PDFCrack - empty debuginfo file
awilliam at redhat.com
Fri Oct 21 19:14:54 UTC 2011
On Fri, 2011-10-21 at 22:06 +0300, Ville Skyttä wrote:
> On 10/21/2011 07:51 PM, Adam Williamson wrote:
> > If it's being stripped there's probably a good *reason* for it to be
> > stripped
> I'd say its much more likely that they're just trying to save some space
> on end user systems where stuff is built from source by users and the
> users are too lazy to strip things him/herself if they happen to want
> that. In my experience, upstreams are receptive to patches that make
> the stripping behavior optional, for example "STRIP = strip" and
> invoking "$(STRIP)" in Makefile which can then be overridden for example
> with "make STRIP=/bin/true ...".
It's being stripped in the spec, not the upstream source.
> BTW rpmbuild will strip things anyway unless a bug or explicit measures
> in specfiles prevent it from doing that, see also
Yes, but it does so in a proper way that doesn't break the debuginfo
packages. So if someone's explicitly stripping the binary in the spec
they're either misunderstanding something and it should be dropped, or
there's some specific reason why that needs to happen in the spec.
Fedora QA Community Monkey
IRC: adamw | Twitter: AdamW_Fedora | identi.ca: adamwfedora
More information about the devel