[releng] PSF issues

[eclipse-ocl-bot]

| --- | --- | | Bugzilla Link | 296182 | | Status | CLOSED FIXED | | Importance | P3 normal | | Reported | Nov 25, 2009 15:09 EDT | | Modified | May 27, 2011 02:48 EDT | | Version | 3.0.0 | | Reporter | Ed Willink |

Description

The project set url meta-data has not been set.

The PSFs reference pserver:dev.eclipse.org rather than pserver:anonymous@dev.eclipse.org.

The PSFs do not include the .edit plugins.

The PSFs presumably needs migrating to the new releng location.

[eclipse-ocl-bot]

By Ed Willink on Nov 25, 2009 16:22

EMF puts its psf in a psf project adjacent to plugins, features, examples etc.

I think this is a better idea than within releng, since developers want psfs but they don't often want releng.

[eclipse-ocl-bot]

By Alexander Igdalov on Dec 02, 2009 09:58

(In reply to comment #1)

EMF puts its psf in a psf project adjacent to plugins, features, examples etc.

I think this is a better idea than within releng, since developers want psfs but they don't often want releng.

Due to bug 294842 there must be http access to the psf. What about putting it to http://www.eclipse.org/modeling/mdt/ocl/project-info/ocl.psf ? Physically it will be in the same folder as our project plan. I think you should raise a bug to the webmaster to have access to that repository path.

[eclipse-ocl-bot]

By Ed Willink on Dec 03, 2009 15:35

The email advertising the PSF URL showed an example from the Rixch Ajax Platform. This has the PSF at: http://eclipse.org/rap/resources/rap.psf

so maybe we should be at http://eclipse.org/modeling/mdt/ocl/resources/ocl.psf

[eclipse-ocl-bot]

By Ed Willink on Dec 05, 2009 13:01

http://eclipse.org/modeling/mdt/ocl/resources/ocl.psf created and referenced from project page. It provides

LPG 1.1\ Basic OCL\ OCL Edit\ OCL Test\ OCL Example Interpreter

no doc, features or other examples

[eclipse-ocl-bot]

By Adolfo Sanchez-Barbudo Herrera on Dec 23, 2009 11:40

Ed, Alex,

How may I access to that file to update it with the new capabilities plugin ?

[eclipse-ocl-bot]

By Ed Willink on Dec 23, 2009 12:15

A new plugin should be added to:

a) some feature hierarchically referenced from the master feature

b) /cvsroot/modeling org.eclipse.mdt/org.eclipse.ocl/releng/maps/ocl.map

(copy a similar line). Order seems irrelevant.

c) /cvsroot/org.eclipse www/modeling/mdt/ocl/resources/ocl.psf

(copy a similar line). Dependencies first is slightly better for clients since concurrent builds work first time.

NB c) is a different repository, which you should be able to access as asanchez as usual.

When you're done you can go to:

https://build.eclipse.org/hudson/job/cbi-mdt-ocl-3.0/

login as asanchez and click on Build Now.

This works in a new private area so you cannot do any damage.

My brief experience is that one build in 10 fails spuriously, so if baffled try again.

---

For completeness the other OCL file of interest is

/cvsroot/callisto /org.eclipse.helios.build/mdt-ocl-3-0.build

which requires a separate request for authorized access.

[eclipse-ocl-bot]

By Adolfo Sanchez-Barbudo Herrera on Dec 23, 2009 12:28

c) /cvsroot/org.eclipse www/modeling/mdt/ocl/resources/ocl.psf

(copy a similar line). Dependencies first is slightly better for clients since concurrent builds work first time.

NB c) is a different repository, which you should be able to access as asanchez as usual.

I think this (c) is I was looking for. Thank you very much, I'll try it.

On the other hand, I think that OCL capabilities plugin is not intended to be included by our specific OCL build, hence, any OCL feature. I think that the global Helios build, have its own feature to handle all the plugins which manage the capabilities (what does the modified files of the attached patch in Bug 273680). So, we only have to keep the org.eclipse.ocl.capabilities in the CVS HEAD, and that should suffice.

Thank you very much for the info. I think it will be helpful in the next future.

Cheers,\ Adolfo.

[eclipse-ocl-bot]

By Ed Willink on May 27, 2011 02:48

Closing after over 18 months in resolved state.