444412 – Update EventBroker usage in PartRenderingEngine

Bug 444412 - Update EventBroker usage in PartRenderingEngine

Summary: Update EventBroker usage in PartRenderingEngine

Status:	RESOLVED FIXED

Alias:	None

Product:	Platform
Classification:	Eclipse Project
Component:	UI (show other bugs)
Version:	4.5
Hardware:	PC Linux

Importance:	P3 normal (vote)
Target Milestone:	4.5 M6
Assignee:	Philip Wenig
QA Contact:

URL:
Whiteboard:
Keywords:	greatfix

Depends on:
Blocks:	462056
	Show dependency tree

Reported:	2014-09-17 18:50 EDT by Lars Vogel
Modified:	2015-06-10 17:32 EDT (History)
CC List:	4 users (show)

See Also:	Gerrit Change Git Commit

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.

Description Lars Vogel

2014-09-17 18:50:52 EDT

Similar to Bug 442285 and Bug 441120 but for PartRenderingEngine.

Comment 1 Philip Wenig

2015-02-07 03:05:18 EST

I'll try to fix this.

Comment 2 Philip Wenig

2015-02-09 07:01:02 EST

https://git.eclipse.org/r/41415

Comment 3 Eclipse Genie

2015-02-09 08:42:38 EST

New Gerrit change created: https://git.eclipse.org/r/41415

Comment 4 Eclipse Genie

2015-03-09 14:05:33 EDT

Gerrit change https://git.eclipse.org/r/41415 was merged to [master].
Commit: http://git.eclipse.org/c/platform/eclipse.platform.ui.git/commit/?id=b523189149eb20a25b055578706cadaf2bab4c9c

Comment 5 Lars Vogel

2015-03-09 14:05:43 EDT

Thanks Philip.

Comment 6 Lars Vogel

2015-03-13 09:53:47 EDT

This did introduce a Bug in CSS switching See Bug 462056.

Comment 7 Patrick Tasse

2015-06-10 17:32:52 EDT

Before this change, the EventBroker would wrap the EventHandler in a UIEventHandler, which would make sure the event is handled in a Runnable on the UI thread.

After this change, the dependency injection callback is wrapped in a DIEventHandler, which executes in whatever thread triggered the event.

In our project we had a few instances where IWorkbenchPage.closeEditor() was (incorrectly) called from a WorkbenchModifyOperation (non-UI thread), and it used to work anyways because the bug was hidden by the use of the UIEventHandler.

Now in 4.5 it triggers a SWTException (Invalid thread access), and we only noticed this today and fixed our code just in time for Mars release.

So I would just like to make sure that the platform is intentionally no longer meant to safeguard against this invalid thread access like the EventBroker used to do when subscribing events?