https://bugzilla.redhat.com/show_bug.cgi?id=1673956
Hirotaka Wakabayashi <hiwkby(a)yahoo.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |hiwkby(a)yahoo.com
--- Comment #1 from Hirotaka Wakabayashi <hiwkby(a)yahoo.com> ---
Hello, this is not a complete review. I will do additional review tomorrow.
Please read this for your reference.
Summary
=======
1. rpmlint results
2. Koji scratch build succeeded
3. License
Details
=======
1. rpmlint results
------------------
One error and one warning on the source rpm and 11 warnings on the binary rpms.
Here are the rpmlint results::
$ rpmlint /home/vagrant/rpmbuild/SRPMS/octave-openems-0.0.35-1.fc29.src.rpm
octave-openems.src:15: W: macro-in-comment %{version}
octave-openems.src: E: specfile-error warning: Macro expanded in comment on
line 15: %{version}.tar.xz --exclude-vcs openEMS-Project
1 packages and 0 specfiles checked; 1 errors, 1 warnings.
$ rpmlint
/home/vagrant/rpmbuild/RPMS/x86_64/octave-ctb-0.0.35-1.fc29.x86_64.rpm
octave-ctb.x86_64: W: dangerous-command-in-%preun mv
1 packages and 0 specfiles checked; 0 errors, 1 warnings.
$ rpmlint
/home/vagrant/rpmbuild/RPMS/x86_64/octave-hyp2mat-0.0.35-1.fc29.x86_64.rpm
octave-hyp2mat.x86_64: W: manual-page-warning
/usr/share/man/man1/hyp2mat.1.gz 230: warning: macro `ni' not defined
octave-hyp2mat.x86_64: W: dangerous-command-in-%preun mv
1 packages and 0 specfiles checked; 0 errors, 2 warnings.
$ rpmlint
/home/vagrant/rpmbuild/RPMS/x86_64/octave-hyp2mat-debuginfo-0.0.35-1.fc29.x86_64.rpm
1 packages and 0 specfiles checked; 0 errors, 0 warnings.
$ rpmlint
/home/vagrant/rpmbuild/RPMS/x86_64/octave-openems-0.0.35-1.fc29.x86_64.rpm
octave-openems.x86_64: W: shared-lib-calls-exit
/usr/lib64/libQCSXCAD.so.0.6.2 exit(a)GLIBC_2.2.5
octave-openems.x86_64: W: shared-lib-calls-exit /usr/lib64/libnf2ff.so.0.1.0
exit(a)GLIBC_2.2.5
octave-openems.x86_64: W: shared-lib-calls-exit
/usr/lib64/libopenEMS.so.0.0.35 exit(a)GLIBC_2.2.5
octave-openems.x86_64: W: no-manual-page-for-binary AppCSXCAD
octave-openems.x86_64: W: no-manual-page-for-binary nf2ff
octave-openems.x86_64: W: no-manual-page-for-binary openEMS
octave-openems.x86_64: W: dangerous-command-in-%preun mv
1 packages and 0 specfiles checked; 0 errors, 7 warnings.
$ rpmlint
/home/vagrant/rpmbuild/RPMS/x86_64/octave-openems-debuginfo-0.0.35-1.fc29.x86_64.rpm
1 packages and 0 specfiles checked; 0 errors, 0 warnings.
$ rpmlint
/home/vagrant/rpmbuild/RPMS/x86_64/octave-openems-debugsource-0.0.35-1.fc29.x86_64.rpm
1 packages and 0 specfiles checked; 0 errors, 0 warnings.
$ rpmlint
/home/vagrant/rpmbuild/RPMS/x86_64/octave-openems-devel-0.0.35-1.fc29.x86_64.rpm
octave-openems-devel.x86_64: W: no-documentation
1 packages and 0 specfiles checked; 0 errors, 1 warnings.
My review on the result above is as followings.
