[gdal-dev] python scripts - create layer problem

[Even Rouault]

Selon Martin Landa <landa.martin at gmail.com>:

> Hi,
>
> 2014-07-11 9:07 GMT+02:00 Martin Landa <landa.martin at gmail.com>:
> > It's random, usually one or two layers are missing in output DB...
>
> I checked also all tables with
>
> psql vfr -c "SELECT * FROM information_schema.tables WHERE
> table_schema = 'public'"
>
> Some tables (one or two) are simply not written for unknown reason.

If you see the CREATE TABLE in the debug log, and it is not created, then it is
really weird. Perhaps check the potential surrounding BEGIN/COMMIT/ROLLBACK just
in case. Might it not be an issue with your PostgreSQL instance ?

Otherwise, if your trunk is >= r27476 (http://trac.osgeo.org/gdal/ticket/5547),
could you try setting OGR_PG_DIFFERED_CREATION=NO as environment variable /
configuration option ?
If that still does not work (if OGR_PG_DIFFERED_CREATION=NO is necessary, then
it would be a bug of trunk), then you could try checking out r27475.

But the randomness issue is quite puzzling.

Even


>
> --
> Martin Landa * http://geo.fsv.cvut.cz/gwiki/Landa
>