Re: Three weeks left until feature freeze

Kris Jurka wrote:
>
>
> On Thu, 13 Jul 2006, Tom Lane wrote:
>
>> The people who think PL/Java is an essential checklist item
>> undoubtedly also think JDBC is an essential checklist item, but I'm
>> not seeing any groundswell of support for putting JDBC back into
>> core. Instead we expect packagers (like the RPM set) to make JDBC
>> available alongside the core postgres packages. That's how PL/Java
>> ought to be handled, too, IMHO.
>
> The fact that the JDBC driver requires no compilation for anyone on any
> platform is one reason for that. Anyone can visit the website and be
> working within minutes with no understanding of the build environment or
> installation. You drop the provided JAR file in your classpath and you
> are done. The same cannot be said for pl/java. Yes, it would be good if
> there were packages for it, but it's very unlikely that pl/java will be
> able to maintain up to date binary packages for every platform.

The problem I pointed in my previous post, is internal changes
impact to several PostgreSQL extensions and components.
They should be developed, tested and shipped more tightly.

The JDBC driver does connect to the backend via well-defined protocol,
so separated development is not a problem. The protocol means
a contract between frontend and backend.

However, several extensions, such as pl/java, strongly depend on
the backend internal functions and arguments. If they are suddenly
changed, the extension XX couldn't be compiled anymore, and the users
will waste their time.

No one gets win if such impedance mismatch exists.
--
NAGAYASU Satoshi <nagayasus(at)nttdata(dot)co(dot)jp>
Phone: +81-3-3523-8122