User:Inkwina/stats.js

Code that you insert on this page could contain malicious content capable of compromising your account. If you import a script from another page with "importScript", "mw.loader.load", "iusc", or "lusc", take note that this causes you to dynamically load a remote script, which could be changed by others. Editors are responsible for all edits and actions they perform, including by scripts. User scripts are not centrally supported and may malfunction or become inoperable due to software changes. A guide to help you find broken scripts is available. If you are unsure whether code you are adding to this page is safe, you can ask at the appropriate village pump.
This code will be executed when previewing this page.
Documentation for this user script can be added at User:Inkwina/stats.
Note: After saving, you have to bypass your browser's cache to see the changes. Google Chrome, Firefox, Microsoft Edge and Safari: Hold down the ⇧ Shift key and click the Reload toolbar button. For details and instructions about other browsers, see Wikipedia:Bypass your cache.
//<pre>
//This function adds a link to the toolbox which, when clicked, displays the size of the page
//and the size of the prose. See the talk page for more details.
//
//To use this function add {{subst:js|User:Dr pda/prosesize.js}} to your monobook.js
//
function loadXMLDocPassingTemplate(url,handler, page)
{
   // branch for native XMLHttpRequest object
   if (window.XMLHttpRequest) {
       var req = new XMLHttpRequest();
   }
   // branch for IE/Windows ActiveX version
   else if (window.ActiveXObject) {
       var req = new ActiveXObject("Microsoft.XMLHTTP");
  }
  if (req) {
    req.onreadystatechange = function () {handler(req, page)};
    req.open("GET", url, true);
    req.send("");
  }
}
function getWikiText(req, page) {
   // only if req shows "loaded"
   if (req.readyState == 4) {
       // only if "OK"
       if (req.status == 200) {
           // ...processing statements go here...
 response = req.responseXML.documentElement;
        var rev = response.getElementsByTagName('rev');
 if(rev.length > 0){
   result = rev[0].getAttribute('size');
          if(result > 10240){
            result = (result/1024).toFixed(0)+'&nbsp;kB';
          }
          else{
            result = result+'&nbsp;B';
          }
          wiki_value = document.createElement("li");
   wiki_value.id = "wiki-size";
   wiki_value.innerHTML = '<b>Wiki text: </b>'+result;
   var output = document.getElementById("document-size-stats");
   prose_value = document.getElementById("prose-size");
   output.insertBefore(wiki_value,prose_value);
 }
 else{
  //alert("There was a problem using the Wikipedia Search to find the wiki text size\nEither the search is not working or the correct article did not appear on the first page of results");
          wiki_value = document.createElement("li");
   wiki_value.id = "wiki-size";
   wiki_value.innerHTML = '<b>Wiki text: </b>Problem getting wiki text size';
   var output = document.getElementById("document-size-stats");
   prose_value = document.getElementById("prose-size");
   output.insertBefore(wiki_value,prose_value);
 }
       } else {
           alert("There was a problem retrieving the XML data:\n" +
               req.statusText);
       }
   }
}
function getFileSize(req, page) {
   // only if req shows "loaded"
   if (req.readyState == 4) {
       // only if "OK"
       if (req.status == 200) {
           // ...processing statements go here...
  var fsize = req.responseText.length;
  window.status = fsize;
  var total_value = document.createElement("li");
  total_value.id = "total-size";
  total_value.innerHTML='<b>File size: </b>'+(fsize/1024).toFixed(0)+'&nbsp;kB';
  var output = document.getElementById("document-size-stats");
  var prose_html_value = document.getElementById("prose-size-html");
  output.insertBefore(total_value,prose_html_value);
        } else {
           alert("There was a problem retrieving the XML data:\n" +
               req.statusText + "\n(" + url + ")");
       }
   }
}
function getLength(id){
var textLength = 0;
for(var i=0;i<id.childNodes.length; i++){
 if(id.childNodes[i].nodeName == '#text'){
  textLength += id.childNodes[i].nodeValue.length;
 }
 else if(id.childNodes[i].id == 'coordinates' || id.childNodes[i].className.indexOf('emplate') != -1){
   //special case for {{coord}} and {{fact}}-like templates
   //Exclude from length, and don't set background yellow
   id.childNodes[i].style.cssText = 'background-color:white';
 }
 else{
   textLength += getLength(id.childNodes[i]);
 }
}
return textLength;
}
function getRefMarkLength(id,html){
var textLength = 0;
for(var i=0;i<id.childNodes.length; i++){
 if(id.childNodes[i].className == 'reference'){
  textLength += (html)? id.childNodes[i].innerHTML.length : getLength(id.childNodes[i]);
 }
}
return textLength;
}

