On Aug 28, 2006, at 10:14 AM, Marc Prud'hommeaux wrote:
If the snapshot deploy directory /x1/www/people.apache.org/repo/
m2-snapshot-repository/org/apache/openjpa/ is available from the
continuum server, I might be able to override the
snapshotRepository to deploy straight to that directly, in which
case we'd have automatic deployment of the snapshot (would would
be very handy). Does that sound plausible to you?
It would be perfect. Unfortunately, it's running in a solaris
zone on a completely different box :(
That's too bad. To anyone's knowledge, has anyone accomplished
automatic nightly deployment of snapshots to the Apache snapshot
repository, or is it always a manual process?
Not to my knowledge for m2. With m1 it was easier as there were no
"index" files that you had to update on every deploy. You could just
rsync and add/update a bunch of jars with no issue.
I'd like to see this solved too.
As an interim solution, continuum itself can be used as a maven2
repository. People would just have to add this to their pom:
<repository>
<id>continuum-m2-snapshot</id>
<name>Continuum Snapshot Repository</name>
<url>http://vmbuild.apache.org/continuum/repository</url>
</repository>
-David
On Aug 24, 2006, at 2:51 PM, David Blevins wrote:
On Aug 24, 2006, at 12:46 PM, Marc Prud'hommeaux wrote:
David-
I added it in (that's the easy part). See http://
vmbuild.apache.org/continuum/servlet/continuum
That machine is terribly slow....
I notice that there is both a "OpenJPA" and a "OpenJPA Project"
build configured in continuum, and "OpenJPA" doesn't seem to
actually be building anything. Was this intended?
In continuum, you only need to add the root pom.xml (parent
pom.xml) and it will set up that project and all the subprojects
(child pom.xmls) for you. You need the parent pom.xml to build
any of the children, so you definitely want that in there.
As an FYI, the names come from each pom.xml file's "<name>"
element as you probably figured out. You may want to adopt the
convention of prefixing "OpenJPA" in all your name elements, such
as "OpenJPA :: JDBC", just so they all show up next to each other
(for example "XML Store" is way down on the page). Whatever you
prefer is fine.
Also as Patrick has said a couple times, you'd like to get a
TeamCity server setup, cool. Just say "when" and I'll yank it
from the continuum setup.
The tricky part with any CI system is the whole question of
"Gee, what credentials do we use to deploy snapshots?"
If the snapshot deploy directory /x1/www/people.apache.org/repo/
m2-snapshot-repository/org/apache/openjpa/ is available from the
continuum server, I might be able to override the
snapshotRepository to deploy straight to that directly, in which
case we'd have automatic deployment of the snapshot (would would
be very handy). Does that sound plausible to you?
It would be perfect. Unfortunately, it's running in a solaris
zone on a completely different box :(
-David
On Aug 23, 2006, at 10:41 PM, David Blevins wrote:
On Aug 23, 2006, at 4:20 PM, Bryan Noll wrote:
So... I saw that David created the following two JIRA's, so I
wanted to follow up on the thread.
http://issues.apache.org/jira/browse/OPENJPA-29
http://issues.apache.org/jira/browse/OPENJPA-30
Regarding OPENJPA-29:
openjpa-0.9.0-full.jar // contains all openjpa code, openjpa-
*.jars merged
openjpa-0.9.0-nodep.jar // contains all openjpa code and all
third party dependency jars
Do you have your descriptions backwards or am I
misunderstanding what you're saying?
I very well could have them backwards. As I said, "There's a
naming convention for this kind of thing, hope I've got it right."
Seems like 'nodep' would mean "no dependencies for this project
are bundled into this jar" to me.
IIUC (which I'm not sure that I do, so please correct me) then...
openjpa-0.9.0-nodep.jar would extract to:
- org/apache/openjpa/*/**.class
- META-INF/* ............ (we still need to resolve the 'merge
multiple files of the same name and locations (eg..
