<%@ page language="java" contentType="text/html; charset=UTF-8" pageEncoding="UTF-8"%>
<%@ page import="org.crosswire.utils.HTTPUtils" %>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en-US" lang="en-US">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8" />
<%
String lhs = request.getParameter("lhs");
if (lhs != null) lhs = new String(lhs.getBytes("iso8859-1"), "UTF-8");
String rhs = request.getParameter("rhs");
if (rhs != null) rhs = new String(rhs.getBytes("iso8859-1"), "UTF-8");
%>
<script type="text/javascript" charset=utf-8" src="jquery-1.6.4.min.js"></script>
<script type="text/javascript">
<!-- charset=utf-8" src="reconciler.js"> -->
var myURL='.';
var file1 = null;
var file2 = null;
var file1String;
var file2String;
var notesRegex = /\{(.*?)\}/g; //vat par check removed from regex
var structureRegex = /[\||<](.*?)[\||>]/g;
var bookRegex = /\|\s*?B\s*?(\d+?)\|/g;
var chapterRegex = /\|\s*?K\s*?(\d+?)\|/g;
var verseRegex = /\|\s*?V\s*?(\d+?)\|/g;
var folioRegex = /^\|F\s*?(\d+?)([rvab])\S{0,1}\||^\|F([rvab])\s*?(\d+?)\S{0,1}\|/i;
var pageRegex = /^\|P\s*?(\d+?)\|/i;
var columnRegex = /^\|C\s*?(\d+?)\|/i;
var linebreakRegex = /^\|L\s*?\d*\|/i;
var file1Background = '#990033;';
var file1TextColour = 'white;';
var file2Background = '#009933;';
var file2TextColour = 'white;';
var checkedState = 'show';
var results = {};
function deletefile(file){
document.getElementById(file + '_details').innerHTML = '';
if (file === 'file1'){
file1 = null;
}
if (file === 'file2'){
file2 = null;
}
}
function uploadFile(){
f = document.getElementById('file').files[0];
if (file1 === null){
file1 = f;
document.getElementById('file1_details').innerHTML = escape(f.name) + '<img class="delete_logo" onclick="javascript:deletefile(\'file1\');" alt="delete" title="delete this file" src="'+myURL+'/delete.png"/>';
}
else {
if (file2 === null){
file2 = f;
document.getElementById('file2_details').innerHTML = escape(f.name)+ '<img class="delete_logo" onclick="javascript:deletefile(\'file2\');" alt="delete" title="delete this file" src="'+myURL+'/delete.png"/>';
}
else {
alert('two files already loaded');
}
}
}
function format_text(text){
text = text.replace(/(\r\n|\n|\r)/gm, ' ');
// text = text.replace(/\u006F/g, '\u03BF');//fix for dodgy omicrons
text = text.replace(notesRegex, '\n{$1}\n');
lines = text.split('\n');
for (var i=0; i<lines.length; i++){
line = lines[i];
if (line.length > 0 && line.search(/\{Status/) != -1){
lines[i] = '';
} else {
if (line.length > 0 && line[0] === '{'){
line = line.replace(/</g, '-');
line = line.replace(/\|/g, '-');
line = line.replace(/>/g, '-');
line = line.replace(/\[/g, '(');
line = line.replace(/\]/g, ')');
line = line.replace(/\u03BF/g, '\u006F'); // put comment os back to Latin not Greek
lines[i] = line;
}
}
}
text = lines.join('\n');
text = text.replace(structureRegex, '\n|$1|\n');
lines = text.split('\n');
return lines;
}
function trim(s) {
s = s.replace(/(^\s*)|(\s*$)/gi,"");
s = s.replace(/[ ]{2,}/gi," ");
s = s.replace(/\n /,"\n");
return s;
}
function clean_verse_text(text, ignore_layout, ignore_comments, ignore_tags, ignore_final_nu, ignore_punctuation, ignore_om, ignore_comm, ignore_lect){
text = text.toLowerCase();
text = text.replace(/\u0305/g, '~');//get rid of combining nomsac lines
text = text.replace(/\u0323/g, '_');//get rid of underdots
if (ignore_final_nu == true){
text = text.replace(/\u0304/g, '\u03BD');//final nu replacement
}
if (ignore_om){
text = text.replace(/\{\s*om\s*\}/g, '');
}
if (ignore_comm){
text = text.replace(/\{\s*comm\s*\}/g, '');
}
if (ignore_lect){
text = text.replace(/\{\s*lect\s*\}/g, '');
}
if (ignore_punctuation){
text = text.replace(/\u0387/g, ''); //ano teleia
text = text.replace(/\u00B7/g, '');//middle dot
text = text.replace(/\u037E/g, '');//greek question mark
text = text.replace(/\./g, '');
text = text.replace(/;/g, '');
text = text.replace(/,/g, '');
text = text.replace(/:/g, '');
}
if (ignore_layout == true){
while (text.search('= ') != -1){
text = text.replace('= ', '');
}
}
if (ignore_comments == true){
text = text.replace(/\{.+?\}/g, '');
}
if (ignore_tags == true){
text = text.replace(/\[.+?\]/g, '');
}
return text;
}
function extract_verse_text(text, ignore_layout, ignore_comments, ignore_tags, ignore_final_nu, ignore_punctuation, ignore_om, ignore_comm, ignore_lect){
lines = format_text(text);
verse_dict = {};
b = null;
k = null;
v = null;
working = [];
inVerse = false;
key = 'preVerse1';
for (var i=0; i<lines.length; i+=1){
line = lines[i];
if (line.search(bookRegex) !== -1){
b = line.replace(bookRegex, '$1');
}
else {
if (line.search(chapterRegex) !== -1){
k = line.replace(chapterRegex, '$1');
if (k.length === 1){
k = '0' + k;
}
}
else {
if (line.search(verseRegex) !== -1){
if (working.length > 0){
if (verse_dict[key] === undefined){
verse_dict[key] = trim(clean_verse_text(working.join(' '), ignore_layout, ignore_comments, ignore_tags, ignore_final_nu, ignore_punctuation, ignore_om, ignore_comm, ignore_lect));
}
}
v = line.replace(verseRegex, '$1');
if (v.length == 1){
v = '0' + v;
}
key = b + '_' + k + ':' + v;
working = [];
}
else {
if (ignore_layout == true && (line.search(linebreakRegex) != -1 || line.search(columnRegex) != -1 || line.search(pageRegex) != -1 || line.search(folioRegex) != -1)){
//squelch
}
else {
working.push(trim(line));
}
}
}
}
}
if (working.length > 0){
verse_dict[key] = trim(clean_verse_text(working.join(' '), ignore_layout));
}
//console.log(verse_dict);
return verse_dict;
}
function extract_structure(text, ignore_layout){
lines = format_text(text);
structure_list = [];
for (var i=0; i<lines.length; i++){
line = lines[i];
if (line.length > 0 && line[0] === '|'){
if (ignore_layout === true){
if (line.search(bookRegex) !== -1){
structure_list.push(trim(line));
}
else {
if (line.search(chapterRegex) !== -1){
structure_list.push(trim(line));
}
else {
if (line.search(verseRegex) !== -1){
structure_list.push(trim(line));
}
}
}
}
else {
structure_list.push(trim(line));
}
}
}
return structure_list;
}
function textToList(text, ignore_structure){
lines = format_text(text);
return [];
}
function getKeys(obj){
var keys = [];
for(var key in obj){
if(obj.hasOwnProperty(key)) {
keys.push(key);
}
}
return keys;
}
function unique(a){
a.sort();
for(var i = 1; i < a.length; ){
if(a[i-1] == a[i]){
a.splice(i, 1);
} else {
i+=1;
}
}
return a;
}
function compareStrings(s1, s2){
if (s1 === s2){
return '';
}
else {
return WDiffString(s1, s2);//diffString(s1, s2);//
}
}
function toggle_checked(id){
if (document.getElementById(id).innerHTML == 'Hide checked verses'){
var checkboxes = document.getElementsByName('marked');
for (var i=0; i<checkboxes.length; i++){
if (checkboxes[i].checked){
checkboxes[i].parentNode.style.display = 'none';
}
}
document.getElementById(id).innerHTML = 'Show hidden verses';
checkedState = 'hide';
} else {
var checkboxes = document.getElementsByName('marked');
for (var i=0; i<checkboxes.length; i++){
if (checkboxes[i].checked){
checkboxes[i].parentNode.style.display = 'block';
}
}
document.getElementById(id).innerHTML = 'Hide checked verses';
checkedState = 'show';
}
}
function toggle_results(id, link_id){
if(document.getElementById(id).style.display == 'none'){
