[gdal-dev] FW: Problem with GDAL perl 1.5 bindings

[Ari Jolma]

Ethan Alpert kirjoitti:
> Ari,
>
> I believe reference counts can be manipulated in perlguts. Would it be
> possible to undef any rasterband variables referencing stale/orphaned
> raster bands when the parent dataset is destroyed/dereferenced?

That could be possible. At least it might be possible to invalidate the 
rasterband object somehow. (but see below)

> This would allow for perl to exit with a perl error rather than dumping core
> as well wouldn't it?
>
> Is there any use of a rasterband that has no parent dataset? 
>   

No.

> If not then trying to use it after it's dataset is destroyed and no
> other datasets reference it would be a programming error and the perl
> "Can't call method "WriteRaster" on an undefined value at ./test.pl line
> 10", for instance, would be very appropriate and much more informative
> than a core dump. It would also be a better outcome than producing a
> bogus tiff.IMHO
>   

Yes, but the objects in Geo::GDAL are just references to the C++ or C 
pointers in the wrappers. At the point when the destruction of a Dataset 
is happening we do have such a reference to the (possible) RasterBand 
pointers. So, we could do something with it, but as it is only a pointer 
(as I believe), the real invalidation should happen in GDAL proper, 
which does not have any mechanism for it -- and I guess Frank's not very 
keen on providing one. ;)

Ari

-- 
Prof. Ari Jolma
Geoinformatiikka / Geoinformatics
Teknillinen Korkeakoulu / Helsinki University of Technology
tel: +358 9 451 3886 address: POBox 1200, 02015 TKK, Finland
Email: ari.jolma at tkk.fi URL: http://www.tkk.fi/~jolma