Shotnest

Shotnest · Library

Find an accented word from an unaccented query

A query for cafe finds Café because Shotnest strips Latin and Vietnamese accent marks before matching, while the record keeps its original spelling.

How accents are folded before matching

Both the search tool on this page and the Shotnest app run the same normalization before comparing text. Letters are lowercased, then a fixed table replaces accented Latin letters with plain ones: é becomes e, ç becomes c, ş becomes s, ø becomes o. A second table does the same for Vietnamese vowels, so Đà Nẵng is compared as da nang. A few letters expand instead of shrinking: ß and ẞ become ss, æ becomes ae and the ligature fi becomes fi.

Only the comparison copy is folded. The record you typed, and in the app the text read from the screenshot, keeps its accents, so the result still shows Café. Folding happens on both sides, which is why the upper-case query CAFÉ returns the same record as cafe.

Try the original example

Supplied records
Café opening notes — Library reference
Query
cafe
Matching record numbers
1

Search a name you only half remember

  1. Type the word the way it comes to mind, with or without accents.
  2. Check the returned record number against the list you entered.
  3. Open the original text and copy the spelling from there, not from your query.
  4. If nothing matches, try one distinctive part of the word instead of the whole phrase.
Query creme brulee in “Crème brûlée recipe”
match
Straße after normalization
strasse
Đà Nẵng after normalization
da nang

Where accent folding stops

Keep the original spelling

A folded match tells you the letters line up once accents are removed. It does not tell you which spelling is correct for a person or a place, so take names from the returned original. In the app, the same rule lets a screenshot of a French menu turn up for a query typed on an English keyboard.

Tools