#!/usr/bin/python 
# -*- coding: utf-8  -*-
# TODO: add per paragraph readability scores
"""
Readabilty.py

(c) Dispenser, 2007
"""

import re, htmlentitydefs, wikipedia
from math import sqrt
import cgi
import cgitb; cgitb.enable()

textFiles = './text/'
ArticleWordStub = {
	'en':	"This article seems to be a stub."
}
ArticleWordLarge = {
	'en':	"This article seems to be too long and probably needs to be split."
}
ArticleLowIndex = {
	'en':	"Longer words and more compound or complex sentences may make this article more interesting, more precise and less wordy."
}
ArticleHighIndex = {
	'en':	"This article seems to have too many long words and sentences for even most university graduates to easily read and understand."
}
ArticleSizeLarge = {
	'en':	"Almost certainly should be divided up"
}
ArticleSizeBig = {
	'en':	"Probably should be divided (although the scope of a topic can sometimes justify the added reading time)"
}
Languages = {
	'en':	"English",
	'simple':	"Simple English",
}
Families = {
	'wikibooks':	"Wikibooks",
	'wikipedia':	"Wikipedia",
	'wikisource':	"Wikisource",
	'wikinews':	"Wikinews",
	'wiktionary':	"Wiktionary",
	'wikiquote':	"Wikiquote",
	'wikiversity':	"Wikiversity",
}
BracketCount	= 0
class wikiParse:
	a = 0
def printu(ustr):
	try:
		print ustr
	except:
		print ustr.encode('utf-8')

def printOptions(dict, default):
	for key in dict:
		if key == default:
			print '<option value="%s" selected="selected">%s</option>' % (key, dict[key])
		else:
			print '<option value="%s">%s</option>' % (key, dict[key])
def printWiki(text, basehref=None, linkClass='extiw'):
	print(wikiToHTML(text, basehref, linkClass).encode('utf-8'))
def wikiLink(text):
	for m in re.finditer(r'\[\[(.*?)\]\]', text):
		text = text.replace(m.group(), r'<a href="http://%s/wiki/%s" class="extiw">%s</a>' % (hostname, m.group(1).replace(' ', '_'), m.group(1)))
	return text
def wikiToHTML(text, basehref = None, linkClass = 'extiw'):
	# Define varibles
	if not basehref:
		basehref = 'http://%s/wiki/' % hostname
	if linkClass:
		linkClass = ' class="%s"' % linkClass
	# Strip HTML comments
	text = re.sub(r"(?s)<!--.*?-->", r'', text)
	text = text.expandtabs()
	
	# Nowiki
	# Nowikicode goes here
	
	# Tempate parser
	stack = [text.find('{{')]
	while 10 > len(stack) > 0:
		start = stack.pop()
		if start == -1:
			continue
		end   = text.find('}}', start)+2
		nextStart = text.find('{{', start+2)
		
		if end > nextStart > start:
			stack.append(start)
			stack.append(nextStart)
		elif end > start:
			tmpl = text[start:end]
			[(title, param)] = re.findall(r'\{\{[{}]*\s*([^{|}\n]+)(.*?)', tmpl)
			text = text.replace(tmpl, '{&#123;[[Template:%s|%s]]&#125;}' % (title, title))
			if stack == []:
				stack.append(text.find('{{', start+2))

	# <math>, <source>
	text = re.sub(r'(?is)<math>(.*?)</math>', r'<img class="tex" alt="Math tag" src="http://upload.wikimedia.org/math/e/e/6/ee6329486d72b1cc1b8a0e2092ddd419.png" />', text)
#	text = text.replace('<source lang=', '<pre class=')
#	text = text.replace('</source', '</pre>')

	# Headers
	text = re.sub(r'(?m)^=====(.*?)===== *$',r'<h5>\1</h5>', text)
	text = re.sub(r'(?m)^====(.*?)==== *$',  r'<h4>\1</h4>', text)
	text = re.sub(r'(?m)^===(.*?)=== *$',    r'<h3>\1</h3>', text)
	text = re.sub(r'(?m)^==(.*?)== *$',      r'<h2>\1</h2>', text)
	text = re.sub(r'(?m)^=(.*?)= *$',        r'<h1>\1</h1>', text)
	
