BUG: sqlalchemy connection string parsing and ldap auth (cloudera#261)

* pass parsed connection string username as user to connect

* pass PLAIN as auth_mech in case of LDAP to sasl_factory

* move auth_mech assignment after user/pass setting

impala/_thrift_api.py

@@ -149,7 +149,7 @@ def get_transport(socket, host, kerberos_service_name, auth_mechanism='NOSASL',
                 # PLAIN always requires a password for HS2.
                 password = 'password'
             log.debug('get_transport: password=%s', password)
-
+        auth_mechanism = 'PLAIN'  # sasl doesn't know mechanism LDAP
     # Initializes a sasl client
     from thrift_sasl import TSaslClientTransport
     try:
@@ -169,6 +169,7 @@ def sasl_factory():
         from impala.sasl_compat import PureSASLClient
 
         def sasl_factory():
-            return PureSASLClient(host, username=user, password=password, service=kerberos_service_name)
+            return PureSASLClient(host, username=user, password=password,
+                                  service=kerberos_service_name)
 
     return TSaslClientTransport(sasl_factory, auth_mechanism, socket)

impala/hiveserver2.py

@@ -755,7 +755,7 @@ def connect(host, port, timeout=None, use_ssl=False, ca_cert=None,
             user=None, password=None, kerberos_service_name='impala',
             auth_mechanism=None):
     log.debug('Connecting to HiveServer2 %s:%s with %s authentication '
-             'mechanism', host, port, auth_mechanism)
+              'mechanism', host, port, auth_mechanism)
     sock = get_socket(host, port, use_ssl, ca_cert)
     if timeout is not None:
         timeout = timeout * 1000.  # TSocket expects millis

impala/sqlalchemy.py

@@ -161,6 +161,17 @@ def dbapi(cls):
         import impala.dbapi
         return impala.dbapi
 
+    def create_connect_args(self, url):
+        kwargs = {
+            'host': url.host,
+            'port': url.port,
+            'user': url.username,
+            'password': url.password,
+            'database': url.database,
+        }
+        kwargs.update(url.query)
+        return ([], kwargs)
+
     def initialize(self, connection):
         self.default_schema_name = connection.connection.default_db