Thread: Binary transfer patches v15

I believe these are now ready for merging to cvs head. The patch set now passes all tests on jdk7u2 + postgres 9.1.0.

Changes from last week:
- Rebased to latest cvs
- Fixed connection leak in jdbc3/DatabaseMetaDataTest.java
    => now it is possible to run the unit tests with max_connections=2 in postgresql.conf
- Fixed ForceBinaryTransfer bug in DatabaseEncodingTest
    => Required a new field to Jdbc2ResultSet to keep the real statement alive
       (Thanks to Oliver for the good analysis on the problem)
- Fixed UpdateableResultTest.testUpdateStreams with binary transfers
    => now ResultSet.refreshRow() informs lower level code that the results are for
       updateable result set and should not use binary transfer

Any comments on what still needs to be done before this is accepted?

-Mikko

Mikko,

Great, I'll try this as soon as possible. I did have a problem with
running out of connections in the test. Did you ever see that ?


Dave Cramer

dave.cramer(at)credativ(dot)ca
http://www.credativ.ca




On Wed, Sep 21, 2011 at 6:22 PM, Mikko Tiihonen
<mikko.tiihonen@nitorcreations.com> wrote:
> I believe these are now ready for merging to cvs head. The patch set now
> passes all tests on jdk7u2 + postgres 9.1.0.
>
> Changes from last week:
> - Rebased to latest cvs
> - Fixed connection leak in jdbc3/DatabaseMetaDataTest.java
>   => now it is possible to run the unit tests with max_connections=2 in
> postgresql.conf
> - Fixed ForceBinaryTransfer bug in DatabaseEncodingTest
>   => Required a new field to Jdbc2ResultSet to keep the real statement alive
>      (Thanks to Oliver for the good analysis on the problem)
> - Fixed UpdateableResultTest.testUpdateStreams with binary transfers
>   => now ResultSet.refreshRow() informs lower level code that the results
> are for
>      updateable result set and should not use binary transfer
>
> Any comments on what still needs to be done before this is accepted?
>
> -Mikko
>
>
>
> --
> Sent via pgsql-jdbc mailing list (pgsql-jdbc@postgresql.org)
> To make changes to your subscription:
> http://www.postgresql.org/mailpref/pgsql-jdbc
>
>

Mikko,

As this is a development branch I have committed as is so that others
can easily build and test it. Thanks for the great work!

Dave Cramer

dave.cramer(at)credativ(dot)ca
http://www.credativ.ca




On Wed, Sep 21, 2011 at 6:22 PM, Mikko Tiihonen
<mikko.tiihonen@nitorcreations.com> wrote:
> I believe these are now ready for merging to cvs head. The patch set now
> passes all tests on jdk7u2 + postgres 9.1.0.
>
> Changes from last week:
> - Rebased to latest cvs
> - Fixed connection leak in jdbc3/DatabaseMetaDataTest.java
>   => now it is possible to run the unit tests with max_connections=2 in
> postgresql.conf
> - Fixed ForceBinaryTransfer bug in DatabaseEncodingTest
>   => Required a new field to Jdbc2ResultSet to keep the real statement alive
>      (Thanks to Oliver for the good analysis on the problem)
> - Fixed UpdateableResultTest.testUpdateStreams with binary transfers
>   => now ResultSet.refreshRow() informs lower level code that the results
> are for
>      updateable result set and should not use binary transfer
>
> Any comments on what still needs to be done before this is accepted?
>
> -Mikko
>
>
>
> --
> Sent via pgsql-jdbc mailing list (pgsql-jdbc@postgresql.org)
> To make changes to your subscription:
> http://www.postgresql.org/mailpref/pgsql-jdbc
>
>

On 09/22/2011 03:54 PM, Dave Cramer wrote:
> Mikko,
>
> As this is a development branch I have committed as is so that others
> can easily build and test it. Thanks for the great work!

Hi,

Here's one missing piece of the binary patches - the sending part was accidentally dropped during my quilt patch
managementexercises. 
It passes all unit tests.

And while looking into it I noticed that the previous binary-send patches did not handle correctly the case of pre 8.1
serverwhere the 
parameter index needs to be adjusted before binding the prepared statement parameters.

-Mikko