Some Eclipse Foundation services are deprecated, or will be soon. Please ensure you've read this important communication.

Bug 328266

Summary:	[launcher] Open up upper bound on fragments' host requirement
Product:	[Eclipse Project] Equinox	Reporter:	Andrew Niefer <aniefer>
Component:	Launcher	Assignee:	Arun Thondapu <arunkumar.thondapu>
Status:	RESOLVED WONTFIX	QA Contact:
Severity:	normal
Priority:	P3	CC:	tjwatson
Version:	unspecified
Target Milestone:	---
Hardware:	PC
OS:	Linux
Whiteboard:

Description Andrew Niefer

2010-10-20 12:22:01 EDT

The launcher fragments currently specify an upper bound on their Fragment-Host directive.  The coupling between the host and the fragment is very loose.

The only real dependency here is on the package for the JNIBridge and Main classes as well as Main#run(String[]) and the default constructor for Main.

We can at least relax the dependency to [1.0.0, 2.0.0)

Bug 323488 is an example of p2 resolution problems that come from restrictions here.

Comment 1 Thomas Watson

2010-10-20 13:28:37 EDT

+1 

I think this would be a fine thing to do for 3.7.  At runtime the relationship is quite artificial since by the time the framework is up and discovers a fragment to attach to the launcher host we are way past the point of loading and executing the launcher (both java and native portions have been loaded already).

Comment 2 Thomas Watson

2011-04-20 12:01:15 EDT

Andrew, I'm not sure if you want to do this at this stage in 3.7.  Feel free to defer it.

Comment 3 Andrew Niefer

2011-05-05 15:10:28 EDT

Defer to 3.8

Comment 4 Thomas Watson

2011-06-08 11:28:26 EDT

Move all 3.8 bugs to Juno.

Comment 5 Andrew Niefer

2011-10-20 14:43:09 EDT

Moving to inbox

Comment 6 Thomas Watson

2012-04-30 11:36:16 EDT

Don't think we should do that right now.

Comment 7 Thomas Watson

2014-04-16 15:13:40 EDT

Assigning to Arun.  Arun feel free to close as wontfix.  But if you see value in doing this I suggest we do it now for M7.  Otherwise I would close as wontfix.

Comment 8 Thomas Watson

2014-05-20 14:52:00 EDT

(In reply to Thomas Watson from comment #7)
> Assigning to Arun.  Arun feel free to close as wontfix.  But if you see
> value in doing this I suggest we do it now for M7.  Otherwise I would close
> as wontfix.

Closing as wontfix.  This bug has gone on long enough.  If it proves to be painful in the future we can consider fixing again.