|
Hi Ed,
The difference is that N-builds fetch the latest CVS
state while I-builds use tags specified in the ocl.map file. What we are
observing now is all related to incorrect tagging or incorrect tag referencing.
I hope I will soon have a stable I-build assembled.
Regards,
- Alex.
Hi Alex
Great the N works, and the I now at least
builds.
Where was the ocl20 reference?
The I tests fail. What is
different about the I and the N. Since N passes and my local fetch passes, I'm
inclined to think that the I build is incorrectly specifying a JUnit 3
context/feature.
Ed
Alexander Igdalov wrote:
Ed,
In short, I don't know why I am getting the message about
ocl20a. I am now investigating this.
As regards the tests, I suppose they failed because the
dependencies on the platform were those of M3. I have just updated
them.
I will now run another N build so that we get clean
results.
Regards,
- Alex.
Hi Alex
The CVS-triggered job on https://build.eclipse.org/hudson/job/cbi-mdt-ocl-3.0/72/
built successsfully but silently failed to run the tests, so
failed.
Your jobs on https://build.eclipse.org/hudson/job/cbi-mdt-ocl-3.0-integration/
keep failing becuase you haven't removed the reference to
org.eclipse.ocl20a.all. Is it in the job configuration that I can't
see?
Could the silent test failures be caused by a failure to import
JUnit4 into the build?
A fresh local install of Platform, EMF, UML2 M4,
with OCL from CVS runs fine, so the [4.5.0 5.0.0) tests range would seem ok
(On Windows with Sun Java 5).
Ed
Ed Willink
wrote:
Hi
Alex
Ready to go.
With the common tests plugin now, the
version is exported to the others, so if the test fails. It's just the one
manifest that needs changing.
Ed
On
16/12/2009 07:39, Alexander Igdalov wrote:
Hi Ed,
I see. Please ping me when you are
finished.
Regards,
- Alex.
Hi Alex
I've just checked our JUnit ranges.
They're going to fail. See https://bugs.eclipse.org/bugs/show_bug.cgi?id=297925.
I'll
commit a change in about 10 minutes that supports [4.5.0 5.0.0) preserving
Galileo functionality. If this fails then we need to go to
[4.7.0).
Regards
Ed
On 16/12/2009 07:30, Alexander Igdalov wrote:
Hi All,
I am very sorry for having confused the
dates. I am starting to build OCL 3.0.0 M4 right
away.
Sorry for the
inconvenience,
- Alex.
Alex,
Please respond.
Thanks,
Kenn
---------- Forwarded message ---------- From:
Anthony Hunter <anthonyh@xxxxxxxxxx>Date: Tue,
Dec 15, 2009 at 5:41 PM Subject: Re: [mdt.dev] Some reminders about
Helios M4 To: MDT developers < mdt.dev@xxxxxxxxxxx>
Hi Team,
Are we going to publish an OCL 3.0.0
M4?
Cheers...
Anthony
--
Anthony Hunter mailto:anthonyh@xxxxxxxxxx
Software
Development Manager
IBM Rational Software: Aurora / Modeling Tools
Phone: 613-270-4613
 Kenn Hussey ---2009/12/13 10:53:36 PM---Team, Please see
the email below, in particular #2. If your project is planning to
participate in the Helios release train, you
Team,Please see the email below, in particular #2. If your project is
planning to participate in the Helios release train, you must declare
this by updating the metadata for your project in the Portal. Please see
http://www.eclipse.org/helios/planning/EclipseSimultaneousRelease.php for the requirements that your project must meet in order to be
on the train.Cheers,Kenn---------- Forwarded message
----------
From: David M
Williams <david_williams@xxxxxxxxxx>
Date: Fri, Dec 11, 2009 at 12:45 PM
Subject:
[cross-project-issues-dev] Some reminders about Helios M4
To:
cross-project-issues-dev@xxxxxxxxxxx
1. its next Friday, 12/18!
2. That's the deadline for expressing intent to
participate in Helios, which is formally done by marking the
simultaneousrelease flag in Eclipse Portal (there's a Helios field to
edit to "1"). The list is currently (understandably) short, so
I'll keep my eye on it, and send more nag notes if expected teams don't
mark soon.
modeling.amalgam
modeling.emf.compare
modeling.emf.eef
modeling.m2t.acceleo rt.jetty rt.swordfish stp.sca technology.egit
technology.subversive tools.ptp webtools
3. I suggest we
not "announce" our individual Project milestones until the Helios
M4 date, 12/18. Of course, we need to announce on mailing lists, etc.,
for our committers and adopters ... I just mean the general news items,
forums, and similar, targeted to general end-users asking for early
testing. That is what we have to do for the final release, and I
suggest we get in the habit of working "simultaneously" now, even during
milestones. Not a huge problem, but I do recall some issue in the past
where there were complaints or questions, because someone tried to
update part of their environment, but couldn't, since some other part
had not release an M4 version yet. Ends up being frustrating and making
us appear unorganized. (And we all know that's not true :)
Thanks,
_______________________________________________
cross-project-issues-dev
mailing list
cross-project-issues-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/cross-project-issues-dev
_______________________________________________
mdt.dev
mailing list
mdt.dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/mdt.dev
_______________________________________________ mdt.dev
mailing list mdt.dev@xxxxxxxxxxxhttps://dev.eclipse.org/mailman/listinfo/mdt.dev
_______________________________________________
mdt-ocl.dev mailing list
mdt-ocl.dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/mdt-ocl.dev
No virus found in this incoming message.
Checked by AVG - www.avg.com
Version: 9.0.716 / Virus Database: 270.14.108/2566 - Release Date: 12/15/09 07:52:00
_______________________________________________
mdt-ocl.dev mailing list
mdt-ocl.dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/mdt-ocl.dev
_______________________________________________
mdt-ocl.dev mailing list
mdt-ocl.dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/mdt-ocl.dev
|
