search

CARTAvis / carta

To CARTA users, this repo holds the CARTA release packages. Please use this repo to log bugs and feature requests. These will be triaged by the development team and prioritised as necessary in the development cycles.
19 stars 0 forks source link

Bug in beta: The coordinates are read incorrectly #86

Closed spectram closed 2 years ago

spectram commented 2 years ago

Attaching the extracted png and header file information of the input file (decam g band image). Issue specific to this release, previous versions parsed the same file correctly.

decam_g_tile22 fits-image-2021-09-04-14-46-50 decam_g_tile22.fits-Header.txt

veggiesaurus commented 2 years ago

@spectram I'm wondering if there is a problem with how the header is actually being reported to the CARTA frontend. Would you be able to attach the raw header (from something like fitshdr or fitsheader)?

spectram commented 2 years ago

@spectram I'm wondering if there is a problem with how the header is actually being reported to the CARTA frontend. Would you be able to attach the raw header (from something like fitshdr or fitsheader)?

@veggiesaurus please find attached decamhdr.txt

kswang1029 commented 2 years ago

I found something odd based on the two fits headers: with decam_g_tile22.fits-Header.txt (header via CARTA),

CTYPE1 = DEC--TAN / WCS projection type for this axis CTYPE2 = RA---TAN / WCS projection type for this axis

but with decamhdr.txt (via fitshdr or fitsheader or else):

CTYPE2 = 'DEC--TAN' / WCS projection type for this axis
CTYPE1 = 'RA---TAN' / WCS projection type for this axis

This might be the reason why the rendered grid in CARTA is odd.

kswang1029 commented 2 years ago

@spectram could you confirm that the correct grid should be something like the following? thanks. file-1

kswang1029 commented 2 years ago

@spectram is it possible to get a copy of your image for debugging and bug fix? A small subimage is fine. We would like to see why via CARTA CTYPE1 and CTYPE2 are swapped.

If it is not convenient that's fine. We will try to make a fake one based on the FITS header you provided.

thank you again.

spectram commented 2 years ago

@spectram is it possible to get a copy of your image for debugging and bug fix? A small subimage is fine. We would like to see why via CARTA CTYPE1 and CTYPE2 are swapped.

If it is not convenient that's fine. We will try to make a fake one based on the FITS header you provided.

thank you again.

Hi @kswang1029 The grid looks correct to me. Please find a small subcube. Do let me know if this is sufficient. I can also extract a bigger one if required. Although, I think this opened correctly for me. subcube.zip

kswang1029 commented 2 years ago

@spectram do you think this is still an issue with v3-beta2 release? Following is a screenshot of the subcube example image loaded in v3-beta2. It seems fine. subcube.fits-image-2022-03-17-11-23-43.png

spectram commented 2 years ago

@kswang1029 The issue has been fixed. Sorry for the delay in confirmation.