[Pljava-dev] pljava on centos 6.5 64bit and postgres 9.? and java ?

As the original creator/author of the code base I'd like to mention that while I have limited time to spend on PL/Java
nowadays, I will help and answer questions about the code if anyone has an interest in improving it. I will also review
and and probably accept patches in the form of pull requests to https://github.com/tada/pljava

Thomas Hallgren

On 2014-11-20 17:36, Hal Hildebrand wrote:
> Well, that hasn't been my experience. I haven't built lately but one of my devs did for our project. I just pointed him at the build and it worked for him. I didn't even have to step in and help him over hurdles.
>
> As to development, it's pretty fricking stable, so there's not a huge passel of work that needs to be constantly done. You can chalk this up to the way java had designed jni. Moving to a different version of Java requires nothing to change.
>
> As to how it works in production, I got to say I haven't had any problems. And I'm running openjpa in my stored procedures. It's an extreme case but it works like a charm.
>
> There's still work to do on supporting the modern jdbc driver and all the Postgres stuff. But that's work that can be done at our leisure. And it's only if you need it.
>
> Anyways, your mileage may vary, of course. But I don't think that one should deduce that this is a useless project just because there's not a whole lotta Dev work going on. Thanks to the jni stability of java, there really isn't all that much that needs to be done.
>
> At least for our project, pl/Java works, and works well.
>
>
>
>> On Nov 19, 2014, at 3:38 PM, Craig Ringer <craig at 2ndquadrant.com> wrote:
>>
>>> On 11/20/2014 05:49 AM, John R Pierce wrote:
>>> I've got a team at another site at $JOB asking me about the state of
>>> pljava on centos 6.x 64bit, postgres 9.whatever and which java to use...
>>>
>>> what IS the state of this project?
>> As far as I can tell from the mailing list and repository activity,
>> essentially abandoned at this point.
>>
>> As someone who likes Java I'd like to see PL/Java be a more viable
>> option than it presently is, despite the mismatch between Java's
>> thread-based heavyweight-startup shared-everything-by-default model and
>> PostgreSQL's lightweight-process-based shared-nothing-by-default model.
>> PL/Java is less useful than it would be on a thread-based RDBMS because
>> you must use remoting to communicate between different PostgreSQL
>> connections within the same PostgreSQL instance, but it remains a
>> powerful tool ... just one that nobody seems to really be doing much
>> with at the moment.
>>
>> PL/Java needs some remediation work in things like the logging
>> integration and its JDBC interface over SPI, so making it usable is not
>> just a matter of updating it for the latest JDK and PostgreSQL.
>>
>> Like so many things I don't think many prospective users are willing to
>> actively contribute to its maintenance, and few people will work on JNI
>> code in their spare time. I have enough trouble getting people to
>> contribute to PgJDBC and finding time to work on that myself.
>>
>> If you're interested in getting PL/Java rolling again it might be worth
>> speaking to someone on
>> http://www.postgresql.org/support/professional_support/
>> about that.
>>
>> Otherwise you could do what everyone else seems to when they see
>> PL/Java's current state and just use an external JDK.
>>
>> --
>> Craig Ringer http://www.2ndQuadrant.com/
>> PostgreSQL Development, 24x7 Support, Training & Services
>> _______________________________________________
>> Pljava-dev mailing list
>> Pljava-dev at lists.pgfoundry.org
>> http://lists.pgfoundry.org/mailman/listinfo/pljava-dev
> _______________________________________________
> Pljava-dev mailing list
> Pljava-dev at lists.pgfoundry.org
> http://lists.pgfoundry.org/mailman/listinfo/pljava-dev