document.getElementById(id).style.display = 'block';
document.getElementById(link_id).innerHTML = document.getElementById(link_id).innerHTML.replace('Show', 'Hide');
}
else {
if(document.getElementById(id).style.display == 'block'){
document.getElementById(id).style.display = 'none';
document.getElementById(link_id).innerHTML = document.getElementById(link_id).innerHTML.replace('Hide', 'Show');
}
}
}
function compareFiles(s1, s2, ignore_structure, ignore_comments, ignore_tags, ignore_final_nu, ignore_punctuation, ignore_om, ignore_comm, ignore_lect){
document.getElementById('results_container').style.display = 'block';
var list1 = extract_structure(s1, ignore_structure);
var list2 = extract_structure(s2, ignore_structure);
var outlineResult = compareStrings(list1.join(' '), list2.join(' '));
document.getElementById('results_summary').innerHTML = 'Comparing <span style="background: ' + file1Background + ' color: ' + file1TextColour + '"> ' + file1.name + ' </span> with <span style="background: ' + file2Background + ' color: ' + file2TextColour + '"> ' + file2.name + ' </span><br/>';
if (outlineResult == ''){
document.getElementById('results_summary').innerHTML += 'Structures identical.<br/>';
}
else {
document.getElementById('results_summary').innerHTML += 'Structures are different.<br/>';
document.getElementById('structure_results').innerHTML += outlineResult.replace(/\|K/g, '\<br/>|K');
document.getElementById('structure_results_toggle').innerHTML += 'Show Structure Results';
}
var dict1 = extract_verse_text(s1, ignore_structure, ignore_comments, ignore_tags, ignore_final_nu, ignore_punctuation, ignore_om, ignore_comm, ignore_lect);
var dict2 = extract_verse_text(s2, ignore_structure, ignore_comments, ignore_tags, ignore_final_nu, ignore_punctuation, ignore_om, ignore_comm, ignore_lect);
var compString1 = '';
var compString2 = '';
var keys = getKeys(dict1).concat(getKeys(dict2));
keys = unique(keys);
for (var i=0; i<keys.length; i++){
if (dict1.hasOwnProperty(keys[i])){
compString1 = dict1[keys[i]];
} else {
compString1 = '';
}
if (dict2.hasOwnProperty(keys[i])){
compString2 = dict2[keys[i]];
} else {
compString2 = '';
}
var result = compareStrings(compString1, compString2);
if (result != ''){
results[keys[i]] = '<input size="100" type="text" name="comments" onchange="update(this);"/><br/><input name="marked" type="checkbox" onchange="update(this)"/>' + result;
}
}
var output = '<input onchange="javascript:toggleAll(this);" type="checkbox"><label>select/deselect all</label><a id="toggle_checked" href="javascript:toggle_checked(\'toggle_checked\')">Hide checked verses</a><br/>';
var keys = getKeys(results);
if (keys.length == 0){
document.getElementById('results_summary').innerHTML += 'Biblical text identical.<br/>';
}
else {
for (var i=0; i<keys.length; i++){
output = output + '<div class="verse_comp">' + keys[i] + ' ' + results[keys[i]] + '</div>';
}
document.getElementById('results_summary').innerHTML += keys.length + ' Biblical verses have differences.<br/>';
document.getElementById('text_results').innerHTML = output;
document.getElementById('text_results_toggle').innerHTML = 'Show Text Results';
}
}
function parse_XML(xml){
try {
xmlDoc=new ActiveXObject("Microsoft.XMLDOM");
xmlDoc.async="false";
xmlDoc.loadXML(xml);
} catch(e) {
parser=new DOMParser();
xmlDoc=parser.parseFromString(xml,"text/xml");
}
return xmlDoc;
}
//try to get a valid dom by correcting once for a known bug in the diff.js library
function get_dom(string){
var xmlDoc = parse_XML(string);
if (xmlDoc.documentElement.nodeName=="parsererror"){
xmlDoc = parse_XML(string.replace(/<\/container>/, '</span></container>'));
if (xmlDoc.documentElement.nodeName=="parsererror"){
return false;
} else {
return xmlDoc;
}
} else {
return xmlDoc;
}
}
//given a difference node find the full word that contains it. Stop at space and ] or [
function get_word_limits(startNode) {
//remove outer if statements for a word of context on one side of a full word change
var following_text = '';
var preceding_text = '';
var following, previous;
if ((startNode.childNodes[0].nodeValue[0] != ' ' && startNode.childNodes[0].nodeValue[0] != '[')
|| ((startNode.childNodes[0].nodeValue[0] == ' ' || startNode.childNodes[0].nodeValue[0] == '[')
&& startNode.childNodes[0].nodeValue.length == 1)){
if (startNode.previousSibling != null && startNode.previousSibling.nodeType === 3){
previous = startNode.previousSibling;
if (previous.nodeValue[previous.nodeValue.length-1] !== ' ' && previous.nodeValue[previous.nodeValue.length-1] !== ']'){
preceding_text = previous.nodeValue.split(' ');
preceding_text = preceding_text[preceding_text.length-1];
preceding_text = preceding_text.substring(preceding_text.lastIndexOf(']')+1);
//console.log(preceding_text)
}
}
}
if ((startNode.childNodes[0].nodeValue[startNode.childNodes[0].nodeValue.length-1] != ' ' && startNode.childNodes[0].nodeValue[startNode.childNodes[0].nodeValue.length-1] != ']')
|| ((startNode.childNodes[0].nodeValue[startNode.childNodes[0].nodeValue.length-1] == ' ' || startNode.childNodes[0].nodeValue[startNode.childNodes[0].nodeValue.length-1] == ']')
&& startNode.childNodes[0].nodeValue.length == 1)){
if (startNode.nextSibling != null && startNode.nextSibling.nodeType === 3){
following = startNode.nextSibling;
//console.log(following)
if (following.nodeValue[0] !== ' ' && following.nodeValue[0] !== '['){
following_text = following.nodeValue.split(' ')[0];
if (following_text.indexOf('[') !== -1){
following_text = following_text.substring(0, following_text.indexOf('['));
}
//console.log(following_text)
}
}
}
return [preceding_text, following_text];
}
//create a list of differences from this verse
//input: the html string of a verse from the text results view
//output: a 2 item list per difference all wrapped up in another list
function process_diffs(string) {
// console.log(string)
var differences = [];
var i, xmlDoc, spans, spanType, surrounding_text, last_span, first_span, pre_string, post_string;
var text1_string = '';
var text2_string = '';
xmlDoc = get_dom(string);
if (xmlDoc == false){
console.log(parse_XML(string));
} else {
spans = xmlDoc.getElementsByTagName('span');
for (i=0; i<spans.length; i+=1) {
//treat as stand alone variant IF:
// the nextSibling is not a span
// OR
// the nextSibling is a span AND the current span ends with a space OR the next span starts with a space
// OR
// there is no nextSibling
if ((spans[i].nextSibling != null && spans[i].nextSibling.nodeName !== 'span')
|| (spans[i].nextSibling != null && spans[i].nextSibling.nodeName == 'span'
&& (spans[i].childNodes[0].nodeValue[spans[i].childNodes[0].nodeValue.length-1] == ' ' || spans[i].nextSibling.childNodes[0].nodeValue[0] == ' '))
|| spans[i].nextSibling == null){
//examples like GK 1:18 could be better
spanType = spans[i].getAttribute('class');
if (spanType === 'wDiffHtmlDelete') {
surrounding_text = get_word_limits(spans[i]);
if (surrounding_text[0] !== '' || surrounding_text[1] !== ''){
differences.push([surrounding_text[0] + spans[i].childNodes[0].nodeValue + surrounding_text[1], surrounding_text[0] + surrounding_text[1]]);
} else {
differences.push([spans[i].childNodes[0].nodeValue, '[BLANK]']);
}
} else {
if (spanType === 'wDiffHtmlInsert') {
