Some Eclipse Foundation services are deprecated, or will be soon. Please ensure you've read this important communication.
Bug 343147 - [ActivityDiagram] Activity diagram plugin does not limit its palette contribution & links to unresolved icons
Summary: [ActivityDiagram] Activity diagram plugin does not limit its palette contribu...
Status: CLOSED FIXED
Alias: None
Product: Papyrus
Classification: Modeling
Component: Core (show other bugs)
Version: unspecified   Edit
Hardware: All All
: P3 critical (vote)
Target Milestone: ---   Edit
Assignee: Vincent Hémery CLA
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2011-04-18 08:46 EDT by Gonzague Reydet CLA
Modified: 2013-05-30 05:13 EDT (History)
6 users (show)

See Also:


Attachments
A patch to fix this issue (633 bytes, patch)
2011-04-18 08:47 EDT, Gonzague Reydet CLA
no flags Details | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Gonzague Reydet CLA 2011-04-18 08:46:10 EDT
Build Identifier: 0.7.3

The activity diagram plug-in makes two paletteProvider contributions. One of these contributions does not limit its scope to the activity diagram editor. This causes side effects on all other GMF based diagram editors.

Moreover some local icons are not resolved in this paletteProvider...

Reproducible: Always
Comment 1 Gonzague Reydet CLA 2011-04-18 08:47:16 EDT
Created attachment 193485 [details]
A patch to fix this issue
Comment 2 Vincent Hémery CLA 2011-04-21 08:55:30 EDT
Hello, we can not reproduce the bug

"This causes side effects on all other GMF based diagram editors."

having no adequate other GMF editor in our environment.

We will make the correction anyway, but can you tell us which editor you have encountered the bug with and the exact steps to reproduce it ?

If, after that, we still can not reproduce (environment too complex to reproduce for us), maybe I will ask you to test our fix to make sure the bug is not reproduced.

Thanks in advance.
Comment 3 Vincent Hémery CLA 2011-04-21 10:08:08 EDT
The part about images has been corrected.

The editor id in the paletteProvider can not be easily added because these extensions are generated.
Hence, the problem is not limited to the activity diagram, but actually to the Papyrus diagram generation.

Could you try and tell us if you can reproduce the same problem with Class diagram for example ? (and how you reproduce it)
Comment 4 Gonzague Reydet CLA 2011-04-26 05:02:19 EDT
(In reply to comment #3)
> The part about images has been corrected.
> 
> The editor id in the paletteProvider can not be easily added because these
> extensions are generated.
> Hence, the problem is not limited to the activity diagram, but actually to the
> Papyrus diagram generation.
> 
> Could you try and tell us if you can reproduce the same problem with Class
> diagram for example ? (and how you reproduce it)

I agree with you that the same kind of problem can occur with other Papyrus plug-ins. I think this only depends on the contribution order of the different Papyrus providers.
Actually Remi has already fixed this kind of issue in the common plugin: https://bugs.eclipse.org/bugs/show_bug.cgi?id=341897
Comment 5 Vincent Hémery CLA 2011-04-27 03:05:26 EDT
(In reply to comment #4)
> I agree with you that the same kind of problem can occur with other Papyrus
> plug-ins. I think this only depends on the contribution order of the different
> Papyrus providers.
> Actually Remi has already fixed this kind of issue in the common plugin:
> https://bugs.eclipse.org/bugs/show_bug.cgi?id=341897

The part that Remi has fixed is not generated. Only diagrams plugins are.

If as you suggest, the contribution order has an impact, this means the first contribution only is taken in account, as they must have similar ids. Hence, the bug will still occur as long as there is at least one contribution.

I think we should modify the Papyrus generators to fix this issue. But as I haven't reproduced this issue, that would help me if you can tell me all the steps you have performed to reproduce the bug :
- which other gmf editor you used ?
- which actions you have performed in this editor ?
- which palette tools pollute your editor ?
Comment 6 Mariot Chauvin CLA 2011-04-27 07:53:24 EDT
(In reply to comment #5)
> (In reply to comment #4)
> > I agree with you that the same kind of problem can occur with other Papyrus
> > plug-ins. I think this only depends on the contribution order of the different
> > Papyrus providers.
> > Actually Remi has already fixed this kind of issue in the common plugin:
> > https://bugs.eclipse.org/bugs/show_bug.cgi?id=341897
> 
> The part that Remi has fixed is not generated. Only diagrams plugins are.
> 
> If as you suggest, the contribution order has an impact, this means the first
> contribution only is taken in account, as they must have similar ids. Hence,
> the bug will still occur as long as there is at least one contribution.
> 
> I think we should modify the Papyrus generators to fix this issue. But as I
> haven't reproduced this issue, that would help me if you can tell me all the
> steps you have performed to reproduce the bug :
> - which other gmf editor you used ?

We reproduce with our GMF runtime based editor, to reproduce simply takes another GMF editor (for instance the logic example), and modify its palette provider priority to be lower than this one.

> - which actions you have performed in this editor ?

No actions is necessary.

> - which palette tools pollute your editor ?

You will remove note and note attachments to all editors. This removal should be strictly limited to your editor id ! 
The entries you provide after will not be contributed because there are only predefined (see "defineOnly="true"). 

See http://help.eclipse.org/helios/index.jsp?topic=/org.eclipse.gmf.doc/tutorials/diagram/paletteConfigurationTutorial.html for more information
Comment 7 Camille Letavernier CLA 2012-08-23 06:10:43 EDT
I see this task is still marked as "Critical", but it seems there where no news from some time now.

Is it fixed?
Comment 8 Arthur Daussy CLA 2012-08-28 06:10:41 EDT
We are unable to reproduce this bug at this point.
So we are closing the bug.