Hi Kai-Uwe,
Neat trick. I am including it in my sources, thanks!
Regards
Marti.
On 15/09/2007 20:54:36, Kai-Uwe Behrmann ([EMAIL PROTECTED]) wrote:
> Marti,
>
> in cmsio1.c:1507 lcms bails out if wcstombs returns with -1, as is the
> case for the mentioned profile.
>
> I tested on x86 linux with libc-2.5. In return of ReadEmbeddedTextTag the
>
> strlen is zero. Probably the glibc implementation is weak, and lcms needs
>
> a work around to wcstombs.
>
> something like (?):
>
> 1504 i = wcstombs(Name, wchar, size_max );
> 1505 if (i == ((size_t) -1)) {
> 1506 // second try at least for latin
> 1507 int i;
> 1508 char *tmp = (char*)wchar;
> 1509 for(i = 0; i < Len/2; ++i)
> 1510 Name[i] = tmp[i*2];
> 1511 //Name[0] = 0; // Error
> 1512 }
>
> A check to ignore multibyte characters would be needed in addition:
>
> /*
> * Index into the table below with the first byte of a UTF-8 sequence to
> * get the number of trailing bytes that are supposed to follow it.
> * Note that *legal* UTF-8 values c
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Lcms-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/lcms-user