surrounding_text = get_word_limits(spans[i]);
if (surrounding_text[0] !== '' || surrounding_text[1] !== ''){
differences.push([surrounding_text[0] + surrounding_text[1], surrounding_text[0] + spans[i].childNodes[0].nodeValue + surrounding_text[1]]);
} else {
differences.push(['[BLANK]', spans[i].childNodes[0].nodeValue]);
}
}
}
} else {
text1_string = '';
text2_string = '';
first_span = spans[i];
last_span = null;
spanType = spans[i].getAttribute('class');
if (spanType === 'wDiffHtmlDelete') {
text1_string += spans[i].childNodes[0].nodeValue;
} else {
if (spanType === 'wDiffHtmlInsert') {
text2_string += spans[i].childNodes[0].nodeValue;
}
}
while (spans[i].nextSibling !== null && spans[i].nextSibling.nodeName === 'span'){
spanType = spans[i].nextSibling.getAttribute('class');
if (spanType === 'wDiffHtmlDelete') {
text1_string += spans[i].nextSibling.childNodes[0].nodeValue;
} else {
if (spanType === 'wDiffHtmlInsert') {
text2_string += spans[i].nextSibling.childNodes[0].nodeValue;
}
}
last_span = spans[i].nextSibling;
i+=1;
}
pre_string = get_word_limits(first_span);
post_string = get_word_limits(last_span);
if (text1_string === ''){
text1_string ='[BLANK]';
}
if (text2_string === ''){
text2_string = '[BLANK]';
}
differences.push([pre_string[0] + text1_string + post_string[1], pre_string[0] + text2_string + post_string[1]]);
//spaces in both strings need to be taken into account for pre/post string
}
}
}
return differences;
}
function saveText(){
newwindow1=open('','name','height=200,width=150,scrollbars=yes');
var tmp = newwindow1.document;
var keys = getKeys(results);
for (var i=0; i<keys.length; i+=1){
tmp.write(keys[i] + '<br/>');
string = results[keys[i]].replace(/ /g, ' ');
diffs = process_diffs('<container>' + string + '</container>');
//console.log(diffs);
for (var j=0; j<diffs.length; j+=1) {
tmp.write(diffs[j][0]);
tmp.write(' ] ');
tmp.write(diffs[j][1]);
tmp.write('<br/>');
}
tmp.write('<br/>');
}
tmp.close();
}
function getReconciliationString() {
var textString = escape(document.getElementById('results_summary').innerHTML) + '|||' + escape(document.getElementById('structure_results').innerHTML) + '|||' + escape(document.getElementById('text_results').innerHTML);
return textString;
}
function saveReconciliation(){
//option 3 - put serialisation in popup window for users to copy and paste into a file
var textString = getReconciliationString();
console.log(textString);
newwindow2=open('','name','height=200,width=150,scrollbars=yes');
var tmp = newwindow2.document;
tmp.write(textString);
tmp.close();
}
function toggleAll(elem){
boxes = document.getElementsByName('marked');
if (elem.checked){
elem.setAttribute('checked', 'checked');
for (var i=0; i<boxes.length; i+=1){
boxes[i].checked = true;
boxes[i].setAttribute('checked', 'checked');
update(boxes[i]);
}
} else {
elem.removeAttribute('checked');
for (var i=0; i<boxes.length; i+=1){
boxes[i].checked = false;
boxes[i].removeAttribute('checked');
update(boxes[i]);
}
}
}
function update(elem){
if (elem.type === 'checkbox'){
if (elem.checked){
elem.setAttribute('checked', 'checked');
if (checkedState === 'hide'){
elem.parentNode.style.display = 'none';
}
} else {
elem.removeAttribute('checked');
if (checkedState === 'hide'){
elem.parentNode.style.display = 'block';
}
}
} else {
elem.setAttribute('value', elem.value);
}
}
function uploadReconciliation (){
f = document.getElementById('reconciliation').files[0];
reader = new FileReader();
reader.onload = ( function (){
return function (evt){
var fullString = evt.target.result;
uploadReconciliationData(fullString);
};
}());
reader.readAsText(f);
return f.name;
}
function uploadReconciliationData(fullString) {
var body = document.getElementsByTagName('body')[0];
body.className = 'waiting';
document.getElementById('results_container').style.display = 'none';
document.getElementById('text_results').style.display = 'none';
document.getElementById('structure_results').style.display = 'none';
document.getElementById('text_results_toggle').innerHTML = '';
document.getElementById('structure_results_toggle').innerHTML = '';
var stringList = fullString.split('|||');
var summaryString = stringList[0];
var structureString = stringList[1];
var textString = stringList[2];
document.getElementById('results_summary').innerHTML = unescape(summaryString);
document.getElementById('structure_results').innerHTML = unescape(structureString);
document.getElementById('text_results').innerHTML = unescape(textString);
document.getElementById('results_container').style.display = 'block';
document.getElementById('structure_results_toggle').innerHTML = 'Show Structure Results';
document.getElementById('text_results_toggle').innerHTML = 'Show Text Results';
body.className = 'active';
document.getElementById('file1_details').innerHTML = '';
document.getElementById('file2_details').innerHTML = '';
file1 = file2 = null;
return;
}
function compare() {
/*
if (file1 == null || file2 == null){
alert('Please select two files before continuing.');
}
*/
var body = document.getElementsByTagName('body')[0];
body.className = 'waiting';
var ignore_structure = document.getElementById('layout').checked;
var ignore_comments = document.getElementById('all_comments').checked;
var ignore_tags = document.getElementById('tags').checked;
var ignore_final_nu = document.getElementById('final_nu').checked;
var ignore_punctuation = document.getElementById('punctuation').checked;
var ignore_om = document.getElementById('comment_om').checked;
var ignore_comm = document.getElementById('comment_comm').checked;
var ignore_lect = document.getElementById('comment_lect').checked;
if (true) {
// if (file1 == '[buffer 1]' && file2 == '[buffer 2]') {
var lhs = decodeURIComponent($('#lhsdata').val());
var rhs = decodeURIComponent($('#rhsdata').val());
compareFiles(lhs, rhs, ignore_structure, ignore_comments, ignore_tags, ignore_final_nu, ignore_punctuation, ignore_om, ignore_comm, ignore_lect);
body.className = 'active';
}
else {
readFiles(function(buffer1, buffer2) {
compareFiles(buffer1, buffer2, ignore_structure, ignore_comments, ignore_tags, ignore_final_nu, ignore_punctuation, ignore_om, ignore_comm, ignore_lect);
body.className = 'active';
});
}
}
function readFiles(callback) {
var reader1 = new FileReader();
var reader2 = new FileReader();
document.getElementById('results_container').style.display = 'none';
document.getElementById('text_results').style.display = 'none';
document.getElementById('structure_results').style.display = 'none';
document.getElementById('text_results_toggle').innerHTML = '';
document.getElementById('structure_results_toggle').innerHTML = '';
reader2.onload = (function () {
return function (evt) {
file2String = evt.target.result;
callback(file1String, file2String);
};
}(file2String));
reader1.onload = (function () {
return function (evt) {
file1String = evt.target.result;
reader2.readAsText(file2);
};
}(file1String));
reader1.readAsText(file1);
}
/*
Name: wDiff.js
Version: 0.9.9 (October 10, 2010)
Info: http://en.wikipedia.org/wiki/User:Cacycle/diff
Code: http://en.wikipedia.org/wiki/User:Cacycle/diff.js
JavaScript diff algorithm by [[en:User:Cacycle]] (http://en.wikipedia.org/wiki/User_talk:Cacycle).
