Hi Jane,
That's caused by a change I made to kaplibs a year ago!
I'm amazed no one has fallen over it before. I've committed a fix to
the starlink repository.
David
2009/5/27 Jane Buckle <[log in to unmask]>:
> Hi All,
>
> If I use kappa/collapse twice with axis=1 to try and get an average spectrum
> from a small region within a datacube, it falls over on the second collaspe
> command:
>
> collapse in=file'(0~64,0~64,)' out=temp axis=1 estimator=mean
> Collapsing pixel axis 1 from pixel -33 to pixel 30 inclusive...
>
> collapse in=temp out=file_reg1_spec axis=1 estimator=mean
> !! AST: Error in routine AST_GETI.
> ! astGet: The CmpFrame given does not have an attribute called
> "lataxis(1)".
> ! COLLAPSE: Unable to collapse an NDF along one axis.
> ! Application exit status AST__BADAT, bad attribute name
> !
>
> I have tried using ndfcopy(0~64,0~64,)' to extract the region, but this
> leads to similar problems. I am sure I have seen this before, but thought
> that it had been solved in the latest version of the software, which I am
> now using.
>
> What is the best method of extracting small regions from a datacube? Should
> ndfcopy and collapse work when used in this way?
>
> cheers,
>
> Jane
>
>
> --
>
> Dr Jane V. Buckle
> Astrophysics Group, Cavendish Lab, J J Thomson Avenue, Cambridge CB3 0HE
> Tel: +44 (0)1223 337298
>
|