ENH: SVO FPS - add method to easily retrieve filter metadata

[cgobat]

This PR adds a new method: SvoFpsClass.get_filter_params(). It follows the same call signature as the existing SvoFpsClass.get_transmission_data(), and returns a dict containing the VOTable PARAMs returned by the FPS. This includes things like the filter zero point, min/max/effective wavelength, description, etc.

For example, one often needs the effective wavelength of a filter. Without this method, the practical options were either to retrieve the full transmission profile and do the numerical integration yourself, or manually go to the FPS website and find it on the webpage, then hardcode it. With this new method, it is as simple as lambda_eff = SvoFps.get_filter_params(my_filter_id)["WavelengthEff"].

[codecov]

Codecov Report

❌ Patch coverage is 8.33333% with 11 lines in your changes missing coverage. Please review.
✅ Project coverage is 72.67%. Comparing base (81f2ac5) to head (459ed3a).
⚠️ Report is 19 commits behind head on main.

Files with missing lines	Patch %	Lines
astroquery/svo_fps/core.py	8.33%	11 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3528      +/-   ##
==========================================
+ Coverage   72.39%   72.67%   +0.27%     
==========================================
  Files         219      219              
  Lines       20432    20485      +53     
==========================================
+ Hits        14792    14887      +95     
+ Misses       5640     5598      -42     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[bsipocz]

Thank you. Overall it looks good, but there are a few minor changes that would make it better.

Also, would you mind adding an example for the narrative docs?

[keflavich]

Could you clarify the aim of this? The suggested motivation is that "Without this method, the practical options were either to retrieve the full transmission profile and do the numerical integration yourself, or manually go to the FPS website and find it on the webpage, then hardcode it. ", but as far as I can tell, this is part of the default returns, e.g.:

from astroquery.svo_fps import SvoFps
filterlist = SvoFps.get_filter_list('JWST')
filterlist[ 'filterID',  'WavelengthEff' ]

gives a list like:

       filterID        WavelengthEff
                          Angstrom
--------------------- ---------------
    JWST/NIRCam.F070W 6988.4272768359
JWST/NIRCam2025.F070W 6996.0621457337
                  ...             ...
     JWST/MIRI.F2100W 205601.06046254
     JWST/MIRI.F2300C  224493.7622614
     JWST/MIRI.F2550W 251515.98944653
Length = 88 rows

[keflavich]

The full set of metadata is accessible:

>>> SvoFps.get_filter_list('JWST').colnames
['FilterProfileService',
 'filterID',
 'WavelengthUnit',
 'WavelengthUCD',
 'PhotSystem',
 'DetectorType',
 'Band',
 'Instrument',
 'Facility',
 'ProfileReference',
 'CalibrationReference',
 'Description',
 'Comments',
 'WavelengthRef',
 'WavelengthMean',
 'WavelengthEff',
 'WavelengthMin',
 'WavelengthMax',
 'WidthEff',
 'WavelengthCen',
 'WavelengthPivot',
 'WavelengthPeak',
 'WavelengthPhot',
 'FWHM',
 'Fsun',
 'PhotCalID',
 'MagSys',
 'ZeroPoint',
 'ZeroPointUnit',
 'Mag0',
 'ZeroPointType',
 'AsinhSoft',
 'TrasmissionCurve']

then you can select by matching on filterID. Since the tables of instruments are pretty small, this isn't much of an overhead

("Trasmission Curve" instead of "Transmission Curve" is apparently an upstream issue?)

[cgobat]

The biggest problem with that proposal is that a number of filters at SVO's FPS don't have an associated "facility", so the get_filter_list method doesn't work at all for them.

Even if that weren't the case, I would still argue that this is a useful addition—it is a very (maybe even more?) common use case to not actually care about the exact structure of a filter's response curve, and really just need to know some defining characteristics like the FWHM, effective width, pivot wavelength, effective wavelength, etc. Given that, there should be a straightforward, non-hacky way to retrieve that kind of information directly, without having to pull down other filters or data you don't need.

[cgobat]

Nice catch about "TrasmissionCurve" though—I had never noticed that. :)

[keflavich]

The biggest problem with that proposal is that a number of filters at SVO's FPS don't have an associated "facility", so the get_filter_list method doesn't work at all for them.

Is this really true? The website is listed facility-first; I didn't think there were any filters without an associated facility, even if it's just a placeholder. Do you have examples?

Even if that weren't the case, I would still argue that this is a useful addition—it is a very (maybe even more?) common use case to not actually care about the exact structure of a filter's response curve, and really just need to know some defining characteristics like the FWHM, effective width, pivot wavelength, effective wavelength, etc. Given that, there should be a straightforward, non-hacky way to retrieve that kind of information directly, without having to pull down other filters or data you don't need.

I agree, this is a more convenient way to get access to those metadata, which I frequently use. I don't think the current approach is hacky, but it is a tiny bit more clunky, so I'm onboard with adding this.

[cgobat]

Yeah, maybe hacky wasn't quite the right word there. Thanks for your approval.

Here are a couple examples of facility-less filters:

• Generic/Bessell.V (along with most of the "Generic" filters)
• Cameras/SQM.B
• Misc/UCAC.R

[bsipocz]

This all looks good to me now; thank you!

[bsipocz]

I interpret Adam's comment above as a thumbs up, so let's merge this now.