Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [orbit-dev] Fw: Ready for de.cau.cs.kieler to be in a build?

Henrik,

Sorry to be the bearer of bad news, but ...

Seeing the Orbit build fail, and looking for the reason, I see that messages such as those below.
These give me pause, and pretty much change my mind about if this is appropriate for Orbit.
I am currently thinking not.

Orbit bundles, besides being from "third parties" are normally low level function that does not depend on other Eclipse projects.

This large set of bundles has so many dependencies on "Modeling" projects, and sounds like it is a "complete system", that I think it better belongs in Modeling. Perhaps proposed as
as a sub-project of gmf or something, if it doesn't fit in to any existing projects.

Missing required plug-in org.eclipse.buckminster.sax_0.0.0.        
Missing required plug-in org.eclipse.acceleo.engine_0.0.0.        
Missing required plug-in org.eclipse.amalgam.discovery.ui_0.0.0.
Missing required plug-in org.eclipse.gmf.runtime.diagram.ui_0.0.0.
Missing required plug-in org.eclipse.gmf.runtime.diagram.ui.render_0.0.0.        
Missing required plug-in org.eclipse.buckminster.sax_0.0.0.        
Missing required plug-in org.eclipse.acceleo.engine_0.0.0.

Sorry this wasn't clearer, sooner. I think we've made one or two exceptions, but its been for "small" things that were "low level" in Eclipse, such as org.eclipse.osgi.resolver (or something ... I forget what exactly).

Thus, I'll comment out your contributions from the Orbit feature, for now, and ask you to discuss the "Modeling project" approach with the Modeling PMC. (Ed Merks CC'd on this note).
Thanks,






From:        David M Williams/Raleigh/IBM@IBMUS
To:        Orbit Developer discussion <orbit-dev@xxxxxxxxxxx>,
Date:        02/06/2013 12:49 PM
Subject:        Re: [orbit-dev] Fw: Ready for de.cau.cs.kieler to be in a build?
Sent by:        orbit-dev-bounces@xxxxxxxxxxx




Actually its the reverse. The feature defines "what to build" (or, what to try and build :) and then once that's known it "looks up" version and location info from the map, so if that "lookup" fails, the build will fail.
Where its fine to be in the map file and if the feature does not say to build it, then that information is simply never looked at.


HTH






From:        
Henrik Rentz-Reichert <hrr@xxxxxxxxx>
To:        
Orbit Developer discussion <orbit-dev@xxxxxxxxxxx>,
Date:        
02/06/2013 11:56 AM
Subject:        
Re: [orbit-dev] Fw: Ready for de.cau.cs.kieler to be in a build?
Sent by:        
orbit-dev-bounces@xxxxxxxxxxx




David,

I have worked over the day to add the KIELER bundles.
I'm almost through, just the IP logs not complete.

I thought I could check in the feature since the map file uses a tag. So I reckoned that the new bundles wouldn't be in the build unless I'd check in the map file as well.

Anyway, thanks for dropping this note.
I think in half an hour from now I'll be done.

-Henrik


Am 06.02.2013 17:40, schrieb David M Williams:

I noticed the
de.cau.cs.kieler*  bundles were added to the feature.xml, but feature project not versioned yet, at least not updated in map file.

So, that makes me wonder if its ready to be built? I want to tag the feature to add org.objectweb.asm 4.0.0 to the build.

I'll go ahead and tag and update map "as is", but if it breaks, will try again just commenting out the "de.cau.cs.kieler*" bundles.

Just wanted to give some "notice" as to what I was doing.

Thanks,





_______________________________________________
orbit-dev mailing list

orbit-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/orbit-dev

_______________________________________________
orbit-dev mailing list
orbit-dev@xxxxxxxxxxx

https://dev.eclipse.org/mailman/listinfo/orbit-dev
_______________________________________________
orbit-dev mailing list
orbit-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/orbit-dev


Back to the top