Richard Feldman (Aug 06 2024 at 17:08): Richard Feldman (Aug 06 2024 at 17:08):Should we have this?
Str.displayUtf8 : List U8 -> Str
...where if it encounters any invalid UTF-8 sequences, it replaces them with the Unicode Replacement Character?
Richard Feldman (Aug 06 2024 at 17:08):it's needed for the Path API, but not sure if it would be helpful elsewhere too :thinking:
Sam Mohr (Aug 06 2024 at 17:18):Maybe Str.sanitizeUtf8?
Sam Mohr (Aug 06 2024 at 17:22):displayUtf8 feels like it's just joining the utf-8 together
Anton (Aug 06 2024 at 17:30):Not sure about the name, but I like the function :)
Isaac Van Doren (Aug 06 2024 at 18:24):This would be great
Richard Feldman (Aug 06 2024 at 18:52):Rust sometimes calls this fromUtf8Lossy but I don't like that because it sounds like it's always going to lose something (like lossy JPEG compression) but that's not really what's happening here
Brendan Hansknecht (Aug 06 2024 at 23:34):It does lose something
Brendan Hansknecht (Aug 06 2024 at 23:34):It loses any character that isn't valid utf8
Brendan Hansknecht (Aug 06 2024 at 23:35):So if you go back to utf8 from that string, you will have lost information. And instead have the urf8 replacement character
Brendan Hansknecht (Aug 06 2024 at 23:36):I definitely think this API should be fromUtf8Something to make it discoverable.
Brendan Hansknecht (Aug 06 2024 at 23:38):You could make it fromUtf8WithReplacement if you want to be really explicit.
Richard Feldman (Aug 06 2024 at 23:49):Brendan Hansknecht said:
It does lose something
It loses any character that isn't valid utf8
yeah I just think of "lossy" meaning "you can expect it to lose something" whereas this is more "in very rare edge cases it can lose something"
Luke Boswell (Aug 06 2024 at 23:54):Str.fromBytesReplaceInvalidUtf8
Last updated: Jun 16 2026 at 16:19 UTC