Eclipse Community Forums: Java Development Tools (JDT) » IWorkbenchPage.getAggregateWorkingSet() has no visibility into JDT's WorkingSetModel.getAllWorkingSe

Home » Language IDEs » Java Development Tools (JDT) » IWorkbenchPage.getAggregateWorkingSet() has no visibility into JDT's WorkingSetModel.getAllWorkingSe

IWorkbenchPage.getAggregateWorkingSet() has no visibility into JDT's WorkingSetModel.getAllWorkingSe [message #250518]

Tue, 08 January 2008 19:59

Eclipse User

In Eclipse 3.2.2, JDT has the following:

package org.eclipse.jdt.internal.ui.workingsets;
import org.eclipse.ui.IWorkingSet;
....

public class WorkingSetModel {
public void addPropertyChangeListener(...) { ... }
public IWorkingSet[] getAllWorkingSets() { ... }
...
}

There is a strange paradox where, in the plug-in perspective
with several working sets defines, we have the following behavior:

IWorkbench wb=PlatformUI.getWorkbench();

IWorkbenchWindow ww=wb.getActiveWorkbenchWindow();
IWorkbenchPage wp=ww.getActivePage();
IWorkingSet agg=wp.getAggregateWorkingSet();

// counter-intuitive:
// agg.isEmpty() == true;

IWorkingSet[] aws = wb.getWorkingSetManager().getAllWorkingSets();

// expected:
// aws.length > 0

Is this intended behavior?

This behavior seems to have some relevance to the following bugs:

https://bugs.eclipse.org/bugs/show_bug.cgi?id=137396
https://bugs.eclipse.org/bugs/show_bug.cgi?id=182978

There are 31 currently pending bugs involving working sets and I didn't
have the time to search through all of them to tell if this is a
duplicate or not.

https://bugs.eclipse.org/bugs/buglist.cgi?query_format=advan ced&short_desc_type=allwordssubstr&short_desc=Workin gSet&classification=Eclipse&version=3.2.2&versio n=3.3&version=3.3.1&version=3.4&long_desc_type=a llwordssubstr&long_desc=&bug_file_loc_type=allwordss ubstr&bug_file_loc=&status_whiteboard_type=allwordss ubstr&status_whiteboard=&keywords_type=allwords& keywords=&bug_status=UNCONFIRMED&bug_status=NEW& bug_status=ASSIGNED&bug_status=REOPENED&emailtype1=s ubstring&email1=&emailtype2=substring&email2=&am p;bugidtype=include&bug_id=&votes=&chfieldfrom=& amp;chfieldto=Now&chfieldvalue=&cmdtype=doit&ord er=Reuse+same+sort+as+last+time&field0-0-0=noop&type 0-0-0=noop&value0-0-0=

As far as the Eclispe 3.2.2 APIs are concerned, it seems to me that
the biggest hurdle against unified access to working sets comes from
the disconnected management of IWorkingSetManager instances which do
not know about each other and do not have peer-level access to
IWorkingSet instances.

Between JDT and the workbench, we have at least 2 distinct
IWorkingSetManager objects:

- JDT's WorkingSetModel.fLocalWorkingSetManager
- IWorkbenchWindow.getWorkingSetManager()

and a pseudo working set manager in each workbench page:

IWorkingSet[] IWorkbenchPage.getAggregateWorkingSet();

It seems to me that there is insufficient protocol-level coordination
and object-level awareness to ensure uniform access to working sets.

Does this characterize the underlying issue behind the confusing
behavior described above or have I missed something?

-- Nicolas.

Previous Topic:	JDT AST question: why does EnumConstantDeclaration uses an anonymous class?
Next Topic:	Get design registry at "developing-time"

Goto Forum:

-=] Back to Top [=-

Current Time: Sat Oct 25 00:17:48 EDT 2025

.:: Contact :: Home ::.

Breadcrumbs

Sign up to our Newsletter