databricks-sql-connector 4.3.0__tar.gz → 4.4.0__tar.gz

{databricks_sql_connector-4.3.0 → databricks_sql_connector-4.4.0}/CHANGELOG.md

@@ -1,5 +1,10 @@
 # Release History
 
+# 4.4.0 (2026-07-22)
+- Raised the minimum supported Python version to 3.10, dropping the end-of-life 3.8/3.9, to update the lockfile and clear CVE-flagged dependencies in the repo (databricks/databricks-sql-python#798)
+- Fix: `REMOVE` staging operations no longer require `staging_allowed_local_path` to be set, since removing a remote file does not touch the local filesystem (databricks/databricks-sql-python#726)
+- Report `cursor.rowcount` for DML on the Thrift backend: INSERT/UPDATE/DELETE/MERGE now set `rowcount` to the server's affected-row count instead of the hardcoded `-1`; SELECT (and statements the server does not report a count for) still return `-1`. `executemany` aggregates the count across all parameter sets per PEP 249 ([#784](https://github.com/databricks/databricks-sql-python/issues/784))
+
 # 4.3.0 (2026-06-12)
 - **New: optional Rust kernel backend (`use_kernel=True`).** Adds an alternative connection path backed by the native [`databricks-sql-kernel`](https://pypi.org/project/databricks-sql-kernel/) client (a Rust core exposed via PyO3), installable with the new `databricks-sql-connector[kernel]` extra. The kernel talks to Databricks over the **SEA (Statement Execution API) HTTP transport** — not Thrift — with CloudFetch and inline-Arrow result fetching, so `use_kernel=True` gives you a modern SEA-native client through the same DB-API surface. Supports PAT, OAuth M2M, and OAuth U2M auth. Requires Python >= 3.10 (the kernel wheel is `cp310-abi3`); on older interpreters the extra is a no-op and `use_kernel=True` raises a clear `ImportError`. The default backend remains Thrift — opt in per connection.
 - Kernel backend behavior is aligned with the Thrift backend so application code works the same either way: consistent cursor-state tracking (`query_id` / `get_query_state`), metadata (catalogs/schemas/tables/columns with JDBC-style filter semantics and case-insensitive `table_types`), DML `rowcount`, server-sourced async execution state, sync `cancel()`, fail-loud staging/volume operations, and structured error context (SQLSTATE, diagnostic info). Kernel logs surface through Python `logging` under the `databricks.sql.kernel` logger (databricks/databricks-sql-python#824, #825, #830, #838, #839 by @vikrantpuppala)

