databricks-sql-connector 3.7.1__tar.gz → 3.7.3__tar.gz

{databricks_sql_connector-3.7.1 → databricks_sql_connector-3.7.3}/CHANGELOG.md

@@ -1,5 +1,15 @@
 # Release History
 
+# 3.7.3 (2025-03-28)
+
+- Fix: Unable to poll small results in execute_async function (databricks/databricks-sql-python#515 by @jprakash-db) 
+- Updated log messages to show the status code and error messages of requests (databricks/databricks-sql-python#511 by @jprakash-db) 
+- Fix: Incorrect metadata was fetched in case of queries with the same alias (databricks/databricks-sql-python#505 by @jprakash-db)
+
+# 3.7.2 (2025-01-31)
+
+- Updated the retry_dela_max and retry_timeout (databricks/databricks-sql-python#497 by @jprakash-db)
+
 # 3.7.1 (2025-01-07)
 
 - Relaxed the number of Http retry attempts (databricks/databricks-sql-python#486 by @jprakash-db)

{databricks_sql_connector-3.7.1 → databricks_sql_connector-3.7.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: databricks-sql-connector
-Version: 3.7.1
+Version: 3.7.3
 Summary: Databricks SQL Connector for Python
 License: Apache-2.0
 Author: Databricks

{databricks_sql_connector-3.7.1 → databricks_sql_connector-3.7.3}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "databricks-sql-connector"
-version = "3.7.1"
+version = "3.7.3"
 description = "Databricks SQL Connector for Python"
 authors = ["Databricks <databricks-sql-connector-maintainers@databricks.com>"]
 license = "Apache-2.0"

{databricks_sql_connector-3.7.1 → databricks_sql_connector-3.7.3}/src/databricks/sql/__init__.py

@@ -68,7 +68,7 @@ DATETIME = DBAPITypeObject("timestamp")
 DATE = DBAPITypeObject("date")
 ROWID = DBAPITypeObject()
 
-__version__ = "3.7.1"
+__version__ = "3.7.3"
 USER_AGENT_NAME = "PyDatabricksSqlConnector"
 
 # These two functions are pyhive legacy

{databricks_sql_connector-3.7.1 → databricks_sql_connector-3.7.3}/src/databricks/sql/auth/retry.py

@@ -290,8 +290,9 @@ class DatabricksRetryPolicy(Retry):
         else:
             proposed_wait = self.get_backoff_time()
 
-        proposed_wait = min(proposed_wait, self.delay_max)
+        proposed_wait = max(proposed_wait, self.delay_max)
         self.check_proposed_wait(proposed_wait)
+        logger.debug(f"Retrying after {proposed_wait} seconds")
         time.sleep(proposed_wait)
         return True
 
@@ -344,23 +345,24 @@ class DatabricksRetryPolicy(Retry):
         if a retry would violate the configured policy.
         """
 
+        logger.info(f"Received status code {status_code} for {method} request")
+
         # Request succeeded. Don't retry.
         if status_code == 200:
             return False, "200 codes are not retried"
 
         if status_code == 401:
-            raise NonRecoverableNetworkError(
-                "Received 401 - UNAUTHORIZED. Confirm your authentication credentials."
+            return (
+                False,
+                "Received 401 - UNAUTHORIZED. Confirm your authentication credentials.",
             )
 
         if status_code == 403:
-            raise NonRecoverableNetworkError(
-                "Received 403 - FORBIDDEN. Confirm your authentication credentials."
-            )
+            return False, "403 codes are not retried"
 
         # Request failed and server said NotImplemented. This isn't recoverable. Don't retry.
         if status_code == 501:
-            raise NonRecoverableNetworkError("Received code 501 from server.")
+            return False, "Received code 501 from server."
 
