RE: Applied [HEAD] RE: [cdt-patch] code reader cache interferes withrefa

["Schorn, Markus" <Markus.Schorn@xxxxxxxxxxxxx>]

As the failures were random, I just guessed that the problem is related to caching.

 

The test-cases delete files without using a WorkspaceModifyOperation. (This occurs

in org.eclipse.cdt.core.tests.BaseTestFramwork.tearDown()). This can well be the

reason that notifications about resource changes do not reach your cache in time

and this should not happen when running the IDE.

 

Markus.

 

---

From: cdt-patch-bounces@xxxxxxxxxxx [mailto:cdt-patch-bounces@xxxxxxxxxxx] On Behalf Of John Camelon
Sent: Montag, 25. April 2005 17:59
To: CDT patches and commits are posted to this list.
Subject: Applied [HEAD] RE: [cdt-patch] code reader cache interferes withrefactoring tests

No what you did was fine.  I have no problem with your patch, I am applying it.  

What I am concerned about is that there is a bug in our CodeReaderCache and that stale information is being served out to clients post save.  
I tried to debug the failing test of yours but its not clear to me how the cache is responsible for what was failing prior to this patch.  
If you can describe to me exactly how you determined it was the cache I'd be grateful.

Thanks
JohnC
www.eclipse.org/cdt

cdt-patch-bounces@xxxxxxxxxxx wrote on 04/25/2005 11:48:07 AM:

> Initially I made calls to CDom.getTranslationUnit(IFile fileToParse)
> which turns out to make
> use of the SavedCodeReaderFactory. In the meanwhile I am using it
> explicitly. Is it better
> to use the FileCodeReaderFactory, what did you expect me to use?
>  
> Markus.
>
> From: cdt-patch-bounces@xxxxxxxxxxx [mailto:cdt-patch-bounces@xxxxxxxxxxx]
> On Behalf Of John Camelon
> Sent: Montag, 25. April 2005 17:24
> To: CDT patches and commits are posted to this list.
> Subject: Re: [cdt-patch] code reader cache interferes with refactoring tests

>
> Is this just a workaround to a bug/race condition?
> Only the DOM/AST service uses these code readers, so I am unsure as
> to why it would affect your test in a way that we should workaround.
>
> JohnC
> www.eclipse.org/cdt
>
> cdt-patch-bounces@xxxxxxxxxxx wrote on 04/25/2005 04:30:37 AM:
>
> > The patch flushes the code reader cache after each testcase.
> > Markus.
> > [attachment "refactoring-tests-patch-20050425.txt" deleted by John
> > Camelon/Ottawa/IBM] _______________________________________________
> > cdt-patch mailing list
> > cdt-patch@xxxxxxxxxxx
> > https://dev.eclipse.org/mailman/listinfo/cdt-patch
> _______________________________________________
> cdt-patch mailing list
> cdt-patch@xxxxxxxxxxx
> https://dev.eclipse.org/mailman/listinfo/cdt-patch