.reset not equivalent to new FL session

[Annie Hughes]

Hi Lajos,

I am using Fawlty Language 0.79.52 (macos amd64 m64).

I have code that calls the mrdfits.pro routine (a staple in the NASA astronomy library used for reading FITS files). The call to mrdfits.pro occurs in a subroutine (~3 levels down from the top-level routine called at the FL> command line).

When I run the code from the FL> command line, the second call to mrdfits from within the subroutine fails, i.e. the critical lines are

file=dir_pacs+'transmissionfile_v2.fits'
wave=mrdfits(file,1,h) ;<<- works
trans=mrdfits(file,2,h) ;<< -- fails with following error:

FL>
% Error: PRODUCT: numeric data required: DIMS
% Execution halted at: FXPOSIT 259 /Users/anniehughes/...pathtoAstroLib.../pro/fits/fxposit.pro
% MRDFITS 2650 /Users/anniehughes/...pathtoAstroLib.../pro/fits/mrdfits.pro

‌

(1) if I run the code snippet above directly from the FL> command line, it works.

and

(2) if I run the code, encounter the above error, and then do FL>.reset and relaunch the code, it works

but

(3) If launch FL, immediately do FL>.reset and then run the code, it produces the same error as above.

Any ideas?

Thanks!

Annie

[Annie Hughes]

After a bit more digging, it seems to be related to a bad relationship between the window command and mrdfits

If I create/plot to a graphics window in the parent subroutine, then mrdfits fails.

If not, it works.

[Lajos Foldy]

Thank you for the report. I’ll investigate it and get back to you.

[Annie Hughes]

Hi Lajos,
We are continuing to do tests to understand the behaviour, and are finding different outcomes depending on the version of the IDL library/routine and the version of fawlty… if you need more information, we will try and send a doc in the next day or two. If you think it would be more efficient to discuss ‘live’, we’d be happy to do so.

Thanks again,

Annie

[Lajos Foldy]

Can I download this transmissionfile_v2.fits file from somewhere?

[Lajos Foldy]

Closed.