#!/usr/bin/env python
# -*- coding: utf-8  -*-
'''
webchecklinks.py
 A web wrapper around the checklinks library
 (c) Dispenser, 2007-2008
'''

import cgi, sys
import checklink, wikipedia
# Don't log since error generally happen in a thread and don't go through the normal handler
import cgitb; cgitb.enable()

def printu(ustr):
	try:
		print ustr
	except:
		print ustr.encode('utf-8')
	
def printFile(name, arg1="$1"):
	f = open('./text/%s.html' % name )
	print f.read().replace('$1', arg1.encode('utf-8'))
	f.close()

def printEntry(self, page, url, refId, context, status, reason, redirect, rank, comment):
	#printu(u'%s %s - %s' % (status, reason, wikilink))
	classes = 'dead-%s'%rank
	if url.encode('utf-8')+'\n' in open('/home/dispenser/webcite_requests.txt','r'):
		classes += ' webcite'
	if redirect:
		classes += ' redirect'
	printu('<tr class="%s"><td>%s</td><td>%s</td><td><abbr title="%s">%s</abbr></td><td>%s</td></tr>' % (classes, refId or '', context.replace('[[', '[&#91;'), reason, status, rank and (comment or reason) or ''))

	# flush to user
	sys.stdout.flush()
	
def textbox(name, value, label = None, attrib=''):
	if label is None:
		label = '%s: ' % name.capitalize()
	print '<label for="%s">%s</label><input type="text" name="%s" value="%s" id="%s" %s/>' % (name, label, name, value, name, attrib)

def checkbox(name, checked, label = None, attr=''):
	if checked:
		attr += ' checked="checked"'
	print '<input type="checkbox" name="%s" id="%s" %s /><label for="%s">%s</label>' % (name, name, attr, name, label or name.capitalize())

def main():
	#global hostname

	form = cgi.FieldStorage()
	htmlmode = bool(form.getfirst('html', False))
	checklink.DEBUG	 = bool(form.getfirst('debug', False))
	checklink.SOURCE = bool(form.getfirst('source', False))
	checklink.config.max_external_links	= int(form.getfirst('threads',   30))
	checklink.config.defaulttimeout 	= int(form.getfirst('timeout',   30))
	checklink.config.httpDebug      	= int(form.getfirst('httpDebug',  0))
	checklink.config.useGET				= bool(form.getfirst('alwaysUseGet', False))
	checklink.config.threaded           = not bool(form.getfirst('nothread', False))
	if form.getfirst('hostname'):
		host = form.getfirst('hostname').split('.')
		site = wikipedia.Site(host[0], host[1])
	else:
		site = wikipedia.getSite()
	page = wikipedia.Page(site, form.getfirst('page', ''))
	#hostname = page.hostname()
	
	wikipedia.startContent(u'Checklinks: %s' % page.title(), form=False, head='''<script src="/~dispenser/resources/checklinks.js" type="text/javascript"></script>''')
	print '<form action="webchecklinks.py" style="text-align:center;" onsubmit="hideViewer()">'
	print '<input name="page" id="page" size="40" value="%s" onchange="fixTitle(this)" />' % page.aslink()[2:-2].encode('utf-8')
	if checklink.DEBUG:
		print '<br/>'
		textbox('httpDebug', checklink.config.httpDebug, 'HTTP debug level: ', ' size="3"')
		textbox('threads', checklink.config.max_external_links, attrib=' size="2"')
		textbox('timeout', checklink.config.defaulttimeout, attrib=' size="3"')
		print '<br/>'
		checkbox('alwaysUseGet', checklink.config.useGET, 'Always download')
		checkbox('debug', checklink.DEBUG, 'Debug')
		checkbox('nothread', not checklink.config.threaded, 'Disable threading')
		checkbox('html', htmlmode, 'HTML input')
		checkbox('source', checklink.SOURCE, 'Print source')
	print '<input type="submit" value="Check page" /></form>'
	try:
		page.get()
	except wikipedia.NoPage, e:
		printu('NoPage error encountered (%s)' % e)
		return
	except wikipedia.IsRedirectPage:
		link = wikipedia.Page(page.site(), page._redirarg).aslink()[2:-2]
		printu('<img src="http://%s/skins-1.5/common/images/redirectltr.png" alt="#REDIRECT " /><a href="webchecklinks.py/%s" class="redirectText">%s</a>' % (page.hostname(), link, link))
		return

	printFile('checklinks-header')
	if checklink.cacheAge(page) < float('inf'):
		print checklink.time.strftime('<p>Previously cached on  %d %B %Y at %H:%M </p>', checklink.time.gmtime(checklink.time.time() - checklink.cacheAge(page)))
	else:
		print ""
	print '<table id="linktable" class="">'
	#import parser
	#printu('<tr class="page"><th colspan="6"><a id="%s" href="http://%s/wiki/%s" title="%s" class="extiw">%s</a></th></tr>' % (parser.escapeId(page.title().encode('utf-8')), page.hostname(), page.urlname(), page.aslink()[2:-2], page.title()))
	printu('<tr class="page"><th colspan="6"><a id="%s" href="http://%s/wiki/%s" title="%s" class="extiw">%s</a></th></tr>' % ('', page.hostname(), page.urlname(), page.aslink()[2:-2], page.title()))
	printFile('checklinks-tableHead')
	try:
		if htmlmode:
			checklink.checkMWhtml(page, printEntry)
		else:
			checklink.checkMediaWikiPage(page, printEntry)
	finally:
		print '</table>'

if __name__ == "__main__" and wikipedia.handleUrlAndHeader():
	try:
		main()
	finally:
		wikipedia.endContent()