	# definition lists
	text = re.sub(r"(?m)^;([^:]*\S)\s*", r'<dl>\n<dt>\1</dt>\n</dl>\n', text)
	text = re.sub(r"(?m)^:(.*)\s*", r'<dl>\n<dd>\1</dd>\n</dl>\n', text)
	text = text.replace(r'\n</dl>\n<dl>', r'')
	
	# bullet and numbered lists
	text = re.sub(r"(?m)^ ?[*#]*\* *(.*)", r'<ul>\n\t<li>\1</li>\n</ul>', text)
	text = re.sub(r"(?m)^ ?[*#]*\# *(.*)", r'<ol>\n\t<li>\1</li>\n</ol>', text)
	# Merge list together
	text = text.replace("</ul>\n<ul>\n", r'')
	text = text.replace("</ol>\n<ol>\n", r'')
	
	# Horizonal line
	text = text.replace("\n----", '<hr />')
	
	# Bold and Illatics
	text = re.sub(r"(?m)'''(.*?)'''", r'<b>\1</b>', text)
	text = re.sub(r"(?m)''(.*?)''",   r'<i>\1</i>', text)	
	
	## space infront to <pre> (simple ver.)
	#text = re.sub(r'\n ([^\n]*)',	r'\n<pre>\1</pre>', text)
	#text = text.replace('</pre>\n<pre>', '\n')
	
	# Internal links
	for l in re.finditer(r'\[\[(?!Image:)(?:([^][|]+)\||)([^][]+)\]\](\w*)', text, re.UNICODE):
		link = l.group(1) or l.group(2)
		title= l.group(2)
		hyperlink = (link[0].upper()+link[1:]).replace(' ', '_').strip('_:')
		text = text.replace(l.group(), '<a href="%s%s" title="%s"%s>%s%s</a>' % (basehref, hyperlink, link, linkClass, title, l.group(3)))
	
	# Images  
	for wimg in re.finditer(r'(?s)\[\[Image:([^\[|\]]+)(\|.*?|)\]\]', text):
		file = 'Image:%s' % wimg.group(1).replace('_', ' ').strip()
		attribs = wimg.group(2).split('|')
		size = ''
		float = None
		thumb = False
		caption = ''
		captionText = ''
		for s in wimg.group(2).split('|'):
			sl = s.lower()
			if sl.endswith('px'):
				try:
					size = int(s[:-2])
				except:
					pass
			elif sl=='left' or sl=='right':
				float = s
			elif sl=='thumb' or sl=='thumbnail':
				thumb = True
				size  = size or 64
				float = float or 'right'
			else:
				caption = s
				captionText = wikipedia.escape(re.sub(r'</?\w+[^<>]*>', '', s))
		if thumb:
			text = text.replace(wimg.group(), """
<div class="thumb" style="float:%s;clear:both;">
<div class="thumbinner" style="width:%spx"><a href="//%s/wiki/%s" class="image" title="%s"><img alt="%s" src="http://upload.wikimedia.org/wikipedia/commons/8/82/Crystal_128_looknfeel.png" class="thumbimage" border="0" width="%s" /></a>
<div class="thumbcaption" style="width:auto;clear:both;">
%s</div>
</div>
</div>""" % (float, size+2, hostname, file.replace(' ', '_'), file, captionText, size, caption ))
		else:
			text = text.replace(wimg.group(), '<a href="%sImage:%s"><img src="http://%s/wiki/Special:Filepath/%s" width="%s" style="float:%s;" alt="%s" /></a>' % (basehref, file, hostname, file, size, float, captionText))

	# External links
	global BracketCount
	text = re.sub(r'\[(\w+://[^][><" \n]+)\s*([^]]*?\]*)\]', r'<a href="\1" class="external">\2</a>', text)
	while '"></a>' in text:
		BracketCount+=1
		text = text.replace('"></a>', '">[%d]</a>' % BracketCount, 1)
		
	text = re.sub(r'(\A|[]<>\n])(\w+://[^][><\s"]*)', r'\1<a href="\2" class="external -breakme"> \2 </a>', text)
	
