Re: Question about “Uppercase” in DerivedCoreProperties.txt

[Philippe Verdy]

Successors to convert strings instead of just isolated "characters" (sorry,
they are NOT what we need to handle "texts", they are not even equivalent
to Unicode characters, they are just code units, most often 8-bit with
"char" or 16-bit only with "wchar_t" !) already exist in all C libraries
(including glibc), under different names unfortunately (this is the main
cause why there are complex header files trying to find the appropriate
name, and providing a "default" basic implementation that just scans
individual characters to filter them with tolower and toupper: this is a
bad practice,

Good libraries should all contain a safe implementation of case conversion
of strings, and softwares should use them (and not reinvent this old bad
trick, just because this works with basic English).


2014-11-10 13:41 GMT+01:00 Steffen Nurpmeso <sdaoden at yandex.com>:

> Philippe Verdy <verdy_p at wanadoo.fr> wrote:
>  |glibc is not more borken and any other C library implementing toupper and
>  |tolower from the legacy "ctype" standard library. These are old APIs that
>  |are just widely used and still have valid contexts were they are simple
> and
>  |safe to use. But they are not meant to convert text.
>
> Hah!  Legacy is good..  I'd wish a usable successor were already
> standardized by ISO C.
>
> --steffen
> _______________________________________________
> Unicode mailing list
> Unicode at unicode.org
> http://unicode.org/mailman/listinfo/unicode
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://unicode.org/pipermail/unicode/attachments/20141110/87559428/attachment.html>