Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [jdt-dev] [platform-dev] Need hint on how to configure launch for testing compiler mod


As I mentioned on the other mailing list, sometimes the API tools just produce bogus errors that can be removed with a clean build. If the IDE was setup with Oomph, it will have a correct baseline (based currently on the 4.13 release).  (Sometimes after a pull there are even phantom syntax errors; a no-op edit and save on the file removes it.)

On 05.10.2019 14:54, Stephan Herrmann wrote:
The versioning issues you are seeing are reported by "API Tooling", which helps us to remain honest in how we evolve JDT's API. You could either disable API analysis in your workspace (as long as you are not modifying JDT API), or try to configure a proper Baseline (2019-09 atm).

For dependencies of your plugin: have you added "org.eclipse.jdt.core" under "Require-Bundle"?

As for a live session: feel free to join us at EclipseCon [1], which is the perfect place to meet some of us in person, and where we could easily arrange a little hands-on session.



On 05.10.19 02:23, Richard Steiger wrote:
Apologies for sending to platform-dev instead of the intended jdt-dev (Stephan clued me in, but hit Reply List without checking).

On 10/4/2019 3:54 PM, Richard Steiger wrote:


Thanks for the helpful remarks.

I've rebuilt my test project from scratch using PDE, and am seeing somewhat similar displays in the various editors to those you shared.  I've also reviewed some plugin dev tutorials and related materials, so have a slightly clearer conceptual map of JDT.

In PDE, I fiddled around with asking the (I think) Manifest Editor to auto-discover the required plugins.  The resulting new project is reporting 25 errors, most are due (its seems) to stale version qualifiers, the following lines are a reasonable proxy for most of the problems (copied from Problems view):

Invalid @since 3.19 tag on newAST(int, boolean); expecting @since 3.20 /org.eclipse.jdt.core/dom/org/eclipse/jdt/core/dom    line 438    @since tag problem Invalid @since 3.19 tag on propertyDescriptors(int, boolean); expecting @since 3.20 /org.eclipse.jdt.core/dom/org/eclipse/jdt/core/dom    line 93    @since tag problem Invalid @since 3.19 tag on RedundantNullCheckOnConstNonNullField; expecting @since 3.20 /org.eclipse.jdt.core/compiler/org/eclipse/jdt/core/compiler line 1806    @since tag problem Invalid @since 3.9 tag on org.eclipse.jface.text.MultiStringMatcher; expecting @since 3.10 /org.eclipse.text/src/org/eclipse/jface/text    line 30 @since tag problem Project has only main sources but depends on project 'org.eclipse.jdt.core.tests.builder' which has only test sources.        Build path    Build Path Problem The minor version should be incremented in version 3.8.700, since new APIs have been added since version 3.8.500 MANIFEST.MF /org.eclipse.ui.console/META-INF    line 5 Version Numbering Problem The minor version should be the same for version 1.6.0, since no new APIs have been added since version 1.5.0    MANIFEST.MF /org.eclipse.core.databinding.beans/META-INF    line 5 Version Numbering Problem

At this point, my blockers are

  * I haven't discovered where the stale jars are specified (can't find
    anything in any project artifacts), and
  * the compiler plugins are still missing, despite my test project importing
    org.eclipse.jdt.core.compiler.CompilationProgress, and
    org.eclipse.jdt.core.compiler.batch.BatchCompiler (which seem should have     a hefty transitive closure, i.e. I expect to see >> 3 dependencies).

Stepping back from such details, my concern at this juncture is that what I'm trying to do is somewhat subtle, involves a lot of moving parts, and my previous test approach has holes in it that will require invoking a lot more JDT and related machinery beyond the compiler.  For example, I suspect I'll need to either load the subject projects in the Runtime Workbench (where the modified compiler is) instead of the Development Workbench (where the unmodified compiler faults their builds due to cp cycles), then build the new projects, then load and execute them, thereby completing the e2e verification.

Given my ponderous pace up to now, extrapolating to getting this working (even taking my forward learning curve into account) could take a lot of hand-holding, i.e. your and/or others' time (given willingness, and I'm not assuming anything).

Let me suggest an alternative: if you or another expert would be willing to get on a video call, at your convenience, and we could essentially pair-up on debugging the launch configuration, it might go pretty fast, and cost significantly less of everyone's time.  Could that work?

Meanwhile, I'll refine the test harness per the above realizations, so I'll have a more complete set of JDT plugin dependencies to work with.

FWIW, I'm also willing to pay it forward by contributing what I've learned (once I've really learned it :-) to (existing or new) documentation.

Thanks again,


On 10/4/2019 1:29 AM, Wim Jongman wrote:
Hi Richard,

If you open the MANIFEST.MF part of your new plugin you notice a green run icon in the top left of the editor. This will launch your bundle in a new eclipse instance AND will create a run configuration.


The run configurations can be edited via an action in the toolbar of the java perspective.


Find your configuration and examine all the tabs. The Plug-ins (bundles) tab lets you select the location from where the bundle should be picked.


On Thu, Oct 3, 2019 at 12:57 PM Richard Steiger <rsteiger@xxxxxxxxxxxxxxxx <mailto:rsteiger@xxxxxxxxxxxxxxxx>> wrote:


    Thanks for the advice.   Using any available UI, how does one control     which bundles are exported for a given launch?   Or, is there a doc that
    explains this?

