ImageSolver.js no longer works in PI 01.08.01.1092

[MikeOates]

When I try and solve an image, almost any image, even raw unprocessed, I get the following error:

Code: [Select]

run --execute-mode=auto "C:/Program Files/PixInsight/src/scripts/AdP/ImageSolver.js"

Processing script file: C:/Program Files/PixInsight/src/scripts/AdP/ImageSolver.js
*** Error [147]: C:/Program Files/PixInsight/include/pjsr/Sizer.jsh, line 56: uncaught exception: Invalid WCS coordinates: Unsuported projection DEF
So I don't get the chance to put any positional data in to search on.

But it does work on integrated images, but not on nearly all subs, calibrated, registered etc.

This was working before the upgrade to the latest version: 01.08.01.1092

If you can't reproduce this error, I will get some files uploaded, just ask.

Thanks,

Mike

[Andres.Pozo]

Hi Mike,

I need a file where the script fails.

Thanks for reporting the problem.

[MikeOates]

Andres,

Sorry to keep you waiting, been away for the Easter break.

Here is a file that does not solve. It's a raw fits sub with no processing.

http://endor.uv.es/files/data/public/f2ed06.php?dl=true

Thank you,

Mike

[Andres.Pozo]

Hi Mike,

your image is already solved and has non-standard FITS keywords. If you remove the keywords CTYPE1 and CTYPE2 the image will solve without problems.

These keywords contains the values 'RA---DEF' and 'DEC--DEF' that mean that the image is solved with the "DEF" projection. This projection is not defined in the document "Representations of celestial coordinates in FITS" that I used as a reference for the implementation.

In any case I will modify the script so it can ignore these non-standard keywords.

[pfile]

i meant to report exactly this some time ago but forgot after i figured out what was happening - i think astrometry.net's solver is using these keywords. do you think it should be reported as a bug to dustin lang?

rob

[Andres.Pozo]

i meant to report exactly this some time ago but forgot after i figured out what was happening - i think astrometry.net's solver is using these keywords. do you think it should be reported as a bug to dustin lang?

rob

It is not really a bug. At least I don't consider not following the "standards" as a bug. Specially anything concerning the FITS format which has very loose specifications. The reference documentation that I am using is only a recommendation and I wouldn't mark as a bug not following it.

I have to change ImageSolver so it can work although the image had been previously solved with an incompatible solver.

In any case, I have solved lots of images with Astrometry.net and the results have always been compatible with my scripts. In this case, I think the problem could be in Sequence Generator Pro.

[pfile]

oh - you are right. it's SGP. i had forgotten what i was doing when i hit that bug.

rob

[MikeOates]

Yes I am using SGP, but using Pinpoint within SGP to solve during image taking. I will let the authors know of the issue in case there is a way to solve it (parden the pun!) within SGP.

Mike

[MikeOates]

Andres,

Just wondering if you have managed to change the script to avoid the errors?

Thanks,

Mike

[Andres.Pozo]

Not yet, but I hope to do it this weekend.