[Bug 222039] Review Request: ogdi - Open Geographic Datastore Interface

bugzilla at redhat.com bugzilla at redhat.com
Sat Feb 10 12:58:58 UTC 2007


Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug report.

Summary: Review Request: ogdi - Open Geographic Datastore Interface


https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=222039





------- Additional Comments From cbalint at redhat.com  2007-02-10 07:58 EST -------
Thanks Patrice !

Well, come again with an update:
Spec: http://openrisc.rdsor.ro/ogdi.spec
SRPMS: http://openrisc.rdsor.ro/ogdi-3.1.5-5.src.rpm

>I may be wrong, but from my reading of config/common.mak and of the makefile
i rennounce. Was useless in fact.
>Also cp -p should in general be better suited than cp -af.
updated.

>make %{?_smp_mflags}
>fails for me.
noticed, disabled.

>am I right in thinking that objects in /usr/lib/ogdi/ are dlopened? In
>that case it would be better if they hadn't a soname.

well, its difficult, i have to mess up make-files, i would prefer to leave it 
like this, and convince upstream people to add -soname suport. I done the 
soname work over to not warrn rpmlint.

>/usr/lib/libecs_tcl.so.1
>is in the main package, I guess this is a mistake.

oh, yes it was fixed.

>README-BIN.TXT is certainly unusefull.
:-)) not included, removed.

>The requires for other subpackages should generally be with full version, ie:
>Requires:       %{name} = %{version}-%{release}
oh. sorry for that.


>Why isn't the sed 's/\"projects.h\"/\<projects.h\>/g' a patch
>instead? It is not a blocker, but I suggest to do patches when possible
>since sed one liners fail without notice.
convert to patch.

>I also suggest using rm instead of rm -f when possible such that it
no force flags anymore.


>In the pkgconfig file, there is certainly some .private requires or
>flags missing, for expat, zlib and proj. And the -devel subpackage
>depends certainly on zlib-devel, expat-devel and proj-devel.
updated.

>There is a scary amount of compiler warnings most of them seem
>serious. Maybe you could try to work patches for upstream?
Well, yes. I would like to clean it up closest to no warrnings,
but not right now, it will take time anyway.

>There is a Requires on pkgconfig for -devel missing.
added.

>It seems to be a matter of style, but the template defattr is
>%defattr(-,root,root,-)
hmm, ok :-)

>If there are some examples, maybe their codes could be in %doc in
>-devel?
added those olso.

can re-review ?


-- 
Configure bugmail: https://bugzilla.redhat.com/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.




More information about the package-review mailing list