News for hackers. Stuff that matters.

A comment on Unicode Normalization Forms: When ö ≠ ö. Parent comment

Re:Unicode Normalization Forms: When ö ≠ ö by vbezhenar on Saturday January 01, 2022 @12:00AM (#29755670)
UTF-16 is a simple encoding. It should take a few dozens of LoC to convert to UTF-8. At least if you don’t need extreme performance with AVX, etc.
[ Parent | on HN ]
Re:Unicode Normalization Forms: When ö ≠ ö by bjoli on Saturday January 01, 2022 @03:13AM (#29756541)
My experience is exactly like what you say. Until it isn't. UTF16 seems like such a neat idea until it meets reality.

Mostly because of weird interactions of different libraries, languages and operating systems.

[ Parent | on HN ]
Re:Unicode Normalization Forms: When ö ≠ ö by saati on Saturday January 01, 2022 @02:35AM (#29756374)
> UTF-16 is a simple encoding

This isn't true, for starters it's two encodings UTF-16LE and UTF-16BE.

[ Parent | on HN ]