelist)
		delLinkFilter(extraLinks)
	if article is not None and u(article.title)[:5] == 'File:':
		# Apply file filters
		content, activeF = sFilter(filters['file'], content, returnActive=True, article=article, redirect=redirect)
		activeFilters.extend(activeF)
	if returnActive:
		return content, activeFilters
	return content
def fixPage(article, **kwargs):
	article = page(article)
	force = False
	if 'force' in kwargs and kwargs['force']:
		force = True
	try:
		catFilter = categoryFilter(article)
	except wikitools.page.NoPage:
		print 'No such page:', article
		return False
	except:
		catFilter = True
	if not force and (not pageFilter(article) or not catFilter):
		print 'Skipping:', article
		return
	originalContent = u(article.getWikiText())
	content, activeFilters = fixContent(originalContent, returnActive=True, article=article)
	if content != originalContent:
		print article, 'needs to be updated.'
		summary = u'Auto: ' + filterRepr(activeFilters)
		if 'reason' in kwargs:
			summary += u' (' + u(kwargs['reason']) + u')'
		if 'fake' in kwargs:
			print '-------- New content is: --------'
			print content
			print '---------------------------------'
		else:
			editPage(article, content, summary=summary)
		return True
	print article, 'is up-to-date.'
	return False
def patrol(change):
	global config
	secondsElapsed = (datetime.datetime.utcnow() - datetime.datetime.fromtimestamp(time.mktime(time.strptime(change['timestamp'], r'%Y-%m-%dT%H:%M:%SZ'))))
	totalTime = secondsElapsed.seconds + secondsElapsed.days * 86400
	if int(change['rcid']) <= config['runtime']['rcid'] or not pageFilter(change['title']) or totalTime <= config['freshnessThreshold']:
		print 'Skipping', change['rcid'], change['title']
		if int(change['rcid']) > config['runtime']['rcid']:
			config['runtime']['rcid'] = int(change['rcid'])
		return
	print 'Patrolling', change['title']
	config['runtime']['rcid'] = int(change['rcid'])
	result = fixPage(change['title'], reason=u'Review RC#' + u(change['rcid']))
	updateRCID()
def loadPage(p):
	p = page(p)
	try:
		code = u(p.getWikiText())
	except:
		error('Couldn\'t grab page', p)
	coderegex = compileRegex(r'^(?: [^\r\n]*(?:[\r\n]+|$))+', re.MULTILINE)
	trimcode = compileRegex(r'^ |</?nowiki>', re.MULTILINE)
	for m in coderegex.finditer(code):
		try:
			exec(trimcode.sub(u'', u(m.group())))
		except:
			error('Error while parsing code: ', m.group())
def patrolChanges():
	try:
		recentChanges = wikitools.api.APIRequest(wiki(), {
			'action':'query',
			'list':'recentchanges',
			'rctoken':'patrol',
			'rclimit':'500'
		}).query(querycontinue=False)[u'query'][u'recentchanges']
		recentChanges.reverse()
	except:
		error('Error while trying to grab recent changes.')
	uniquePages = []
	uniqueChanges = {}
	for change in recentChanges:
		if change['title'] not in uniquePages:
			uniquePages.append(change['title'])
		if change['title'] not in uniqueChanges or uniqueChanges[change['title']]['rcid'] < change['rcid']:
			uniqueChanges[change['title']] = change
			# Move page to end of queue
			uniquePages.remove(change['title'])
			uniquePages.append(change['title'])
	for title in uniquePages:
		change = uniqueChanges[title]
		try:
			patrol(change)
		except KeyboardInterrupt:
			error('Interrupted:', change)
		except:
			warning('Failed to patrol change:', change)
	print 'Done patrolling.'
