Vitali Diatchkov ha scritto: > Performance benchmark is necessary thing. > If to concern things that are new here in comparison to what I have seen in > old OracleDataStore module: > > New Oracle Spatial datastore implementation is based on prepared statements > where geometries are handled as STRUCT objects. The latter takes away lot of > constraints for geometries processing. In all kinds of SQL queries (SELECT, > UPDATE, INSERT ) geometry objects are passed as prepared objects. It is > impossible to pass to Oracle a geometry with 5000 vertices, e.g., encoded > into text-based SQL...
As far as I know even the current Oracle datastore uses prepared statements for geometry insertion/update? Look at the one in 2.3.x or 2.4.x. > > We should not be in hurry with making this stuff as a default one. I am able > to spend time and get this to the shape based on community decisions and > proposals. > > So next step: I will split JDBC and Oracle code into unsupported GeoTools > module. > One of the main issues right now is how to determine CRS of geometry > attribute types from Oracle in a general way. What is kept in Oracle > metadata tables is far from standard WKT CRS represention. Vitali, during the last month I did quite some work on the current Oracle data store to make it work and this was one of the changes I made. It's a little hacky, but that can't be helped I fear, since as you have noticed Oracle does not give a damn about proper standard implementation. Cheers Andrea ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Geotools-devel mailing list Geotools-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/geotools-devel
