[bug #4066] NSWorkspace launchApplicaion: misbehaved

[nobody]

=================== BUG #4066: LATEST MODIFICATIONS ==================
http://savannah.gnu.org/bugs/?func=detailbug&bug_id=4066&group_id=99

Changes by: David Ayers <d.ayers@inode.at>
Date: Sun 06/22/2003 at 12:39 (Europe/Vienna)

------------------ Additional Follow-up Comments ----------------------------
I think the price on performance (time waiting on an app to finish all of its 
initialization) would be to high for the avarage case.  Maybe we could augment 
the API with a method, that actually waits?  Or send a distributed notification 
before/after posting appDidFinishLaunching:?



=================== BUG #4066: FULL BUG SNAPSHOT ===================


Submitted by: stefanu                 Project: GNUstep                      
Submitted on: Sun 06/22/2003 at 01:52
Category:  Gui/AppKit                 Severity:  5 - Major                  
Bug Group:  Bug                       Resolution:  None                     
Assigned to:  None                    Status:  Open                         

Summary:  NSWorkspace launchApplicaion: misbehaved

Original Submission:  NSWorkspace launchApplication:'name' does not behave as 
it should. 

>From apple docs:

Returns YES if the application is successfully launched...

However, it seems to return immediately after running application process. I 
think that application is 'successfulyl lanchued' when all app initialisation 
priocess ends and app's delegate receives applicationDidFinishLaunching: 
notification. Or ... application is sucessfully launched when i can connect to 
it. I wouls expect this method to wait for all application initialisation and 
for some distant notification about successfull launch.

Current misbehaviour requires workarounds to make it work.


Follow-up Comments
*******************

-------------------------------------------------------
Date: Sun 06/22/2003 at 12:39       By: ayers
I think the price on performance (time waiting on an app to finish all of its 
initialization) would be to high for the avarage case.  Maybe we could augment 
the API with a method, that actually waits?  Or send a distributed notification 
before/after posting appDidFinishLaunching:?

-------------------------------------------------------
Date: Sun 06/22/2003 at 12:36       By: cehardin

Well, I do "#1" in my dock.app, I launch apps "just for fun"
What I do is wait for 30 seconds, if I don't get a 
NSWorkspaceDidLaunchApplicationNotification, then I assume the App did not 
launch an act appropriately.

The Dock in OS X does the exact same thing AFAIK, the AppIcon just keeps on 
bouncing until it gets the NSWorkspaceDidLaunchApplicationNotification.  After 
a certain amount of time the Dock just assumes the App did not start.

So...  the implementation seems correct, just wait for the proper notification.

-------------------------------------------------------
Date: Sun 06/22/2003 at 11:56       By: stefanu
I disagree and I would like to open this to discussion.

Ask a question: Why would you like to launch an application in most of the 
cases?

1. Just to launch it and let it be launched ('just for fun')
or
2. Launch it and do something with it

I think that 2. is the case. Then I think if OS/X behaves same as gnustep at 
this time, then the behaviour is wrong. This method should wait until I can 
communicate with the application, that means, the method should wait until the 
application is successfully lanched.


-------------------------------------------------------
Date: Sun 06/22/2003 at 11:07       By: CaS
I've just spent a couple of hours checking this.
The current GNUstep behavior appears to be the same as the MacOS-X and OpenStep 
behavior ... ie the interpretation of the word 'launched' means that a process 
was started, not that it is completely initialised/working.  I have updated the 
documentation to be (I hope) clearer about this than the MacOS-X documentation.


CC list is empty


No files currently attached


For detailed info, follow this link:
http://savannah.gnu.org/bugs/?func=detailbug&bug_id=4066&group_id=99

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/