Text Diff Checker

Comparing two versions of a document or a piece of code by eye is slow and error-prone past a few lines — the eye is bad at spotting a single changed word buried in a paragraph, or a one-line change in a hundred-line file. A proper diff algorithm solves this precisely: it computes the smallest possible set of additions and removals that transforms one text into the other, so what's highlighted is genuinely what changed, not an artifact of a naive line-by-line comparison.

This tool implements the longest common subsequence (LCS) algorithm directly — the same foundational approach behind GNU diff and, in spirit, Git's diff engine (which uses the faster but equivalent-output Myers algorithm). LCS finds the longest sequence of lines or words that both texts share, in order, and everything not part of that shared sequence is marked as added or removed. This guarantees a minimal, genuinely accurate diff rather than a rough approximation.

Line view is built for comparing documents, config files, or code, where the meaningful unit of change is usually a whole line. Word view instead diffs at the word level within a block of prose, which is far more useful for catching a small wording change in a paragraph than a line-level diff would be, since it highlights only the specific words that changed rather than flagging the entire surrounding line.

Ignore-case and ignore-whitespace options matter for a common false-positive: a line that's functionally identical but differs only in capitalization or trailing spaces. Toggling these treats those differences as non-changes, so the diff reflects meaningful content changes only. The result can be copied as a standard unified diff — the same +/- format used by Git and most version control tools — ready to paste into a commit message, code review comment, or bug report.