What is a unified diff?

A unified diff is the standard patch format used by git and the patch tool. It shows changed regions as hunks with an @@ -l,s +l,s @@ header, prefixing removed lines with minus, added lines with plus, and unchanged context lines with a space.

How are the changes computed?

The tool runs a longest-common-subsequence diff over the lines of both inputs to find the minimal set of insertions and deletions, then groups them into hunks and pads each hunk with your chosen number of context lines.

What do the context lines do?

Context lines are the unchanged lines printed around each change so the patch can be located reliably even if surrounding code shifts. Three lines is the git default; increase it for fragile files, decrease it for a terser patch.

Can I apply the output with git?

Yes. The output uses git-compatible --- a/file and +++ b/file headers and standard @@ hunk markers, so git apply patch-file or patch -p1 < patch-file will apply it, assuming the original matches.

No. Both inputs are diffed entirely in your browser with JavaScript. Nothing is sent to a server, so you can safely diff private or proprietary text.

What is the Unified Diff to Patch File Generator?

Free unified-diff generator. Paste an original and a modified text and get a standard unified-diff .patch file with hunk headers and configurable context lines, ready for git apply or patch. Computed in your browser; nothing is uploaded. It runs free in your browser on Gera Tools, with nothing uploaded.

Unified Diff to Patch File Generator

Name: Unified Diff to Patch File Generator
Creator: Gera Tools
License: https://creativecommons.org/licenses/by/4.0/

Get one useful tool a week

Like this tool? Enter your email and we'll send you one genuinely useful Gera tool a week — plus a link to come back to this one. No spam, one-click unsubscribe any time.

A unified diff is the patch format git and the patch tool understand: it records exactly which lines were removed and added, wrapped in hunks with location headers and a few lines of surrounding context. This generator diffs two pasted texts and emits a clean .patch file you can share or apply offline.

When to use a patch file

A patch file is useful any time you need to share a precise text change that can be applied exactly — a code fix that needs to be applied to many files, a configuration change to distribute to a team, a contribution to a project you cannot push to directly, or a change you want to store alongside a file rather than in its git history. The .patch format is also the standard for email-based patch workflows in open-source projects, where contributors send patches to a maintainer who applies them with git am or git apply.

How it works

Both inputs are split into lines.
A longest-common-subsequence (LCS) diff finds the minimal edit script — the smallest set of line deletions (-) and insertions (+) that turns the original into the modified text; matching lines become unchanged context ( ).
Consecutive changes are grouped into hunks, each padded with up to N lines of context (default 3, matching git’s default) and separated into distinct hunks when the gap between change regions exceeds 2×N lines.
Each hunk gets an @@ -start,len +start,len @@ header locating it in both the old and new file, and the whole patch opens with --- a/<name> and +++ b/<name> headers.

Reading a hunk header

@@ -12,7 +12,7 @@

This means: starting at line 12 in the original file, 7 lines are shown (6 context + 1 removed); in the new file, also starting at line 12, 7 lines are shown (6 context + 1 added). When a change inserts or deletes lines, the two numbers in the header differ.

Worked example

An original file contains .btn { color: red; }. Changing red to blue with 1 context line:

@@ -1,3 +1,3 @@
 .btn {
-  color: red;
+  color: blue;
 }

The space-prefixed line is unchanged context. The - line was removed, the + line added.

Applying the patch

Save the output as fix.patch, then:

git apply fix.patch          # inside a git repo
patch -p1 < fix.patch        # with the patch utility (GNU patch)

The -p1 strips the leading a/ from file paths, which is what git-format patches use. Line numbers in the hunk header are 1-based, and lengths count context plus changed lines — both exactly matching what git diff emits. Trailing-newline differences are handled so the patch applies cleanly. Everything is computed locally; your text never leaves the browser.