On Thu, Nov 26, 2009 at 8:41 AM, Jeff Johnson <n3npq@mac.com> wrote:

>
> On Nov 26, 2009, at 9:41 AM, Matthew Dawkins wrote:
>
>
>> But look at what the original bug report says. Shouldn't the missing
>> sources also show up in the errors if that is the correct logic?
>>
>
> I did (and have) looked.
>
> The bug report states either ... or ... as a preference and the
> context pretty clearly says the expectation is
>        No error message is desired.
>
> Meanwhile that's @rpm.org's issue, not mine.
>
> You have stated a behavior, and added a URL and have yet
> to state what your expectations is.
>
> Which goes around and around and around. The --specfile
> (and --specsrpm) options are perhaps the highest maintenance
> implementations in rpmbuild, wired in perl bindings and URPMI, and mostly
> _NOT_ what rpmbuild or --query was intended for.
>
> 73 de Jeff
> ______________________________________________________________________
> RPM Package Manager                                    http://rpm5.org
> Developer Communication List                        rpm-devel@rpm5.org
>


Ohh sorry, I thought that was clear. I would prefer to not have it show the
errors. The query is handy to list what packages are to be built from a
specfile, but I not sure that was it's original intention.

If it is, then the other strange behavior would be that it lists the rpm for
Name: even if the orignal Name may not have a %files list to be built.