Sort Lines

Paste a list, then build the clean-up you need out of steps: sort it, remove the duplicates, trim stray spaces, number it, pull one column out of a table, find and replace, join it into one line, or turn every line into a URL slug. The steps run top to bottom, each on the output of the one above, and you can reorder them with the arrow buttons. The result updates as you type, and each step shows how many lines it left, so you can see which one removed what.

A worked example

The page opens on ten messy lines: banana with leading spaces, Apple and apple, file10.txt and file2.txt, a blank line, a line of spaces, and a second banana. Four steps clean it up. Trimming leaves ten lines but removes the stray spaces. Removing empty lines takes it to eight. Removing duplicates with ignore case on takes it to six, because apple repeats Apple and the two bananas are now identical. A natural sort then gives Apple, banana, cherry, Date, file2.txt, file10.txt.

Order matters. Trim before you remove duplicates, or banana and   banana count as different lines. Number after you sort, or the numbers are shuffled along with the lines.

Which sort to pick

• A–Z is dictionary order and ignores case, so apple and Apple sit together. Lines that tie keep the order they came in. Accented letters sort beside their plain ones: éclair comes after cherry, not after Zebra.
• Natural is A–Z, except that a run of digits counts as a number. A–Z puts file10.txt before file2.txt, because the character 1 comes before 2. Natural puts file2.txt first. Use it for file names, versions and chapter numbers.
• Numeric orders by the number each line starts with. It reads a minus sign, decimals, an exponent and thousands commas, so 1,250 users is 1,250 and 1e3 is 1,000. Lines that do not start with a number, like n/a, go to the end in their original order. They stay at the end when you reverse the sort, too, so reversing only reverses the numbers.
• Case-sensitive sorts by character code, the way sort does on the command line under the C locale. Every capital letter comes before every lowercase one, so Zebra comes before apple.

Every sort can be reversed with its switch, for Z to A or largest first.

Duplicates and case

Removing duplicates keeps the first copy of each line and drops the later ones, so the list keeps its order. By default it is exact: Apple and apple are different lines. Turn on ignore case and they count as the same, and the first one wins. Spaces always count, which is why a trim step usually goes first.

Find and replace

Find and replace works on plain text unless you ask otherwise: a dot is a dot and a dollar sign is a dollar sign. Switch on regular expression for JavaScript patterns. They run line by line, so ^ and $ mean the start and end of each line, and the replacement can use $1 for a captured group. Finding (\d+)-(\d+)-(\d+) and replacing it with $3/$2/$1 turns 2024-06-01 into 01/06/2024. A pattern that does not compile shows a message and the step is skipped, so the rest of the chain still runs.

Columns, joining and slugs

Extract column splits each line on a delimiter and keeps one field, counting from 1. Text copied from a spreadsheet is tab-separated, so pick the Tab preset. Follow it with Join lines and a comma to turn a column into a comma-separated list. In the separator and delimiter boxes, \t means a tab and \n a new line. Slugify lowercases each line, drops accents and turns everything else into hyphens, so Crème Brûlée! becomes creme-brulee and Straße 5 becomes strasse-5.

Sharing a set of steps

The address bar carries your steps and their settings, so Copy link to these steps gives you a link that opens the same clean-up for someone else, or for you next week. The link carries the steps, never the text. Lists are often private, so the person who opens it pastes their own. The text is not saved either: it stays in this tab and is gone when you close it. Nothing you paste is sent anywhere.

What it does not do

Column extraction splits on the delimiter and does not read CSV quoting, so a comma inside quotes still splits the field. For real CSV, the CSV to JSON converter parses it properly. For longer chains that mix line operations with decoding, hashing or date reformatting, CyberChef chains the same kind of steps from more than 500 operations, and its sort adds IP-address and length orders. It runs in your browser too, but its page loads Google Analytics.