def populateValue(g, datasetId, ds, data, p, o, iriCache):
## Skipping following IRI's as they are handled separately (getResearcher, getProtocols, etc.)
skipIri = [
term.URIRef('http://uri.interlex.org/temp/uris/contributorTo'),
term.URIRef('http://www.w3.org/1999/02/22-rdf-syntax-ns#type'),
term.URIRef('http://uri.interlex.org/temp/uris/hasUriApi'),
term.URIRef('http://uri.interlex.org/temp/uris/hasUriHuman'),
term.URIRef('http://uri.interlex.org/temp/uris/hasProtocol'),
term.URIRef('http://uri.interlex.org/temp/uris/wasUpdatedAtTime')]
key = strip_iri(p.strip())
if p in skipIri:
return
if isinstance(o, term.URIRef):
value = iri_lookup(g, o.strip(), iriCache)
if value:
if isinstance(value, dict) and 'curie' in value:
ds['term'][value['curie']] = value
value = value['curie']
# if isinstance(value, dict) and 'iri' in value:
# key = strip_iri(value['iri'])
# ds['term'][key] = value
# value = key
if key in arrayProps:
array = data.setdefault(key, [])
array.append(value)
else:
if key in data:
log.warning('Unexpected creation of array for: %s - %s - %s', datasetId, key, value)
log.warning('Existing value for this key : %s - %s - %s', datasetId, key, data[key])
log.warning('----- Will use the shortest value -----')
if len(value) < len(data[key]):
data[key] = value
else:
data[key] = value
elif isinstance(o, term.Literal):
value = strip_iri(o.strip())
if key in arrayProps:
array = data.setdefault(key, [])
array.append(value)
else:
if key in data:
log.warning('Unexpected creation of array for: %s - %s - %s', datasetId, key, value)
log.warning('Existing value for this key : %s - %s - %s', datasetId, key, data[key])
log.warning('----- Will use the shortest value -----')
if len(value) < len(data[key]):
data[key] = value
else:
data[key] = value
elif isinstance(o, term.BNode):
data[key] = parseMeasure(datasetId, g, o, {'value': '', 'unit': ''})
else:
raise Exception('Unknown RDF term: %s' % type(o))
def getResearchers(gNew, gDelta, output, iriCache):
# Iterate over Researchers
for s, o in gNew.subject_objects(URIRef('http://uri.interlex.org/temp/uris/contributorTo')):
m = re.search(r".*(?P<ds>N:dataset:[:\w-]+)", o)
datasetId = strip_iri(m.group(0).strip())
user = strip_iri(s)
# user = s #s.split('/')[-1] # either a blackfynn user id or "Firstname-Lastname"
newEntry = {}
for p2, o2 in gDelta.predicate_objects(s):
populateValue(gDelta, datasetId, output[datasetId], newEntry, p2, o2, iriCache)
if newEntry:
output[datasetId]['researcher'][user] = newEntry
def parseMeasure(dsId, g, node, values):
if (node, None, URIRef('http://uri.interlex.org/tgbugs/uris/readable/sparc/Measurement')) in g:
# Current BNode is a measurement
# preds = g.predicates(subject=node)
# for v in preds:
# print('pred: {}'.format(v))
# values['unit'] = strip_iri(v)
unit = strip_iri(g.value(subject=node, predicate=URIRef('http://uri.interlex.org/temp/uris/hasUnit')))
values['unit'] = unit
value = g.value(subject=node, predicate=URIRef('http://www.w3.org/1999/02/22-rdf-syntax-ns#value'))
values['value'] = str(value)
if values['unit'] == 'dimensionless':
log.warning("Measurement with no unit (value: {}) in {}".format(values['value'], dsId))
elif (node, None, URIRef('http://www.w3.org/2000/01/rdf-schema#Datatype')) in g:
# Current BNode is a rdfs:Datatype
unit = strip_iri(g.value(subject=node, predicate=URIRef('http://www.w3.org/2002/07/owl#onDatatype')))
values['unit'] = strip_iri(unit)
value = g.value(subject=node, predicate=URIRef('http://www.w3.org/2002/07/owl#withRestrictions'))
# Get Lower Bound Range
first = g.value(subject=value, predicate=URIRef('http://www.w3.org/1999/02/22-rdf-syntax-ns#first'))
min_incl = g.value(subject=first, predicate=URIRef('http://www.w3.org/2001/XMLSchema#minInclusive'))
#Get Higher Bound Range
rest = g.value(subject=value, predicate=URIRef('http://www.w3.org/1999/02/22-rdf-syntax-ns#rest'))
rest_first = g.value(subject=rest, predicate=URIRef('http://www.w3.org/1999/02/22-rdf-syntax-ns#first'))
max_incl = g.value(subject=rest_first, predicate=URIRef('http://www.w3.org/2001/XMLSchema#maxInclusive'))
values['value'] = "{}-{}".format(str(min_incl), str(max_incl))
if values['unit'] == 'dimensionless':
log.warning("Measurement with no unit (value: {}) in {}".format(values['value'], dsId))
else:
log.warning("Encountered a B-Node that is not a measurement in {}".format(dsId))
return values
def getProtocols(gNew, gDelta, output, iriCache):
# Iterate over Protocols
for s, o in gNew.subject_objects(URIRef('http://uri.interlex.org/temp/uris/hasProtocol')):
m = re.search(r".*(?P<ds>N:dataset:[:\w-]+)", s)
datasetId = strip_iri(m.group(0).strip())
url = str(o)
newEntry = {}
for p2, o2 in gDelta.predicate_objects(o):
populateValue(gDelta, datasetId, output[datasetId], newEntry, p2, o2, iriCache)
if newEntry:
output[datasetId]['protocol'][url] = newEntry
def getDatasets(gNew, gDelta, output, iriCache):
# Iterate over Datasets
for ds in gNew.subjects(RDF.type, URIRef('http://uri.interlex.org/tgbugs/uris/readable/sparc/Dataset')):
log.info(ds)
m = re.search(r".*(?P<ds>N:dataset:[:\w-]+)", ds)
datasetId = strip_iri(m.group(0).strip())
addEntry(output, datasetId)
for p, o in gDelta.predicate_objects(ds):
if p == URIRef("http://uri.interlex.org/temp/uris/hasAwardNumber"):
getAwards(o, datasetId, output)
populateValue(gDelta, datasetId, output[datasetId], output[datasetId]['summary'], p, o, iriCache)
def getTags(gNew, gDelta, output, iriCache):
# Iterate over Protocols
for s, o in gNew.subject_objects(URIRef('http://purl.obolibrary.org/obo/IAO_0000136')):
m = re.search(r".*(?P<ds>N:dataset:[:\w-]+)", s)
if m:
if isinstance(o, term.URIRef):
t = iri_lookup(gNew, o, iriCache)
if t:
tag = t['labels'][0]
else:
continue
else:
tag = str(o)
datasetId = strip_iri(m.group(0).strip())
if tag not in output[datasetId]['tag']:
output[datasetId]['tag'].append(tag)
def getAwards(awardIdURI, dsId, output):
# Iterate over awards
awardId = strip_iri(awardIdURI)
output[dsId]['award'][awardId] = {
'awardId': awardId
}