User:Cacycle/diff.js: Difference between revisions

Revision as of 10:34, 21 September 2014

Revision as of 14:32, 21 September 2014

Line 3:

// ==UserScript==

// @name wDiff

// @version 1.1.12

// @date September 2122, 2014

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

// @homepage https://en.wikipedia.org/wiki/User:Cacycle/diff

Line 969:

// TextDiff.splitRefineChars(): split tokens into chars in the following unresolved regions (gaps):

// - one token became connected or separated by space, or dash, (or any stringtoken)

// - same number of tokens in gap and strong similarity of all tokens:

// - addition or deletion of flanking strings in tokens

Line 1,064:

// one word became separated by space, dash, or any string

if ( (gaps[gap].newTokens == 1) && (gaps[gap].oldTokens == 3) ) {

ifvar token = (this.newText.tokens[ gaps[gap].newFirst ].token;
var tokenFirst != this.oldText.tokens[ gaps[gap].oldFirst ].token;
var +tokenLast = this.oldText.tokens[ gaps[gap].oldLast ].token ) {;

if ( (token.indexOf(tokenFirst) !== 0) || (token.indexOf(tokenLast) != token.length - tokenLast.length) ) {

continue;

}

}

else if ( (gaps[gap].oldTokens == 1) && (gaps[gap].newTokens == 3) ) {

ifvar token = (this.oldText.tokens[ gaps[gap].oldFirst ].token;
var tokenFirst != this.newText.tokens[ gaps[gap].newFirst ].token;
var +tokenLast = this.newText.tokens[ gaps[gap].newLast ].token ) {;

if ( (token.indexOf(tokenFirst) !== 0) || (token.indexOf(tokenLast) != token.length - tokenLast.length) ) {

continue;

}

Line 1,076 ⟶ 1,082:

continue;

}

gaps[gap].charSplit = charSplittrue;

}

// cycle trough new text tokens list and set charSplit

else {
var i = gaps[gap].newFirst;

var j = gaps[gap].oldFirst;

while (i !== null) {

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

var oldToken = this.oldText.tokens[j].token;

// get shorter and longer token

var shorterToken;

var longerToken;

if (newToken.length < oldToken.length) {

shorterToken = newToken;

longerToken = oldToken;

}

else {

shorterToken = oldToken;

longerToken = newToken;

}

// not same token length

if (newToken.length != oldToken.length) {

// test for addition or deletion of internal string in tokens

// find number of identical chars from left

var left = 0;

while (left < shorterToken.length) {

if (newToken.charAt(left) != oldToken.charAt(left)) {

break;

}

left ++;

}

left ++;

}

// find number of identical chars from right

var right = 0;

while (right < shorterToken.length) {

if (newToken.charAt(newToken.length - 1 - right) != oldToken.charAt(oldToken.length - 1 - right)) {

break;

}

right ++;

}

right ++;

}

// no simple insertion or deletion of internal string

if (left + right != shorterToken.length) {

// not addition or deletion of flanking strings in tokens (smaller token not part of larger token)

if (longerToken.indexOf(shorterToken) == -1) {

// same text at start or end shorter than different text

if ( (left < shorterToken.length / 2) && (right < shorterToken.length / 2) ) {

// do not split into chars this gap

charSplit = false;

break;

}

}

}

}

}

// same token length

else if (newToken != oldToken) {

// tokens less than 50 % identical

var ident = 0;

for (var pos = 0; pos < shorterToken.length; pos ++) {

if (shorterToken.charAt(pos) == longerToken.charAt(pos)) {

ident ++;

}

break; }

if (ident/shorterToken.length < 0.49) {

// do not split into chars this gap

charSplit = false;

break;

}

}

if (ident/shorterToken.length < 0.49) {

// donext notlist split into chars this gapelements

if (i == gaps[gap].newLast) {

charSplit = false;

break;

}

i = this.newText.tokens[i].next;

j = this.oldText.tokens[j].next;

}

gaps[gap].charSplit = charSplit;

// next list elements

if (i == gaps[gap].newLast) {

break;

}

i = this.newText.tokens[i].next;

j = this.oldText.tokens[j].next;

}

gaps[gap].charSplit = charSplit;

}

Line 1,172 ⟶ 1,181:

if (gaps[gap].charSplit === true) {

// cycle trough new text tokens list, link spaces, and split into chars

var i = gaps[gap].newFirst;

var j = gaps[gap].oldFirst;

var newGapLength = i - gaps[gap].newLast;

while (i !== null) {

var newTokenoldGapLength = this.newText.tokensj - gaps[igap].tokenoldLast;

while ( (i !== null) || (j !== null) ) {

var oldToken = this.oldText.tokens[j].token;

// link identical tokens (spaces) to keep char refinement to words

if ( (newGapLength == oldGapLength) && (this.newText.tokens[i].token == this.oldText.tokens[j].token) ) {

if (newToken == oldToken) {

this.newText.tokens[i].link = j;

this.oldText.tokens[j].link = i;

}

// refine different words into chars

else {

if (i !== null) {
this.newText.split('character', i);

}

this.oldText.split('character', j);

if (j !== null) {

this.oldText.split('character', j);

}

}

// next list elements

if (i == gaps[gap].newLast) {

breaki = null;

}

if (j == gaps[gap].oldLast) {

j = null;

}

while if (i !== null) {

i = this.newText.tokens[i].next;

}

if (newTokenj !== oldTokennull) {

j = this.oldText.tokens[j].next;

}

i = this.newText.tokens[i].next;

j = this.oldText.tokens[j].next;

}

}

Line 1,209 ⟶ 1,229:

this.slideGaps = function (text, textLinked) {

// cycle through tokens list

var i = text.first;