Thank you for your contribution to Astropy! 🌌 This checklist is meant to remind the package maintainers who will review this pull request of some common things to look for.

• Do the proposed changes actually accomplish desired goals?
• Do the proposed changes follow the Astropy coding guidelines?
• Are tests added/updated as required? If so, do they follow the Astropy testing guidelines?
• Are docs added/updated as required? If so, do they follow the Astropy documentation guidelines?
• Is rebase and/or squash necessary? If so, please provide the author with appropriate instructions. Also see instructions for rebase and squash.
• Did the CI pass? If no, are the failures related? If you need to run daily and weekly cron jobs as part of the PR, please apply the "Extra CI" label. Codestyle issues can be fixed by the bot.
• Is a change log needed? If yes, did the change log check pass? If no, add the "no-changelog-entry-needed" label. If this is a manual backport, use the "skip-changelog-checks" label unless special changelog handling is necessary.
• Is this a big PR that makes a "What's new?" entry worthwhile and if so, is (1) a "what's new" entry included in this PR and (2) the "whatsnew-needed" label applied?
• Is a milestone set? Milestone must be set but we cannot check for it on Actions; do not let the green checkmark fool you.
• At the time of adding the milestone, if the milestone set requires a backport to release branch(es), apply the appropriate "backport-X.Y.x" label(s) before merge.

👋 Thank you for your draft pull request! Do you know that you can use [ci skip] or [skip ci] in your commit messages to skip running continuous integration tests until you are ready?

The FITS compression code is being heavily refactored in #15474, so let's ping @astrofrog :)

Oh, I wasn't aware, thanks for noticing ! if it helps, maybe we should switch this one to draft mode for now ?

The bug this PR is fixing was reported back in 5.x. Looks like this PR is good to backport, but #15474 is not till v7 at the earliest, so if Tom R is okay with this patch, we should accept it just so we do not have to wait till v7 for a fix?

I rebased to refresh CI. Let's actually ping @astrofrog while I'm here.

@astrofrog does this need to wait for #15474 ?

Ideally yes otherwise it will not be a trivial conflict to resolve.

you got it

@astrofrog - do you think those tests would be useful ?

I guess we might as well, though not sure if we need it to be that extensive - maybe we can just check the compression options in the compressed header that was written out? (Or maybe it wouldn't be enough to illustrate the problem?)

would the following assertion match your suggestion @astrofrog ? It appears to pass on main and fail on v6.1.4, which I think indicates that it captures the regression ?

assert hdul[1].header["BITPIX"] == hdul2[1].header["BITPIX"]

Yes partially but I am thinking more along the lines of checking e.g. ZVAL?/ZNAME?/ZCMPTYPE/ZQUANTIZ which are the actual options you are trying to set.

None of these options seemed to appear in the headers I considered while looking into this. Maybe I'm not looking in the right place ?

@neutrinoceros - are you opening the file disabling the compression in order to access the header? (these will just be in the compressed header):

In [1]: import numpy as np

In [2]: from astropy.io import fits

In [3]: hdu  = fits.CompImageHDU(np.random.random((128, 128)))

In [5]: hdu.writeto('test.fits')

In [6]: hdu2 = fits.open('test.fits', disable_image_compression=True)[1]

In [7]: hdu2.header
Out[7]: 
XTENSION= 'BINTABLE'           / binary table extension                         
BITPIX  =                    8 / array data type                                
NAXIS   =                    2 / number of array dimensions                     
NAXIS1  =                   32 / width of table in bytes                        
NAXIS2  =                  128 / number of rows in table                        
PCOUNT  =                14585 / number of group parameters                     
GCOUNT  =                    1 / number of groups                               
TFIELDS =                    4 / number of fields in each row                   
TTYPE1  = 'COMPRESSED_DATA'    / label for field 1                              
TFORM1  = '1PB(122)'           / data format of field: variable length array    
TTYPE2  = 'GZIP_COMPRESSED_DATA' / label for field 2                            
TFORM2  = '1PB(0)  '           / data format of field: variable length array    
TTYPE3  = 'ZSCALE  '           / label for field 3                              
TFORM3  = '1D      '           / data format of field: 8-byte DOUBLE            
TTYPE4  = 'ZZERO   '           / label for field 4                              
TFORM4  = '1D      '           / data format of field: 8-byte DOUBLE            
ZIMAGE  =                    T / extension contains compressed image            
ZTENSION= 'IMAGE   '           / Image extension                                
ZBITPIX =                  -64 / data type of original image                    
ZNAXIS  =                    2 / dimension of original image                    
ZNAXIS1 =                  128 / length of original image axis                  
ZNAXIS2 =                  128 / length of original image axis                  
ZPCOUNT =                    0 / number of parameters                           
ZGCOUNT =                    1 / number of groups                               
ZTILE1  =                  128 / size of tiles to be compressed                 
ZTILE2  =                    1 / size of tiles to be compressed                 
ZCMPTYPE= 'RICE_1  '           / compression algorithm                          
ZNAME1  = 'BLOCKSIZE'          / compression block size                         
ZVAL1   =                   32 / pixels per block                               
ZNAME2  = 'BYTEPIX '           / bytes per pixel (1, 2, 4, or 8)                
ZVAL2   =                    4 / bytes per pixel (1, 2, 4, or 8)                
ZNAME3  = 'NOISEBIT'           / floating point quantization level              
ZVAL3   =                 16.0 / floating point quantization level              
ZQUANTIZ= 'NO_DITHER'          / No dithering during quantization               
EXTNAME = 'COMPRESSED_IMAGE'   / name of this binary table extension  

Thanks ! I cleaned up this PR and now it's all about adding my tests

Thanks! I updated the labels and also moved the milestone to match #15474

Ugh the wheel jobs got stuck. Maybe I need to prioritize fixing the labeled event for that job.