OpenEdge Architect Startup

Posted by Matt Baker on 11-Nov-2009 19:03

Mostly going through the r-code referenced in your PROPATH and reading it looking for signature information for procedures and classes. It does keep a cache of what it finds so it the next time you start it reads the cache if OEA was previously shutdown properly.

Under window->preferences->Openedge->Advanced->Class cache there are a couple of options that you can adjust to change where it looks.

Posted by Admin on 11-Nov-2009 19:11

That is what I expected it was doing. While it is doing this, I get the following exception it the logfile, has this behaviour been noticed before?

java.lang.ArrayIndexOutOfBoundsException: 65521
at com.openedge.core.reflect.SignatureBlock.getString(SignatureBlock.java:797)
.

.

.
at com.openedge.pdt.text.codeinfo.CodeModelManager$CodeModelBuildJob.run(CodeModelManager.java:632)
at org.eclipse.core.internal.jobs.Worker.run(Worker.java:55)

Posted by Matt Baker on 12-Nov-2009 07:36

You need to call tech support and report this to them.

Posted by Admin on 12-Nov-2009 15:53

I will log an issue with supoprt for this.

Are there any implications on manually stopping this task when starting Eclipse? (What features are disabled when this is cancelled)

Thanks.

Posted by Matt Baker on 12-Nov-2009 16:09

"Disabled" might be too strong of a word. The set of classes and signature of procedures would incomplete. The biggest affect would be code completion and quick outline wouldn't show the right data if you're using classes. Class browser probably wouldn't show the right information either for any of those classes that were skipped. Maybe a few other things.

What is "suddenly taking a long time" mean? Did you install a newer service pack? Did you change your PROPATH? Did you try changing any of the preference settings and see if that helps? Are any of your folders on a network drive? A new virus scanner setting that doesn't skip .p and .r and .pl files?

Posted by Admin on 12-Nov-2009 16:20

"Suddenly" means, since I've started compiling the codebase and running of r-code. Also it does not happen every time I start Eclipse. Everything is fine for a couple of days and the next day it take 5-10 minutes to complete "Initializing OpenEdge ...".

Class Cache is set to: Propath and to NOT scan the installation directories.

Virus scanner: is configures to not scan the entire workspace

Description:

When starting Eclipse it takes a very long time to “Initializing OpenEdge tooling”. When checking the .log file in the Eclipse metadata directory there is the following exception: java.lang.ArrayIndexOutOfBoundsException: 65521.

Environment:

Progress:

Progress OpenEdge ArchitectVersion: 10.2.0.02OpenEdge Release: 10.2A02

Eclipse:

Eclipse for RCP/Plug-in DevelopersBuild id: 20090920-1017 Project settings:

r-code: 16316

p-code:16145

incude files: ~8000

PROPATH: 23 entries (including the standard paths)

Stack:

java.lang.ArrayIndexOutOfBoundsException: 65521

at com.openedge.core.reflect.SignatureBlock.getString(SignatureBlock.java:797)

at com.openedge.core.reflect.SignatureBlock.parseProcedureSignature(SignatureBlock.java:716)

at com.openedge.core.reflect.SignatureBlock.getInternalSignatures(SignatureBlock.java:387)

at com.openedge.core.reflect.SignatureBlock.read(SignatureBlock.java:173)

at com.openedge.core.reflect.RCodeObject.readSignature(RCodeObject.java:323)

at com.openedge.core.reflect.RCodeObject.readSignature(RCodeObject.java:303)

at com.openedge.core.reflect.RCodeObject.getSignature(RCodeObject.java:489)

at com.openedge.pdt.text.codeinfo.reader.RProcedureCodeReader.readPartialInfo(RProcedureCodeReader.java:86)

at com.openedge.pdt.text.codeinfo.reader.RCodeReader.readFileContent(RCodeReader.java:139)

at com.openedge.pdt.text.codeinfo.reader.CodeReaderUtil.createCodeContainerInfo(CodeReaderUtil.java:177)

at com.openedge.pdt.text.codeinfo.reader.CodeReaderUtil.createCodeContainerInfo(CodeReaderUtil.java:127)

at com.openedge.pdt.text.codeinfo.reader.ResourceFolderCodeReader$ResourceVisitor.visit(ResourceFolderCodeReader.java:123)

at org.eclipse.core.internal.resources.Resource$1.visitElement(Resource.java:60)

at org.eclipse.core.internal.watson.ElementTreeIterator.doIteration(ElementTreeIterator.java:82)

at org.eclipse.core.internal.watson.ElementTreeIterator.doIteration(ElementTreeIterator.java:86)

at org.eclipse.core.internal.watson.ElementTreeIterator.iterate(ElementTreeIterator.java:127)

at org.eclipse.core.internal.resources.Resource.accept(Resource.java:70)

at com.openedge.pdt.text.codeinfo.reader.ResourceFolderCodeReader.buildLibrary(ResourceFolderCodeReader.java:56)

at com.openedge.pdt.text.codeinfo.reader.CodeReaderUtil.createLibraryInfo(CodeReaderUtil.java:60)

at com.openedge.pdt.text.codeinfo.CodeModelManager.buildLibraryInfo(CodeModelManager.java:309)

at com.openedge.pdt.text.codeinfo.CodeModelManager.buildTOCforProject(CodeModelManager.java:271)

at com.openedge.pdt.text.codeinfo.CodeModelManager.buildCodeModel(CodeModelManager.java:236)

at com.openedge.pdt.text.codeinfo.CodeModelManager.access$700(CodeModelManager.java:121)

at com.openedge.pdt.text.codeinfo.CodeModelManager$CodeModelBuildJob.run(CodeModelManager.java:632)

at org.eclipse.core.internal.jobs.Worker.run(Worker.java:55)

Posted by Admin on 12-Nov-2009 19:09

Tech Support: W911120179

Thanks,

Sascha

Posted by jquerijero on 17-Feb-2010 15:44

Any news on this being fixed in 10.2A?

Posted by rbf on 17-Feb-2010 16:07

jquerijero wrote:

Any news on this being fixed in 10.2A?

It is certainly a lot faster and more stable in 10.2B if that is what you are asking.

Posted by jquerijero on 17-Feb-2010 16:12

I really mean 10.2A.

Posted by Admin on 17-Feb-2010 16:21

I really mean 10.2A.

If you need this fixed on 10.2A, I'd discuss that with tech support. However I believe that your reasons not to upgrade to 10.2B must be solid to be succesful.

This thread is closed

All Replies