function countSyllables(oneWord) {
  syllableCnt=0;
  oneWord = oneWord.toLowerCase();

// Specific common exceptions that don't follow the rule set below are handled individually
// Array of problem words (with word as key, syllable count as value)

  var problemWords = { "simile"    : 3,
                       "forever"   : 3,
                       "shoreline" : 2
                      };

  if( oneWord in problemWords ) {
    return problemWords[oneWord];
  }
  
  // These syllables would be counted as two but should be one
  var subSyllables = new Array(
                 'cial'
                ,'tia'
                ,'cius'
                ,'cious'
                ,'giu'
                ,'ion'
                ,'iou'
                ,'sia$'
                ,'[^aeiuoyt]{2,}ed$'
                ,'.ely$'
                ,'[cg]h?e[rsd]?$'
                ,'rved?$'
                ,'[aeiouy][dt]es?$'
                ,'[aeiouy][^aeiouydt]e[rsd]?$'
                ,'^[dr]e[aeiou][^aeiou]+$' // Sorts out deal, deign etc
                ,'[aeiouy]rse$' // Purse, hearse
            );

  // These syllables would be counted as one but should be two
  var addSyllables = new Array(
                 'ia'
                ,'riet'
                ,'dien'
                ,'iu'
                ,'io'
                ,'ii'
                ,'[aeiouym]bl$'
                ,'[aeiou]{3}'
                ,'^mc'
                ,'ism$'
                ,'([^aeiouy])\1l$'
                ,'[^l]lien'
                ,'^coa[dglx].'
                ,'[^gq]ua[^auieo]'
                ,'dnt$'
                ,'uity$'
                ,'ie(r|st)$'
            );

  // Single syllable prefixes and suffixes
  var prefixSuffix = new Array(
                 '/^un/'
                ,'/^fore/'
                ,'/ly$/'
                ,'/less$/'
                ,'/ful$/'
                ,'/ers?$/'
                ,'/ings?$/'
            );

  // Remove prefixes and suffixes and count how many were taken
  //php $strWord = preg_replace($arrPrefixSuffix, '', $strWord, -1, $intPrefixSuffixCount);
  var psCount=0;
  for (i=0; i<prefixSuffix.length; i++) {
      psCount += oneWord.match(prefixSuffix[i]+'gi').length;
      oneWord =  oneWord.replace(prefixSuffix[i]+'gi',"");
   }

// Removed non-word characters from word

  return syllableCnt;
}


function calcARI(sentence_count,word_count,letter_count) {
 return ( ( 4.71*(letter_count/word_count) ) + ( 0.5*(word_count/sentence_count) ) - 21.43 ).toFixed(2);
}

function getReadability() {
  readabilityString="";
  readabilityString +="<i>Automated Readability Index: </i>"+calcARI(sentence_count,word_count,letter_count);

