Shotnest

Shotnest · Library

Find a full-width Latin code

The query abc12 finds Code ABC12 because Shotnest shifts full-width Latin letters and digits (U+FF01 to U+FF5E) back to ordinary width before comparing.

How full-width characters are mapped

East Asian fonts and some apps display Latin letters and digits in full-width forms such as ABC12. They look like ordinary characters but have different code points, from U+FF01 to U+FF5E. Shotnest's normalization subtracts a fixed offset from every character in that range, which turns A into A, then lowercases the result. The ideographic space U+3000 is replaced with an ordinary space at the same step.

Because the mapping runs on both the record and the query, it does not matter which width you type. A query typed in ordinary letters finds the full-width code, and in the app a query typed with a Japanese keyboard in full-width mode finds the same screenshot.

Full-width punctuation sits in the same block, so the full-width hyphen, colon and brackets are converted as well. A code shown as AB-12 is searched as ab-12.

Try the original example

Supplied records
Code ABC12 — Code XYZ99
Query
abc12
Matching record numbers
1

Find a code you saw in an Asian-language screen

  1. Type the code with an ordinary keyboard; capital letters are not needed.
  2. Check the returned record: here only 1, because Code XYZ99 shares no part of abc12.
  3. Copy the code from the original text or image when the exact width matters, for example in a form that rejects full-width digits.
  4. If no result appears, search the digits alone.
ABC12 after normalization
abc12
Query ABC12 in the Shotnest app
record 1

Look-alikes that are not mapped

Width is folded, typography is not checked

A match means the characters agree after a fixed width shift. It does not confirm which form the original document uses, so keep the screenshot when you need to reproduce the code exactly.

Tools