Sort Lines Online - A to Z, Numeric, by Length, Reverse or Shuffle

Lines are split at every line break: Windows, Unix and old Mac endings, and the Unicode line and paragraph separators. A single break at the very end of the text ends the last line. It is not sorted as an empty line, and it is put back at the end of the result. In A to Z order empty lines come first unless you tick 'Remove empty lines'. Lines that compare as equal keep their original order, because JavaScript's sort has been required to be stable since ECMAScript 2019.

Text is ordered with Intl.Collator, the language-aware comparison built into your browser. By default it compares base letters first, then accents, then case. For English and most other languages it puts lower case before upper case when everything else is equal (a before A), as the default table of the Unicode Collation Algorithm does (UTS #10, section 6.6, gives 'a' and 'A' different tertiary weights with a first). A few languages, such as Danish, reverse that and put capitals first, and UTS #10 section 9.6 notes that this varies by language. Pick a language to change the rules: in Swedish the letters å, ä and ö come after z, while in German ä sorts with a. 'Ignore upper and lower case' makes a and A equal. If your browser has no rules for the language you pick, the tool says so and uses its default language, and if it cannot build a collator at all it falls back to code point order.

'None: Unicode code point order' compares plain Unicode values. Every capital letter comes before every lower case letter (Z before a), accented letters come after z, and with 'Ignore spaces' and 'Ignore case' off the result matches what the Unix sort command gives in the C locale. It is predictable but not what a reader expects of an alphabetical list. This mode compares real code points. JavaScript's own default sort compares UTF-16 code units, which puts characters outside the Basic Multilingual Plane, such as most emoji, before some characters that have a lower code point.

'Numbers inside text in numeric order' compares each run of digits by its value, so item2 comes before item10 (natural sort order). The two 'Numbers' orders read the number at the start of each line instead: an optional sign, thousands commas, a decimal point and an exponent, so -10 comes before -5 and 1,200 comes after 999. Lines with no number at the start go last in their original order. Numbers are read as ordinary decimals, so digits beyond about 15 lose precision; for long IDs use the numeric option on text. 'Shortest first' and 'Longest first' count Unicode characters, so an emoji is one, and lines of equal length are in A to Z order. Shuffle is a Fisher-Yates shuffle driven by the seed you can see, so a shuffle can be repeated; it uses a fast 32-bit generator, which is fine for rearranging a list and not for a draw that must be unpredictable.