Outputs html/css-formatted new text with highlighted deletions, inserts, and block moves.
For newline highlighting the following style rules have to be added to the document:
.wDiffParagraph:before { content: "¶"; };
The program uses cross-browser code and should work with all modern browsers. It has been tested with:
* Mozilla Firefox 1.5.0.1
* Mozilla SeaMonkey 1.0
* Opera 8.53
* Internet Explorer 6.0.2900.2180
* Internet Explorer 7.0.5730.11
This program is also compatible with Greasemonkey
An implementation of the word-based algorithm from:
Communications of the ACM 21(4):264 (1978)
http://doi.acm.org/10.1145/359460.359467
With the following additional feature:
* Word types have been optimized for MediaWiki source texts
* Additional post-pass 5 code for resolving islands caused by adding
two common words at the end of sequences of common words
* Additional detection of block borders and color coding of moved blocks and their original position
* Optional "intelligent" omission of unchanged parts from the output
This code is used by the MediaWiki in-browser text editors [[en:User:Cacycle/editor]] and [[en:User:Cacycle/wikEd]]
and the enhanced diff view tool wikEdDiff [[en:User:Cacycle/wikEd]].
Usage: var htmlText = WDiffString(oldText, newText);
This code has been released into the public domain.
Datastructures (abbreviations from publication):
text: an object that holds all text related datastructures
.newWords: consecutive words of the new text (N)
.oldWords: consecutive words of the old text (O)
.newToOld: array pointing to corresponding word number in old text (NA)
.oldToNew: array pointing to corresponding word number in new text (OA)
.message: output message for testing purposes
symbol table:
symbols[word]: associative array (object) of detected words for passes 1 - 3, points to symbol[i]
symbol[i]: array of objects that hold word counters and pointers:
.newCtr: new word occurences counter (NC)
.oldCtr: old word occurences counter (OC)
.toNew: first word occurrence in new text, points to text.newWords[i]
.toOld: last word occurrence in old text, points to text.oldWords[i]
block: an object that holds block move information
blocks indexed after new text:
.newStart: new text word number of start of this block
.newLength: element number of this block including non-words
.newWords: true word number of this block
.newNumber: corresponding block index in old text
.newBlock: moved-block-number of a block that has been moved here
.newLeft: moved-block-number of a block that has been moved from this border leftwards
.newRight: moved-block-number of a block that has been moved from this border rightwards
.newLeftIndex: index number of a block that has been moved from this border leftwards
.newRightIndex: index number of a block that has been moved from this border rightwards
blocks indexed after old text:
.oldStart: word number of start of this block
.oldToNew: corresponding new text word number of start
.oldLength: element number of this block including non-words
.oldWords: true word number of this block
*/
// css for change indicators
if (typeof(wDiffStyleDelete) == 'undefined') { wDiffStyleDelete = 'font-weight: normal; text-decoration: none; color: #fff; background-color: #990033;'; }
if (typeof(wDiffStyleInsert) == 'undefined') { wDiffStyleInsert = 'font-weight: normal; text-decoration: none; color: #fff; background-color: #009933;'; }
if (typeof(wDiffStyleMoved) == 'undefined') { wDiffStyleMoved = 'font-weight: bold; color: #000; vertical-align: text-bottom; font-size: xx-small; padding: 0; border: solid 1px;'; }
if (typeof(wDiffStyleBlock) == 'undefined') { wDiffStyleBlock = [
'color: #000; background-color: #ffff80;',
'color: #000; background-color: #c0ffff;',
'color: #000; background-color: #ffd0f0;',
'color: #000; background-color: #ffe080;',
'color: #000; background-color: #aaddff;',
'color: #000; background-color: #ddaaff;',
'color: #000; background-color: #ffbbbb;',
'color: #000; background-color: #d8ffa0;',
'color: #000; background-color: #d0d0d0;'
]; }
// html for change indicators, {number} is replaced by the block number
// {block} is replaced by the block style, class and html comments are important for shortening the output
if (typeof(wDiffHtmlMovedRight) == 'undefined') { wDiffHtmlMovedRight = '<input class="wDiffHtmlMovedRight" type="button" value=">" style="' + wDiffStyleMoved + ' {block}"><!--wDiffHtmlMovedRight-->'; }
if (typeof(wDiffHtmlMovedLeft) == 'undefined') { wDiffHtmlMovedLeft = '<input class="wDiffHtmlMovedLeft" type="button" value="<" style="' + wDiffStyleMoved + ' {block}"><!--wDiffHtmlMovedLeft-->'; }
if (typeof(wDiffHtmlBlockStart) == 'undefined') { wDiffHtmlBlockStart = '<span class="wDiffHtmlBlock" style="{block}">'; }
if (typeof(wDiffHtmlBlockEnd) == 'undefined') { wDiffHtmlBlockEnd = '</span><!--wDiffHtmlBlock-->'; }
if (typeof(wDiffHtmlDeleteStart) == 'undefined') { wDiffHtmlDeleteStart = '<span class="wDiffHtmlDelete" style="' + wDiffStyleDelete + '">'; }
if (typeof(wDiffHtmlDeleteEnd) == 'undefined') { wDiffHtmlDeleteEnd = '</span>'; }
if (typeof(wDiffHtmlInsertStart) == 'undefined') { wDiffHtmlInsertStart = '<span class="wDiffHtmlInsert" style="' + wDiffStyleInsert + '">'; }
if (typeof(wDiffHtmlInsertEnd) == 'undefined') { wDiffHtmlInsertEnd = '</span>'; }
// minimal number of real words for a moved block (0 for always displaying block move indicators)
if (typeof(wDiffBlockMinLength) == 'undefined') { wDiffBlockMinLength = 3; }
// exclude identical sequence starts and endings from change marking
if (typeof(wDiffWordDiff) == 'undefined') { wDiffWordDiff = false; }
// enable recursive diff to resolve problematic sequences
if (typeof(wDiffRecursiveDiff) == 'undefined') { wDiffRecursiveDiff = true; }
// enable block move display
if (typeof(wDiffShowBlockMoves) == 'undefined') { wDiffShowBlockMoves = true; }
// remove unchanged parts from final output
// characters before diff tag to search for previous heading, paragraph, line break, cut characters
if (typeof(wDiffHeadingBefore) == 'undefined') { wDiffHeadingBefore = 1500; }
if (typeof(wDiffParagraphBefore) == 'undefined') { wDiffParagraphBefore = 1500; }
if (typeof(wDiffLineBeforeMax) == 'undefined') { wDiffLineBeforeMax = 1000; }
if (typeof(wDiffLineBeforeMin) == 'undefined') { wDiffLineBeforeMin = 500; }
if (typeof(wDiffBlankBeforeMax) == 'undefined') { wDiffBlankBeforeMax = 1000; }
if (typeof(wDiffBlankBeforeMin) == 'undefined') { wDiffBlankBeforeMin = 500; }
if (typeof(wDiffCharsBefore) == 'undefined') { wDiffCharsBefore = 500; }
// characters after diff tag to search for next heading, paragraph, line break, or characters
if (typeof(wDiffHeadingAfter) == 'undefined') { wDiffHeadingAfter = 1500; }
if (typeof(wDiffParagraphAfter) == 'undefined') { wDiffParagraphAfter = 1500; }
if (typeof(wDiffLineAfterMax) == 'undefined') { wDiffLineAfterMax = 1000; }