{databricks_sql_connector-4.3.0 → databricks_sql_connector-4.4.0}/PKG-INFO

@@ -1,16 +1,14 @@
 Metadata-Version: 2.4
 Name: databricks-sql-connector
-Version: 4.3.0
+Version: 4.4.0
 Summary: Databricks SQL Connector for Python
 License: Apache-2.0
 License-File: LICENSE
 Author: Databricks
 Author-email: databricks-sql-connector-maintainers@databricks.com
-Requires-Python: >=3.8.0,<4.0.0
+Requires-Python: >=3.10,<4.0
 Classifier: License :: OSI Approved :: Apache Software License
 Classifier: Programming Language :: Python :: 3
-Classifier: Programming Language :: Python :: 3.8
-Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
@@ -19,21 +17,20 @@ Classifier: Programming Language :: Python :: 3.14
 Provides-Extra: kernel
 Provides-Extra: pyarrow
 Requires-Dist: databricks-sql-kernel (>=0.2.0,<0.3.0) ; (python_version >= "3.10") and (extra == "kernel")
-Requires-Dist: lz4 (>=4.0.2,<5.0.0) ; python_version >= "3.8" and python_version < "3.14"
+Requires-Dist: lz4 (>=4.0.2,<5.0.0) ; python_version >= "3.10" and python_version < "3.14"
 Requires-Dist: lz4 (>=4.4.5,<5.0.0) ; python_version >= "3.14"
 Requires-Dist: oauthlib (>=3.1.0,<4.0.0)
 Requires-Dist: openpyxl (>=3.0.10,<4.0.0)
-Requires-Dist: pandas (>=1.2.5,<4.0.0) ; python_version >= "3.8" and python_version < "3.13"
+Requires-Dist: pandas (>=1.2.5,<4.0.0) ; python_version >= "3.10" and python_version < "3.13"
 Requires-Dist: pandas (>=2.2.3,<4.0.0) ; python_version >= "3.13"
 Requires-Dist: pyarrow (>=14.0.1) ; (python_version >= "3.10" and python_version < "3.13") and (extra == "pyarrow")
-Requires-Dist: pyarrow (>=14.0.1,<23) ; (python_version >= "3.8" and python_version < "3.10") and (extra == "pyarrow")
 Requires-Dist: pyarrow (>=18.0.0) ; (python_version == "3.13") and (extra == "pyarrow")
 Requires-Dist: pyarrow (>=22.0.0) ; (python_version >= "3.14") and (extra == "pyarrow")
 Requires-Dist: pybreaker (>=1.0.0,<2.0.0)
 Requires-Dist: pyjwt (>=2.0.0,<3.0.0)
 Requires-Dist: python-dateutil (>=2.8.0,<3.0.0)
 Requires-Dist: requests (>=2.18.1,<3.0.0)
-Requires-Dist: thrift (>=0.22.0,<0.23.0)
+Requires-Dist: thrift (>=0.24.0,<0.25.0)
 Requires-Dist: urllib3 (>=1.26)
 Project-URL: Bug Tracker, https://github.com/databricks/databricks-sql-python/issues
 Project-URL: Homepage, https://github.com/databricks/databricks-sql-python

{databricks_sql_connector-4.3.0 → databricks_sql_connector-4.4.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "databricks-sql-connector"
-version = "4.3.0"
+version = "4.4.0"
 description = "Databricks SQL Connector for Python"
 authors = ["Databricks <databricks-sql-connector-maintainers@databricks.com>"]
 license = "Apache-2.0"
@@ -9,14 +9,28 @@ packages = [{ include = "databricks", from = "src" }]
 include = ["CHANGELOG.md"]
 
 [tool.poetry.dependencies]
-python = "^3.8.0"
-thrift = "~=0.22.0"
+python = "^3.10"
+# Floor is 0.24.0 -- the first release that both clears the open Apache Thrift
+# CVEs (CVE-2025-48431 + the CVE-2026-41602..41636 set, all fixed in 0.23.0)
+# AND is safe to install on DBR LTS. History: 0.23.0 fixed the CVEs but ships
+# sdist-only and its setup.py calls sys.exit(0) on the build-success path,
+# killing the PEP 517 backend on the OLD setuptools bundled by DBR LTS -- the
+# SEV0 ES-1960554 (4.2.7 widened to <0.24.0 and was yanked; PR #840), which is
+# why we held at ~=0.22.0. thrift 0.24.0 (THRIFT-6067) resolves this: it ships
+# prebuilt manylinux2014 wheels (cp310-cp314) + macOS/musl/Windows, so pip uses
+# a wheel and never runs setup.py on DBR LTS -- the build-time break cannot
+# trigger. The `DBR LTS Install` CI check (.github/workflows/dbr-lts-install.yml)
+# installs the built artifact on real DBR LTS clusters and is the authoritative
+# gate for this. Cap at <0.25.0: thrift is pre-1.0, each 0.x minor can carry
+# breaking changes or packaging regressions (see 0.23.0), so bump this
+# deliberately once a new minor ships and the DBR-LTS gate proves it safe.
+thrift = "~=0.24.0"
 pandas = [
-    { version = ">=1.2.5,<4.0.0", python = ">=3.8,<3.13" },
+    { version = ">=1.2.5,<4.0.0", python = ">=3.10,<3.13" },
     { version = ">=2.2.3,<4.0.0", python = ">=3.13" }
 ]
 lz4 = [
-    { version = "^4.0.2", python = ">=3.8,<3.14" },
+    { version = "^4.0.2", python = ">=3.10,<3.14" },
     { version = "^4.4.5", python = ">=3.14" }
 ]
 requests = "^2.18.1"
@@ -25,15 +39,6 @@ openpyxl = "^3.0.10"
 urllib3 = ">=1.26"
 python-dateutil = "^2.8.0"
 pyarrow = [
-    # The <3.10 band is capped at <23 because pyarrow>=23 dropped
-    # Python 3.9 (it requires >=3.10). Without the cap, poetry tries to
-    # unify this entry with the kernel's transitive pyarrow>=23.0.1,<24
-    # across the 3.8–3.10 slice and `poetry lock` fails ("pyarrow is
-    # forbidden"). The cap removes no installable version — the newest
-    # pyarrow with a 3.9 wheel is 21.x — it just makes that explicit to
-    # the solver so the optional [kernel] extra (python>=3.10) can
-    # coexist. See the kernel dep + [kernel] extra below.
-    { version = ">=14.0.1,<23", python = ">=3.8,<3.10", optional=true },
     { version = ">=14.0.1", python = ">=3.10,<3.13", optional=true },
     { version = ">=18.0.0", python = ">=3.13,<3.14", optional=true },
     { version = ">=22.0.0", python = ">=3.14", optional=true }
@@ -77,15 +82,21 @@ pyarrow = ["pyarrow"]
 kernel = ["databricks-sql-kernel"]
 