         # Request failed and this method is not retryable. We only retry POST requests.
         if not self._is_method_retryable(method):
@@ -399,8 +401,9 @@ class DatabricksRetryPolicy(Retry):
             and status_code not in self.status_forcelist
             and status_code not in self.force_dangerous_codes
         ):
-            raise UnsafeToRetryError(
-                "ExecuteStatement command can only be retried for codes 429 and 503"
+            return (
+                False,
+                "ExecuteStatement command can only be retried for codes 429 and 503",
             )
 
         # Request failed with a dangerous code, was an ExecuteStatement, but user forced retries for this

{databricks_sql_connector-3.7.1 → databricks_sql_connector-3.7.3}/src/databricks/sql/auth/thrift_http_client.py

@@ -198,6 +198,12 @@ class THttpClient(thrift.transport.THttpClient.THttpClient):
         self.message = self.__resp.reason
         self.headers = self.__resp.headers
 
+        logger.info(
+            "HTTP Response with status code {}, message: {}".format(
+                self.code, self.message
+            )
+        )
+
     @staticmethod
     def basic_proxy_auth_headers(proxy):
         if proxy is None or not proxy.username:

{databricks_sql_connector-3.7.1 → databricks_sql_connector-3.7.3}/src/databricks/sql/client.py

@@ -737,6 +737,7 @@ class Cursor:
         self,
         operation: str,
         parameters: Optional[TParameterCollection] = None,
+        enforce_embedded_schema_correctness=False,
     ) -> "Cursor":
         """
         Execute a query and wait for execution to complete.
@@ -801,6 +802,7 @@ class Cursor:
             use_cloud_fetch=self.connection.use_cloud_fetch,
             parameters=prepared_params,
             async_op=False,
+            enforce_embedded_schema_correctness=enforce_embedded_schema_correctness,
         )
         self.active_result_set = ResultSet(
             self.connection,
@@ -822,6 +824,7 @@ class Cursor:
         self,
         operation: str,
         parameters: Optional[TParameterCollection] = None,
+        enforce_embedded_schema_correctness=False,
     ) -> "Cursor":
         """
 
@@ -862,6 +865,7 @@ class Cursor:
             use_cloud_fetch=self.connection.use_cloud_fetch,
             parameters=prepared_params,
             async_op=True,
+            enforce_embedded_schema_correctness=enforce_embedded_schema_correctness,
         )
 
         return self

{databricks_sql_connector-3.7.1 → databricks_sql_connector-3.7.3}/src/databricks/sql/thrift_backend.py

@@ -66,7 +66,7 @@ DEFAULT_SOCKET_TIMEOUT = float(900)
 # - 900s attempts-duration lines up w ODBC/JDBC drivers (for cluster startup > 10 mins)
 _retry_policy = {  # (type, default, min, max)
     "_retry_delay_min": (float, 1, 0.1, 60),
-    "_retry_delay_max": (float, 30, 5, 3600),
+    "_retry_delay_max": (float, 60, 5, 3600),
     "_retry_stop_after_attempts_count": (int, 30, 1, 60),
     "_retry_stop_after_attempts_duration": (float, 900, 1, 86400),
     "_retry_delay_default": (float, 5, 1, 60),
@@ -883,6 +883,7 @@ class ThriftBackend:
         use_cloud_fetch=True,
         parameters=[],
         async_op=False,
+        enforce_embedded_schema_correctness=False,
     ):
         assert session_handle is not None
 
@@ -898,8 +899,12 @@ class ThriftBackend:
             sessionHandle=session_handle,
             statement=operation,
             runAsync=True,
-            getDirectResults=ttypes.TSparkGetDirectResults(
-                maxRows=max_rows, maxBytes=max_bytes
+            # For async operation we don't want the direct results
+            getDirectResults=None
+            if async_op
+            else ttypes.TSparkGetDirectResults(
+                maxRows=max_rows,
+                maxBytes=max_bytes,
             ),
             canReadArrowResult=True if pyarrow else False,
             canDecompressLZ4Result=lz4_compression,
@@ -910,6 +915,7 @@ class ThriftBackend:
             },
             useArrowNativeTypes=spark_arrow_types,
             parameters=parameters,
+            enforceEmbeddedSchemaCorrectness=enforce_embedded_schema_correctness,
         )
         resp = self.make_request(self._client.ExecuteStatement, req)