Text Reverser

"Reverse this text" is ambiguous without saying what unit you want reversed, and that ambiguity is exactly why this tool offers three distinct modes rather than one. Reversing a string character by character is the most literal interpretation and is genuinely useful for things like checking if a word or phrase is a palindrome, creating a simple text puzzle or novelty message, or working around software that displays text backwards. Reversing word order, by contrast, keeps every individual word spelled correctly and just flips which word comes first — useful for a specific kind of wordplay, or for sanity-checking sentence structure by reading it end to first. Reversing line order keeps every line's content untouched and only flips which line comes first, which is genuinely practical: log files are often written oldest-entry-first, and flipping them to newest-first (or vice versa) without touching the content of any individual log line is a common small task that this mode handles directly. Character reversal is implemented with `Array.from(text).reverse().join('')` rather than the more naive (and broken) approach of splitting a string with `.split('')` and reversing that. The naive approach operates on UTF-16 code units, and any character outside the Basic Multilingual Plane — most emoji, for instance — is stored internally as a surrogate pair of two code units. Reversing those two units independently corrupts the character into two meaningless replacement glyphs instead of reversing it as a whole. `Array.from` iterates a string by Unicode code point instead of code unit, so a single emoji, and the vast majority of accented Latin letters (é, ș, ö and similar precomposed characters are each a single code point) survive character reversal intact and simply move to their new position in the reversed string. That said, code-point-safe reversal is not the same as grapheme-cluster-safe reversal, and it is worth being precise about the difference. Some visual "characters" are actually built from more than one Unicode code point joined together — a combining diacritical mark applied to a base letter, an emoji with a skin-tone modifier, or a family emoji built from multiple person emoji joined with invisible zero-width-joiner characters are all, technically, several code points that render as one glyph. `Array.from`-based reversal treats each of those code points as a separate unit and reverses their order independently within the string, which can visually break a multi-code-point emoji or separate a combining accent from the letter it was attached to. Achieving fully correct grapheme-cluster reversal for every possible Unicode sequence requires the `Intl.Segmenter` API (or an equivalent grapheme-segmentation library) to first group code points into user-perceived characters before reversing those groups — this tool intentionally keeps things simple and dependency-free with code-point-level reversal, which correctly handles the overwhelming majority of real-world text (including all normal accented letters and simple, single-code-point emoji) while acknowledging this specific edge case rather than silently getting it wrong. Reverse Words and Reverse Lines do not have this problem at all, since neither one reverses the characters within a word or line — they only reorder whole units (words, separated by whitespace and rejoined with single spaces; or lines, split and rejoined on newline characters) whose internal content is left completely untouched.