[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: iconv() subsystem headaches...



"Petr Vandrovec Ing. VTEI" <VANDROVE@vc.cvut.cz> writes:

> Synopsis: iconv_open() dumps core under some circumstances when feed with legal parameters

Correct.  I've applied your patch.

> Description:
>         I did not found encoding which returns string in format available
>     by using L"Text" command of gcc; which is also only format accepted
>     by wcs* functions.
>         On big-endian, you can use UCS4, but on little-endian, you are
>     out of luck. For UCS2, there is UNICODEBIG and UNICODELITTLE.
>         Why there is not UCS4LITTLE?
>         Or did I miss something important?

I know about this problem but I'm not yet sure how to handle it.  I
don't really like the approach.

In glibc 2.2 there will be all these wide char stream functions which
normally should be used.

-- 
---------------.      drepper at gnu.org  ,-.   1325 Chesapeake Terrace
Ulrich Drepper  \    ,-------------------'   \  Sunnyvale, CA 94089 USA
Cygnus Solutions `--' drepper at cygnus.com   `------------------------


Reply to: