Text Compare & Diff Checker

A diff is not simply a line-by-line comparison — that would fall apart the moment a single line is inserted, marking everything after it as changed. What you actually want is the longest common subsequence: the largest set of lines appearing in both versions in the same order. Everything in that set is unchanged; everything else was added or removed. This is the algorithm behind `git diff` and every serious comparison tool, and it is what runs here.

The result is that inserting one line at the top of a thousand-line file produces exactly one change, not a thousand. That property is what makes a diff useful for review.

The two views serve different purposes. Side by side keeps the original and the changed version in their own columns with line numbers, which is easier when you are comparing whole documents or want the surrounding context. Unified interleaves them with `+` and `−` markers, which is the format you already know from pull requests and patch files, and it is more compact for scattered small edits.

The two ignore options exist because not every difference is meaningful. Ignoring whitespace collapses runs of spaces and trims line ends, which strips out the noise from a reformat or a change of editor settings so you can see the real edits underneath. Ignoring case is useful for comparing lists of names, identifiers or config keys where capitalisation has drifted. Both options affect only the comparison — the text shown is always exactly what you pasted.

The comparison is capped at 3,000 lines per side. The algorithm builds a table proportional to the two lengths multiplied together, so beyond that point a browser tab would become unresponsive; the tool tells you rather than freezing.

Everything runs locally in JavaScript, which makes it appropriate for comparing contract drafts, unreleased copy, or source code you cannot paste into a hosted service.