On Thu, 26 Feb 2004, David Berry wrote:
> I suppose the testing required for a perl interface is likely to be more
> closely related to the testing required for JNIAST than to AST itself.
> Regression testing of AST is by no means as extensive as I would like it
> to be. I have a series of "critically tricky" plots which I print out at
> each release and compare by eye with the previous version. I also have
> an application which converts FrameSets from one encoding to another
> and checks that the results are the same as before. I could really do
> to put some serious time into tightening up these tests, but other things
> are always more pressing...
If you send me that application for testing FrameSet conversion I can see
about perlifying it. I'm mainly looking for reproducible code for testing
the interface but since I don't know enough ast to verify that everything
works I was going to transcode some C to perl.
--
Tim Jenness
JAC software
http://www.jach.hawaii.edu/~timj
|