    Thanks again,


    On 10/3/2019 2:41 AM, Wim Jongman wrote:
    Hey Richard,

    Maybe it is caused by the fact that Eclipse is using split packages.
    This enables multiple bundles (plugins) to contribute to the same
    package. I'm guessing that parts of the org.eclipse.core.runtime package     are exported by other bundles. You may need to include all bundles that
    contribute to this package from the workspace.



    On Thu, Oct 3, 2019 at 10:41 AM Richard Steiger
    <rsteiger@xxxxxxxxxxxxxxxx <mailto:rsteiger@xxxxxxxxxxxxxxxx>> wrote:

        As I've discussed in prior messages to platform-dev@, I'm working on         prototyping a proposed new ejc feature, to surface an option in the
        IDE's Java compiler preferences, allowing setting
        JavaCore.*/CORE_CIRCULAR_CLASSPATH/*, and adding an IGNORE value to
        it's range (in addition to ERROR and WARN), with the ultimate
        purpose of allowing building systems that intentionally leverage         mutually-recursive modules (where I use /module /in the classic
        sense, not the JDK's Module subsystem since 9).

        As I've also promised, I'll be creating a feature bug that explains         the background, rationale, and proposed code changes, hopefully by         this Friday.  That said, prior to doing so, I feel it's essential         for me to empirically verify that this mod works end-to-end as I         intend, and, for example, doesn't spawn a class-loading or other         runtime hornet's nest (that's not straightforwardly avoided|addressed).

        To this verification end, I'm trying to run a test harness that

         1. selects the IGNORE option,
         2. builds 2 (/test/) /subject /projects that are mutually-dependent,          3. assuming that the subject projects successfully build, launches             the main (entry-point) project (which triggers loading the
            second one, etc), and
         4. assuming no hornets appear, verifies that at least a minimal
            circular classpath case is correctly handled, end-to-end.

        I've spent quite a bit of time reading eclipse docs, how-tos, the         /Eclipse Plugins/ bible (3rd ed), and browsed a fair amount of JDT         code, trying to figure out how to get the test to work, and at this         point, my pick is bent.  I could really use a hint how to proceed.

        To set context, the test project, and both subject projects, all         reside in the jdt-master workspace (btw, is there a vernacular         shorthand for this workspace?).  My first attempt to get the test to         work was simply to either run or debug it as a Java app.  It fails
        with the backtrace

            Exception in thread "main" java.lang.SecurityException: class             "org.eclipse.core.runtime.IStatus"'s signer information does not             match signer information of other classes in the same package                 at java.lang.ClassLoader.checkCerts(                 at java.lang.ClassLoader.preDefineClass(                 at java.lang.ClassLoader.defineClass(
                at                 at                 at$100(                 at$                 at$                 at Method)                 at                 at java.lang.ClassLoader.loadClass(                 at sun.misc.Launcher$AppClassLoader.loadClass(                 at java.lang.ClassLoader.loadClass(
                at net.ess.jcore.CompilerRunner.main(

        Just prior to throwing the SecurityException, the JVM's classpath is


        My understanding is that the launchConfiguration that's being
        generated and applied is erroneous: as can be seen, the classpath         includes references to plugin jars in my base eclipse installation         (paths starting with C:\Users\rsteiger\.p2\pool\plugins\), and not
        from the just-built jdt-master workspace (paths starting with
        T:\eclipse\SDK\jdt-master\git\). I've tried several approaches to
        fix this:

          * made the test project into a plugin (not sure I didn't goof this
            up, since am a plugin-noob);
          * discovering that the project's /Plug-in Dependencies/ is a             read-only classpath container containing the above incorrect
            jars, tried removing <classpathentry kind="con"
path="org.eclipse.pde.core.requiredPlugins"/> from the project's
            .classpath, then adding kind="lib" entries for (what I thought
            are) the correct jars;
          * revisiting various references describing care and feeding of             launchConfigs, finding nothing covering how to configure configs             from the jdt workspace UI, just how to do so programmatically
            (which is of course totally useless in this context); and
          * gnashing my teeth and swearing.

        None of these approaches have helped.

        I turn to y'all to suggest a viable approach.

        Thanks in advance,


        platform-dev mailing list
        platform-dev@xxxxxxxxxxx <mailto:platform-dev@xxxxxxxxxxx>
        To change your delivery options, retrieve your password, or
        unsubscribe from this list, visit

    platform-dev mailing list
    platform-dev@xxxxxxxxxxx <mailto:platform-dev@xxxxxxxxxxx>
    To change your delivery options, retrieve your password, or unsubscribe from this list, visit
    platform-dev mailing list
    platform-dev@xxxxxxxxxxx <mailto:platform-dev@xxxxxxxxxxx>
    To change your delivery options, retrieve your password, or unsubscribe
    from this list, visit

platform-dev mailing list
To change your delivery options, retrieve your password, or unsubscribe from this list, visit

platform-dev mailing list
To change your delivery options, retrieve your password, or unsubscribe from this list, visit

jdt-dev mailing list
To change your delivery options, retrieve your password, or unsubscribe from this list, visit

jdt-dev mailing list
To change your delivery options, retrieve your password, or unsubscribe from this list, visit

Back to the top