[postgis-devel] does PostGIS CHECK_FOR_INTERRUPTS? Should it?

[Nicklas Avén]

ST_MemUnion seems to be more mortal.

Maybe it is a specific ST_Union problem?

/Nicklas


On Thu, 2011-11-17 at 10:12 +0100, Nicklas Avén wrote:
> Hallo and happy Post GIS day
> 
> Quite often I run into the problem that I can't cancel a query. This
> time I was trying to union to many polygons and I just want to kill it.
> 
> If I understand this thread right:
> http://archives.postgresql.org/pgsql-admin/2009-05/msg00009.php
> 
> we should have CHECK_FOR_INTERRUPTS in the code to handle canceled
> queries.
> 
> Do we have that? 
> In my case with ST_Union, with GEOS involved I have two questions:
> 1) can PostGIS kill a geos task
> 2) when running a huge st_union for instace, is the control handled back
> to PostGIS during the execution or is it just a huge collection or is it
> just a big geometry collection handled over to geos to do the union on?
> 
> 
> /Nicklas
> 
> _______________________________________________
> postgis-devel mailing list
> postgis-devel at postgis.refractions.net
> http://postgis.refractions.net/mailman/listinfo/postgis-devel
>