Re: Interpretation of non-UTF8 strings
[email protected] (Nick Ing-Simmons)
| Newsgroups | perl.unicode |
|---|---|
| Message-ID | <[email protected]> |
Dominic Mitchell <[email protected]> writes: >Marcin 'Qrczak' Kowalczyk wrote: >> This leaves chr() ambiguous, so there should be some other function for >> making Unicode code points, as chr should probably be kept for >> compatibility to mean the default encoding. > >In the past when I've needed to guarantee Unicode code points, I've used > unpack("U",300). chr() violates the principle of least astonishment >(for me anyway) by producing single byte output for input between 0x7f < >n < 0x100. That is so legacy code that used chr() in non-ASCII locale "works" same as it always did.