def parsePageRequest(l, links=[]):
	l = u(l)
	content = []
	selfContent = u'* [[:' + l + u']]'
	if l.find(u':'):
		if l[:l.find(u':')].lower() == 'category':
			subpages = wikitools.category.Category(wiki(), l[l.find(u':')+1:]).getAllMembers(titleonly=True)
			for s in subpages:
				if s not in links:
					links.append(s)
					newLink, links = parsePageRequest(s, links=links)
					content.append(newLink)
	if len(content):
		selfContent += u'\r\n' + u'\r\n'.join(content)
	return selfContent, links
def doPageRequests(force=False):
	global config
	print 'Executing page requests. Force =', force
	if force:
		requestPageTitle = config['pages']['pagerequestsforce']
	else:
		requestPageTitle = config['pages']['pagerequests']
	requestPage = page(requestPageTitle)
	reqre = compileRegex(r'^\*[\t ]*\[\[:?([^][]+)\]\]', re.MULTILINE)
	originalRequests = u(requestPage.getWikiText())
	requests = originalRequests
	matches = []
	links = []
	for m in reqre.finditer(requests):
		matches.append(m)
		l = u(m.group(1))
		if l not in links:
			links.append(l)
	matches.reverse()
	for m in matches:
		pagelink, links = parsePageRequest(u(m.group(1)), links=links)
		requests = requests[:m.start()] + pagelink + requests[m.end():]
	requests = regSub({r'^[ \t]*(\*[^\r\n]+)[\r\n]+(?=^[ \t]*\*)':'$1\r\n'}, requests)
	reqre2 = compileRegex(r'^\*[\t ]*\[\[:?([^][]+)\]\]\s*', re.MULTILINE)
	matches2 = []
	requestsDone = 0
	tooMany = False
	for m in reqre2.finditer(requests):
		requestsDone += 1
		if requestsDone > config['maxrequests']:
			tooMany = True
			break
		matches2.append(m)
	matches2.reverse()
	tofix = []
	for m in matches2:
		tofix.append(u(m.group(1)))
		requests = requests[:m.start()] + requests[m.end():]
	tofix.reverse()
	for p in tofix:
		fixPage(p, reason=u'Requested on [[:' + u(requestPageTitle) + u']]', force=force)
	requests = regSub({r'^[ \t]*(\*[^\r\n]+)[\r\n]+(?=^[ \t]*\*)':'$1\r\n'}, requests)
	if len(tofix) and originalRequests != requests:
		if tooMany:
			editPage(requestPage, requests, summary=u'Processed: [[:' + u']], [[:'.join(tofix) + u']]')
		else:
			editPage(requestPage, requests, summary=u'Finished all requests. Processed: [[:' + u']], [[:'.join(tofix) + u']]')
def parseLocaleFile(content, language='english', languages={}):
	content = u(content)
	language = u(language)
	if content.find('Tokens') != -1:
		content = content[content.find('Tokens')+6:]
	regexSplit = compileRegex('\n(?=\s*")', re.IGNORECASE | re.MULTILINE)
	content = regexSplit.split(content)
	regexLang = compileRegex(r'^"\[([-\w]+)\]([^"\s]+)"\s+"([^"]*)"', re.IGNORECASE | re.MULTILINE)
	regexNoLang = compileRegex(r'^"([^[][^"\s]+)"\s+"([^"]*)"', re.IGNORECASE | re.MULTILINE)
	for l in content:
		l = u(l.strip())
		curlang = None
		key, value = None, None
		langRes = regexLang.search(l)
		if langRes:
			curlang = u(langRes.group(1))
			key, value = langRes.group(2), langRes.group(3)
		else:
			langRes = regexNoLang.search(l)
			if langRes:
				curlang = language
				key, value = langRes.group(1), langRes.group(2)
		if curlang is not None:
			if u(key) not in languages:
				languages[u(key)] = {}
			languages[u(key)][curlang] = u(value)
		else:
			pass
	return languages
def languagesFilter(languages, commonto=None, prefix=None, suffix=None, exceptions=[]):
	filtered = {}
	for k in languages:
		if k in exceptions:
			continue
		if commonto is not None:
			doit = True
			for i in commonto:
				if i not in languages[k]:
					doit = False
					break
			if not doit:
				continue
		if prefix is not None:
			doit = False
			for i in prefix:
				if k.lower()[:len(i)] == i.lower():
					doit = True
					break
			if not doit:
				continue
		if suffix is not None:
			doit = False
			for i in suffix:
				if k.lower()[-len(i):] == i.lower():
					doit = True
					break
			if not doit:
				continue
		filtered[u(k)] = languages[k]
	return filtered
def readLocaleFile(f):
	return u(f.decode('utf16'))
def associateLocaleWordFilters(languages, fromLang, toLang, targetPageLang=None):
	for a in languages:
		f = wordFilter(languages[a][toLang], languages[a][fromLang])
		if targetPageLang is None:
			addSafeFilter(f)
		else:
			addSafeFilter(f, language=targetPageLang)
def getRandBits():
	return random.getrandbits(128)
def getFileHash(filename):
	h = hashlib.md5()
	f = open(filename, 'rb')
	for i in f.readlines():
		h.update(i)
	f.close()
	return u(h.hexdigest())
def deleteFile(*fs):
	for f in fs:
		try:
			os.remove(f)
		except:
			pass
def programExists(programName):
	try:
		result = subprocess.call(['which', programName])
		return result == 0
	except:
		return False
def run():
	global config
	print 'Bot started.'
	loadPage(config['pages']['filters'])
	for p in sys.argv[1:]:
		print 'Forced update to', p, '...'
		fixPage(p)
	loadBlacklist()
	patrolChanges()
	updateRCID()
	doPageRequests(force=True)
	doPageRequests(force=False)
	updateEditCount()
	runScheduledTasks()
	try:
		import rcNotify
		rcNotify.main(once=True)
	except:
		pass
	try:
		subprocess.Popen(['killall', 'cpulimit']).communicate()
	except:
		pass
	print 'All done.'
if __name__ == '__main__':
	run()