'referenceValues'][ref_retrieved_index]
return employee_dict
# -----------------
# Begin script
# -----------------
# Calculate the reference date retrieved value for all statements
whole_time_string_z = datetime.datetime.utcnow().isoformat(
) # form: 2019-12-05T15:35:04.959311
dateZ = whole_time_string_z.split('T')[0] # form 2019-12-05
ref_retrieved = '+' + dateZ + 'T00:00:00Z' # form +2019-12-05T00:00:00Z as provided by Wikidata
filename = deptShortName + '-employees-with-wikidata.csv'
employees = vbc.readDict(filename)
# create a list of the employees who have Wikidata qIDs
qIds = []
for employee in employees:
if employee['wikidataId'] != '':
qIds.append(employee['wikidataId'])
# ------------------------------------------------------
# get all of the ORCID data that is already in Wikidata
#prop = 'P496' # ORCID ID
#value = '' # since no value is passed, the search will retrieve the value
#refProps = ['P813'] # retrieved
# The script determines what's being tracked with respect to references by whether these field name strings are empty or not.
field_name = 'orcid'
'string': 'J.D.',
'value': 'J.D.'
}]
# NCBI identification requirements:
# tool name and email address should be sent with all requests
# see https://www.ncbi.nlm.nih.gov/books/NBK25499/#chapter4.ESearch
emailAddress = '*****@*****.**' # put your email address here
toolName = 'VanderBot' # give your application a name here
# -----------------
# Begin script
# -----------------
filename = deptShortName + '-employees.csv'
employees = vbc.readDict(filename)
filename = 'orcid_data.csv'
orcidData = vbc.readDict(filename)
testRatio = 90
departmentTestRatio = 90
for employeeIndex in range(0, len(employees)):
matched = False
for row in orcidData:
name = row['givenNames'] + ' ' + row['familyName']
#ratio = fuzz.ratio(name, employees[employeeIndex][0])
#partialRatio = fuzz.partial_ratio(name, employees[employeeIndex][0])
#sortRatio = fuzz.token_sort_ratio(name, employees[employeeIndex][0])
output = ''