if (typeof(wDiffLineAfterMin) == 'undefined') { wDiffLineAfterMin = 500; }
if (typeof(wDiffBlankAfterMax) == 'undefined') { wDiffBlankAfterMax = 1000; }
if (typeof(wDiffBlankAfterMin) == 'undefined') { wDiffBlankAfterMin = 500; }
if (typeof(wDiffCharsAfter) == 'undefined') { wDiffCharsAfter = 500; }
// maximal fragment distance to join close fragments
if (typeof(wDiffFragmentJoin) == 'undefined') { wDiffFragmentJoin = 1000; }
if (typeof(wDiffOmittedChars) == 'undefined') { wDiffOmittedChars = '…'; }
if (typeof(wDiffOmittedLines) == 'undefined') { wDiffOmittedLines = '<hr style="height: 2px; margin: 1em 10%;">'; }
if (typeof(wDiffNoChange) == 'undefined') { wDiffNoChange = '<hr style="height: 2px; margin: 1em 20%;">'; }
// compatibility fix for old name of main function
StringDiff = WDiffString;
// WDiffString: main program
// input: oldText, newText, strings containing the texts
// returns: html diff
function WDiffString(oldText, newText) {
// IE / Mac fix
oldText = oldText.replace(/\r\n?/g, '\n');
newText = newText.replace(/\r\n?/g, '\n');
var text = {};
text.newWords = [];
text.oldWords = [];
text.newToOld = [];
text.oldToNew = [];
text.message = '';
var block = {};
var outText = '';
// trap trivial changes: no change
if (oldText == newText) {
outText = newText;
outText = WDiffEscape(outText);
outText = WDiffHtmlFormat(outText);
return(outText);
}
// trap trivial changes: old text deleted
if ( (oldText == null) || (oldText.length == 0) ) {
outText = newText;
outText = WDiffEscape(outText);
outText = WDiffHtmlFormat(outText);
outText = wDiffHtmlInsertStart + outText + wDiffHtmlInsertEnd;
return(outText);
}
// trap trivial changes: new text deleted
if ( (newText == null) || (newText.length == 0) ) {
outText = oldText;
outText = WDiffEscape(outText);
outText = WDiffHtmlFormat(outText);
outText = wDiffHtmlDeleteStart + outText + wDiffHtmlDeleteEnd;
return(outText);
}
// split new and old text into words
WDiffSplitText(oldText, newText, text);
// calculate diff information
WDiffText(text);
//detect block borders and moved blocks
WDiffDetectBlocks(text, block);
// process diff data into formatted html text
outText = WDiffToHtml(text, block);
// IE fix
outText = outText.replace(/> ( *)</g, '> $1<');
return(outText);
};
//
//
//// WDiffSplitText: split new and old text into words
//// input: oldText, newText, strings containing the texts
//// changes: text.newWords and text.oldWords, arrays containing the texts in arrays of words
//
function WDiffSplitText(oldText, newText, text) {
// convert strange spaces
oldText = oldText.replace(/[\t\u000b\u00a0\u2028\u2029]+/g, ' ');
newText = newText.replace(/[\t\u000b\u00a0\u2028\u2029]+/g, ' ');
// split old text into words
// / | | | | | | | | | | | | | | /
//var pattern=/[\w]+|\[\[|\]\]|\{\{|\}\}|\n+| +|&\w+;|'''|''|=+|\{\||\|\}|\|\-|./g;
var pattern = /[\w]+|[\u03B1-\u03C9\u0391-\u03A9~_]+|\[\[|\]\]|\{\{|\}\}|\n+| +|&\w+;|'''|''|=+|\{\||\|\}|\|\-|./g;
var result;
do {
result = pattern.exec(oldText);
if (result != null) {
text.oldWords.push(result[0]);
}
} while (result != null);
// split new text into words
do {
result = pattern.exec(newText);
if (result != null) {
text.newWords.push(result[0]);
}
} while (result != null);
// console.log(text.newWords)
// console.log(text.oldWords)
return;
};
// WDiffText: calculate diff information
// input: text.newWords and text.oldWords, arrays containing the texts as arrays of words
// optionally for recursive calls: newStart, newEnd, oldStart, oldEnd, recursionLevel
// changes: text.newToOld and text.oldToNew, arrays pointing to corresponding words
function WDiffText(text, newStart, newEnd, oldStart, oldEnd, recursionLevel) {
var symbol = [];
var symbols = {};
// set defaults
if (typeof(newStart) == 'undefined') { newStart = 0; }
if (typeof(newEnd) == 'undefined') { newEnd = text.newWords.length; }
if (typeof(oldStart) == 'undefined') { oldStart = 0; }
if (typeof(oldEnd) == 'undefined') { oldEnd = text.oldWords.length; }
if (typeof(recursionLevel) == 'undefined') { recursionLevel = 0; }
// limit recursion depth
if (recursionLevel > 10) {
return;
}
//
// pass 1: Parse new text into symbol table
//
for (var i = newStart; i < newEnd; i ++) {
var word = text.newWords[i];
// preserve the native method
if (word.indexOf('hasOwnProperty') == 0) {
word = word.replace(/^(hasOwnProperty_*)$/, '$1_');
}
// add new entry to symbol table
if (symbols.hasOwnProperty(word) == false) {
var last = symbol.length;
symbols[word] = last;
symbol[last] = { newCtr: 1, oldCtr: 0, toNew: i, toOld: null };
}
// or update existing entry
else {
// increment word counter for new text
var hashToArray = symbols[word];
symbol[hashToArray].newCtr ++;
}
}
//
// pass 2: parse old text into symbol table
//
for (var i = oldStart; i < oldEnd; i ++) {
var word = text.oldWords[i];
// preserve the native method
if (word.indexOf('hasOwnProperty') == 0) {
word = word.replace(/^(hasOwnProperty_*)$/, '$1_');
}
// add new entry to symbol table
if (symbols.hasOwnProperty(word) == false) {
var last = symbol.length;
symbols[word] = last;
symbol[last] = { newCtr: 0, oldCtr: 1, toNew: null, toOld: i };
}
// or update existing entry
else {
// increment word counter for old text
var hashToArray = symbols[word];
symbol[hashToArray].oldCtr ++;
// add word number for old text
symbol[hashToArray].toOld = i;
}
}
//
// pass 3: connect unique words
//
for (var i = 0; i < symbol.length; i ++) {
// find words in the symbol table that occur only once in both versions
if ( (symbol[i].newCtr == 1) && (symbol[i].oldCtr == 1) ) {
var toNew = symbol[i].toNew;
var toOld = symbol[i].toOld;
// do not use spaces as unique markers
if (/^\s+$/.test(text.newWords[toNew]) == false) {
// connect from new to old and from old to new
text.newToOld[toNew] = toOld;
text.oldToNew[toOld] = toNew;
}
}
}
//
// pass 4: connect adjacent identical words downwards
//
for (var i = newStart; i < newEnd - 1; i ++) {
// find already connected pairs
if (text.newToOld[i] != null) {
var j = text.newToOld[i];
// check if the following words are not yet connected
if ( (text.newToOld[i + 1] == null) && (text.oldToNew[j + 1] == null) ) {
// connect if the following words are the same
if (text.newWords[i + 1] == text.oldWords[j + 1]) {
text.newToOld[i + 1] = j + 1;
text.oldToNew[j + 1] = i + 1;
}
}
}
}
//
// pass 5: connect adjacent identical words upwards
//
for (var i = newEnd - 1; i > newStart; i --) {
// find already connected pairs
if (text.newToOld[i] != null) {
var j = text.newToOld[i];
// check if the preceeding words are not yet connected
if ( (text.newToOld[i - 1] == null) && (text.oldToNew[j - 1] == null) ) {
// connect if the preceeding words are the same
if ( text.newWords[i - 1] == text.oldWords[j - 1] ) {
text.newToOld[i - 1] = j - 1;
text.oldToNew[j - 1] = i - 1;
}
}
}
}
//
// "pass" 6: recursively diff still unresolved regions downwards
//
if (wDiffRecursiveDiff == true) {
