Remove numbers

When you need the digits gone

The most frequent case is a list copied out of a PDF or a web page that drags its numbering along: «1. Apples», «2. Pears», «3. Kiwis». If that list is going into a document that numbers items by itself, the old digits become a duplicate. Same story with transcripts carrying a timecode in front of every line, with catalogues where each entry starts with a product code, and with texts being prepared for language analysis, where numbers are just noise.

What exactly gets removed

Every number disappears whole: the digits and whatever sits inside the number, that is the thousands and decimal separators and the sign in front. «1,234.56» goes entirely, instead of leaving a comma and a full stop lying around, and so do «-5» and «1'000». The numbering at the start of a line goes with its mark, so «1. Apples», «2) Pears» and «00:01 Hello» become «Apples», «Pears» and «Hello», and the keycap 1️⃣ goes with the digit it carries. Letters, accents, punctuation and the other emojis stay where they are, and the lines do not move.

The thing to keep in mind is that numbers disappear wherever they are, even when attached to a word: «H2O» becomes «HO», «VAT at 22%» becomes «VAT at %» and «Windows 11» becomes «Windows». A hyphen attached to a word stays instead, because there it is not a minus sign: «COVID-19» becomes «COVID-». If the text contains chemical formulas, acronyms, product models or percentages, check them before copying the result.

«Remove extra spaces» is ticked by default as well, because a number taken out of the middle of a sentence would leave two spaces in a row: in «I have 3 apples» they would sit between «have» and «apples». If you need the spaces as they were, for instance for the indentation of a list, untick that box.

Which numbers, and which not

The digits of every script count as numbers, not only 0 to 9: Arabic and Devanagari digits, superscripts such as ², fractions such as ½, circled numbers such as ① and Roman numerals written with their own characters, such as Ⅻ.

Numbers written as words («twenty-two») and Roman numerals written with ordinary letters («XIV») stay instead: removing them would mean guessing which words are numbers, and «ten» inside «often» is not one.

Several clean-ups at once, in a fixed order

Below the box you will find all ten clean-ups of this family, and you can tick as many as you like: each page starts with its own already ticked. The program is the same on all ten, so the same text with the same boxes gives the same result on whichever page you paste it.

The order does not depend on which boxes you tick, and it is the order you see them in: first the characters (emojis, accents, numbers, punctuation), then the spaces inside each line (tabs, indentation, extra spaces), then whole lines (blank and duplicate) and last of all joining everything into one line. So a line that is left empty once the emojis are gone disappears with the blank lines, and two lines that become identical once the spaces are gone count as duplicates.

The sister pages, each with its own box already ticked: Remove extra spaces, Remove indentation, Remove blank lines, Remove duplicate lines, Remove emojis, Remove accents, Remove punctuation, Text on one line and Tabs to spaces.

Nearby tools

If instead of losing the numbers you need to keep only those, the opposite direction is Extract numbers from text; for phone numbers alone there is Extract phone numbers. To drop punctuation marks there is Remove punctuation, and to take out anything at all, deciding for yourself what, Find and replace.