Re: absolute positioning - Mailing list pgsql-jdbc

From Indra Heckenbach
Subject Re: absolute positioning
Date
Msg-id 406138E0.8090501@macnica.com
Whole thread Raw
In response to Re: absolute positioning  (Oliver Jowett <oliver@opencloud.com>)
Responses Re: absolute positioning
Re: absolute positioning
List pgsql-jdbc
Hi Oliver, <br /><br /> Thanks for pointing me to this discussion.  From what I can tell, work must have been done on
thepostgres backend to improve support for cursors.  I have been using cursors with fairly complex queries (joins and
nestedsubqueries and order bys) without any problems.  <br /><br /> I went ahead and patched the JDBC jar from the CVS
headon gborg.  This patch should enable scrollable cursor support, and provide absolute() w/o loading all results into
theJDBC buffer.  Unfortunately, the patch involves changing some core aspects of the driver.  I have not reviewed the
existingcode thoroughly, so I could have introduced problems.  I have also not tested extensively.  However, the patch
worksfor me, and I could use the patched driver with existing applications.  Is there a test suite for the driver?  At
thispoint I hope to get some feedback.<br /><br /> Here is a summary of the patch:<br />     AbstractJdbc1ResultSet<br
/>        - added "buffer_row" and "cursor_row" to track cursor and buffer position.  remove "current_row", and use
theseinstead.<br />         - use a new variable "needsRefresh" to control reload state explicitly<br />         -
convertnext() method to load on demand instead of pre-load.  the former method does not work with cursor positioning,
orwould require a hack to load data twice, and support multiple modes of loading.<br /><br />    
Abstract1Jdbc1Statement<br/>         - to prevent pre-load, I made the while loop over reallyResultSet() occur only if
cursorsare disabled.  I don't think this is needed in this context for cursors, but I'm not certain.<br />         -
addedDECLARE SCROLL CURSOR to handle scrollable case.<br /><br />     AbstractJdbc2ResultSet<br />         - use a new
variable"internalIndex" in AbstractJdbc2ResultSet to track absolute position, to avoid overlapping repositioning.<br />
       - implement absolute() as FETCH ABSOLUTE.  relative() uses this code, so it should be supported as well.<br
/><br/> thanks,<br /> Indra<br /><br /> Oliver Jowett wrote:<br /><blockquote cite="mid405E5CDF.4000308@opencloud.com"
type="cite">IndraHeckenbach wrote: <br /><blockquote type="cite">Thanks for the info.  So the implementation is
functional,but does so by preventing cursors.  Unfortunately, I am dealing with a large result set, so I need to use
cursors. I started modifying the jdbc driver.  Any other suggestions? <br /></blockquote><br /> There was a discussion
aboutimplementing this a couple of months ago that might be useful to you, starting at: <a
class="moz-txt-link-freetext"
href="http://archives.postgresql.org/pgsql-jdbc/2004-01/msg00288.php">http://archives.postgresql.org/pgsql-jdbc/2004-01/msg00288.php</a><br
/><br/> -O <br /><br /></blockquote><br /> 

pgsql-jdbc by date:

Previous
From: dinakar
Date:
Subject: Re: Postmaster not dropping connections on tomcat restart.
Next
From: Oliver Jowett
Date:
Subject: Re: absolute positioning