  return readabilityString;
}
function getDocumentSize(){
contentDivName = '';
if(skin == 'vector' || skin == 'monobook' || skin == 'chick' || skin == 'myskin' || skin == 'simple'){
  contentDivName = 'bodyContent';
}
else if (skin == 'modern'){
  contentDivName = 'mw_contentholder';
}
else if (skin == 'standard' || skin == 'cologneblue' || skin == 'nostalgia'){
  contentDivName = 'article';
}
else{
  //fallback case; the above covers all currently existing skins
  contentDivName = 'bodyContent';
}
//Same for all skins if previewing page
if(wgAction == 'submit') contentDivName = 'wikiPreview';
var bodyContent = document.getElementById(contentDivName);
if(document.getElementById("document-size-stats")){
  //if statistics already exist, turn them off and remove highlighting
  var output = document.getElementById("document-size-stats");
  var oldStyle = output.className;
  var pList = bodyContent.getElementsByTagName("p");
  for(var i=0;i<pList.length; i++){
    if( (pList[i].parentNode.parentNode.id == contentDivName && pList[i].parentNode.className != 'previewnote')|| pList[i].parentNode.parentNode.nodeName == "BLOCKQUOTE") pList[i].style.cssText = oldStyle;
  }
  output.parentNode.removeChild(output);
  var header = document.getElementById("document-size-header");
  header.parentNode.removeChild(header);
}
else{
var output = document.createElement("ul");
output.id = "document-size-stats";
var prose_html_value = document.createElement("li");
prose_html_value.id = "prose-size-html";
output.appendChild(prose_html_value);
var ref_html_value = document.createElement("li");
ref_html_value.id = "ref-size-html";
output.appendChild(ref_html_value);
var prose_value = document.createElement("li");
prose_value.id = "prose-size";
output.appendChild(prose_value);
output.className = bodyContent.getElementsByTagName("p").item(0).style.cssText;
var ref_value = document.createElement("li");
ref_value.id = "ref-size";
output.appendChild(ref_value);
var readability_value = document.createElement("li");
readability_value.id = "readbility-scores";
output.appendChild(readability_value);
bodyContent.insertBefore(output, bodyContent.firstChild);
var header = document.createElement("span");
header.id = "document-size-header";
header.innerHTML = '<br/>Document statistics: <small><i>(See <a href="//en.wikipedia.org/wiki/User_talk:Dr_pda/prosesize.js">here</a> for details.)<i></small>';
bodyContent.insertBefore(header,output);
//File size not well defined for preview mode or section edit
if(wgAction != 'submit'){
 loadXMLDocPassingTemplate(location.pathname,getFileSize,'')
}