	# Encode & -> &amp;
	text = text.replace('&', '&amp;')
	text = re.sub(r'\&amp;(#\d+|#x\w+|\w{1,6});', r'&\1;', text)

	# Remove Reference (no good way to parse)
	text = re.sub(r'(?is)<ref[^/>]*>.*?</ref>|<ref[^>]*/>', '<sup style="color:green;">[ref]</sup>', text)
	
	# Wiki markup table paser
	stack = [text.find('{|')]
	while 10 > len(stack) > 0:
		start = stack.pop()
		if start == -1:
			continue
		end   = text.find('|}', start)+2
		nextStart = text.find('{|', start+2)
		
		if end > nextStart > start:
			stack.append(start)
			stack.append(nextStart)
		else:
			table = text[start:end]
			table = re.sub(r'(?m)^\{\| *(.*)',               r'<table \1><tr >', table)
			table = table.replace('\n|}',                      '\n</tr></table>')
			table = re.sub(r'(?m)^\|- *(.*)',                r'</tr></tr \1>', table)
			table = re.sub(r'(?m)^\|\+ *([^][\n|]*\||)(.*)', r'<caption \1>\2</caption>', table)
			while re.findall(r'(?m)^([!|])(.*?)(\|\||!!)', table):
				for r in re.finditer(r'(?m)^([!|])(.*?)(\|\||!!)', table):
					table = table.replace(r.group(), r.group(1)+r.group(2)+'\n'+r.group(1))
			table = re.sub(r'(?m)^\! *([^][<>\n|]*\||)(.*)',  r'<th \1>\2</th>', table)
			table = re.sub(r'(?m)^\| *([^][<>\n|]*\||)(.*)',  r'<td \1>\2</td>', table)
			# Optimize <tr>s
			table = table.replace('<tr ></tr>', '')
			text = text[:start] + table + text[end:]
			if stack == []:
				stack.append(text.find('{|', start+2))

	
	# New paragraphs
	text = re.sub(r"\n\n+([^\n<>][^\n]*)", r'\n\n<p>\1</p>', text)
	
	return text
#########################################
##
#########################################
def grade2age(grade):
	return int(grade+5.3)
def rowPrint(*cells):
	print '<tr>'
	for cell in cells:
		print('<td>%s</td>' % wikiLink(str(cell)))
	print '</tr>'
def average(list):
	if list == []:
		return 0
	else:
		return float(sum(list))/float(len(list))
class ReadingNumbers(object):
  def __init__(self):
  	# How AWB does it
	# Remove tables and templates
	# Count instances using r"[a-zA-Z]+"
	flags = re.UNICODE
	self.WordCount     = re.compile(r'[^\W\d]+', flags)
	self.SentenceCount = re.compile(r'\w*\w\w[ \'"\])]*[\.\!\?]+[\s\'"\]()]', flags)
#IgnoreWords	= re.split(r'\W', page.title().lower())
#IgnoreWords.append(u'wikipedia')

  def populateStats(self, text):
  	self.markedup  = text
	self.words     = len(self.WordCount.findall(text))     or 1
	self.sentences = len(self.SentenceCount.findall(text)) or 1

	print '<!-- senetence counting-->'
	for s in self.SentenceCount.finditer(text):
		report = report.replace(s.group(), '<span style="background:#fee;">%s</span>%s' % (s[:-1],s[-1]))
	
	# The actual staticstics stuff, should be branched off into an object/class
	PropNouns	= re.findall(r'(?<!\w.[\.\n])\s+\b((?:Mc|)[A-Z][a-z\-]+)\b', prose)
	printu("<!-- Proper nouns:\n %r\n -->" % PropNouns)
	
	letters = 0
	HardWords = 0
	syllable_list	= []
	
	print '<!-- word analysis -->'
	for word in self.WordCount.finditer(text):
		s = word
	#	if s.lower() in IgnoreWords:
	#		continue
		if len(s) > 1 and (s.endswith('e')):
			s = s[:-1]
		elif len(s) > 2 and (s.endswith('es') or s.endswith('ed')):
			s = s[:-2]
		elif len(s) > 3 and (s.endswith('ing') or s.endswith('ies')):
			s = s[:-3]
		elif len(s) > 3 and (s.endswith('ely')):
			s = s[:-3]+s[-2:]
		vowels = re.findall(r'(?i)[aeiouy]+', s)
		if vowels:
			letters += len(word)	# Make sure we're adding words
			syllable_list.append(len(vowels))
		if len(vowels) >= 3 and not word in PropNouns:
			HardWords += 1
			report = re.sub(r'(%s)\b([^<>])' % word, r'<u>\1</u>\2', report)
	self.letters = letters
	#print syllable_list
	if debug:
		print '<pre>'
		print ( 0.39 * float(self.words) / float(self.sentences) ) 
		print ( 11.8 * average(syllable_list) )
		print 'syllable list length: %d' % len(syllable_list)
		print 'sum of syllable:%d' % sum(syllable_list)
		print '</pre>'