 [tool.poetry.group.dev.dependencies]
-pytest = "^7.1.2"
+pytest = "^9.0.3"
 mypy = "^1.10.1"
 pylint = ">=2.12.0"
-black = "^22.3.0"
+black = "^26.3.1"
 pytest-dotenv = "^0.5.2"
 pytest-cov = "^4.0.0"
 pytest-xdist = "^3.0.0"
+# tests/unit/test_parameters.py imports pytz directly. It used to arrive
+# transitively via pandas, but pandas 3.0 (resolved on Python >=3.11)
+# dropped pytz as a required dependency -- it is now only a pandas extra.
+# Declare it explicitly here so the test suite does not depend on pandas's
+# transitive graph, which differs across the Python matrix.
+pytz = ">=2020.1"
 numpy = [
-    { version = ">=1.16.6", python = ">=3.8,<3.11" },
+    { version = ">=1.16.6", python = ">=3.10,<3.11" },
     { version = ">=1.23.4", python = ">=3.11" },
 ]
 

{databricks_sql_connector-4.3.0 → databricks_sql_connector-4.4.0}/src/databricks/sql/__init__.py

@@ -71,7 +71,7 @@ DATETIME = DBAPITypeObject("timestamp")
 DATE = DBAPITypeObject("date")
 ROWID = DBAPITypeObject()
 
-__version__ = "4.3.0"
+__version__ = "4.4.0"
 USER_AGENT_NAME = "PyDatabricksSqlConnector"
 
 # These two functions are pyhive legacy

{databricks_sql_connector-4.3.0 → databricks_sql_connector-4.4.0}/src/databricks/sql/auth/auth.py

@@ -102,7 +102,7 @@ def get_python_sql_connector_auth_provider(hostname: str, http_client, **kwargs)
     # TODO : unify all the auth mechanisms with the Python SDK
 
     auth_type = kwargs.get("auth_type")
-    (client_id, redirect_port_range) = get_client_id_and_redirect_port(
+    client_id, redirect_port_range = get_client_id_and_redirect_port(
         auth_type == AuthType.AZURE_OAUTH.value
     )
 
@@ -124,9 +124,11 @@ def get_python_sql_connector_auth_provider(hostname: str, http_client, **kwargs)
         azure_client_secret=kwargs.get("azure_client_secret"),
         azure_tenant_id=kwargs.get("azure_tenant_id"),
         azure_workspace_resource_id=kwargs.get("azure_workspace_resource_id"),
-        oauth_redirect_port_range=[kwargs["oauth_redirect_port"]]
-        if kwargs.get("oauth_client_id") and kwargs.get("oauth_redirect_port")
-        else redirect_port_range,
+        oauth_redirect_port_range=(
+            [kwargs["oauth_redirect_port"]]
+            if kwargs.get("oauth_client_id") and kwargs.get("oauth_redirect_port")
+            else redirect_port_range
+        ),
         oauth_persistence=kwargs.get("experimental_oauth_persistence"),
         credentials_provider=kwargs.get("credentials_provider"),
         identity_federation_client_id=kwargs.get("identity_federation_client_id"),

