How to Turn café into cafe (and Which Letters Will Not Change)
The same name, José in one system and Jose in another, gets treated as two people. Stripping accents makes them match, but it also discards the original spelling, so it helps to know what changes and what does not.
When stripping accents helps
Importing into an old system often runs into field length limits, a narrow character set, or a validation rule that turns café into caf? or rejects it outright. Reducing the text to base Latin letters is usually the quickest way around that.
Name matching is the other common case: one spreadsheet says José, another says Jose, and a lookup in Excel or a database finds nothing. Search keywords, URL slugs, usernames and email prefixes raise the same question.
- Importing into systems that reject accented characters
- Matching one name across different spellings
- Generating slugs, usernames and email prefixes
- Normalising search keywords
How it works: split the letter, drop the mark
An accented letter usually has two Unicode representations: a single composed character, or a base letter followed by a combining mark. The tool normalises the text into decomposed form, then deletes the combining marks in the U+0300 to U+036F range, so é becomes e, ñ becomes n, ü becomes u, ç becomes c and Ā becomes A.
Some characters survive because they are not a base letter plus a mark: German ß, Danish ø, Polish ł, Vietnamese đ, and the ligatures æ and œ all stay as they are. Chinese characters, kana and other scripts are outside the rule and remain readable afterwards.
- é, ñ, ü, ç and Ā reduce to their base letters
- ß, ø, ł, đ, æ and œ are unchanged
- Chinese characters and kana are unaffected
- Only combining marks in U+0300–U+036F are removed
It is irreversible, and it is not transliteration
The accent information is gone once the mark is deleted: from cafe there is no way back to café, cafè or cafê. Keep the original and save the result separately rather than overwriting the source file.
It is also not a transliteration tool. German ü is often written ue in formal transcription and Danish ø as oe, but this tool produces u and leaves ø alone. Meeting a national transliteration standard needs a dedicated table or an internationalisation library.
- Irreversible — always keep the original
- No ü to ue transliteration
- Cyrillic and Greek are untouched
- Letter case is preserved
FAQ
Can the accents be restored afterwards?
Not automatically. Deleting the mark turns é, è and ê all into e, and there is no way to tell which one it was. Keep the source file, and save the stripped version under another name if it is only for search or import.
Why did ø not become o?
Because ø is not a base letter plus a combining mark in Unicode; it is a single character with nothing to decompose. The same applies to ß, ł, đ, æ and œ, which do not become ss, l, d, ae or oe unless you replace them by hand.
Do Chinese and Japanese change?
No. Only combining marks on Latin letters are removed, so Chinese characters, kana and other scripts stay readable rather than turning into blanks or question marks. Mixed text can be processed directly; just check the spelling afterwards.
Related tool：Strip Accents