  def makeIndexes():
	# Calculate stats
	self.Flesch_Ease		= 206.835 - (1.015 * self.words/ float(self.sentences)) - (84.6 * average(syllable_list))
	self.ARI				= ( 4.71 * self.letters/ float(self.words)) + ( 0.5 * float(self.words) / float(self.sentences) ) - 21.43
	self.Flesch_Kincaid	= ( 0.39 * self.words   / float(self.sentences) ) + ( 11.8 * average(syllable_list) ) - 15.59
	self.Cloeman_Liau	= ( 5.89 * self.letters/ float(self.words) ) - ( 0.3 * float(self.sentences) / float(self.words) ) - 15.8
	self.Gunning_fog		= 0.4 * ( (float(self.words) / float(self.sentences)) + 100 * (float(HardWords) / float(self.words)) )
	self.SMOG			= 1.0430 * sqrt(30.0 * HardWords / float(self.sentences) ) + 3.1291
	#LIX = W/S+(100*LW)/W

def doStats(page):
	try:
		wikiText = page.get()
		wikiText = wikiText.replace('\r\n', '\n')
	except wikipedia.NoPage:
		print '<div style="margin-left:2em;">Page is empty/non-existant</div>'
		return
	except wikipedia.IsRedirectPage:
		print '<img src="http://%s/skins-1.5/common/images/redirectltr.png" alt="#REDIRECT " /><a href="?page=%s:%s" class="redirectText">%s</a>' % (page.hostname(), page.site().language(), page._redirarg, page._redirarg)
		return
	if not wikiText.strip():
		print 'Page does not exist'
		return
	elif '{{disambig}}' in wikiText:
		printWiki(wikiText, basehref='?page=', linkClass = '')
		return
	print """
<!--[if lt IE 5.5000]><style type="text/css">@import "http://en.wikipedia.org/skins-1.5/monobook/IE50Fixes.css?116";</style><![endif]-->
<!--[if IE 5.5000]><style type="text/css">@import "http://en.wikipedia.org/skins-1.5/monobook/IE55Fixes.css?116";</style><![endif]-->
<!--[if IE 6]><style type="text/css">@import "http://en.wikipedia.org/skins-1.5/monobook/IE60Fixes.css?116";</style><![endif]-->
<!--[if IE 7]><style type="text/css">@import "http://en.wikipedia.org/skins-1.5/monobook/IE70Fixes.css?116";</style><![endif]-->
<!--[if lt IE 7]><script type="text/javascript" src="http://en.wikipedia.org/skins-1.5/common/IEFixes.js?116"></script>
<meta http-equiv="imagetoolbar" content="no" /><![endif]-->
"""

