How do I use pyodbc.connect() with authorization=ActiveDirectoryIntegrated?

Question

Trying to access Azure SQL through a python function in VS code, with Authentication set to Active Directory Integrated. Using pyodbc to connect.

Works fine when run locally but get an error after deploying to Azure. Also works fine if I use SQL login but I want to use Active Directory Integrated. I have already set myself as an AD admin.

What I am trying:

cnxn = pyodbc.connect("Driver={ODBC Driver 17 for SQL Server};Server=tcp:khawajaserver1.database.windows.net,1433;Database=KhawajaDB1;Encrypt=yes;TrustServerCertificate=no;Connection Timeout=30;Authentication=ActiveDirectoryIntegrated")

Error I get:

Result: Failure
Exception: Error: ('HY000', '[HY000] [Microsoft][ODBC Driver 17 for SQL Server]MAX_PROVS: Error code 0x57 (87) (SQLDriverConnect)')
Stack:   File "/usr/local/lib/python3.6/site-packages/azure/functions_worker/dispatcher.py", line 308, in _handle__invocation_request

Answer 1

I know Active Directory Password as authentication type works.

db_list = [TEST_DB1, TEST_DB2]
sql_conn = None
for db in db_list:
    try:
        conn_string = 'DRIVER={ODBC Driver 17 for SQL Server};' \
                      'SERVER=' + <db_url> + \
                      ';DATABASE=' + <db_name> + \
                      ';UID=' + <db_username> + \
                      ';PWD=' + <db_password> + \
                      ';Authentication=ActiveDirectoryPassword'
        print conn_string
        sql_conn = pyodbc.connect(conn_string)
except Exception as e:
    print "Exception:::", e
    print 'Cannot connect to DB' + str(sys.exc_info()[0])
    return None
sql_conn.cursor().execute(<some SQL Query>)
sql_conn.close()

Answer 2

The following code worked for me

# Connection to SQL Server using AADIntegrated

import pyodbc 

server = 'data1.database.windows.net' 
database = 'MyTestDB' 
authentication = 'ActiveDirectoryIntegrated'
kpi_server_connection = pyodbc.connect('DRIVER={ODBC Driver 17 for SQL Server};SERVER='+server+';DATABASE='+database+';Authentication='+authentication+';TrustServerCertificate='+ 'no')

query_string = '''
    select top 10 * from [SomeTable] 
    '''
df = pd.read_sql(query_string, kpi_server_connection)
df