Exemple #1
0
    def test_rows(self, urlopen):
        """ Sending a valid query returns the correct response. """
        # Set the API key
        dc.set_api_key('TEST-API-KEY')

        # Create the SPARQL query
        query_string = ('''
SELECT  ?name ?dcid
WHERE {
  ?a typeOf Place .
  ?a name ?name .
  ?a dcid ("geoId/06" "geoId/21" "geoId/24") .
  ?a dcid ?dcid
}
''')
        selector = lambda row: row['?name'] != 'California'

        # Issue the query
        results = dc.query(query_string)
        selected_results = dc.query(query_string, select=selector)

        # Execute the query and iterate through the results.
        for idx, row in enumerate(results):
            if idx == 0:
                self.assertDictEqual(row, {
                    '?name': 'California',
                    '?dcid': 'geoId/06'
                })
            if idx == 1:
                self.assertDictEqual(row, {
                    '?name': 'Kentucky',
                    '?dcid': 'geoId/21'
                })
            if idx == 2:
                self.assertDictEqual(row, {
                    '?name': 'Maryland',
                    '?dcid': 'geoId/24'
                })

        # Verify that the select function works.
        for idx, row in enumerate(selected_results):
            if idx == 0:
                self.assertDictEqual(row, {
                    '?name': 'Kentucky',
                    '?dcid': 'geoId/21'
                })
            if idx == 1:
                self.assertDictEqual(row, {
                    '?name': 'Maryland',
                    '?dcid': 'geoId/24'
                })
Exemple #2
0
def main(): 
    response =  make_response({'error' : f'Please specify url query like this: ...url...?desired_method=args'})
    print("Got request args:")
    print(request.args)
    # The requestor would like to perform a SPARQL query
    if (request.args and 'sparql' in request.args) :
        sparql_query = request.args.get('sparql')
        print("Got query:")
        print(sparql_query)
        response = make_response(dc.query(sparql_query))
    # The requestor would like to request all triples associated with given dcids
    elif (request.args and 'triples' in request.args) :
        dcids = json.loads(request.args.get('triples'))
        print("Got dcids: ")
        print(dcids)
        #limit = dc.utils._MAX_LIMIT
        #if ('limit' in request.args)  :
        #    limit = request.args.get('limit')
        #print(f"limiting to: {limit}")
        response = make_response(dc.get_triples(dcids))

    # test endpoint for debugging
    elif request.args and 'test' in request.args :
        response = make_response({'message' : f'Test acknowledged and received data: {request.args.get("test")}' })
    else:
        pass # return_msg already set

    return response
 def test_query_w_api_key(self, urlopen):
     """ Handles row-less response. """
     # Set the API key
     dc.set_api_key('make_sure_I_am_replaced')
     dc.set_api_key(_TEST_API_KEY)
     # Issue a dummy SPARQL query that tells the mock to expect a key
     self.assertEqual(dc.query(_SPARQL_W_KEY), [])
    def test_no_rows(self, urlopen):
        """ Handles row-less response. """
        # Create a SPARQL query
        query_string = ('''
SELECT  ?name ?dcid
WHERE {
  ?a typeOf Place .
  ?a name ?name .
  ?a dcid ("geoId/DNE") .
  ?a dcid ?dcid
}
''')
        # Issue the query
        self.assertEqual(dc.query(query_string), [])
Exemple #5
0
def main():
  # Create a SPARQL query querying for the name of some states
  query = ('''
SELECT  ?name ?dcid
WHERE {
  ?a typeOf Place .
  ?a name ?name .
  ?a dcid ("geoId/06" "geoId/21" "geoId/24") .
  ?a dcid ?dcid
}
''')
  print('> Issuing query.\n{}'.format(query))

  # Iterate through all the rows in the results.
  print('> Printing results.\n')
  for row in dc.query(query_string=query):
    print('  {}'.format(row))
Exemple #6
0
def write_place_url(place_type):
    logging.info(place_type)
    sparql = '''
      SELECT ?dcid
      WHERE {{
        ?a typeOf {} .
        ?a dcid ?dcid
      }}
      Order By ASC(?dcid)
    '''.format(place_type)
    try:
        data = dc.query(sparql)
    except Exception:
        logging.exception('Got an error while query %s', place_type)
        return
    index = 0
    for places in chunks(data, LINE_LIMIT):
        file_name = '{}{}.{}.txt'.format(SAVE_PATH, place_type, index)
        with open(file_name, 'w') as f:
            logging.info('writing to file: %s', file_name)
            for p in places:
                f.write(SITE_PREFIX + p['?dcid'] + '\n')
        index += 1
        time.sleep(10)
st.write(duplicated_region_codes)

################################################################################
##### Query datacommons to get dcids                                       #####
################################################################################

query_str = '''
SELECT ?name ?dcid ?isoCode
WHERE {
  ?a typeOf Place .
  ?a name ?name .
  ?a isoCode ?isoCode .
  ?a dcid ?dcid
  }
'''
results = dc.query(query_str)

iso_level_2 = list(
    filter(lambda elem: elem['?dcid'].startswith('iso'), results))
iso_level_2_df = pd.DataFrame(iso_level_2)
iso_level_2_df = iso_level_2_df.rename(
    columns={
        '?name': 'datacommons_name',
        '?dcid': 'datacommons_id',
        '?isoCode': 'region_code'
    })

# Filter out dcids that aren't ISO-3166-2 codes
# (These appear to be freebase ids from wikidata)
iso_level_2_df['real_iso'] = iso_level_2_df.apply(
    lambda x: '/m/' not in x['region_code'], axis=1)
Exemple #8
0
################################################################################
##### Query datacommons to get dcids                                       #####
################################################################################

datacommons_query_str = '''
SELECT ?name ?dcid ?geoId ?wikidataId
WHERE {
  ?a typeOf County .
  ?a name ?name .
  ?a geoId ?geoId .
  ?a wikidataId ?wikidataId .
  ?a dcid ?dcid
  }
'''
dc_results = dc.query(datacommons_query_str)
dc_df = pd.DataFrame(dc_results)
dc_df = dc_df.rename(
    columns={
        '?name': 'datacommons_name',
        '?dcid': 'datacommons_id',
        '?wikidataId': 'wikidata_id',
        '?geoId': 'fips_code'
    })

dc_df = dc_df.drop_duplicates()

# New York County has two wikidata ids in datacommons:
# Q500416 (correct)
# Q11299 (incorrect - this is Manhattan not the fips county)
dc_df = dc_df[dc_df['wikidata_id'] != 'Q11299']
Exemple #9
0
import datacommons as dc

print("Running query...")

query_str = '''
	SELECT ?name ?dcid
	 WHERE {
	  ?a typeOf Place .
	  ?a name ?name .
	  ?a dcid ("geoId/06" "geoId/21" "geoId/24") .
	   ?a dcid ?dcid
	 }
'''

result = dc.query(query_str)

print("Done:")

for r in result:
    print(r)
 def test_query_no_api_key(self, urlopen):
     del os.environ[utils._ENV_VAR_API_KEY]
     # Issue a dummy SPARQL query that tells the mock to not expect a key
     self.assertEqual(dc.query(_SPARQL_NO_KEY), [])