var i = newStart;
var j = oldStart;
while (i < newEnd) {
if (text.newToOld[i - 1] != null) {
j = text.newToOld[i - 1] + 1;
}
// check for the start of an unresolved sequence
if ( (text.newToOld[i] == null) && (text.oldToNew[j] == null) ) {
// determine the ends of the sequences
var iStart = i;
var iEnd = i;
while ( (text.newToOld[iEnd] == null) && (iEnd < newEnd) ) {
iEnd ++;
}
var iLength = iEnd - iStart;
var jStart = j;
var jEnd = j;
while ( (text.oldToNew[jEnd] == null) && (jEnd < oldEnd) ) {
jEnd ++;
}
var jLength = jEnd - jStart;
// recursively diff the unresolved sequence
if ( (iLength > 0) && (jLength > 0) ) {
if ( (iLength > 1) || (jLength > 1) ) {
if ( (iStart != newStart) || (iEnd != newEnd) || (jStart != oldStart) || (jEnd != oldEnd) ) {
WDiffText(text, iStart, iEnd, jStart, jEnd, recursionLevel + 1);
}
}
}
i = iEnd;
}
else {
i ++;
}
}
}
//
// "pass" 7: recursively diff still unresolved regions upwards
//
if (wDiffRecursiveDiff == true) {
var i = newEnd - 1;
var j = oldEnd - 1;
while (i >= newStart) {
if (text.newToOld[i + 1] != null) {
j = text.newToOld[i + 1] - 1;
}
// check for the start of an unresolved sequence
if ( (text.newToOld[i] == null) && (text.oldToNew[j] == null) ) {
// determine the ends of the sequences
var iStart = i;
var iEnd = i + 1;
while ( (text.newToOld[iStart - 1] == null) && (iStart >= newStart) ) {
iStart --;
}
if (iStart < 0) {
iStart = 0;
}
var iLength = iEnd - iStart;
var jStart = j;
var jEnd = j + 1;
while ( (text.oldToNew[jStart - 1] == null) && (jStart >= oldStart) ) {
jStart --;
}
if (jStart < 0) {
jStart = 0;
}
var jLength = jEnd - jStart;
// recursively diff the unresolved sequence
if ( (iLength > 0) && (jLength > 0) ) {
if ( (iLength > 1) || (jLength > 1) ) {
if ( (iStart != newStart) || (iEnd != newEnd) || (jStart != oldStart) || (jEnd != oldEnd) ) {
WDiffText(text, iStart, iEnd, jStart, jEnd, recursionLevel + 1);
}
}
}
i = iStart - 1;
}
else {
i --;
}
}
}
return;
};
// WDiffToHtml: process diff data into formatted html text
// input: text.newWords and text.oldWords, arrays containing the texts in arrays of words
// text.newToOld and text.oldToNew, arrays pointing to corresponding words
// block data structure
// returns: outText, a html string
function WDiffToHtml(text, block) {
var outText = text.message;
var blockNumber = 0;
var i = 0;
var j = 0;
var movedAsInsertion;
// cycle through the new text
do {
var movedIndex = [];
var movedBlock = [];
var movedLeft = [];
var blockText = '';
var identText = '';
var delText = '';
var insText = '';
var identStart = '';
// check if a block ends here and finish previous block
if (movedAsInsertion != null) {
if (movedAsInsertion == false) {
identStart += wDiffHtmlBlockEnd;
}
else {
identStart += wDiffHtmlInsertEnd;
}
movedAsInsertion = null;
}
// detect block boundary
if ( (text.newToOld[i] != j) || (blockNumber == 0 ) ) {
if ( ( (text.newToOld[i] != null) || (i >= text.newWords.length) ) && ( (text.oldToNew[j] != null) || (j >= text.oldWords.length) ) ) {
// block moved right
var moved = block.newRight[blockNumber];
if (moved > 0) {
var index = block.newRightIndex[blockNumber];
movedIndex.push(index);
movedBlock.push(moved);
movedLeft.push(false);
}
// block moved left
moved = block.newLeft[blockNumber];
if (moved > 0) {
var index = block.newLeftIndex[blockNumber];
movedIndex.push(index);
movedBlock.push(moved);
movedLeft.push(true);
}
// check if a block starts here
moved = block.newBlock[blockNumber];
if (moved > 0) {
// mark block as inserted text
if (block.newWords[blockNumber] < wDiffBlockMinLength) {
identStart += wDiffHtmlInsertStart;
movedAsInsertion = true;
}
// mark block by color
else {
if (moved > wDiffStyleBlock.length) {
moved = wDiffStyleBlock.length;
}
identStart += WDiffHtmlCustomize(wDiffHtmlBlockStart, moved - 1);
movedAsInsertion = false;
}
}
if (i >= text.newWords.length) {
i ++;
}
else {
j = text.newToOld[i];
blockNumber ++;
}
}
}
// get the correct order if moved to the left as well as to the right from here
if (movedIndex.length == 2) {
if (movedIndex[0] > movedIndex[1]) {
movedIndex.reverse();
movedBlock.reverse();
movedLeft.reverse();
}
}
// handle left and right block moves from this position
for (var m = 0; m < movedIndex.length; m ++) {
// insert the block as deleted text
if (block.newWords[ movedIndex[m] ] < wDiffBlockMinLength) {
var movedStart = block.newStart[ movedIndex[m] ];
var movedLength = block.newLength[ movedIndex[m] ];
var str = '';
for (var n = movedStart; n < movedStart + movedLength; n ++) {
str += text.newWords[n];
}
str = WDiffEscape(str);
str = str.replace(/\n/g, '<span class="wDiffParagraph"></span><br/>');
blockText += wDiffHtmlDeleteStart + str + wDiffHtmlDeleteEnd;
}
// add a placeholder / move direction indicator
else {
if (movedBlock[m] > wDiffStyleBlock.length) {
movedBlock[m] = wDiffStyleBlock.length;
}
if (movedLeft[m]) {
blockText += WDiffHtmlCustomize(wDiffHtmlMovedLeft, movedBlock[m] - 1);
}
else {
blockText += WDiffHtmlCustomize(wDiffHtmlMovedRight, movedBlock[m] - 1);
}
}
}
// collect consecutive identical text
while ( (i < text.newWords.length) && (j < text.oldWords.length) ) {
if ( (text.newToOld[i] == null) || (text.oldToNew[j] == null) ) {
break;
}
if (text.newToOld[i] != j) {
break;
}
identText += text.newWords[i];
i ++;
j ++;
}
// collect consecutive deletions
while ( (text.oldToNew[j] == null) && (j < text.oldWords.length) ) {
delText += text.oldWords[j];
j ++;
}
// collect consecutive inserts
while ( (text.newToOld[i] == null) && (i < text.newWords.length) ) {
insText += text.newWords[i];
i ++;
}
// remove leading and trailing similarities between delText and ins from highlighting
var preText = '';
var postText = '';
if (wDiffWordDiff) {
if ( (delText != '') && (insText != '') ) {
// remove leading similarities
while ( delText.charAt(0) == insText.charAt(0) && (delText != '') && (insText != '') ) {
preText = preText + delText.charAt(0);
delText = delText.substr(1);
insText = insText.substr(1);
}
// remove trailing similarities
while ( delText.charAt(delText.length - 1) == insText.charAt(insText.length - 1) && (delText != '') && (insText != '') ) {
postText = delText.charAt(delText.length - 1) + postText;
delText = delText.substr(0, delText.length - 1);
insText = insText.substr(0, insText.length - 1);
}
}
}
// output the identical text, deletions and inserts
// moved from here indicator
if (blockText != '') {
outText += blockText;
}
// identical text
if (identText != '') {
outText += identStart + WDiffEscape(identText);
}
outText += preText;
// deleted text
if (delText != '') {
delText = wDiffHtmlDeleteStart + WDiffEscape(delText) + wDiffHtmlDeleteEnd;
delText = delText.replace(/\n/g, '<span class="wDiffParagraph"></span><br/>');
outText += delText;
}
// inserted text
if (insText != '') {
insText = wDiffHtmlInsertStart + WDiffEscape(insText) + wDiffHtmlInsertEnd;
