make_header missing keywords for imaging pipelines

[aliciacanipe]

Hello,

I've done some preliminary tests on the make_header script for NIRCam imaging (level 1, 2, and 3) data. I was trying to add simulated data to the output FITS file and run it through the pipeline, but so far have been unsuccessful. Here is a list of missing keywords I noticed when comparing it to other exposures, though some keywords are known already, e.g., FILTER, READPATT, etc. Here is my list (most are probably not used by the pipeline algorithms):

(PRIMARY extension) FASTAXIS SLOWAXIS FILTER PUPIL PATTTYPE PATT_NUM NUMDTHPT PATTSIZE SUBPXTYP SUBPXNUM SUBPXPNS XOFFSET YOFFSET READPATT TSAMPLE EFFINTTM EFFEXPTM EXPEND EXPMID DURATION

(SCI extension) NAXIS3 NAXIS4 BUNIT BSCALE BZERO RA_V1 DEC_V1 PA_V3 RADESYS ASNPOOL

[stscicrawford]

Thanks for the list -- can you make the data file you used available along with the command you used to run it through the pipeline so we are able to reproduce the results and test it ourselves?

[aliciacanipe]

Sure. My files are here:

/grp/jwst/wit/nircam/canipe/make_header_testing/

In this directory you have baseline__header.fits files, which are the outputs from make_header. NOTES.txt has the commands I used to create them. I also included the simulated image exposure I used to compare the headers. The .dat tables are just lists of headers I noticed were missing.

[bernie-simon]

The purpose of the make_header script is to list those header keywords that are used in a specific pipeline. For this reason the output will not contain all the header keywords typically found in the input image headers for the pipelines. For example, a search of the jwst source code shows that fastaxis and slowaxis are only used in fits_generator/create_dms_data.py, which is not called by any of the pipelines.

What's more interesting to me is your effort to create input images and run them through the pipelines. If you can supply information about the problems you are having and give me the files you are using, I will investigate as well.

[aliciacanipe]

Thanks, Bernie. I assumed most of the headers were not necessary for the specific pipelines, it was just faster to make a table of all the missing headers rather than searching individually through the pipeline source code to see which ones are required. I am working on testing the spec2 pipeline headers, but will also let you all know about progress with running the make_header outputs through the pipeline. I'll put all the files in the same directory mentioned above (/grp/jwst/wit/nircam/canipe/make_header_testing/), with an explanation of the files in NOTES.txt.

[bernie-simon]

"it was just faster to make a table of all the missing headers rather than searching individually through the pipeline source code to see which ones are required."

That's all that make_headers does, search through the source code to find the required headers.