[jbosstools-issues] [JBoss JIRA] (JBDS-4176) Useless-provides detected by RPMlint

[Nick Boldt (JIRA)]

    [ https://issues.jboss.org/browse/JBDS-4176?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13322102#comment-13322102 ] 

Nick Boldt commented on JBDS-4176:
----------------------------------

You can safely ignore all the nested jars inside forge plugins when checking for rpmlint. 

Sure, we *SHOULD* have the *SAME* versions everywhere, but Forge is special and needs to embed lots of redundant/old runtime jars. (I'm not sure why, perhaps [~koen.aers] [~gastaldi] can explain. Last time I asked I was told "These JARs are part of [an] addon and unfortunately cannot be reused from  the installed footprint.")

Not sure what to do about the extra aether.impl inside m2e 1.7.1. Maybe [~fbricon] can explain why there's a nested jar there instead of just having a dependency on Orbit/Neon/target platform/rh-eclipse46-* packages.

For this specific example of aether.impl, your TP includes the version from Neon.1 (1.0.1). If there's a reason to move to the one from Neon.2 (1.0.2), we could update the devstudio TP accordingly. Is there a reason other than "so rpmlint doesn't complain"?  :D

> Useless-provides detected by RPMlint
> ------------------------------------
>
>                 Key: JBDS-4176
>                 URL: https://issues.jboss.org/browse/JBDS-4176
>             Project: Red Hat JBoss Developer Studio (devstudio)
>          Issue Type: Bug
>          Components: rpm
>    Affects Versions: 10.2.0.AM3
>         Environment: RHEL7
>            Reporter: Lukáš Valach
>            Assignee: Nick Boldt
>             Fix For: 10.2.0.GA
>
>         Attachments: opt-rh-list.2.txt, opt-rh-list.2.txt.sh, opt-rh-list.txt, opt-rh-list.txt.sh
>
>
> The rpmlint detects some problems in rpm package:
> {code}
> rpmlint -f ./rpmlint.conf ./rh-eclipse46-devstudio-10.2-0.20161111.1151.el7.x86_64.rpm
> rh-eclipse46-devstudio.x86_64: E: useless-provides rh-eclipse46-osgi(com.google.guava)
> rh-eclipse46-devstudio.x86_64: E: useless-provides rh-eclipse46-osgi(com.google.inject)
> rh-eclipse46-devstudio.x86_64: E: useless-provides rh-eclipse46-osgi(javax.wsdl)
> rh-eclipse46-devstudio.x86_64: E: useless-provides rh-eclipse46-osgi(org.apache.commons.collections)
> rh-eclipse46-devstudio.x86_64: E: useless-provides rh-eclipse46-osgi(org.apache.commons.io)
> rh-eclipse46-devstudio.x86_64: E: useless-provides rh-eclipse46-osgi(org.apache.commons.lang3)
> rh-eclipse46-devstudio.x86_64: E: useless-provides rh-eclipse46-osgi(org.apache.commons.logging)
> rh-eclipse46-devstudio.x86_64: E: useless-provides rh-eclipse46-osgi(org.eclipse.aether.api)
> rh-eclipse46-devstudio.x86_64: E: useless-provides rh-eclipse46-osgi(org.eclipse.aether.connector.basic)
> rh-eclipse46-devstudio.x86_64: E: useless-provides rh-eclipse46-osgi(org.eclipse.aether.impl)
> rh-eclipse46-devstudio.x86_64: E: useless-provides rh-eclipse46-osgi(org.eclipse.aether.spi)
> rh-eclipse46-devstudio.x86_64: E: useless-provides rh-eclipse46-osgi(org.eclipse.aether.util)
> rh-eclipse46-devstudio.x86_64: E: useless-provides rh-eclipse46-osgi(com.fasterxml.classmate)
> rh-eclipse46-devstudio.x86_64: E: useless-provides rh-eclipse46-osgi(org.jboss.logging.jboss-logging)
> rh-eclipse46-devstudio.x86_64: E: useless-provides rh-eclipse46-osgi(org.yaml.snakeyaml)
> rh-eclipse46-devstudio.x86_64: E: useless-provides rh-eclipse46-osgi(slf4j.api)
> rh-eclipse46-devstudio.x86_64: E: useless-provides rh-eclipse46-osgi(org.objectweb.asm)
> 1 packages and 0 specfiles checked; 17 errors, 0 warnings.
> {code}
> There are only important problems in the list above, another errors/warnings were suppressed according to discussion with [~vkadlcik].
> There is configuration of rpmlint filter: 
> {code}
> addFilter("W: ") 				# Warnings can be suppressed
> addFilter("E: dir-or-file-in-opt") 		# False positive
> addFilter("E: devel-dependency")		# False positive
> addFilter("E: explicit-lib-dependency")		# May be improved but is not important for now
> addFilter("E: non-executable-script")		# 99% of these errors are false positive comming from upstream, there is a very small chance to get relevant error
> addFilter("E: wrong-script-interpreter")	# The same as above
> addFilter("E: zero-length")			# The same as above
> {code}
> [~vkadlcik] please correct me if I'm wrong.
> RPMlint test has been automated, you can find it in [jenkins|https://dev-platform-jenkins.rhev-ci-vms.eng.rdu2.redhat.com/view/DevStudio%20RPM%20Tests/job/rpm_rpmlint_test/]



--
This message was sent by Atlassian JIRA
(v7.2.3#72005)