insText = insText.replace(/\n/g, '<span class="wDiffParagraph"></span><br/>');
outText += insText;
}
outText += postText;
} while (i <= text.newWords.length);
outText += '\n';
outText = WDiffHtmlFormat(outText);
return(outText);
};
// WDiffEscape: replaces html-sensitive characters in output text with character entities
function WDiffEscape(text) {
text = text.replace(/&/g, '&');
text = text.replace(/</g, '<');
text = text.replace(/>/g, '>');
text = text.replace(/"/g, '"');
return(text);
};
// HtmlCustomize: customize indicator html: replace {number} with the block number, {block} with the block style
function WDiffHtmlCustomize(text, block) {
text = text.replace(/\{number\}/, block);
text = text.replace(/\{block\}/, wDiffStyleBlock[block]);
return(text);
};
// HtmlFormat: replaces newlines and multiple spaces in text with html code
function WDiffHtmlFormat(text) {
text = text.replace(/ {2}/g, ' ');
text = text.replace(/\n/g, '<br/>');
return(text);
};
// WDiffDetectBlocks: detect block borders and moved blocks
// input: text object, block object
function WDiffDetectBlocks(text, block) {
block.oldStart = [];
block.oldToNew = [];
block.oldLength = [];
block.oldWords = [];
block.newStart = [];
block.newLength = [];
block.newWords = [];
block.newNumber = [];
block.newBlock = [];
block.newLeft = [];
block.newRight = [];
block.newLeftIndex = [];
block.newRightIndex = [];
var blockNumber = 0;
var wordCounter = 0;
var realWordCounter = 0;
// get old text block order
if (wDiffShowBlockMoves) {
var j = 0;
var i = 0;
do {
// detect block boundaries on old text
if ( (text.oldToNew[j] != i) || (blockNumber == 0 ) ) {
if ( ( (text.oldToNew[j] != null) || (j >= text.oldWords.length) ) && ( (text.newToOld[i] != null) || (i >= text.newWords.length) ) ) {
if (blockNumber > 0) {
block.oldLength[blockNumber - 1] = wordCounter;
block.oldWords[blockNumber - 1] = realWordCounter;
wordCounter = 0;
realWordCounter = 0;
}
if (j >= text.oldWords.length) {
j ++;
}
else {
i = text.oldToNew[j];
block.oldStart[blockNumber] = j;
block.oldToNew[blockNumber] = text.oldToNew[j];
blockNumber ++;
}
}
}
// jump over identical pairs
while ( (i < text.newWords.length) && (j < text.oldWords.length) ) {
if ( (text.newToOld[i] == null) || (text.oldToNew[j] == null) ) {
break;
}
if (text.oldToNew[j] != i) {
break;
}
i ++;
j ++;
wordCounter ++;
if ( /\w/.test( text.newWords[i] ) ) {
realWordCounter ++;
}
}
// jump over consecutive deletions
while ( (text.oldToNew[j] == null) && (j < text.oldWords.length) ) {
j ++;
}
// jump over consecutive inserts
while ( (text.newToOld[i] == null) && (i < text.newWords.length) ) {
i ++;
}
} while (j <= text.oldWords.length);
// get the block order in the new text
var lastMin;
var currMinIndex;
lastMin = null;
// sort the data by increasing start numbers into new text block info
for (var i = 0; i < blockNumber; i ++) {
currMin = null;
for (var j = 0; j < blockNumber; j ++) {
curr = block.oldToNew[j];
if ( (curr > lastMin) || (lastMin == null) ) {
if ( (curr < currMin) || (currMin == null) ) {
currMin = curr;
currMinIndex = j;
}
}
}
block.newStart[i] = block.oldToNew[currMinIndex];
block.newLength[i] = block.oldLength[currMinIndex];
block.newWords[i] = block.oldWords[currMinIndex];
block.newNumber[i] = currMinIndex;
lastMin = currMin;
}
// detect not moved blocks
for (var i = 0; i < blockNumber; i ++) {
if (block.newBlock[i] == null) {
if (block.newNumber[i] == i) {
block.newBlock[i] = 0;
}
}
}
// detect switches of neighbouring blocks
for (var i = 0; i < blockNumber - 1; i ++) {
if ( (block.newBlock[i] == null) && (block.newBlock[i + 1] == null) ) {
if (block.newNumber[i] - block.newNumber[i + 1] == 1) {
if ( (block.newNumber[i + 1] - block.newNumber[i + 2] != 1) || (i + 2 >= blockNumber) ) {
// the shorter one is declared the moved one
if (block.newLength[i] < block.newLength[i + 1]) {
block.newBlock[i] = 1;
block.newBlock[i + 1] = 0;
}
else {
block.newBlock[i] = 0;
block.newBlock[i + 1] = 1;
}
}
}
}
}
// mark all others as moved and number the moved blocks
j = 1;
for (var i = 0; i < blockNumber; i ++) {
if ( (block.newBlock[i] == null) || (block.newBlock[i] == 1) ) {
block.newBlock[i] = j++;
}
}
// check if a block has been moved from this block border
for (var i = 0; i < blockNumber; i ++) {
for (var j = 0; j < blockNumber; j ++) {
if (block.newNumber[j] == i) {
if (block.newBlock[j] > 0) {
// block moved right
if (block.newNumber[j] < j) {
block.newRight[i] = block.newBlock[j];
block.newRightIndex[i] = j;
}
// block moved left
else {
block.newLeft[i + 1] = block.newBlock[j];
block.newLeftIndex[i + 1] = j;
}
}
}
}
}
}
return;
};
// WDiffShortenOutput: remove unchanged parts from final output
// input: the output of WDiffString
// returns: the text with removed unchanged passages indicated by (...)
function WDiffShortenOutput(diffText) {
// html <br/> to newlines
diffText = diffText.replace(/<br[^>]*>/g, '\n');
// scan for diff html tags
var regExpDiff = /<\w+ class="(\w+)"[^>]*>(.|\n)*?<!--\1-->/g;
var tagStart = [];
var tagEnd = [];
var i = 0;
var found;
while ( (found = regExpDiff.exec(diffText)) != null ) {
// combine consecutive diff tags
if ( (i > 0) && (tagEnd[i - 1] == found.index) ) {
tagEnd[i - 1] = found.index + found[0].length;
}
else {
tagStart[i] = found.index;
tagEnd[i] = found.index + found[0].length;
i ++;
}
}
// no diff tags detected
if (tagStart.length == 0) {
return(wDiffNoChange);
}
// define regexps
var regExpHeading = /\n=+.+?=+ *\n|\n\{\||\n\|\}/g;
var regExpParagraph = /\n\n+/g;
var regExpLine = /\n+/g;
var regExpBlank = /(<[^>]+>)*\s+/g;
// determine fragment border positions around diff tags
var rangeStart = [];
var rangeEnd = [];
var rangeStartType = [];
var rangeEndType = [];
for (var i = 0; i < tagStart.length; i ++) {
var found;
// find last heading before diff tag
var lastPos = tagStart[i] - wDiffHeadingBefore;
if (lastPos < 0) {
lastPos = 0;
}
regExpHeading.lastIndex = lastPos;
while ( (found = regExpHeading.exec(diffText)) != null ) {
if (found.index > tagStart[i]) {
break;
}
rangeStart[i] = found.index;
rangeStartType[i] = 'heading';
}
// find last paragraph before diff tag
if (rangeStart[i] == null) {
lastPos = tagStart[i] - wDiffParagraphBefore;
if (lastPos < 0) {
lastPos = 0;
}
regExpParagraph.lastIndex = lastPos;
while ( (found = regExpParagraph.exec(diffText)) != null ) {
if (found.index > tagStart[i]) {
break;
}
rangeStart[i] = found.index;
rangeStartType[i] = 'paragraph';
}
}
// find line break before diff tag
if (rangeStart[i] == null) {
lastPos = tagStart[i] - wDiffLineBeforeMax;
if (lastPos < 0) {
lastPos = 0;
}
regExpLine.lastIndex = lastPos;
while ( (found = regExpLine.exec(diffText)) != null ) {
if (found.index > tagStart[i] - wDiffLineBeforeMin) {
break;
}
rangeStart[i] = found.index;
rangeStartType[i] = 'line';
}
}
// find blank before diff tag
if (rangeStart[i] == null) {
lastPos = tagStart[i] - wDiffBlankBeforeMax;
if (lastPos < 0) {
