Integrate ORNL DAAC Identify calls

[cmungall]

The goal is to implement methods within the NMDC sample-annotator framework to access DAAC APIs, and transform the results into json objects conforming to the JSON schema. Broadly there are two types of calls, those that produce (1) QUantityValues, e.g. elevation (2) qualitative, e.g. soil types.

@StantonMartin started on code here: https://github.com/StantonMartin/Identify/

However, this functionality will not be implemented as a standalone script, but in sample-annotator framework. Before attempting to implement this ticket, the implementor must familiarize themselves with the code in this repo, see the README and the unit tests.

As a stub example, the existing codebase uses the google maps API to query for elevation.

See code:

sample-annotator/sample_annotator/geolocation/geotools.py

Lines 34 to 36 in bc17c12

	def get_elevation(self, latlon: LATLON = (40.714224, -73.961452)):
	results = self.get_client().elevation(latlon)
	return results

and the unit test:

sample-annotator/tests/test_elevation.py

Lines 16 to 26 in bc17c12

	def test_elevation(self):
	ge = GeoEngine()
	if os.path.exists(KEYPATH):
	ge.load_key(KEYPATH)
	mtEverestCoord = (27.9881, 86.9250)
	e = ge.get_elevation(mtEverestCoord)
	print(e)
	# Wikipedia says 8,848.86m
	e0 = e[0]
	assert e0.get('elevation') > 8800
	assert e0.get('elevation') < 8900

You can see how this is implemented within the main sample annotator here:

sample-annotator/sample_annotator/sample_annotator.py

Lines 213 to 251 in bc17c12

	def perform_geolocation_inference(self, sample: SAMPLE, report: AnnotationReport):
	"""
	Performs inference using geolocation information
	"""
	# TODO: Stan to populate
	ll_str = sample.get(KEY_LAT_LON, None)
	if ll_str is None:
	report.add_message(f'No lat_long specified',
	severity=3,
	category=Category.MissingCore)
	return
	try:
	lat_lon = tuple([float(x.strip()) for x in ll_str.split(' ')])
	except:
	report.add_message(f'Incorrect format for lat_lon: {ll_str}', severity=3)
	return
	sample[KEY_LAT_LON] = {'latitude': lat_lon[0], 'longitude': lat_lon[1]}
	ge = self.geoengine
	if ge is None:
	report.add_message('Skipping geo-checks', severity=0)
	return
	logging.info('Using geoengine')
	elevs = ge.get_elevation(lat_lon)
	if len(elevs) != 1:
	report.add_message(f'Something went wrong, elevs = {elevs}')
	if len(elevs) > 0:
	elev = elevs[0].get('elevation')
	res = elevs[0].get('resolution')
	if KEY_ELEV in sample:
	curr = sample.get(KEY_ELEV)
	if curr.has_unit == 'meter':
	if abs(curr.has_value - elev) > res:
	report.add_message(f'Conflicting values for elevation; current: {curr} Googlemaps: {elev} +/- {res}')
	else:
	report.add_message(f'Filling in missing value for elevation {elev}',
	was_repaired=True)
	sample[KEY_ELEV] = {'has_unit': 'meter',
	'has_numeric_value': elev}

[ADDING MORE LATER...]

For qualitative data we will curate mapping tables, see my PR here: StantonMartin/Identify#1

[ssarrafan]

@cmungall @dehays let me know if this should be re-assigned to Stan or if I should create a new issue for Stan related to this and the meeting you had with him last week.

[ssarrafan]

@cmungall @StantonMartin should this issue be closed, moved to the backlog or moved to the October sprint?

[ssarrafan]

No updates on this since August so I've added the backlog label and removed from the project board.
FYI @cmungall @emileyfadrosh

[StantonMartin]

Integration complete. Closing