Skip to main content
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [platform-releng-dev] 4.17 I-Build: I20200611-1800 - BUILD FAILED
Alex, by choice I made my maven project use toolchains.xml to configure in a Java 7 SDK to compile against and I made the project compile against the latest OSGi R8 APIs that have Optional in some of the methods in FrameworkUtils.  It then has a class that uses the old getBundle method from FrameworkUtils (which does not return Optional) and it compiles fine.
 
Are you saying this should not work? It seems to work for me.  Or are you saying what I did with that maven project is different than what Tycho does and that Tycho has additional checks above and beyond just compiling against a specific BREE library.
 
Tom
----- Original message -----
From: Aleksandar Kurtakov <akurtako@xxxxxxxxxx>
Sent by: platform-releng-dev-bounces@xxxxxxxxxxx
To: "Eclipse platform release engineering list." <platform-releng-dev@xxxxxxxxxxx>
Cc:
Subject: [EXTERNAL] Re: [platform-releng-dev] 4.17 I-Build: I20200611-1800 - BUILD FAILED
Date: Thu, Jun 18, 2020 1:39 PM
 
 
 
On Thu, Jun 18, 2020 at 9:15 PM Thomas Watson <tjwatson@xxxxxxxxxx> wrote:
Sorry for resurfacing this old thread.  I did try to compile my simple project [1] against Java 7 SDK and it works for me.  It also works if I configure the maven build of my project to use ECJ.  I'm not sure why it fails in our Eclipse builds with tycho but not with a simple maven project that targets to Java 7 and is compiled against a Java 7 libraries.
 
Note that if I try to compile against one of the new methods in FrameworkUtil that returns an Optional it then fails as expected.  But compiling against old methods with no Optional in their signature compile fine when also compiling against the Java 7 SDK libraries.
 
This only concerns me because it would appear Tycho will aggressively force all bundles that use the existing FrameworkUtil.getBundle(Object) method to move up to Java 8 in order to compile.  Some will see this as a good thing, nobody should be targeting Java 7 at all anymore!  But others will see this as an unnecessary change they have to make just to work around some build issue.
 
In the end this is not really an Eclipse Project issue, we can easily move all our bundles to Java 8 or later.  I only worry about consumers of the new release that may run into this when they start compiling against our next release.
 
This is project choice and not default in Tycho [1]. Projects have to enable it explicitly to compile their plugin against the BREE set in it. And in this case compiling against latest osgi.core that uses Optional on Java 7 can't work. So these projects have the choice to not use BREE toolchain (with all possible implications from that), compile against old osgi.core or move their dependent bundles to Java 8.
I don't think there is more we can do to help projects.
 
 

Tom
 
 
 
----- Original message -----
From: "Daniel Megert" <daniel_megert@xxxxxxxxxx>
Sent by: platform-releng-dev-bounces@xxxxxxxxxxx
To: "Eclipse platform release engineering list." <platform-releng-dev@xxxxxxxxxxx>
Cc:
Subject: [EXTERNAL] Re: [platform-releng-dev] 4.17 I-Build: I20200611-1800 - BUILD FAILED
Date: Sun, Jun 14, 2020 11:04 AM
 
Tom,

my guess is that you don't/didn't have a JRE 1.7 installed and hence a newer JRE is/was bound to the specified EE. There's a warning for that case on the project.

Dani



From:        Aleksandar Kurtakov <akurtako@xxxxxxxxxx>
To:        "Eclipse platform release engineering list." <platform-releng-dev@xxxxxxxxxxx>
Date:        12.06.2020 15:49
Subject:        [EXTERNAL] Re: [platform-releng-dev] 4.17 I-Build: I20200611-1800 - BUILD        FAILED
Sent by:        platform-releng-dev-bounces@xxxxxxxxxxx





On Fri, Jun 12, 2020 at 4:00 PM Thomas Watson <tjwatson@xxxxxxxxxx> wrote:
Why would everything have to move up in their execution environment just because the internals of FrameworkUtil reference Optional?  Is this using ECJ or javac to compile.  I'm fairly certain javac would not flag this as an error.

Our build system build against the JVM version as specified in BREE. Thus when the bundle has BREE 1.7 it fails to compile as its dependent bundle uses Optional and it's not found.
 

Tom
 
 
 
----- Original message -----
From: Aleksandar Kurtakov <akurtako@xxxxxxxxxx>
Sent by: platform-releng-dev-bounces@xxxxxxxxxxx
To: "Eclipse platform release engineering list." <platform-releng-dev@xxxxxxxxxxx>
Cc:
Subject: [EXTERNAL] Re: [platform-releng-dev] 4.17 I-Build: I20200611-1800 - BUILD FAILED
Date: Fri, Jun 12, 2020 1:18 AM
 
 
 
On Fri, Jun 12, 2020 at 2:30 AM <genie.releng@xxxxxxxxxxx> wrote:

Please go to https://ci-staging.eclipse.org/releng/job/I-build-4.17/21/consoleand check the build failure.


Error is:
[ERROR] /home/jenkins/agent/workspace/I-build-4.17/eclipse.platform.releng.aggregator/eclipse.platform.releng.aggregator/cje-production/gitCache/eclipse.platform.releng.aggregator/rt.equinox.bundles/bundles/org.eclipse.equinox.http.servlet.tests/src/org/eclipse/equinox/http/servlet/testbase/BaseTest.java:[182]
02:30:31  [ERROR]         return FrameworkUtil.getBundle(BaseTest.class).getBundleContext();
02:30:31  [ERROR]                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
02:30:31  [ERROR] The type java.util.Optional cannot be resolved. It is indirectly referenced from required .class files
02:30:31  [ERROR] 1 problem (1 error)
This makes me think it's related to https://git.eclipse.org/c/equinox/rt.equinox.framework.git/commit/?id=ce9a3f27b26b7bbde23b747a0aa7ad0c470b55d2 . I'm looking further into it.


_______________________________________________
platform-releng-dev mailing list
platform-releng-dev@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/platform-releng-dev




--
Alexander Kurtakov
Red Hat Eclipse Team
_______________________________________________
platform-releng-dev mailing list
platform-releng-dev@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/platform-releng-dev
 


_______________________________________________
platform-releng-dev mailing list
platform-releng-dev@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/platform-releng-dev




--
Alexander Kurtakov
Red Hat Eclipse Team_______________________________________________
platform-releng-dev mailing list
platform-releng-dev@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/platform-releng-dev


 
_______________________________________________
platform-releng-dev mailing list
platform-releng-dev@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/platform-releng-dev
 

_______________________________________________
platform-releng-dev mailing list
platform-releng-dev@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/platform-releng-dev


--
Alexander Kurtakov
Red Hat Eclipse Team
_______________________________________________
platform-releng-dev mailing list
platform-releng-dev@xxxxxxxxxxx
To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/platform-releng-dev
 

Back to the top