How the diff checker works

Paste the earlier version of your text on the left and the newer version on the right, then read the result panel. Under the hood, the tool splits both texts into individual lines and looks for the longest sequence of lines the two versions share in the same order — a longest-common-subsequence diff, the same approach behind the diff view in Git. Lines that appear only in the original are reported as removed, lines that appear only in the changed text as added, and the shared lines as unchanged. The comparison re-runs as you type, and nothing is uploaded: the whole calculation happens in your browser tab.

How to read the result

Each row of the result is one line. Green rows marked with a + exist only in the changed text — they were added. Red rows marked with a − exist only in the original — they were removed. Plain rows are unchanged and appear in both. The two narrow columns on the left are line numbers: the first counts lines in the original, the second in the changed text, and a gap in a column shows where the other version has a line this one does not. The pills at the top total the added, removed and unchanged lines. One pattern to learn: an edited line shows up as a red row immediately followed by a green row — old wording removed, new wording added.

Line-based, not word-based — stated plainly

This checker compares whole lines. It does not pick out the single word that changed inside a line, and it does not do the character-by-character highlighting some code editors offer. For most real jobs — checking what changed between two drafts of an article, a contract, a config file or a list — the line is the natural unit, and a line-level view is faster to scan than a page peppered with tiny highlights. The trade-off is honest: change one word in a long line and you will see the whole line flagged as a removed/added pair, and it is up to your eye to spot the word. If you need to study one sentence in fine detail, counting and comparing it with the Character Counter alongside this checker works well.

When to use the ignore options

Text that has travelled — copied out of an email, a PDF or a web page — often picks up stray spaces at line ends or loses its indentation, and every such line then looks “changed”, drowning the real edits in noise. The ignore-spaces option trims the leading and trailing spaces of each line before matching, so purely cosmetic spacing differences stop flagging. The ignore-case option does the same for capitalisation drift, which is common in lists of names, product codes and headings that were retyped. Leave both options off when spacing or case genuinely matters — in most code, indentation is part of the meaning, and a password or identifier is case-sensitive. Either way, the options only change how lines are matched; the result always displays your lines exactly as pasted.

What people use it for

  • Editorial checks. Seeing exactly what an editor, client or collaborator changed in an article, essay or report draft — and what they quietly deleted.
  • Config and code files. Comparing two versions of a settings file, an .htaccess file or a script before deploying, so a one-line change does not hide among hundreds of identical lines.
  • Contracts and policies. Spotting altered clauses between drafts of an agreement, a privacy policy or terms of service, where a single swapped word can matter.
  • Lists and data. Reconciling two versions of a price list, inventory sheet or name list to find which entries were added, dropped or edited.
  • Documents that “only had small changes”. Verifying that claim in seconds instead of re-reading both versions side by side.

Limits, privacy and practical tips

Each side is capped at 5,000 lines. If either side is longer, the tool compares the first 5,000 lines of each and shows a notice in the result, so a partial comparison is never passed off as complete. Privacy is simpler: your text is never transmitted or stored, and closing the tab erases it — so download the report first if you need a record. Two tips prevent most confusion. Compare like with like: if one version was hard-wrapped (broken into short lines by an email program) and the other was not, almost every line will differ, because the lines themselves are different. And if literally everything shows as changed, check for invisible differences first — trailing spaces are the usual culprit, and the ignore-spaces option will confirm it in one click.