CoDICE: CAVA metadata support

[tech3371]

Subject: Re: CoDICE in CAVA

I grabbed imap_codice_l2_hi-omni_20260104_v001.cdf off the website.

There are a lot of invalid FILLVALs...should be -1e31, and are 1e31

Despite 20260104 as the name, there are several records from 2026-01-03

The epoch_delta variables are doubles not ints.

A lot of the energy uncertainties have values under the VALIDMIN (maybe
the values for the energy center channels were applied to the
uncertainties?)

Most format strings are (incorrectly) C-style instead of Fortran-style.

There's some SI_CONVERSION attrs that make no sense (" > ", i.e.
"multiply by blank to convert this to SI units of blank")

I'm not sure of all the "N/A". There aren't LABLAXIS attributes on those
variables, but technically it's not required because they're using
LABL_PTR_1. That might be a CAVA bug--if we're normally displaying
LABLAXIS below there, we should probably use FIELDNAM, or put nothing
there, if there is a LABLAXIS. (This is a part of the spec that's really
unclear IMO). I can't find the code in CAVA that populates that field,
and there is no "N/A" actually in any of the attributes.

The epoch is compressed and shouldn't be.

So it's got a lot of the required metadata but needs some work to be
compliant.

--
Jonathan Niehof, Ph.D.
UNH Space Science Center
106 Morse Hall, 8 College Rd., Durham, NH 03824
(603) 862-0649