Bug 447533
| Summary: | Review Request: minirpc - an RPC library for stream oriented transports | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Adam Goode <adam> |
| Component: | Package Review | Assignee: | Patrice Dumas <pertusus> |
| Status: | CLOSED NEXTRELEASE | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
| Severity: | medium | Docs Contact: | |
| Priority: | medium | ||
| Version: | rawhide | CC: | fedora-package-review, notting, pertusus |
| Target Milestone: | --- | Flags: | pertusus:
fedora-review+
kevin: fedora-cvs+ |
| Target Release: | --- | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2008-05-21 21:13:10 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: | |||
|
Description
Adam Goode
2008-05-20 13:04:32 UTC
I suggest addint INSTALL='install -p' to the make install line to keep header file timestamps. Given that the date appears in doxygen generated files, you could have a look at http://fedoraproject.org/wiki/PackagingDrafts/MultilibTricks to avoid conflicts in multilib settings. Everyting else looks good. (In reply to comment #1) > I suggest addint INSTALL='install -p' to the make install line to > keep header file timestamps. > I could do this, but it doesn't seem to matter with multilib. Plus, one of the headers is a generated file anyway, so the timestamps don't mean much. Is there a reason for doing this? > Given that the date appears in doxygen generated files, you could > have a look at > http://fedoraproject.org/wiki/PackagingDrafts/MultilibTricks > to avoid conflicts in multilib settings. > Ok, I have fixed this: http://www.cs.cmu.edu/~agoode/fedora/minirpc.spec http://www.cs.cmu.edu/~agoode/fedora/minirpc-0.3-2.fc9.src.rpm (In reply to comment #2) > I could do this, but it doesn't seem to matter with multilib. Plus, one of the > headers is a generated file anyway, so the timestamps don't mean much. Is there > a reason for doing this? For the file which is not generated the timestamp is an interesting information in any case. For the generated file, it does matter if you install both arch and do a rpm -V. there is also some simple related hints on http://fedoraproject.org/wiki/PackagingDrafts/MultilibTricks But since the doxygen files are generated too, I won't make it a blocker. The timestamps could be kept with (untested): touch -c -r minirpc/minirpc.h.in minirpc/minirpc.h touch -r CHANGES doc/html/* (In reply to comment #3) > > Is there a reason for doing this? No there isn't. > For the file which is not generated the timestamp is an interesting > information in any case. The time-stamps are technically completely irrelevant in the vast majority of cases (like yours) - Some people refuse to understand this. Anyway these are not blockers, only suggestions, so * rpmlint is silent * follow guidelines * free software license included * match upstream: 892e83621ee9f840a99f2707adbd2aa3 minirpc-0.3.tar.gz * %files section right * library properly packaged I don't think it is useful to duplicate the README/CHANGES/COPYING files in the -devel subpackage since it depends on the main package and these files are already in the main package. I also won't make it a blocker. APPROVED Thanks for the review. I think your other suggestions are good, I will try to incorporate them. New Package CVS Request ======================= Package Name: minirpc Short Description: RPC library for stream transports Owners: agoode Branches: F-7 F-8 F-9 InitialCC: Cvsextras Commits: yes cvs done. Package Change Request ====================== Package Name: minirpc New Branches: EL-5 Owners: agoode cvs done. |