User:Cacycle/diff.js: Difference between revisions

Revision as of 23:32, 20 September 2014

Revision as of 10:34, 21 September 2014

Line 3:

// ==UserScript==

// @name wDiff

// @version 1.1.001

// @date September 21, 2014

// @description improved word-based diff library with block move detection

Line 190:

}

// regExps for sliding gaps: newlines and space/word breaks

if (wDiff.regExpSlideStop === undefined) { wDiff.regExpSlideStop = new RegExp('[\\n\\r' + wDiff.newLines + ']$'); }

if (wDiff.regExpSlideBorder === undefined) { wDiff.regExpSlideBorder = new RegExp('[^ \\t' + wDiff.lettersnewLinesAll + wDiff.newParagraph + '\\x0C\\x0b]$'); }

// regExps for counting words

Line 1,209:

this.slideGaps = function (text, textLinked) {

// cycle through tokens list

var i = text.first;

Line 1,222 ⟶ 1,221:

// find gap end

else if ( (gapStart !== null) && (text.tokens[i].link !== null) ) {

var backTestgapFront = nullgapStart;

backvar gapBack = text.tokens[backi].prev;

// slide down as deep as possible

var front = gapStartgapFront;

var back = itext.tokens[gapBack].next;

)if {(

var frontTest = null;

var backTest = null;

while (

(front !== null) && (back !== null) &&

(text.tokens[front].link === null) && (text.tokens[back].link !== null) &&

Line 1,236 ⟶ 1,235:

textLinked.tokens[ text.tokens[front].link ].link = front;

text.tokens[back].link = null;

frontTest = front;

backTestgapFront = backtext.tokens[gapFront].next;

text.tokens[back].linkgapBack = text.tokens[frontgapBack].linknext;

front = text.tokens[front].next;

back = text.tokens[back].next;

Line 1,243 ⟶ 1,244:

// test slide up, remember last line break or word border

var front = text.tokens[gapFront].prev;

var back = gapBack;

var frontStop = null;

while (

(frontTestfront !== null) && (backTestback !== null) &&

(text.tokens[frontTestfront].link !== null) && (text.tokens[backTestback].link === null) &&

(text.tokens[frontTestfront].token == text.tokens[backTestback].token)

) {

if (wDiff.regExpSlideStop.test(text.tokens[frontTest].token) === true) {

frontStop// =stop frontTest;at line break

if (wDiff.regExpSlideStop.test(text.tokens[frontTestfront].token) === true) {

frontTest frontStop = front;

break;

}

else if ( (frontStop === null) && (wDiff.regExpSlideBorder.test(text.tokens[frontTest].token) === true) ) {

// stop at first space/word break

frontStop = frontTest;

else if ( (frontStop === null) && (wDiff.regExpSlideBorder.test(text.tokens[frontTestfront].token) === true) ) {

frontStop = frontTestfront;

}

frontTestfront = text.tokens[frontTestfront].prev;

backTestback = text.tokens[backTestback].prev;

}

// actually slide up to line break or, if absent, word borderstop

var front = text.tokens[gapFront].prev;

if (frontStop !== null) {

var whileback (= gapBack;

while (

(front !== null) && (back !== null) && (front !== frontStop) &&

(text.tokens[front].link !== null) && (text.tokens[back].link =!== null) && (front !== frontStop) &&

(text.tokens[front].tokenlink !== null) && (text.tokens[back].tokenlink === null) &&

(text.tokens[front].token == text.tokens[back].token)

) {

}) {

text.tokens[back].link = text.tokens[front].link;

textLinked.tokens[ text.tokens[back].link = text.tokens[front].link = back;

textLinked.tokens[ text.tokens[frontback].link ].link = nullback;

front = text.tokens[front].prevlink = null;

back = text.tokens[back].prev;

front = text.tokens[front].prev;

}

back = text.tokens[back].prev;

}

gapStart = null;

Line 1,413 ⟶ 1,421:

// check if token contains unique word

if ( (recursionLevel === 0) && (level != 'character') ) {

var unique = false;

if (level == 'character') {

var token = this.newText.tokens[newToken].token;

var words = (token.match(wDiff.regExpWord) || []).concat(token.match(wDiff.regExpChunk) || []);

// unique if longer than min block length

if (words.length >= wDiff.blockMinLength) {

unique = true;

}

// unique if it contains at least one unique word

else {

for (var wordtoken = 0; word < wordsthis.lengthnewText.tokens[newToken].token; word ++) {

var words = (token.match(wDiff.regExpWord) || []).concat(token.match(wDiff.regExpChunk) || []);

if ( (this.oldText.words[ words[word] ] == 1) && (this.newText.words[ words[word] ] == 1) ) {

unique = true;

// unique if longer than min block length

break;

if (words.length >= wDiff.blockMinLength) {

unique = true;

}

// unique if it contains at least one unique word

else {

for (var word = 0; word < words.length; word ++) {

if ( (this.oldText.words[ words[word] ] == 1) && (this.newText.words[ words[word] ] == 1) ) {

unique = true;

break;

}

}

}

Line 1,732 ⟶ 1,745:

if (wDiff.debug === true) {

this.oldText.debugText('Old text');

this.newText.debugText('New text');

}