I suppose that Denny faced the issue similar to the issue described in
https://jira.jboss.org/jira/browse/JBIDE-3733 .
A classpath container can't be updated using
JavaCore.setClasspathContainer(...). The
AbstractClasspathContainer.refresh() method must be used.
This method creates a new classpath container and informs JDT that the
classpath container has been changed.
Snjeza
Max Rydahl Andersen wrote:
On 20-02-2009 20:55, Snjezana Peco wrote:
> I have seen the changes related to JBIDE-3836 and think they will
> break performance.
okey looking at it I agree.
Even more important we don't start forking of different classpath
container behavior.
Why is that fix necessary to always calculate it instead of just
caching it ?
afaik other classpath containers just gets refreshed forcefully when
something related to changes instead of on every calculation.
Could we get these things aligned before GA ?
/max
> Namely, IClasspathContainer.getClasspathEntries() is often called
> which is the reason entries are cached within a classpath container.
> This isn't specific for our containers. Almost all of Eclipse/WTP
> containers work this way.
>
> Denny,
>
> Could you tell us the steps to reproduce the issue mentioned in
> JBIDE-3836?
>
> Snjeza
>
> Max Rydahl Andersen wrote:
>> Just saw
https://jira.jboss.org/jira/browse/JBIDE-3836 which shows
>> how bad it is that we now have the classpath containers duplicated.
>>
>> That issue I see as being just as relevant for WS, AS and ESB....not
>> something that should just be fixed in the ESB classpath container.
>>
>> Could we please move the base set of functionality (source/java doc
>> attachement and basic CPC func) back into AS core and all base
>> on that at least for GA ?
>>
>> /max
>> _______________________________________________
>> jbosstools-dev mailing list
>> jbosstools-dev(a)lists.jboss.org
>>
https://lists.jboss.org/mailman/listinfo/jbosstools-dev
>