[AstroPy] Request for Spectra files (in FITS format)

[Derek Homeier]

Hi Shailesh,

>> I am working on the Google summer of code project for reading/writing spectra. To successfully cover all use cases, I need your help to provide me Spectra files in FITS format. At this stage we are most interested files that contain 1-dimensional extracted spectra (either long slit, multi-object or echelle) from optical or infrared spectrographs. But feel free to send any spectra that you would like to able to read easily with Astropy.
>> 
> thanks for your work on this! Is this intended to become part of the specutils package,
> or going directly into Astropy “core”?
> I assume you’d be considering model spectra as well; I have one specific case however
> in the high-resolution spectra here
> http://phoenix.astro.physik.uni-goettingen.de/?page_id=10
> which require the wavelengths to be read from a separate file (I suppose to deal with such
> a case would require a new keyword for the FITS reader).

sorry for getting back at this after such a long time; I don’t know if you are still working on the
project. But I’ve just uploaded some sample files for the above format, and also another set
of model spectra in IMAGE_HDU and BINTABLE format, respectively.
The img spectrum can be read already, but wavelengths are not recognised; the table format
generally can be read as an astropy Table.
I have also added another file which is read in by Table.read() but raises several

UnitsWarning: 'ANGSTROMS' did not parse as fits unit: At col 0, Unit 'ANGSTROMS' not supported by the FITS standard. Did you mean Angstrom or angstrom? [astropy.units.core]

and consequently does not have a wavelength unit. This applies for probably every single spectrum in  
ftp://ftp.stsci.edu/cdbs/calspec/

so we should perhaps consider if we can tolerate the use of ‘ANGSTROMS’ or want to educate the
folks at STScI on the use of proper FITS headers…

Best,
						Derek