error 1 on virtual point layer

[Frank Warmerdam]

Mark Adams wrote:
> What is interesting is this: prior to going back to the earlier 
> mapserver executable, I tried to simply swap out the gdal13.dll for the 
> earlier version. This resulted in the same errors ("[MapServer Error]: 
> msOGRFileNextShape(): SetAttributeFilter() not supported on ExecuteSQL() 
> results."), so no change. It was when I swapped out the 
> mapserver/mapscript executables that I was able to get the virtual layer 
> to work. So does that imply that it's not gdal itself that is causing 
> the problem, but in how mapserver/mapscript is using it? (Apologies if 
> that's an egghead question.)

Mark,

I suspect this is the change that is trigging problems:

  * $Log: mapogr.cpp,v $
  * Revision 1.96  2006/04/08 03:36:48  frank
  * Ensure that an error in GetNextFeature() will be properly translated
  * into a MapServer error in GetNextShape().

This is the OGR interface code from MapServer.  Basically, in the past,
errors were getting ignored and people couldn't figure out what was
going wrong. On the other hand, some errors were not really fatal (such
as yours) and ignoring them works out well.

I think I need to go into the OGR ODBC driver and not have it issue
that driver.  Instead it should actually implement the attribute filtering
as a post processing step.   I believe I did this for one of the other
underlying non-spatial drivers.

If you can indicate the detailed configuration you are using, I might
be able to prepare some fixes for you.  I'm not sure how easy it will
be for you to try and use them though.   Perhaps you will need an updated
build from Assefa.

Best regards,
-- 
---------------------------------------+--------------------------------------
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | President OSGF, http://osgeo.org