	printu('<a href="http://%s/wiki/%s" class="extiw" style="font-size:200%%;">%s</a>' % (page.hostname(), page.urlname(), page.title()))
	printu('<div style="font-size:x-small; clear:both;">Old version <a href="readability-old.py?page=%s:%s">still available</a></div>' % (page.site().language(),page.urlname()))
	print '<table class="wikitable" style="min-width:225px;float:left;">'
	rowPrint('<img src="http://upload.wikimedia.org/wikipedia/commons/thumb/5/54/Gartoon-Galeon.png/24px-Gartoon-Galeon.png" alt="" />', 
		'Links', len(re.findall(r'\[\[[^][:]+\]\]', wikiText)))
	rowPrint('<img src="http://upload.wikimedia.org/wikipedia/commons/thumb/5/5f/Image-x-generic.svg/24px-Image-x-generic.svg.png" alt="" />',
		'Images', len(re.findall(r'\[\[ *[Ii]mage:',	wikiText)))
	rowPrint('<img src="http://upload.wikimedia.org/wikipedia/commons/thumb/c/c7/Crystal_Clear_app_virus_detected.png/24px-Crystal_Clear_app_virus_detected.png" alt="" />', 
		'Categories',	len(re.findall(r'\[\[ *[Cc]ategory:',	wikiText)))
	rowPrint('<img src="http://upload.wikimedia.org/wikipedia/commons/thumb/b/bf/Preferences-desktop-locale.svg/24px-Preferences-desktop-locale.svg.png" alt="" />', 
		'Interwiki links',	len(re.findall(r'\[\[ *[a-z\-]{2,4}:',	wikiText)))
	rowPrint('<img src="http://upload.wikimedia.org/wikipedia/commons/thumb/2/25/External.svg/16px-External.svg.png" style="padding:4px;" alt="" />', 
		'External links', len(re.findall(r'\w+://\w+\.\w+',	wikiText)))
	rowPrint('<img src="http://upload.wikimedia.org/wikipedia/commons/thumb/1/1b/X-office-document.svg/24px-X-office-document.svg.png" alt="" />', 
		'References', wikiText.lower().count('</ref>'))
	rowPrint('<img src="http://upload.wikimedia.org/wikipedia/commons/thumb/3/3b/X-office-drawing-template.svg/24px-X-office-drawing-template.svg.png" alt="" />', 
		'Templates', wikiText.lower().count('}}'))
	print '</table>'
	
	text = wikipedia.removeDisabledParts(wikiText)
	print '<!-- removed disabled parts -->'
	# Remove links
	while re.search(r'\[\[[^][]+\]\]', text):
		text = re.sub(r'\[\[(Image|Category|[A-Za-z]{2,3}(-[A-Za-z]+)*|simple):[^][]*\]\]', '', text, re.IGNORECASE | re.DOTALL | re.UNICODE)
		text = re.sub(r'\[\[([^]|[]+\|)?([^][]+)\]\]', r'\2', text, re.IGNORECASE|re.DOTALL)
		text = re.sub(r'\[\[Image:[^][]*\]\]', '', text, re.IGNORECASE|re.DOTALL)
	print '<!-- wikilink removal complete -->'

    # Remove templates
	#while '{{' in text and '}}' in text:
	while re.findall(r'\{\{[^{}]*?\}\}', text, re.DOTALL):
		text = re.compile(r'\{\{[^{}]*?\}\}', re.DOTALL).sub(r'', text)
	print '<!-- template removal complete -->'

	# Remove italics and bolding
	text = re.sub(r"(?m)'''(.*?)'''", r'<b>\1</b>', text)
	text = re.sub(r"(?m)''(.*?)''",   r'<i>\1</i>', text)
	
	# Remove External links
	text = re.sub(r'\[\w+:/*[^][<>\s"]* *(.*?)\]', r'\1', text)

	# Get "readable prose"
	prose = re.compile('|'.join((
                r'\{\|.*?\|\}',
                r'<ref[^<>/]*>.*?</ref>',
                r'\{\{.*?\}\}',
                r'^=+[^\n]+=+ *$',
                r'^ *\*[^\n]*$',
                r'^ *\#[^\n]*$',
                r'^ [^\n]+$',
                r'^;+[^\n]*$',
                #r'^:+[^\n]{1,200}$', # can do full definition list
                r"^:+ *<i>\w[^\n]*$", # remove dab like lines
                r'&[^;]{2,6};')), re.IGNORECASE | re.DOTALL | re.MULTILINE | re.UNICODE).sub(r'', text)
	# Remove HTML
	text = re.sub(r'</?\w+[^<>]*>', r'', text)
	prose= re.sub(r'</?\w+[^<>]*>', r'', prose)

	# Size Comparisions
	print '<table class="wikitable" style="min-width:225px;float:left;z-index:1000"><tr><td>'
	print '<div style="width:200px;"><!-- min width HACK --></div>' 
	print 'Wikitext: %#.3g KB' % (len(wikiText)/1024.)
	print '<div style="width:%dpx; height:10px; background:lightblue;"><!-- IE6 Hack --></div>' % (len(wikiText)/350)
	print 'Plain text: %#.3g KB' % (len(text)/1024.)
	print '<div style="width:%dpx; height:10px; background:LightCoral;"><!-- IE6 Hack --></div>' % (len(text)/350)
	print 'Readabible proses: %#.3g KB' % (len(prose)/1024.)
	print '<div style="width:%dpx; height:10px; background:gold;"><!-- IE6 Hack --></div>' % (len(prose)/350)
	print '</td></tr></table>'

