[Fedora-packaging] Re: iconcache proposal, v2

[Axel Thimm]

On Thu, Dec 14, 2006 at 06:27:07PM -0600, Rex Dieter wrote:
> Axel Thimm wrote:
> > c) why is xdg-icon-resource's exit code thrown away? We do know it
> >    will exist (explicit direct or indirect Requires:) contrary to
> >    gtk-update-icon-cache, so if it fails it probably indicates some
> >    system error that should propagate up to the user and get hos
> >    attention (e.g. write failure to the cache/fs full etc.)
> 
> failed scriptlets = aborted rpm transaction, which is bad bad bad.

Yes, I'm aware of that, aborted transaction = user is notified of
somethng broken. So the question is if xdg-icon-resource is always
present and is known (is it?) to not fail unless the file system was
remounted read-only or the binary is of the wrong elf format or other
catastrophies, then better fail early than late.

Of course the actual work not having been done by this script is
negligible, but I just don't want to silence away any failures in all
scriplets for the sake of (presumably) allowing rpm to install
further on.
-- 
Axel.Thimm at ATrpms.net
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/fedora-packaging/attachments/20061215/3371600c/attachment.sig>