{databricks_sql_connector-4.3.0 → databricks_sql_connector-4.4.0}/src/databricks/sql/auth/authenticators.py

@@ -33,12 +33,10 @@ class CredentialsProvider(abc.ABC):
     for authenticating requests to Databricks REST APIs"""
 
     @abc.abstractmethod
-    def auth_type(self) -> str:
-        ...
+    def auth_type(self) -> str: ...
 
     @abc.abstractmethod
-    def __call__(self, *args, **kwargs) -> HeaderFactory:
-        ...
+    def __call__(self, *args, **kwargs) -> HeaderFactory: ...
 
 
 # Private API: this is an evolving interface and it will change in the future.
@@ -109,7 +107,7 @@ class DatabricksOAuthProvider(AuthProvider):
             if self._access_token and self._refresh_token:
                 self._update_token_if_expired()
             else:
-                (access_token, refresh_token) = self.oauth_manager.get_tokens(
+                access_token, refresh_token = self.oauth_manager.get_tokens(
                     hostname=self._hostname, scope=self._scopes_as_str
                 )
                 self._access_token = access_token
@@ -231,9 +229,9 @@ class AzureServicePrincipalCredentialProvider(CredentialsProvider):
             }
 
             if self.azure_workspace_resource_id:
-                headers[
-                    self.DATABRICKS_AZURE_WORKSPACE_RESOURCE_ID_HEADER
-                ] = self.azure_workspace_resource_id
+                headers[self.DATABRICKS_AZURE_WORKSPACE_RESOURCE_ID_HEADER] = (
+                    self.azure_workspace_resource_id
+                )
 
             return headers
 

{databricks_sql_connector-4.3.0 → databricks_sql_connector-4.4.0}/src/databricks/sql/auth/oauth.py

@@ -45,7 +45,7 @@ class Token:
             exp_time = decoded_token.get("exp")
             current_time = time.time()
             buffer_time = 30  # 30 seconds buffer
-            return exp_time and (exp_time - buffer_time) <= current_time
+            return exp_time is not None and (exp_time - buffer_time) <= current_time
         except Exception as e:
             logger.error("Failed to decode token: %s", e)
             raise e
@@ -134,7 +134,7 @@ class OAuthManager:
             try:
                 with HTTPServer(("", port), handler) as httpd:
                     redirect_url = OAuthManager.__get_redirect_url(port)
-                    (auth_req_uri, _, _) = client.prepare_authorization_request(
+                    auth_req_uri, _, _ = client.prepare_authorization_request(
                         authorization_url=auth_url,
                         redirect_url=redirect_url,
                         scope=scope,
@@ -269,7 +269,7 @@ class OAuthManager:
         auth_url = self.idp_endpoint.get_authorization_url(hostname)
 
         state = OAuthManager.__token_urlsafe(16)
-        (verifier, challenge) = OAuthManager.__get_challenge()
+        verifier, challenge = OAuthManager.__get_challenge()
         client = oauthlib.oauth2.WebApplicationClient(self.client_id)
 
         try:

{databricks_sql_connector-4.3.0 → databricks_sql_connector-4.4.0}/src/databricks/sql/auth/retry.py

@@ -245,7 +245,7 @@ class DatabricksRetryPolicy(Retry):
         return self._command_type
 
     @command_type.setter
-    def command_type(self, value: CommandType) -> None:
+    def command_type(self, value: Optional[CommandType]) -> None:
         self._command_type = value
 
     @property

{databricks_sql_connector-4.3.0 → databricks_sql_connector-4.4.0}/src/databricks/sql/auth/thrift_http_client.py

