remove uses of deprecated asdf.resolver

[braingram]

asdf.resolver will be deprecated in 2.15: asdf-format/asdf#1362

A few uses were found in stdatamodels

1. fits_support adds a resolver for a non-existent path:

   stdatamodels/src/stdatamodels/fits_support.py

   Lines 362 to 366 in 740477c

   	FITS_SCHEMA_URL_MAPPING = resolver.Resolver(
   	[
   	('http://stsci.edu/schemas/fits-schema/',
   	'file://' + META_SCHEMA_PATH + '/{url_suffix}.yaml')
   	] + resolver.DEFAULT_URL_MAPPING, 'url')

2. jwst/datamodels/multiexposure provides resolvers that are equivalent to the defaults

   stdatamodels/src/stdatamodels/jwst/datamodels/multiexposure.py

   Lines 89 to 98 in 740477c

   	schema = asdf_schema.load_schema(
   	self.schema_url,
   	resolver=AsdfFile().resolver,
   	resolve_references=True
   	)
   	core_schema = asdf_schema.load_schema(
   	self.core_schema_url,
   	resolver=AsdfFile().resolver,
   	resolve_references=True
   	)

3. jwst/datamodels/schema_editor uses the default resolver to resolve schema references

   stdatamodels/src/stdatamodels/jwst/datamodels/schema_editor.py

   Lines 397 to 415 in 740477c

   	resolver = get_default_resolver()
   	def resolve_refs(node, json_id):
   	if json_id is None:
   	json_id = url
   	if isinstance(node, dict) and '$ref' in node:
   	suburl = generic_io.resolve_uri(json_id, node['$ref'])
   	parts = urlparse(suburl)
   	fragment = parts.fragment
   	if len(fragment):
   	suburl_path = suburl[:-(len(fragment) + 1)]
   	else:
   	suburl_path = suburl
   	suburl_path = resolver(suburl_path)
   	try:
   	subschema = aschema.load_schema(suburl_path,
   	resolver,
   	True)

1 and 2 are unnecessary and this PR removes these uses.

schema_editor duplicates some of the code internal to asdf see asdf.schema._load_schema_cached. The uses of resolver appear unnecessary as the call to generic_io.resolve_uri and schema.load_schema appear sufficient to resolve the schema references (as tested with the jwst regression test: https://github.com/spacetelescope/jwst/blob/master/jwst/regtest/test_schema_editor.py).

A complete run of the jwst regression tests with these changes will be needed. EDIT: jwst regression tests passed here: #128 (comment)

Resolves JP-nnnn
Resolves RCAL-nnnn

Closes #

This PR addresses ...

Checklist

• added entry in CHANGES.rst (either in Bug Fixes or Changes to API)
• updated relevant tests
• updated relevant documentation
• updated relevant milestone(s)
• added relevant label(s)

[braingram]

JWST regression tests run against this branch passed without errors: https://plwishmaster.stsci.edu:8081/job/RT/job/JWST/2243

JWST CI failure I believe is due to: #129

[codecov]

Codecov Report

Patch coverage: 100.00% and project coverage change: -0.01 ⚠️

Comparison is base (dec52f9) 65.17% compared to head (f4cb3a5) 65.17%.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #128      +/-   ##
==========================================
- Coverage   65.17%   65.17%   -0.01%     
==========================================
  Files         101      101              
  Lines        5445     5438       -7     
==========================================
- Hits         3549     3544       -5     
+ Misses       1896     1894       -2     

Flag	Coverage Δ
unit	65.17% <100.00%> (-0.01%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
src/stdatamodels/fits_support.py	90.28% <ø> (-0.09%)	⬇️
src/stdatamodels/jwst/datamodels/schema_editor.py	10.23% <ø> (-0.09%)	⬇️
src/stdatamodels/jwst/datamodels/multiexposure.py	57.89% <100.00%> (ø)

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[braingram]

@tapastro @stscieisenhamer mentioned you may have run the schema_editor for the past few builds. Would you have time to look over the changes in this PR? Let me know if you have any questions.

[tapastro]

Sorry it took me a while to get back to you, @braingram! I don't know if there is good test coverage of the schema_editor, but I can try running it off this branch if you think this change may have affected functionality. To be honest, while I have used the code multiple times, the lines you've modified look completely unfamiliar and I'm not sure what would change on the face of it.

EDIT: I needed to read the whole post... I see the link to the test now. I'll review that and see if it covers the general usage.

[braingram]

The new JWST CI failure is due to a version conflict (jwst requires stdatamodels <1.2, this repo is currently >=1.2 so the stpipe entry points fail to load with a VersionConflict):
https://github.com/spacetelescope/stdatamodels/actions/runs/4315086232/jobs/7528977635#step:12:848
A PR for stpipe is open that should fix this:
spacetelescope/stpipe#84

EDIT: given that the issue appears to be stpipe and the CI job runs with a pip installed stpipe this job is not expected to pass until a version of stpipe is released that has the fix mentioned above.

[eslavich]

I wonder if we can drop this whole method and replace with the asdf resolve_references=True flag? the try/catch that was added here seems like a kludge for some long ago schema URI problem that may not exist anymore.

If not, it's worth comparing this to the asdf implementation and updating the code here where it's fallen behind. It looks like a copy of an older version of the resolve_refs callback in asdf.schema.

[braingram]

It's possible this isn't needed but given the lack of test coverage and that it sounds like this is a tool that's manually run I'm less inclined to make changes unless they're absolutely necessary. Perhaps @tapastro has some thoughts on this.