Some Eclipse Foundation services are deprecated, or will be soon. Please ensure you've read this important communication.
Bug 337521 - AbstractSimpleNavigationNodeProvider should provide a way to build nodes asynchronous
Summary: AbstractSimpleNavigationNodeProvider should provide a way to build nodes asyn...
Status: RESOLVED FIXED
Alias: None
Product: Riena
Classification: RT
Component: navigation (show other bugs)
Version: unspecified   Edit
Hardware: PC All
: P3 normal (vote)
Target Milestone: 3.0.0.M7   Edit
Assignee: Nobody - feel free to take it CLA
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2011-02-18 03:08 EST by Nobody - feel free to take it CLA
Modified: 2011-05-03 10:16 EDT (History)
3 users (show)

See Also:

Attachments
Patch for async-buildNode(..) (12.94 KB, patch)
2011-02-18 03:30 EST, Nobody - feel free to take it CLA 		steffen.kriese: iplog+
Details | Diff
New patch for async navigation support (16.17 KB, patch)
2011-02-24 03:25 EST, Steffen Kriese CLA 		steffen.kriese: iplog+
Details | Diff
Updated patch with fix for LnfUpdater (14.32 KB, patch)
2011-02-25 07:07 EST, Nobody - feel free to take it CLA 		no flags Details | Diff
Show Obsolete (1) View All 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 Nobody - feel free to take it CLA 2011-02-18 03:08:47 EST 
At the moment the buildNode(..) method of INavigationAssembler is allways called on the user interface thread. As the implementation of buildNode(..) can be time consuming there should be a way (API?) to build the node on a parallel worker thread for a more responsive user interface.
Comment 1 Nobody - feel free to take it CLA 2011-02-18 03:30:35 EST 
Created attachment 189255 [details]
Patch for async-buildNode(..)

Patch for async-buildNode(..) + Example
Comment 2 Holger Hoch CLA 2011-02-23 02:10:01 EST 
The patch can lead to a high cpu cycle consumtion for at least one processor. It should be modified as follows:

while (!shell.isDisposed()) {
   if (!display.readAndDispatch()) {
      display.sleep();
   }
}

Sending an explicit
   display.wake();
should not be required as #isFinished is set at the end of the #runJob() method and #finalUpdateUI() is executed on the display thread afterwards thus quitting the above dispatch loop.
Comment 3 Steffen Kriese CLA 2011-02-24 03:25:21 EST 
Created attachment 189675 [details]
New patch for async navigation support

- adds NavigationNode.createNodeAsync() method
- reduces the performance problem, while dispatching the ui-events
Comment 4 Nobody - feel free to take it CLA 2011-02-25 07:07:52 EST 
Created attachment 189788 [details]
Updated patch with fix for LnfUpdater
Comment 5 Steffen Kriese CLA 2011-03-21 04:53:03 EDT 
Patch applied