@@ -122,9 +122,11 @@ class THttpClient(thrift.transport.THttpClient.THttpClient):
             pool_class = HTTPSConnectionPool
             _pool_kwargs.update(
                 {
-                    "cert_reqs": ssl.CERT_REQUIRED
-                    if self._ssl_options.tls_verify
-                    else ssl.CERT_NONE,
+                    "cert_reqs": (
+                        ssl.CERT_REQUIRED
+                        if self._ssl_options.tls_verify
+                        else ssl.CERT_NONE
+                    ),
                     "ca_certs": self._ssl_options.tls_trusted_ca_file,
                     "cert_file": self._ssl_options.tls_client_cert_file,
                     "key_file": self._ssl_options.tls_client_cert_key_file,

{databricks_sql_connector-4.3.0 → databricks_sql_connector-4.4.0}/src/databricks/sql/auth/token_federation.py

@@ -111,9 +111,9 @@ class TokenFederationProvider(AuthProvider):
         """Add authentication headers to the request."""
 
         if self._cached_token and not self._cached_token.is_expired():
-            request_headers[
-                "Authorization"
-            ] = f"{self._cached_token.token_type} {self._cached_token.access_token}"
+            request_headers["Authorization"] = (
+                f"{self._cached_token.token_type} {self._cached_token.access_token}"
+            )
             return
 
         # Get the external headers first to check if we need token federation

{databricks_sql_connector-4.3.0 → databricks_sql_connector-4.4.0}/src/databricks/sql/backend/sea/utils/filters.py

@@ -227,10 +227,8 @@ class ResultSetFilter:
         if not case_sensitive:
             allowed_values = [v.upper() for v in allowed_values]
         # Helper lambda to get column value based on case sensitivity
-        get_column_value = (
-            lambda row: row[column_index].upper()
-            if not case_sensitive
-            else row[column_index]
+        get_column_value = lambda row: (
+            row[column_index].upper() if not case_sensitive else row[column_index]
         )
 
         # Filter rows based on allowed values

{databricks_sql_connector-4.3.0 → databricks_sql_connector-4.4.0}/src/databricks/sql/backend/sea/utils/http_client.py

@@ -161,9 +161,11 @@ class SeaHttpClient:
             pool_class = HTTPSConnectionPool
             pool_kwargs.update(
                 {
-                    "cert_reqs": ssl.CERT_REQUIRED
-                    if self.ssl_options.tls_verify
-                    else ssl.CERT_NONE,
+                    "cert_reqs": (
+                        ssl.CERT_REQUIRED
+                        if self.ssl_options.tls_verify
+                        else ssl.CERT_NONE
+                    ),
                     "ca_certs": self.ssl_options.tls_trusted_ca_file,
                     "cert_file": self.ssl_options.tls_client_cert_file,
                     "key_file": self.ssl_options.tls_client_cert_key_file,

{databricks_sql_connector-4.3.0 → databricks_sql_connector-4.4.0}/src/databricks/sql/backend/thrift_backend.py

@@ -12,7 +12,6 @@ from databricks.sql.common.unified_http_client import UnifiedHttpClient
 from databricks.sql.result_set import ThriftResultSet
 from databricks.sql.telemetry.models.event import StatementType
 
-
 if TYPE_CHECKING:
     from databricks.sql.client import Cursor
     from databricks.sql.result_set import ResultSet
@@ -680,7 +679,10 @@ class ThriftDatabricksClient(DatabricksClient):
                 num_rows,
             ) = convert_column_based_set_to_arrow_table(t_row_set.columns, description)
         elif t_row_set.arrowBatches is not None:
-            (arrow_table, num_rows,) = convert_arrow_based_set_to_arrow_table(
+            (
+                arrow_table,
+                num_rows,
+            ) = convert_arrow_based_set_to_arrow_table(
                 t_row_set.arrowBatches, lz4_compressed, schema_bytes
             )
         else:
@@ -801,7 +803,9 @@ class ThriftDatabricksClient(DatabricksClient):
             for col in t_table_schema.columns
         ]
 
