[gdal-dev] sqlite dialect and MySQL spatial input

Jukka Rahkonen jukka.rahkonen at mmmtike.fi
Mon Sep 30 23:58:22 PDT 2013


Even Rouault <even.rouault <at> mines-paris.org> writes:

> 
> I've just pushed in trunk a change to make SQLITE dialect available to all
OGR 
> drivers that have a specialized ExecuteSQL() implementation.

Hi,

Is this a correct interpretation:

- The default SQL dialect is "OGR SQL", except for databases when the
default is to use the native SQL dialect of the RDBMS
- The default dialect can be overridden by the -dialect parameter, which can
take values "OGR SQL" or "SQLITE" and after your change this works for all
drivers in a similar way.

If this is right I will have a look at the documentation and see if there is
something to improve. At least I would make a little addition into
http://www.gdal.org/ogr2ogr.html
Now there reads

-dialect dialect:
    SQL dialect. In some cases can be used to use (unoptimized) OGR SQL
instead of the native SQL of an RDBMS by passing OGRSQL. Starting with GDAL
1.10, the "SQLITE" dialect can also be used with any datasource. 

It would be good to give the alternatives and a hint about how to give OGR
SQL so that the space character will not cause trouble as

-dialect dialect:
    SQL dialect. Can take values "OGR SQL" or "SQLITE". 

Should we do something also for the statement "Starting with GDAL 1.10, the
"SQLITE" dialect can also be used with any datasource" which is totally true
with databases only after your commit http://trac.osgeo.org/gdal/changeset/26506

-Jukka Rahkonen- 



More information about the gdal-dev mailing list