lastPos = 0;
}
regExpBlank.lastIndex = lastPos;
while ( (found = regExpBlank.exec(diffText)) != null ) {
if (found.index > tagStart[i] - wDiffBlankBeforeMin) {
break;
}
rangeStart[i] = found.index;
rangeStartType[i] = 'blank';
}
}
// fixed number of chars before diff tag
if (rangeStart[i] == null) {
rangeStart[i] = tagStart[i] - wDiffCharsBefore;
rangeStartType[i] = 'chars';
if (rangeStart[i] < 0) {
rangeStart[i] = 0;
}
}
// find first heading after diff tag
regExpHeading.lastIndex = tagEnd[i];
if ( (found = regExpHeading.exec(diffText)) != null ) {
if (found.index < tagEnd[i] + wDiffHeadingAfter) {
rangeEnd[i] = found.index + found[0].length;
rangeEndType[i] = 'heading';
}
}
// find first paragraph after diff tag
if (rangeEnd[i] == null) {
regExpParagraph.lastIndex = tagEnd[i];
if ( (found = regExpParagraph.exec(diffText)) != null ) {
if (found.index < tagEnd[i] + wDiffParagraphAfter) {
rangeEnd[i] = found.index;
rangeEndType[i] = 'paragraph';
}
}
}
// find first line break after diff tag
if (rangeEnd[i] == null) {
regExpLine.lastIndex = tagEnd[i] + wDiffLineAfterMin;
if ( (found = regExpLine.exec(diffText)) != null ) {
if (found.index < tagEnd[i] + wDiffLineAfterMax) {
rangeEnd[i] = found.index;
rangeEndType[i] = 'break';
}
}
}
// find blank after diff tag
if (rangeEnd[i] == null) {
regExpBlank.lastIndex = tagEnd[i] + wDiffBlankAfterMin;
if ( (found = regExpBlank.exec(diffText)) != null ) {
if (found.index < tagEnd[i] + wDiffBlankAfterMax) {
rangeEnd[i] = found.index;
rangeEndType[i] = 'blank';
}
}
}
// fixed number of chars after diff tag
if (rangeEnd[i] == null) {
rangeEnd[i] = tagEnd[i] + wDiffCharsAfter;
if (rangeEnd[i] > diffText.length) {
rangeEnd[i] = diffText.length;
rangeEndType[i] = 'chars';
}
}
}
// remove overlaps, join close fragments
var fragmentStart = [];
var fragmentEnd = [];
var fragmentStartType = [];
var fragmentEndType = [];
fragmentStart[0] = rangeStart[0];
fragmentEnd[0] = rangeEnd[0];
fragmentStartType[0] = rangeStartType[0];
fragmentEndType[0] = rangeEndType[0];
var j = 1;
for (var i = 1; i < rangeStart.length; i ++) {
if (rangeStart[i] > fragmentEnd[j - 1] + wDiffFragmentJoin) {
fragmentStart[j] = rangeStart[i];
fragmentEnd[j] = rangeEnd[i];
fragmentStartType[j] = rangeStartType[i];
fragmentEndType[j] = rangeEndType[i];
j ++;
}
else {
fragmentEnd[j - 1] = rangeEnd[i];
fragmentEndType[j - 1] = rangeEndType[i];
}
}
// assemble the fragments
var outText = '';
for (var i = 0; i < fragmentStart.length; i ++) {
// get text fragment
var fragment = diffText.substring(fragmentStart[i], fragmentEnd[i]);
var fragment = fragment.replace(/^\n+|\n+$/g, '');
// add inline marks for omitted chars and words
if (fragmentStart[i] > 0) {
if (fragmentStartType[i] == 'chars') {
fragment = wDiffOmittedChars + fragment;
}
else if (fragmentStartType[i] == 'blank') {
fragment = wDiffOmittedChars + ' ' + fragment;
}
}
if (fragmentEnd[i] < diffText.length) {
if (fragmentStartType[i] == 'chars') {
fragment = fragment + wDiffOmittedChars;
}
else if (fragmentStartType[i] == 'blank') {
fragment = fragment + ' ' + wDiffOmittedChars;
}
}
// add omitted line separator
if (fragmentStart[i] > 0) {
outText += wDiffOmittedLines;
}
// encapsulate span errors
outText += '<div>' + fragment + '</div>';
}
// add trailing omitted line separator
if (fragmentEnd[i - 1] < diffText.length) {
outText = outText + wDiffOmittedLines;
}
// remove leading and trailing empty lines
outText = outText.replace(/^(<div>)\n+|\n+(<\/div>)$/g, '$1$2');
// convert to html linebreaks
outText = outText.replace(/\n/g, '<br />');
return(outText);
};
$(document).ready(function () {
var lhs = decodeURIComponent($('#lhsdata').val());
var rhs = decodeURIComponent($('#rhsdata').val());
file1 = {};
file2 = {};
file1.name = '[buffer 1]';
file2.name = '[buffer 2]';
compare();
});
</script>
<link rel="stylesheet" type="text/css" href="reconciler.css" />
</head>
<body class="active">
<h1>ITSEE Transcription Reconciler</h1>
<div id="header_container">
<div class="column">
<h3>Create New Reconciliation</h3>
<label>Select File: </label><input type="file" id="file" name="file" onchange="javascript:uploadFile()"/>
<div id="list">
<ul>
<li><span style="background: #990033; color: white;">File 1: </span><span class="file_details" id="file1_details"></span></li>
<li><span style="background: #009933; color: white;">File 2: </span><span class="file_details" id="file2_details"></span></li>
</ul>
</div>
<span><b>Options</b></span>
<br/>
<label>Ignore layout information</label><input type="checkbox" id="layout" name="layout" checked="checked"/>
<label>Ignore all tags [ ]</label><input type="checkbox" id="tags" name="tags"/>
<br/>
<label>Treat final nu mark as nu</label><input type="checkbox" id="final_nu" name="final_nu" checked="checked"/>
<label>Ignore punctuation</label><input type="checkbox" id="punctuation" name="punctuation" checked="checked"/>
<br/>
<label>Ignore all comments</label><input type="checkbox" id="all_comments" name="all_comments"/>
<label>Ignore Status Note</label><input type="checkbox" id="status_note" name="status_note" checked="checked"/>
<br/>
<span>Ignore comments containing only:</span>
<label>OM</label><input type="checkbox" id="comment_om" name="comment_om" checked="checked"/>
<label>comm</label><input type="checkbox" id="comment_comm" name="comment_comm" checked="checked"/>
<label>lect</label><input type="checkbox" id="comment_lect" name="comment_lect" checked="checked"/>
<br/>
<br/>
<input type="button" onclick="javascript:compare()" value="Compare"/>
</div>
<div class="column">
<h3>Save as Text</h3>
<input type="button" onclick="javascript:saveText()" value="SaveText"/>
<br/>
<h3>Save Current State</h3>
<input type="button" onclick="javascript:saveReconciliation()" value="Save"/>
<br/>
<h3>Load Saved Reconciliation</h3>
<select id="savedRecons">
<option value="">Select</option>
</select>
<button onclick="loadRecon();return false;">Load</button>
</div>
<div class="column">
<h3>Export Current State</h3>
<button onclick="exportRecon();return false;">Export</button>
<h3>Import Saved Reconciliation</h3>
<label>Upload: </label><input type="file" id="reconciliation" name="reconciliation" onchange="javascript:setReconName(uploadReconciliation())"/>
</div>
</div>
<div id="results_container">
<h2>Results</h2>
<div id="results_summary"></div>
<a id="structure_results_toggle" href="javascript:toggle_results('structure_results', 'structure_results_toggle')"></a>
<a id="text_results_toggle" href="javascript:toggle_results('text_results', 'text_results_toggle')"></a>
<div id="structure_results" class="results" style="display:none;"></div>
<div id="text_results" class="results" style="display:none;"></div>
</div>
<textarea id="lhsdata" style="display:none;">
<%=HTTPUtils.canonize(lhs)%>
</textarea>
<textarea id="rhsdata" style="display:none;">
<%=HTTPUtils.canonize(rhs)%>
</textarea>
</body>
</html>