PR#20: Added rolling suffix to release - rpms/java-openjdk

rpms / java-openjdk

#20 Added rolling suffix to release

Merged 4 years ago by jvanek. Opened 4 years ago by jvanek.

Unknown source master into master

Added rolling suffix to release

Jiri Vanek • 4 years ago

fdb1a56

java-openjdk.spec

file modified

+7 -1

		`@@ -857,7 +857,10 @@`

		`Name: java-%{origin}`
		`Version: %{newjavaver}.%{buildver}`
		`- Release: 7%{?dist}`
		+ # This package needs `.rolling` as part of Release so as to not conflict on install with
		`+ # java-X-openjdk. I.e. when latest rolling release is also an LTS release packaged as`
		`+ # java-X-openjdk. See: https://bugzilla.redhat.com/show_bug.cgi?id=1647298`
		`+ Release: 8.rolling%{?dist}`
		`# java-1.5.0-ibm from jpackage.org set Epoch to 1 for unknown reasons`
		`# and this change was brought into RHEL-4. java-1.5.0-ibm packages`
		`# also included the epoch in their virtual provides. This created a`
		`@@ -1794,6 +1797,9 @@`


		`%changelog`
		`+ * Fri Nov 30 2018 Jiri Vanek <jvanek@redhat.com> - 1:11.0.1.13-8`
		`+ - added rolling suffix to release (before dist) to prevent conflict with java-11-openjdk which now have same major version`
		`+`
		`* Mon Nov 12 2018 Jiri Vanek <jvanek@redhat.com> - 1:11.0.1.13-6`
		`- fixed tck failures of arraycopy and process exec with shenandoah on`
		`- added patch585 rh1648995-shenandoah_array_copy_broken_by_not_always_copy_forward_for_disjoint_arrays.patch`

no initial comment

jvanek commented 4 years ago

yup. fixed the colission issue

jvanek commented 4 years ago

https://bugzilla.redhat.com/show_bug.cgi?id=1647298

jerboaa commented 4 years ago

Thanks. I'll have a look at this tomorrow.

jerboaa commented on line 6 of java-openjdk.spec 4 years ago

Please consider rephrasing this comment to something like this:

# This package needs `.rolling` as part of Release so as to not conflict on install with
# java-X-openjdk. I.e. when latest rolling release is also an LTS release packaged as
# java-X-openjdk. See: https://bugzilla.redhat.com/show_bug.cgi?id=1647298

jerboaa commented 4 years ago

I've tested this and it works fine. Thanks! Consider an updated comment before you push, though.

Ship it!

jvanek commented 4 years ago

Few more comments please :)
Is releaase really bets place?
isnt vrersion better? - imho not
Isnt uniquesuffix better place? It can be...

Still I vote for this part of release. WDYT?

jerboaa commented 4 years ago

Few more comments please :)
Is releaase really bets place?
isnt vrersion better? - imho not

I don't think there is a big difference between version and release. The packaged OpenJDK version is the same when latest JDK == LTS. Putting it in version isn't great for that reason.

Isnt uniquesuffix better place? It can be...

uniquesuffix won't be sufficient. For example jre-lnk would conflict if only uniquesuffix was changed. It would end up with something like
https://src.fedoraproject.org/rpms/java-openjdk/pull-request/18

Still I vote for this part of release. WDYT?

It seems an acceptable solution. One concern for me are provides:

# rpm -qa --qf '%{NAME}\n' | grep openjdk | sort
java-11-openjdk
java-11-openjdk-devel
java-11-openjdk-headless
java-openjdk
java-openjdk-devel
java-openjdk-headless
# rpm -q --whatprovides java-11-openjdk-headless
java-11-openjdk-headless-11.0.1.13-7.fc30.x86_64
java-openjdk-headless-11.0.1.13-7.rolling.fc30.x86_64

If we go with this, and somebody requires java-11-openjdk one might get java-openjdk or java-11-openjdk.