//Get size of images only if browser supports filesize property (IE)
var iList = bodyContent.getElementsByTagName("img");
if(iList.length >0 && iList[0].fileSize){
//Get size of images included in document
  var image_size = 0;
  var first_magnify = true;
  for (var i=0;i<iList.length; i++){
   var im = iList[i];
   if(im.getAttribute("src").indexOf("magnify-clip.png") != -1){
     if(first_magnify){
       image_size += im.fileSize*1;
       first_magnify = false;
     }
   }
   else{
     image_size += im.fileSize*1;
   }
  }
  var image_value = document.createElement("li");
  image_value.id = "image-size";
  image_value.innerHTML='<b>Images: </b>'+(image_size/1024).toFixed(0)+'&nbsp;kB';
  output.appendChild(image_value);
 }
//Calculate prose size and size of reference markers ([1] etc)
var pList = bodyContent.getElementsByTagName("p");
prose_size = 0;
prose_size_html = 0;
refmark_size = 0;
refmark_size_html = 0;
word_count = 0;
letter_count = 0;
sentence_count = 0;
for(var i=0;i<pList.length; i++){
  var para = pList[i];
  if( (para.parentNode.parentNode.id == contentDivName && para.parentNode.className != 'previewnote' ) || para.parentNode.parentNode.nodeName == "BLOCKQUOTE"){
   prose_size += getLength(para);
   prose_size_html += para.innerHTML.length;
   refmark_size += getRefMarkLength(para,false);
   refmark_size_html += getRefMarkLength(para,true);
   clean_text = para.innerHTML.replace(/(<([^>]+)>)/ig,"");
   word_count += clean_text.split(' ').length;
   sentence_count += clean_text.split(/[\.!?]/).length;
   letter_count += clean_text.replace(/[^a-z]+/ig,"").length;
   para.style.cssText = "background-color:yellow";
  }
}
if((prose_size-refmark_size)>10240){
  prose_value.innerHTML='<b>Prose size (text only): </b>'+((prose_size-refmark_size)/1024).toFixed(0)+'&nbsp;kB ('+sentence_count+' sentences, '+word_count+' words, '+letter_count+' letters) "readable prose size"';
}
else{
  prose_value.innerHTML='<b>Prose size (text only): </b>'+(prose_size-refmark_size)+'&nbsp;B ('+word_count+' words) "readable prose size"';
}
if((prose_size_html-refmark_size_html)>10240){
  prose_html_value.innerHTML='<b>Prose size (including all HTML code): </b>'+((prose_size_html-refmark_size_html)/1024).toFixed(0)+'&nbsp;kB';
}
else{
  prose_html_value.innerHTML='<b>Prose size (including all HTML code): </b>'+(prose_size_html-refmark_size_html)+'&nbsp;B';
}
//Calculate size of references (i.e. output of <references/>)
var rList = bodyContent.getElementsByTagName("ol");
var ref_size = 0;
var ref_size_html = 0;
for (var i=0; i<rList.length; i++){
  if(rList[i].parentNode.className == "references"){
    ref_size = getLength(rList[i]);
    ref_size_html = rList[i].innerHTML.length;
  }
}
if((ref_size+refmark_size)>10240){
  ref_value.innerHTML='<b>References (text only): </b>'+((ref_size+refmark_size)/1024).toFixed(0)+'&nbsp;kB';
}
else{
  ref_value.innerHTML='<b>References (text only): </b>'+(ref_size+refmark_size)+'&nbsp;B';
}
if((ref_size_html+refmark_size_html)>10240){
  ref_html_value.innerHTML='<b>References (including all HTML code): </b>'+((ref_size_html+refmark_size_html)/1024).toFixed(0)+'&nbsp;kB';
}
else{
  ref_html_value.innerHTML='<b>References (including all HTML code): </b>'+(ref_size_html+refmark_size_html)+'&nbsp;B';
}
readability_value.innerHTML='<b>Readbility: </b>'+getReadability();
//get correct name of article from wikipedia-defined global variables
var pageNameUnderscores = wgPageName;
var pageNameSpaces = pageNameUnderscores.replace(/_/g,' ')
//if page is a permalink, diff, etc don't try to search
if(!location.pathname.match('/w/index.php')){ 
 //Get revision size from API
 var searchURL = wgScriptPath + '/api.php?action=query&prop=revisions&rvprop=size&format=xml&revids=' + wgCurRevisionId;
 loadXMLDocPassingTemplate(searchURL,getWikiText,pageNameSpaces);
}
else if(wgAction == 'submit'){
  //Get size of text in edit box
  result = document.getElementById('wpTextbox1').value.length;
  if(result > 10240){
    result = (result/1024).toFixed(0)+'&nbsp;kB';
  }
  else{
    result = result+'&nbsp;B';
  }
  wiki_value = document.createElement("li");
  wiki_value.id = "wiki-size";
  wiki_value.innerHTML = '<b>Wiki text: </b>'+result;
  var output = document.getElementById("document-size-stats");
  prose_value = document.getElementById("prose-size");
  output.insertBefore(wiki_value,prose_value);
}
}
}
$(function () {
 if(wgAction == 'edit' || (wgAction == 'submit' && document.getElementById('wikiDiff')) ){
   mw.util.addPortletLink('p-tb', 'javascript:alert("You need to preview the text for the prose size script to work in edit mode.")', 'Page size', 't-page-size', 'Calculate page and prose size', '', '');
   document.getElementById("t-page-size").firstChild.style.cssText = "color:black;"
 }
 else if(wgAction == 'view' || wgAction == 'submit' || wgAction == 'historysubmit' || wgAction == 'purge'){
   mw.util.addPortletLink('p-tb', 'javascript:getDocumentSize()', 'Page size', 't-page-size', 'Calculate page and prose size', '', '');
 }
});
//</pre>