Bug 427832 - Spurious org.eclipse.jdt.annotation dependecy
Summary: Spurious org.eclipse.jdt.annotation dependecy
Status: RESOLVED WORKSFORME
Alias: None
Product: JDT
Classification: Eclipse Project
Component: UI (show other bugs)
Version: 4.4   Edit
Hardware: PC Windows NT
: P3 normal (vote)
Target Milestone: ---   Edit
Assignee: JDT-UI-Inbox CLA
QA Contact:
URL:
Whiteboard:
Keywords: needinfo
Depends on:
Blocks:
 
Reported: 2014-02-10 13:39 EST by Ed Willink CLA
Modified: 2014-04-11 07:19 EDT (History)
2 users (show)

See Also:

Attachments
Picture with quick fixes (4.17 KB, image/png)
2014-04-11 06:12 EDT, Dani Megert CLA 		no flags Details
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 Ed Willink CLA 2014-02-10 13:39:18 EST 
At least M2 to M5. I keep having to maintain many MANIFEST.MF files to eliminate spurious org.eclipse.jdt.annotation Require-Bundle dependencies.

I think they are being produced by a bad refactoring that should only be adding a source classpath addition, and should only be adding it for one plugin.
Comment 1 Dani Megert CLA 2014-02-11 04:15:02 EST 
Do you have steps to reproduce this?

> At least M2 to M5.
Are you using plain SDK? Did you install the Java 8 BETA?
Comment 2 Ed Willink CLA 2014-02-11 05:13:03 EST 
No precise repro; though I'm keeping my eyes open.

SDK.

No Java 8.
Comment 3 Dani Megert CLA 2014-03-04 04:17:49 EST 
I can't reproduce this on our latest M6 candidate [1]. Ed, is this still a problem for you? If so, can you try to find steps to reproduce it? Otherwise there isn't much we can/will do here.

[1] http://download.eclipse.org/eclipse/downloads/drops4/I20140303-2000/
Comment 4 Ed Willink CLA 2014-03-04 05:52:57 EST 
Could you reproduce prior to M6?

I'm still keeping my open for a repro.
Comment 5 Dani Megert CLA 2014-03-04 10:28:39 EST 
(In reply to Ed Willink from comment #4)
> Could you reproduce prior to M6?

No. And I haven't heard about this bug from any other source so far.
Comment 6 Ed Willink CLA 2014-03-09 11:21:33 EDT 
I just used the QuickFix again on M5.

Studying the offering carefully, the 'correct' fix "Add library" is disconcertingly far down the suggestions aggravated by the bad suggestion ordering bug. It's spelling is also disconcertinly different to "Add ... to imported packages", "Add ... to required bundles".

Suggest: "Add ... to build path" for symmetry and consistent ordering once the ordering bug fixed.

I had not seen "Add library with default annotations to build path" before; it worked perfectly. Even Undo works!

So my problems of multi-MANIFEST change were due to using "Add ... to imported packages". ?? since the quick fix affects an API, might that explain why it affects other MANIFESTs perhaps all bundles sharing the API??
Comment 7 Dani Megert CLA 2014-04-11 05:45:58 EDT 
Tried again to reproduce, but no "luck". Given we didn't get any other bug reports, it looks like some special setup on your side. We can't do anything here without more detailed steps. Sorry.
Comment 8 Ed Willink CLA 2014-04-11 05:51:28 EDT 
(In reply to Ed Willink from comment #6)
> Suggest: "Add ... to build path" for symmetry and consistent ordering once
> the ordering bug fixed.
Comment 9 Dani Megert CLA 2014-04-11 06:12:54 EDT 
Created attachment 241881 [details]
Picture with quick fixes

That's what I see. If you see different Quick Fixes and ordering, then please open separate bugs regarding the order and the naming of the Quick Fixes, with exact steps to reproduce and see the problems.
Comment 10 Ed Willink CLA 2014-04-11 07:19:37 EDT 
(In reply to Dani Megert from comment #9)
> please open separate bugs regarding the order

Bug 400031

> and the naming of the Quick Fixes

Bug 432611