Note: This is a public test instance of Red Hat Bugzilla. The data contained within is a snapshot of the live data so any changes you make will not be reflected in the production Bugzilla. Email is disabled so feel free to test any aspect of the site that you want. File any problems you find or give feedback at bugzilla.redhat.com.
Bug 192052 (bitgtkmm)
Summary: | Review Request: bitgtkmm (Gtkmm widgets for the bit library) | ||
---|---|---|---|
Product: | [Fedora] Fedora | Reporter: | Rick L Vinyard Jr <rvinyard> |
Component: | Package Review | Assignee: | Paul F. Johnson <paul> |
Status: | CLOSED NEXTRELEASE | QA Contact: | Fedora Package Reviews List <fedora-package-review> |
Severity: | medium | Docs Contact: | |
Priority: | medium | ||
Version: | rawhide | ||
Target Milestone: | --- | ||
Target Release: | --- | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Fixed In Version: | Doc Type: | Bug Fix | |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2006-08-02 03:51:08 UTC | Type: | --- |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: | |||
Bug Depends On: | 183953 | ||
Bug Blocks: | 163779 |
Description
Rick L Vinyard Jr
2006-05-17 04:49:29 UTC
Spec URL: http://miskatonic.cs.nmsu.edu/pub/bitgtkmm.spec SRPM URL: http://miskatonic.cs.nmsu.edu/pub/fedora/5/srpms/bitgtkmm-0.2.1-2.src.rpm Changes: - Added AUTHORS and COPYING to bitgkmm main package - Changed prdownloads.sf.net to download.sf.net Spec URL: http://miskatonic.cs.nmsu.edu/pub/bitgtkmm.spec SRPM URL: http://miskatonic.cs.nmsu.edu/pub/fedora/5/srpms/bitgtkmm-0.2.2-1.src.rpm Changes: - New upstream release - Removed *.md5, *.map and *.dot in upstream package - Changed mv of docs to cp %build %configure --enable-static=no %{__make} %{?_smp_mflags} %install %{__rm} -rf %{buildroot} make DESTDIR=%{buildroot} install You need to be consistant with using macros. Either constantly use %{__make} or just plain make I'll look some more later Ahh, good catch. %{__make} is probably better. I'll change it in papyrus too. Builds fine in mock. However. Blocker : For the devel, you need to add bitgtkmm to the Requires list Blocker : run rpm -qa --requires on the installed bitgtkmm. Quite a lot of the output are for standard libs, but you cannot rely on people having gtkmm installed (or a few of the others) Fix these and we should be good to go (more or less!) > Blocker : For the devel, you need to add bitgtkmm to the Requires list It's there. It's in the standard devel line, right above the specific requires that I add for bitgtkmm: Requires: %{name} = %{version}-%{release} Puts a specific dependency on the main package and the specific version and release that the headers belong to. How about if I change it to: Requires: bitgtkmm = %{version}-%{release} > Blocker : run rpm -qa --requires on the installed bitgtkmm. Quite a lot of the > output are for standard libs, but you cannot rely on people having gtkmm > installed (or a few of the others) Those are all added by rpmbuild. 8-->
It's there. It's in the standard devel line, right above the specific requires
that I add for bitgtkmm:
Requires: %{name} = %{version}-%{release}
<--8
D'oh!
8-->
> Blocker : run rpm -qa --requires on the installed bitgtkmm. Quite a lot of the
> output are for standard libs, but you cannot rely on people having gtkmm
> installed (or a few of the others)
Those are all added by rpmbuild.
<--8
No. These are packages required to run. Say I didn't have atkmm on my machine
(just as an example). Without the R atkmm being explictly states in the spec
file, the package would know no better until it gets to a point that atkmm is
needed and then it falls over dead.
I had a similar problem with Anjuta-1.2.4a a while back in that gtkmm was not in
as an explicit R. However, when you came to try and create a gtkmm package, the
software complained like crazy. Add the R line for gtkmm and everything was
happy again.
> No. These are packages required to run. Say I didn't have atkmm on my machine > (just as an example). Without the R atkmm being explictly states in the spec > file, the package would know no better until it gets to a point that atkmm is > needed and then it falls over dead. Check out: http://fedoraproject.org/wiki/Packaging/Guidelines#Requires rpmbuild adds the requires, such as (continuing with the atkmm example): libatkmm-1.6.so.1()(64bit) If you try and install bitgtkmm, rpm will complain that libatkmm-1.6.so.1 is required. If you're using yum, it will look up libatkmm-1.6.so.1 as a library in atkmm and add atkmm to the dependency installs. The situation you ran into with Anjuta was a little different. The reason why you had to explicitly add it was that the Anjuta runtime didn't have a dependency on the gtkmm runtime libraries, so rpmbuild didn't add it to the rpm requires. Yep. You're right. Okay - let's get on with this! Good ---- Builds fine in mock (x86) Spec file checks out fine - no dupes, no problems with permissions, documentation included in subpackages, consistent use of macros, no BR problems Package installs fine, rpmlint shows nothing of importance on the packages (installed or as rpms) Niggles ------- Unhappy with the *.so things in the %files and %files devel, but this is down to me and nothing that can be blocked. I can't see any other problems with this. APPROVED The -devel package includes a .pc file, which is useless without pkgconfig, ergo the -devel package should Require: pkgconfig See also: http://fedoraproject.org/wiki/Packaging/IRCLog20060706 Thanks - rpm -qa --provides though didn't show this. I'll remember this for future spec files. > Unhappy with the *.so things in the %files and %files devel > The -devel package includes a .pc file, which is useless without pkgconfig, ergo > the -devel package should Require: pkgconfig Fixed in the first build. |