#!/usr/bin/python 
# -*- coding: utf-8  -*-
'''
JSON version of webchecklinks.py
webchecklinks.py
 A web wrapper around the checklinks library
 (c) Dispenser, 2007-2008
'''
import cgi
import checklink, wikipedia
import parser
import cgitb; cgitb.enable(logdir='./logs/')

def printu(ustr):
	print ustr.encode('utf-8')
def quote(s):
	return unicode(s).replace('"', '\\"').replace('\n', '\\n')
def list2json(list):
	if list:
		return ','.join([u'"%s"'%quote(value) for value in list])
	else:
		return ''
def dict2json(dict):
	if dict:
		return ',\n\t\t\t\t'.join([(u'"%s": "%s"'%(key, quote(value))) for (key, value) in dict.items()])
	else:
		return ''

def addEntry(page, url, title, status, reason, rank, comment, headers, redirect, refId, metadata = {}, statsData=''):
	#wikipedia.output(page.aslink())
	#return 
    # Do nothing with statsData
	if len(reason) >= 37:
		reason = reason[0:25] + '...'
	if rank >= threshold:
		# "([^\n]*)", "([^\n]*)", "([^\n]*)", "([^\n]*)", "([^\n]*)", "([^\n]*)", "([^\n]*)", "([^\n]*)"
		# \1\t\5\t\3\t\4\t\6\t\7\t\2\t\8
		printu('''\
		{
			"url": "%s",
			"title": "%s",
			"metadata": {
				%s
				},
			"status": %d,
			"reason": "%s",
			"headers": {
				%s
				},
			"redirect": "%s",
			"rank": %d,
			"comment": "%s"
		},''' % (quote(url), quote(parser.parser(title or "", page.hostname())), dict2json(metadata), status, quote(reason), dict2json(headers), quote(redirect or ''), rank, quote(parser.parser(comment, page.hostname())),))

def main():
	global threshold
	form = cgi.FieldStorage()
	checklink.DEBUG = False
	global callback
	callback     = form.getfirst('callback','')
	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))
	alwaysUseGet = bool(form.getfirst('alwaysUseGet',	False))
	nothread     = bool(form.getfirst('nothread',		False))
	threshold    = int( form.getfirst('threshold',      0))
	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', ''))
	
	if bool(form.getfirst('debug', False)):
		import cgitb; cgitb.enable()
		print 'Content-type: text/plain'
		print 
		print page.get().encode('utf-8')

	# Check to make sure we can get the page
	try:page.get()
	except:pass

	print 'Content-Type: text/javascript; charset=utf-8'
	print
	if callback:
		print callback + '({'
	else:
		print '{'
	print '"error":{"info":"This tool has been temporaly disabled"}})'
	print '/*'
	raise
	try:
		page.get()
	except Exception, e:
		print '"error": {"info":"%s"}' % quote(e)
		if callback:	print '})'
		else:			print '}'
		return

	try:
		printu('	"title":"%s",' % page.title())
		print '	"hostname":"%s",' % page.hostname()
		print '	"links": ['
		checklink.checkMediaWikiPage(page, addEntry, alwaysUseGet, not nothread)
	finally:
		print '		null/* hack for last , */'
		print '	]'
		if callback: print '})'
		else:
			print '}'
#			print 'if(checklinksError)checklinksError("Please update to the new API")'
	#	print '/* Callback function */'
	#	print 'if(checklinksDone)checklinksDone();'

if __name__ == "__main__":
	main()