Why use DiffMill?
Text comparison runs in your browser — paste API responses, config files, SQL schemas, or any two text blocks without transmitting them anywhere.
Sensitive content belongs on your device
Environment files, database configs, JWT payloads, and API responses regularly contain credentials or proprietary data. Server-side diff tools receive and process this content on infrastructure you don't own or audit. DiffMill runs in your browser — nothing you paste is transmitted or logged.
Three view modes for different tasks
Split view works well when two files are substantially different. Unified view shows a linear patch format that code reviewers and version control tools commonly use. Character-level diff surfaces the single character that changed within a long line — the kind of subtle change that's invisible in word-level comparison.
No comparison limits, no account gate
Some diff tools restrict daily comparisons or cap file size without a paid account. DiffMill has no such limits — paste two blocks of any size and compare. The only ceiling is your browser's available memory.
Clean output, no injected content
Every addition and deletion is marked, and nothing else competes for your attention. No interstitial ads between changed lines, no promoted content in the output gutter, no watermark on anything you export or screenshot.
How text diffing works
A diff algorithm finds the minimal edit script — the smallest set of insertions and deletions — that transforms one text into another. DiffMill uses the Myers algorithm, the same one Git uses, and presents results either as a split view (original left, modified right, changes aligned horizontally) or a unified view (single column with + and − markers, matching the output of git diff). Paste any two text blocks — API responses, configuration files, database schemas, Markdown documents, SQL migrations — and see immediately what was added, what was removed, and what stayed the same.
Split view vs. unified diff
The split view places original and modified versions side by side, aligning changed lines horizontally so you can read before-and-after in one glance. Most code review tools default to this layout because the eye can scan both columns in parallel. The unified view collapses both versions into a single column, marking removed lines with − and added lines with +. It is more compact — better for large files where most content is unchanged — and matches the format that Git, patch, and CI pipelines emit. DiffMill lets you switch between both views at any time without re-pasting your inputs.
Character-level diff: finding changes inside long lines
Line-level diff marks the entire line as changed even when only a single character was edited. For long lines — SQL WHERE clauses, minified JSON, URL query strings, or concatenated config values — the changed character can be buried anywhere in a wall of text. DiffMill's character-level mode highlights the exact characters that differ within each changed line, coloring individual additions and deletions so the precise change is impossible to miss even in dense content. This is the view to reach for when the line-level result is technically correct but practically unreadable.
Practical uses for text comparison
Code review without a Git host: paste two versions of a migration script or config template and share the diff. API debugging: compare a successful response against a failing one to find exactly which field changed. Documentation audit: check what actually changed between two versions of a contract or specification delivered as plain text. Environment debugging: paste log output from two servers to isolate an environment-specific difference. Schema comparison: diff two CREATE TABLE statements to see which columns, indexes, or constraints differ. DiffMill works on any plain text with no file upload, no account required, and no size limit beyond browser memory.
What a diff is and why developers can’t live without it
A diff shows you exactly what changed between two versions of something — not “the file changed” but “line 47 changed from this to that, and lines 52–54 were added.” It’s how Git shows you changes before you commit, how code review tools highlight what’s different in a pull request, and how you can compare a working API response to a broken one and instantly spot which field changed. DiffMill does this for any two blocks of text you paste in, no Git required.
Two ways to read a diff
The split view puts the old version on the left and the new version on the right, with changed lines side by side. Your eye can scan both columns simultaneously — good when changes are scattered throughout. The unified view stacks everything in one column, marking removed lines with a minus sign and added lines with a plus sign, so the overall flow stays readable — better when most content is unchanged and you want context around the edits. Switch between them based on whichever is easier for what you’re looking at.
When a whole line is marked changed but you need to find the exact character
A standard diff marks entire lines as changed or unchanged. But if a line is 200 characters long and only one character changed — a space that was added, a letter that was capitalized — the whole line is highlighted and you still have to hunt for the needle. Character-level mode highlights the exact characters that differ within a changed line. Useful for URLs (where did the query param change?), SQL (which column name is slightly different?), or any dense line where the change is subtle.
When DiffMill actually saves you hours
Reviewing a config change without a Git history. Comparing two API responses to find which field is causing a failure. Checking that two environments’ database schemas match before running a migration. Verifying that a “no changes” deploy actually has no changes. Auditing what changed between two versions of a document delivered as an email attachment. Any time you have two versions of something and need to know the difference precisely, DiffMill is faster than reading them side by side yourself.