ProductDerivations) problem for this)
openjpa-0.9.0-full.jar would extract to:
- org/apache/openjpa/*/**.class
- thirdPartyJars.jar (many of these)
- META-INF/* .......... (same issue as above)
The classes from the third party jars would also have been
extracted into the jar.
Would the openjpa-*.jar files be there instead of the unpacked
classes?
Classes.
Should the names actually be openjpa-nodep-0.9.0-SNAPSHOT.jar
and openjpa-full-0.9.0-SNAPSHOT.jar?
That'd be possible. You'd have to have one maven module for
each jar and poke at the assembly plugin setup to not use the
default name of <artifactId>-<version>-
<assemblyName>.<archiveType>. The maven guys may know a better
way.
Regarding OPENJPA-30:
- I saw that Marc got the newly named SNAPSHOT jars up to a
repo.... good stuff.
Yep. Thanks again, Marc! Very appreciated!
- The next step would have one jar somewhere like http://
people.apache.org/repo/m2-snapshot-repository/org/apache/
openjpa/openjpa/0.9.0-SNAPSHOT/ for people to depend on,
instead of many, correct?
We (Geronimo and OpenEJB) use maven 2 which has transitive
dependencies, so it's not critical for us. But it'd be real
handy for anyone using maven 1 or ant. Or anyone really who
just wants to throw that jar in their webapp and say "go!"
- David... if you catch this... I would be glad to help
(translation... observe and learn) you get the thing publishing
nightly from the apache continuum server you spoke of.
I added it in (that's the easy part). See http://
vmbuild.apache.org/continuum/servlet/continuum
That machine is terribly slow....
The tricky part with any CI system is the whole question of
"Gee, what credentials do we use to deploy snapshots?"
To skirt the issue of what user to "push" jars as, I've
typically setup an rsync on the destination box and "pulled" the
jars down. That worked fine with maven 1, but maven 2 has these
wonderful index files in the repo that make that impossible to
my knowledge. Maybe Brett can confirm.
-David
Thanks...
Bryan
David Blevins wrote:
On Aug 15, 2006, at 7:42 AM, Craig L Russell wrote:
On Aug 15, 2006, at 6:38 AM, Patrick Linskey wrote:
Check out the Assembly plugin and it's predefined
"jar-with-dependencies" descriptor[1]. I'm pretty sure that
most people
using Maven will just have a runtime dependency on the core
OpenJPA
library and be done with it, but it's always nice for Ant
users.
Exactly -- mvn users will use the dependencies, but people
who want to
download and use the distribution directly may not want a
billion little
jars.
Yes, my experience says it's ok to have 10 dependencies on
e.g. commons-logger, antlr, etc. but having 10 dependencies
on openjpa-kernel4, openjpa-kernel5, openjpa-api, openjpa-
blah is annoying.
So if possible, I'd also like to have a jar with all the
stuff in it.
That'd be nice. Some jar that contains all the openjpa-*.jar
files is perfect for my needs. The good news is that the
maven-assembly-plugin allows you to define several assembly
xml files and build multiple archives containing whatever you
like. So if people want, we could have:
openjpa-0.9.0-full.jar // contains all openjpa code, openjpa-
*.jars merged
openjpa-0.9.0-nodep.jar // contains all openjpa code and all
third party dependency jars
There's a naming convention for this kind of thing, hope I've
got it right.
-David
Craig
-Patrick
_______________________________________________________________
________
Notice: This email message, together with any attachments,
may contain
information of BEA Systems, Inc., its subsidiaries and
affiliated
entities, that may be confidential, proprietary,
copyrighted and/or
legally privileged, and is intended solely for the use of
the individual
or entity named in this message. If you are not the intended
recipient,
and have received this message in error, please immediately
return this
by email and then delete it.
Craig Russell
Architect, Sun Java Enterprise System http://java.sun.com/
products/jdo
408 276-5638 mailto:[EMAIL PROTECTED]
P.S. A good JDO? O, Gasp!
