Bug #4324
closedptII/module-info/revision.txt not being updated
0%
Description
The bug is that ptII/module-info/revision.txt has not been updated since
August 3. The problem is that the Kepler tree and the Ptolemy tree are
diverging. The fix is to fix the script that is supposed to update
this file.
The enhancement is that if this file is not updated, then email should
be sent to, probably to David, Chad and I.
On 8/17, I wrote:
Hi David,
It looks like ptII/module-info/revision.txt was updated once since
7/23:bash-3.2$ svn log ptII/module-info/revision.txt
------------------------------------------------------------------------
r20164 | welker | 2009-08-03 17:01:43 -0700 (Mon, 03 Aug 2009) | 1 line[build-system] Checking in changes to ptII...
------------------------------------------------------------------------
r20102 | berkley | 2009-07-23 13:48:38 -0700 (Thu, 23 Jul 2009) | 1 linereverting to the change before the modules directory got nuked.
------------------------------------------------------------------------
r20043 | welker | 2009-07-21 00:02:45 -0700 (Tue, 21 Jul 2009) | 1 lineChecking in changes to ptII...
------------------------------------------------------------------------
r20026 | welker | 2009-07-18 00:02:40 -0700 (Sat, 18 Jul 2009) | 1 lineChecking in changes to ptII...
------------------------------------------------------------------------The reason this is an issues is that it means that Kepler is using
an older version of ptII, which will make merging more difficult.I'm not sure how the update of revision.txt is handled automatically.
Could you include a README.txt in that directory that describes the
setup?One solution would be to have the process that is supposed to increment
the revision send email if it decides not to increment the revision._Christopher
David Welker wrote:
Interesting. I will look into it.
I don't think it is related to the modules directory. Probably the
nightly process has been shut down.I am glad you noticed!
David
On Aug 3, 2009, at 4:47 PM, Christopher Brooks wrote:
Hi David,
It looks like trunk/modules/ptII/module-info has not been updated since
the modules directory was accidentally removed.bash-3.2$ svn log module-info
------------------------------------------------------------------------
r20102 | berkley | 2009-07-23 13:48:38 -0700 (Thu, 23 Jul 2009) | 1 linereverting to the change before the modules directory got nuked.
------------------------------------------------------------------------
r20054 | berkley | 2009-07-21 13:26:35 -0700 (Tue, 21 Jul 2009) | 1 lineadded the .classpath.default file to the installer
------------------------------------------------------------------------
r20043 | welker | 2009-07-21 00:02:45 -0700 (Tue, 21 Jul 2009) | 1 lineChecking in changes to ptII...
------------------------------------------------------------------------
r20026 | welker | 2009-07-18 00:02:40 -0700 (Sat, 18 Jul 2009) | 1 lineChecking in changes to ptII...
------------------------------------------------------------------------Is there are problem with ptII that is preventing Kepler from using
the stable (yesterday's) version of ptII?Below is the last update message I received, which corresponds to your
last auto update.Thanks,
_Christopher
-------- Original Message --------
To: kepler-cvs@kepler-project.org
From: welker@ecoinformatics.org
Message-Id: <20090721070248.214EAE34756@ceres.ecoinformatics.org>
Date: Tue, 21 Jul 2009 00:02:47 -0700 (PDT)
X-ASG-Orig-Subj: [kepler-code] r20043 - trunk/modules/ptII/module-info
Subject: [kepler-code] r20043 - trunk/modules/ptII/module-info
Author: welker
Date: 2009-07-21 00:02:45 -0700 (Tue, 21 Jul 2009)
New Revision: 20043Modified:
trunk/modules/ptII/module-info/revision.txt
Log:
Checking in changes to ptII...Modified: trunk/modules/ptII/module-info/revision.txt ===================================================================
--- trunk/modules/ptII/module-info/revision.txt 2009-07-21
03:11:57 UTC (rev 20042)
+++ trunk/modules/ptII/module-info/revision.txt 2009-07-21
07:02:45 UTC (rev 20043)@ -1 +1
@
-55097
+55143--
Christopher Brooks (cxh at eecs berkeley edu) University of California
CHESS Executive Director US Mail: 337 Cory Hall
Programmer/Analyst CHESS/Ptolemy/Trust Berkeley, CA 94720-1774
ph: 510.643.9841 fax:510.642.2718 (Office: 545Q Cory)
home: (F-Tu) 707.665.0131 (W-F) 510.655.5480
Related issues
Updated by Chad Berkley over 15 years ago
The revision.txt file needs to be updated each night if the ptII build succeeds. I was under the impression that this was running as a cron job on a server. This is def. something that should be automated not done manually. If the revision.txt file gets too out of date, forking will ensue between ptII and kepler.
Updated by Chad Berkley over 15 years ago
I created a script that lives on code.kepler-project.org (ceres) to do this. It is set to run in a cron job every day at 5am PDT. I will monitor it to make sure it keeps running.
Updated by David Welker over 15 years ago
Excellent. For whatever reason, the cron job on Pantara was not running. Hopefully this works.
