Javascript line diff breaks beyond 65K lines

Author: tkruseCreated Dec 4, 2018Updated Feb 12, 2024

I try using The google diff-match-path library from nodejs for line diffs: https://github.com/google/diff-match-patch/wiki/Line-or-Word-Diffs. I get wrong patches when in sum the lines of both inputs goes beyond 65,536 (2^16) lines.

Is that a bug (in my code or diff-match-patch), or am I hitting a known limitation of javascript/nodejs? Anything I can do to use d-m-p with larger files?

This script reproduces the problem

var diff_match_patch = require("diff-match-patch")

// function copied from google wiki 
// https://github.com/google/diff-match-patch/wiki/Line-or-Word-Diffs
function diff_lineMode(text1, text2) {
  var dmp = new diff_match_patch();
  var a = dmp.diff_linesToChars_(text1, text2);
  var lineText1 = a.chars1;
  var lineText2 = a.chars2;
  var lineArray = a.lineArray;
  var diffs = dmp.diff_main(lineText1, lineText2, false);
  dmp.diff_charsToLines_(diffs, lineArray);
  return diffs;
}

// reproduce problem by diffing string with many lines to "abcd"
for (let size = 65534; size < 65538; size += 1) {
  let text1 = "";
  for (let i = 0; i < size; i++) {
    text1 += i + "\n";
  }

  var patches = diff_lineMode(text1, "abcb")
  console.log("######## Size: " + size + ": patches " + patches.length)
  for (let i = 0; i < patches.length; i++) {
    // patch[0] is action, patch[1] is value
    var action = patches[i][0] < 0 ? "remove" : (patches[i][0] > 0 ? "add" : "keep")
    console.log("patch" + i + ": " + action + "\n" + patches[i][1].substring(0, 10))
  }
}

Giving these outputs (using substring in code above to shorten outputs):

######## Size: 65534: patches 2
patch0: remove
0
1
2
3
4

patch1: add
abcb
######## Size: 65535: patches 2
patch0: remove
0
1
2
3
4

patch1: add

######## Size: 65536: patches 2
patch0: keep
0

patch1: remove
1
2
3
4
5

######## Size: 65537: patches 3
patch0: remove
0

patch1: keep
1

patch2: remove
2
3
4
5
6

Using

$ node --version v6.3.1
cat package.json
{
  "name": "dmp_bug",
  "version": "1.0.0",
  "description": "reproduce issue with diff match patch",
  "main": "dmpbug.js",
  "scripts": {
    "test": "echo \"Error: no test specified\" && exit 1"
  },
  "author": "",
  "license": "ISC",
  "dependencies": {
    "diff-match-patch": "^1.0.4"
  }
}