1.1. octave-openems.src:15: W: macro-in-comment %{version}
You can escape a macro in comment in the specfile by adding another leading %
to suppress this warning. Macros in comments can be a problem because they are
expanded everywhere.::
$ diff octave-openems.spec.orig octave-openems.spec
15c15
< # tar cJvf openems-%{version}.tar.xz --exclude-vcs openEMS-Project
---
# tar cJvf openems-%%{version}.tar.xz --exclude-vcs openEMS-Project
1.2. octave-openems.src: E: specfile-error warning: Macro expanded in comment
on line 15: %{version}.tar.xz --exclude-vcs openEMS-Project
I think the reason is same with the 1.1's one.
1.3. octave-ctb.x86_64: W: dangerous-command-in-%preun mv
I think this warning is probably because modifying the file system by root.
Executing "rpmspec -P octave-openems.spec" will show what it is doing.
1.4. octave-hyp2mat.x86_64: W: manual-page-warning
/usr/share/man/man1/hyp2mat.1.gz 230: warning: macro `ni' not defined
The "ni" macro is undefined.
1.5. octave-hyp2mat.x86_64: W: dangerous-command-in-%preun mv
I think this warning is probably because modifying the file system by root.
Executing "rpmspec -P octave-openems.spec" will show what it is doing.
1.6. octave-openems.x86_64: W: shared-lib-calls-exit
/usr/lib64/libQCSXCAD.so.0.6.2 exit(a)GLIBC_2.2.5
Functions in this library should return success or error so that calling
program can handle the result. The library might not return nothing and
call the "exit" function that causes normal process termination.
1.7. octave-openems.x86_64: W: shared-lib-calls-exit
/usr/lib64/libnf2ff.so.0.1.0 exit(a)GLIBC_2.2.5
The reason is same with the 1.6's one.
1.8. octave-openems.x86_64: W: shared-lib-calls-exit
/usr/lib64/libopenEMS.so.0.0.35 exit(a)GLIBC_2.2.5
The reason is same with the 1.6's one.
1.9. octave-openems.x86_64: W: no-manual-page-for-binary AppCSXCAD
The package should contain the man page for "AppCSXCAD" [1]. You might know
that
help2man [2] is a useful tool.
[1]
https://docs.fedoraproject.org/en-US/packaging-guidelines/#_manpages
[2]
https://www.gnu.org/software/help2man/
1.10. octave-openems.x86_64: W: no-manual-page-for-binary nf2ff
The reason is same with the 1.9's one.
1.11. octave-openems.x86_64: W: no-manual-page-for-binary openEMS
The reason is same with the 1.9's one.
1.12. octave-openems.x86_64: W: dangerous-command-in-%preun mv
I think this warning is probably because modifying the file system by root.
Executing "rpmspec -P octave-openems.spec" will show what it is doing.
1.13. octave-openems-devel.x86_64: W: no-documentation
The package should include documentation like README if you have.
2. Koji scratch build succeeded
--------------------------------
Here is the result of "koji build --scratch rawhide
octave-openems-0.0.35-1.fc29.src.rpm"
https://koji.fedoraproject.org/koji/taskinfo?taskID=33656976
Here is the reference to run a koji scratch build.
https://fedoraproject.org/wiki/Using_the_Koji_build_system#Scratch_Builds
3. License
-----------
The packaging guidelines say maintainers must make every possible effort to be
accurate when filling the License: field [1]. If QCSXCAD is licensed under
LGPL-3.0, License: field should contain "LGPLv3". See the Fedora Software
License List [2].
The packaging guidelines say multiple Licensing scenario is highly encouraged
to
be avoided whenever reasonably possible [3]. If multiple Licensing scenario
happens, the package must contain a comment explaining the multiple licensing
breakdown [3].
[1]
https://docs.fedoraproject.org/en-US/packaging-guidelines/LicensingGuidel...
[2]
https://fedoraproject.org/wiki/Licensing:Main?rd=Licensing#SoftwareLicenses
[3]
https://docs.fedoraproject.org/en-US/packaging-guidelines/LicensingGuidel...
Thanks in advance,
Hirotaka Wakabayashi
--
You are receiving this mail because:
You are on the CC list for the bug.
You are always notified about changes to this product and component