Updated by Christopher Brooks about 15 years ago
revisions.txt has not been updated since 9/17.
bash-3.2$ svn log ptII/module-info/revision.txt
svn log ptII/module-info/revision.txt
------------------------------------------------------------------------
r20713 | welker | 2009-09-17 07:55:40 -0700 (Thu, 17 Sep 2009) | 1 line
[build-system] Checking in changes to ptII...
------------------------------------------------------------------------
r20691 | berkley | 2009-09-16 10:58:45 -0700 (Wed, 16 Sep 2009) | 1 line
[build-system] Checking in changes to ptII...
We run the real risk of forking the ptII tree and having a hard time merging.
Updated by Chad Berkley about 15 years ago
My cron job has, for some reason, decided not to find ant. I'm working with Nick on figuring this out.
Updated by Chad Berkley about 15 years ago
Working again. I'll monitor this to make sure it keeps working.
Updated by Christopher Brooks over 14 years ago
ptolemy/revisions.txt is not being automatically updated so I'm reopening this.
From the svn logs, it looks like the last time it was updated by
the build system was 2/04/10.
I'm marking this as P3/Normal because we run the risk of Kepler
using an out of sync version of ptII and thus forking ptII.
Updated by Christopher Brooks over 14 years ago
It is important that Kepler developers use the most recent Ptolemy sources
so I'm marking this as targeting Kepler 2.1.0. As of today, 7/6/2010,
kepler/ptolemy/module-info/revision.txt has not been updated since 6/7/2010.
Ptolemy is now at revision 58488, Kepler developers are using 57948. Obviously,
not each revision is a bug fix that is relevant to Kepler.
I'm estimating that it will take 6 hours to get the nightly build to
update ptolemy/module-info/revisions.txt file because it will take
time over several days to keep on top of this.
Updated by Derik Barseghian over 14 years ago
I just tried to manually move kepler trunk up to ptolemy head, but I can't compile:
[compile] /Users/derik/dev2/kepler.modules.sensorview/ptolemy/src/ptolemy/data/ontologies/InfiniteConcept.java:50: cannot find symbol
[compile] symbol : constructor Concept(ptolemy.kernel.util.NamedObj,java.lang.String)
[compile] location: class ptolemy.data.ontologies.Concept
[compile] super(finiteConcept.getContainer(), finiteConcept.getName());
[compile] ^
[compile] /Users/derik/dev2/kepler.modules.sensorview/ptolemy/src/ptolemy/data/ontologies/lattice/adapters/monotonicityAnalysis/MonotonicityCounterexamples.java:97: method does not override a method from its superclass
[compile] @Override
[compile] ^
[compile] /Users/derik/dev2/kepler.modules.sensorview/ptolemy/src/ptolemy/data/ontologies/lattice/adapters/monotonicityAnalysis/MonotonicityCounterexamples.java:105: method does not override a method from its superclass
[compile] @Override
[compile] ^
[compile] /Users/derik/dev2/kepler.modules.sensorview/ptolemy/src/ptolemy/data/ontologies/lattice/adapters/monotonicityAnalysis/MonotonicityCounterexamples.java:114: method does not override a method from its superclass
[compile] @Override
[compile] ^
r58807 is the most recent version I can compile Kepler with, but on import to eclipse I get 114 errors, to do w/ classes in:
/ptolemy/src/ptolemy/domains/jogl/
This look like because of problems with these missing imports:
import javax.media.opengl.*
com.sun.opengl.*
I see we're already excluding domains/gr in PtolemyClasspathStart. Is domains/jogl something we want to exclude too?
Updated by Christopher Brooks over 14 years ago
I think the @override warnings are from using Java 1.5.
Adding jogl to the excludes seems reasonable.
data/ontologies could also be added to the excludes.
Updated by jianwu jianwu over 14 years ago
If excluding those from building system, please make sure it's also updated for eclipse building. To update it, just modify the PtolemyClasspathStart file at build-area/resources/eclipse.
We probably also need to update corresponding files at build-area/resources/netbeans and build-area/resources/idea.
Updated by Derik Barseghian over 14 years ago
Trunks now using ptolemy r58807, and I added the jogl exclude for eclipse.
idea's PtolemyModuleStart already excludes domains/jogl and I couldn't find any exclude section for netbeans.
Updated by Sean Riddle almost 14 years ago
A cron job has been started on lore.genomecenter.ucdavis.edu to run 'ant validate-ptolemy' every midnight. The revision.txt commits should be a few minutes after, by SVN user 'riddle'. There were some issues with making the ant target send an email when the command fails. Where to place code that is executed in response to an error at any point along the pipeline (updating modules, compilation, committing revision.txt, etc.) and taking uniform action is not clear.
The code to send the email is, however, already written, so if anyone wants to investigate ant and figure out how to process the different types of errors, they are welcome to.
Since this cron job approach has had problems before, logs are maintained and rotated with logrotate. The logs are located at /home/sean/ptolemy-validation/ (on lore), and can be viewed by any user in the sudo group (Sean Riddle, Daniel Zinn, Sven Koehler, at the moment).
Updated by Sean Riddle almost 14 years ago
I also added the same information as above to the Kepler Project wiki: https://kepler-project.org/developers/reference/validate-ptolemy-cron-job