SQL Tools

Diff two SQL queries side by side — normalize formatting first to surface only real changes.

Samples:

Original

Modified

Differences

Paste SQL into both panes to see the differences

About this tool

SQL Compare diffs two SQL queries and highlights exactly what changed — added, removed and modified lines — in a side-by-side or unified view. It's built for reviewing query edits, migrations and pull-request changes where you need to see the real difference, not noise.

Its defining feature is SQL-aware normalization: before diffing, both queries can be run through the SQL formatter, so two statements that differ only in spacing, line breaks or keyword casing are recognized as identical. Toggle normalization off to compare raw text exactly, and layer on 'ignore case' and 'ignore whitespace' to focus purely on structural changes.

Like the rest of the SQL toolkit, comparison happens entirely in your browser — nothing is uploaded. If one side isn't valid SQL for the chosen dialect, that side falls back to a raw-text comparison so you always get a useful result.

How to use

  1. Paste both queries

    Put the original query on the left and the modified one on the right (or load a sample pair).

  2. Normalize to cut the noise

    Keep 'Normalize' on so formatting-only differences disappear and only meaningful changes remain. Pick the dialect that matches your SQL.

  3. Read the diff

    Green marks additions, red marks removals; the +N / −N badges summarize the change. Switch between Side by side and Unified views.

  4. Fine-tune the comparison

    Use 'Ignore case' and 'Ignore whitespace' to further narrow the diff to structural changes.

Comparison options

OptionEffect
Normalize (format both first)Formatting-only differences are ignored — compares meaning, not layout
Ignore caseTreats differently-cased text as equal
Ignore whitespaceIgnores indentation and spacing changes
Side by side / UnifiedAligned two-column view, or a single combined column
DialectUsed when normalizing so each side formats correctly

Frequently asked questions