	WordCounter 	= re.compile(r'\b[^\W\d]+\b', re.UNICODE)
	sentenceFinder	= re.compile(r'\w*\w\w[ \'"\])]*[\.\!\?]+[\s\'"\]()]', re.UNICODE)

	wordsText	= len(	WordCounter.findall(text    )) or 1
	wordsProse	= len(	WordCounter.findall(prose   )) or 1
	sentencesText	= len(	sentenceFinder.findall(text    )) or 1
	sentencesProse	= len(	sentenceFinder.findall(prose   )) or 1

	report		= prose
	# hightlight end of sentences
	for s in sentenceFinder.findall(prose):
		report = report.replace(s, '<span style="background:#fee;">%s</span>%s' % (s[:-1],s[-1]))
	print '<!-- senetence counter complete -->'
	
	# The actual staticstics stuff, should be branched off into an object/class
	IgnoreWords	= re.split(r'\W', page.title().lower())
	IgnoreWords.append(u'wikipedia')
	PropNouns	= re.findall(r'(?<!\w.[\.\n])\s+\b((?:Mc|)[A-Z][a-z\-]+)\b', prose)
	HardWords	= 0
	lettersProse	= 0
	syllable_list	= []
	
	printu("<!-- Proper nouns:\n %r\n -->" % PropNouns)
	print '<!-- polysyllabic identified:'
	for s in WordCounter.findall(prose):
		word = s
	#	if s.lower() in IgnoreWords:
	#		continue
		if len(s) > 1 and (s.endswith('e')):
			s = s[:-1]
		elif len(s) > 2 and (s.endswith('es') or s.endswith('ed')):
			s = s[:-2]
		elif len(s) > 3 and (s.endswith('ing') or s.endswith('ies')):
			s = s[:-3]
		elif len(s) > 3 and (s.endswith('ely')):
			s = s[:-3]+s[-2:]
		vowels = re.findall(r'(?i)[aeiouy]+', s)
		if vowels:
			lettersProse += len(word)	# Make sure we're adding words
			syllable_list.append(len(vowels))
		if len(vowels) >= 3 and not s.lower() in IgnoreWords and not word in PropNouns:
			HardWords += 1
			report = re.sub(r'(%s)\b([^<>])' % word, r'<u>\1</u>\2', report)
	#print syllable_list
	if debug:
		print ( 0.39 * float(wordsProse) / float(sentencesProse) ) 
		print ( 11.8 * average(syllable_list) )
		print 'syllable list length: %d' % len(syllable_list)
		print 'sum of syllable:%d' % sum(syllable_list)
	print '-->'
	
	# Calculate stats
	Flesch_Ease		= 206.835 - (1.015 * wordsProse / float(sentencesProse)) - (84.6 * average(syllable_list))
	ARI				= ( 4.71 * lettersProse / float(wordsProse)) + ( 0.5 * float(wordsProse) / float(sentencesProse) ) - 21.43
	Flesch_Kincaid	= ( 0.39 * wordsProse   / float(sentencesProse) ) + ( 11.8 * average(syllable_list) ) - 15.59
	Cloeman_Liau	= ( 5.89 * lettersProse / float(wordsProse) ) - ( 0.3 * float(sentencesProse) / float(wordsProse) ) - 15.8
	Gunning_fog		= 0.4 * ( (float(wordsProse) / float(sentencesProse)) + 100 * (float(HardWords) / float(wordsProse)) )
	SMOG			= 1.0430 * sqrt(30.0 * HardWords / float(sentencesProse) ) + 3.1291
	#LIX = W/S+(100*LW)/W
	
	print '<table class="wikitable" style="min-width:225px;float:left;">'
	print '<tr><th></th><th>Plain text</th><th>Prose</th></tr>'
	rowPrint('Words',       wordsText, wordsProse)
	rowPrint('Sentences',sentencesText,sentencesProse)
	rowPrint('Letters',            '', lettersProse)
	rowPrint('Polysyllabic words', '', HardWords)
	rowPrint('Characters per word','', '%.2f' % (float(lettersProse)/float(wordsProse)))
	rowPrint('Syllables per word', '', '%.2f' % average(syllable_list))
	rowPrint('Words per sentence', '', '%.2f' % (float(wordsProse)/float(sentencesProse)))
	print '</table>'
	
