Hi Ludovic and Frank, As I understand from discussions with Evan Rouault, Destroy() is an obsolete method in the Python bindings that should not be used. See http://trac.osgeo.org/gdal/wiki/PythonGotchas#CertainobjectscontainaDestroym ethodbutyoushouldneveruseit.
If that understanding is not correct, please let me know. Rather than calling Destroy(), it might be better to set all of the variables to None (i.e. joinlayer, layer, and datasource) to hopefully cause Python to destruct these instances. But I see from Ludovic's latest email that he tried it and it did not work. So it is time for a new idea... In any case, please let me know if there is a scenario in which Destroy() is supposed to be called from the Python bindings, contrary to Evan's advice. Thanks, Jason -----Original Message----- From: gdal-dev-boun...@lists.osgeo.org [mailto:gdal-dev-boun...@lists.osgeo.org] On Behalf Of Frank Warmerdam Sent: Monday, November 22, 2010 9:24 AM To: Ludovic Granjon Cc: gdal-dev@lists.osgeo.org Subject: Re: [gdal-dev] Delete a sqlite database Ludovic Granjon wrote: > Hi all > I try in a python script to delete a sqlite database that I use in an > OGR ExecuteSQL. > Let me give you an example : > If I do : > > >>> from osgeo import * > >>> import os > >>> datasource = ogr.Open("decoupe.shp") > >>> layer = datasource.GetLayer() > >>> joinLayer=datasource.ExecuteSQL("SELECT decoupe.FID as ligne_id, > c.lFid as ligne_id2, c.count as count FROM decoupe LEFT JOIN > 'project.sqlite'.comptage c ON decoupe.FID=c.lFid") > >>> datasource.Destroy() > > It seems that there still a connection to my sqlite database and I can't > delete it. > > When I try on MS Windows (There's no error on linux) : > >>> os.remove('project.sqlite') > > I have : > Traceback (most recent call last): > File "<pyshell#5>", line 1, in <module> > os.remove('project.sqlite') > WindowsError: [Error 13] Le processus ne peut pas accéder au fichier car > ce fichier est utilisé par un autre processus: 'project.sqlite' > > > Do you know a way to do that ? Ludovic, I think newish versions of the Python bindings attempt to keep the datasource alive as long as their are references to layers from it. I would suggest setting layer and joinLayer variables to None before calling destroy on the datasource or even just setting the datasource to None. I'm not positive that this is the issue - perhaps you could let us know if it works. Best regards, -- ---------------------------------------+------------------------------------ -- I set the clouds in motion - turn up | Frank Warmerdam, warmer...@pobox.com light and sound - activate the windows | http://pobox.com/~warmerdam and watch the world go round - Rush | Geospatial Programmer for Rent _______________________________________________ gdal-dev mailing list gdal-dev@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/gdal-dev _______________________________________________ gdal-dev mailing list gdal-dev@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/gdal-dev
