On 02/17/2016 03:09 PM, Dominik 'Rathann' Mierzejewski wrote:
On Wednesday, 17 February 2016 at 15:00, Richard Shaw wrote:
On Wed, Feb 17, 2016 at 7:54 AM, Dominik 'Rathann' Mierzejewski <
dominik@xxxxxxxxxxxxxx> wrote:
Hello,
I'm orphaning the dx and dx-samples in rawhide.
It is FTBFS in rawhide due to stricter checks in gcc-6, the upstream
is dead (or actually it was never there, ever since IBM released the
code), and the code itself is ancient (1999). I don't use it and I don't
know anyone who does, so it's not really worth my effort to fix it
to build with gcc-6. I've been keeping the code building (and running,
mostly) for almost 10 years now and I think it's time for someone
else to take over, if they're interested.
If no one is using it then it's not worth the effort but in the packages
I've had that have the narrowing conversion issues they have been
relatively easy to fix.
Well, in this case, an array of char is initialised with values > 128
and the passed to XCreatePixmapFromBitmapData, which requires a char *
parameter, so it's not trivial (at least to me).
AFAIS, this is Xt/Xm code. Xm/Xt is C, with all types and parameters
being C. Also, Xm/Xt's APIs predate the invention of "const" and had
never actually cared about it.
I.e. this is a situation, where spraying C-type-casts into
Xm/Xt-API-calls is the solution - seriously!
I haven't looked into this particular package yet, but in the situation
you line out, something along the lines of the snippet below, is the
solution:
unsigned char xbm[] = { 0xff, .... }
XCreatePixmapFromBitmapData( ...., (char*) xbm,...)
[Similar considerations apply to tcl/tk and other "old" C/C++-SDKs]
Ralf
--
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxxx
http://lists.fedoraproject.org/admin/lists/devel@xxxxxxxxxxxxxxxxxxxxxxx