	# Readability Indexes (print)
	print '<table class="wikitable" style="min-width:225px;float:left;">'
	print '<tr><th>Test</th><th>U.S. grade level</th><th>Reader age</th><th>Scale</th></tr>'
	rowPrint('[[Flesch Reading Ease]]',        '', '', '%.1f' % Flesch_Ease)
	rowPrint('[[Automated Readability Index]]','%.2f'%ARI,            grade2age(ARI))
	rowPrint('[[Flesch-Kincaid Grade Level]]', '%.2f'%Flesch_Kincaid, grade2age(Flesch_Kincaid))
	rowPrint('[[Coleman-Liau Index]]',         '%.2f'%Cloeman_Liau,   grade2age(Cloeman_Liau))
	rowPrint('[[Gunning fog index]]',          '%.2f'%Gunning_fog,    grade2age(Gunning_fog))
	rowPrint('[[SMOG Index]]',                 '%.2f'%SMOG,           grade2age(SMOG))
	rowPrint('[[Laesbarhedsindex]] (LIX)',     '', '')
	rowPrint('[[Linsear Write]]',              '', '')
	print '</table>'
		
	print '<div style="clear:both;"></div>'
	# Suggestion
	if wordsProse < 300:
		suggest = ArticleWordStub
	elif wordsProse > 10000:
		suggest = ArticleWordLarge
	elif SMOG <= 8:
		suggest = ArticleLowIndex
	elif SMOG >= 16:
		suggest = ArticleHighIndex
	elif len(text) > 75e3:
		suggest = ArticleSizeLarge
	elif len(text) > 50e3:
		suggest = ArticleSizeBig
	else:
		suggest = None
	if suggest:
		print '<dl>'
		print '<dt>Notes</dt>'
		print '<dd>%s</dd>' % wikipedia.translate(page.site().language(), suggest)
		print '</dl>'

	printWiki('<div>The follow is what the software to recognized as "[[Wikipedia:Article length#What is and is not included as .22readable prose.22|Readable Prose]]", underlined words are polysyllabic words (syllables &#8805; 3) and highlighed words indicate puncuation.  No direct action should result from the information provided here.  Redistrubuted under the [[Wikipedia:Text of the GNU Free Documentation License|GNU Free Documentation License]]</div>')

	print '<table style="clear:both; table-layout:fixed;"><tr valign="top"><td style="overflow:scroll;">'
	print '<h1>Wikitext</h1>'
	printWiki(wikiText, basehref="?page=", linkClass="")
	print '</td><td style="overflow:scroll;">'
	print '<h1>Text</h1>'
	printWiki(text, linkClass="")
	print '</td><td>' 
	print '<h1>Proses (Marked up)</h1>'
	printWiki(report, linkClass="")
	print '</td></tr></table>'

def main():
	form = cgi.FieldStorage()
	global debug
	debug	= bool(form.getfirst('debug', False))
	host    = form.getfirst('hostname', 'en.wikipedia.org').split('.')
	lang    = form.getfirst('lang',     host[0])
	family  = form.getfirst('family',   host[1])
	site    = wikipedia.Site(lang, family)
	page    = wikipedia.Page(site, form.getfirst('page', ''))
	global hostname
	hostname= page.hostname()
	
	wikipedia.startContent(u'Readability for %s' % page.title(), form=False)
	print '<form action="" style="text-align:center;"><input name="page" id="page" size="40" value="%s" />' % page.title().encode('utf-8')
	print '<select name="lang">'
	printOptions(Languages, lang)
	print '</select>'
	print '<select name="family">'
	printOptions(Families, family)
	print '</select>'
	print '<input type="submit" value="Check page" /></form>'
	if page.title():
		doStats(page)
		
if __name__ == "__main__" and wikipedia.handleUrlAndHeader():
	try:
		main()
	finally:
		wikipedia.endContent()
		wikipedia.stopme()