Using ResultSet positional access instead of name based one
------------------------------------------------------------
Key: GEOT-2251
URL: http://jira.codehaus.org/browse/GEOT-2251
Project: GeoTools
Issue Type: Improvement
Components: data jdbc-ng
Affects Versions: 2.5.2
Reporter: Andrea Aime
Fix For: 2.5.3
Following up from GEOT-2250, the other part of the slowdown was due to use of
names instead of integer positions to access elements of the resultset. Again
the profile showed the driver to lose a significant amount of time looking up
the position from the name on each attribute access.
Changing the code to use position access fixed the slowdown, thought it
introduces some positional dependencies that have to be explained in the
javadocs at the very least:
- the primary key columns are always first
- all of the other columns follow and are specified in the same order as the
feature type that's being built
The SQLDialect already has a decodeGeometryValue using positional access, but
it's final and it's using the RS metadata (ugh, for some database that is
retrieved with a separate call...).
We should deprecate the name based one and have the dialects implement the
position based one instead.
Justin, opinions?
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.codehaus.org/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
------------------------------------------------------------------------------
_______________________________________________
Geotools-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-devel