-    def _results_message_to_execute_response(self, resp, operation_state):
+    def _results_message_to_execute_response(
+        self, resp, operation_state, num_modified_rows=None
+    ):
         if resp.directResults and resp.directResults.resultSetMetadata:
             t_result_set_metadata_resp = resp.directResults.resultSetMetadata
         else:
@@ -864,6 +868,7 @@ class ThriftDatabricksClient(DatabricksClient):
             is_staging_operation=t_result_set_metadata_resp.isStagingOperation,
             arrow_schema_bytes=schema_bytes,
             result_format=t_result_set_metadata_resp.resultFormat,
+            num_modified_rows=num_modified_rows,
         )
 
         return execute_response, has_more_rows
@@ -945,6 +950,7 @@ class ThriftDatabricksClient(DatabricksClient):
             self._check_command_not_in_error_or_closed_state(
                 op_handle, initial_operation_status_resp
             )
+        final_status_resp = initial_operation_status_resp
         operation_state = (
             initial_operation_status_resp
             and initial_operation_status_resp.operationState
@@ -956,7 +962,10 @@ class ThriftDatabricksClient(DatabricksClient):
             poll_resp = self._poll_for_status(op_handle)
             operation_state = poll_resp.operationState
             self._check_command_not_in_error_or_closed_state(op_handle, poll_resp)
-        return operation_state
+            final_status_resp = poll_resp
+        # Return the terminal status response (not just the state) so callers
+        # can read ``numModifiedRows`` — the DML affected-row count — from it.
+        return operation_state, final_status_resp
 
     def get_query_state(self, command_id: CommandId) -> CommandState:
         thrift_handle = command_id.to_thrift_handle()
@@ -1046,11 +1055,13 @@ class ThriftDatabricksClient(DatabricksClient):
             statement=operation,
             runAsync=True,
             # For async operation we don't want the direct results
-            getDirectResults=None
-            if async_op
-            else ttypes.TSparkGetDirectResults(
-                maxRows=max_rows,
-                maxBytes=max_bytes,
+            getDirectResults=(
+                None
+                if async_op
+                else ttypes.TSparkGetDirectResults(
+                    maxRows=max_rows,
+                    maxBytes=max_bytes,
+                )
             ),
             canReadArrowResult=True if pyarrow else False,
             canDecompressLZ4Result=lz4_compression,
@@ -1274,12 +1285,21 @@ class ThriftDatabricksClient(DatabricksClient):
         cursor.active_command_id = command_id
         self._check_direct_results_for_error(resp.directResults, self._host)
 
-        final_operation_state = self._wait_until_command_done(
+        final_operation_state, final_status_resp = self._wait_until_command_done(
             resp.operationHandle,
             resp.directResults and resp.directResults.operationStatus,
         )
 
-        return self._results_message_to_execute_response(resp, final_operation_state)
+        # ``numModifiedRows`` is populated by the server for DML statements
+        # (INSERT/UPDATE/DELETE/MERGE) and is None for SELECT. Surface it so it
+        # can flow to ``cursor.rowcount``.
+        num_modified_rows = (
+            final_status_resp.numModifiedRows if final_status_resp else None
+        )
+
+        return self._results_message_to_execute_response(
+            resp, final_operation_state, num_modified_rows
+        )
 
     def _handle_execute_response_async(self, resp, cursor):
         command_id = CommandId.from_thrift_handle(resp.operationHandle)

{databricks_sql_connector-4.3.0 → databricks_sql_connector-4.4.0}/src/databricks/sql/backend/types.py

@@ -425,3 +425,7 @@ class ExecuteResponse:
     is_staging_operation: bool = False
     arrow_schema_bytes: Optional[bytes] = None
     result_format: Optional[Any] = None
+    # Number of rows modified by a DML statement (INSERT/UPDATE/DELETE/MERGE),
+    # surfaced as ``cursor.rowcount``. ``None`` for SELECT and any statement
+    # for which the server does not report a count → ``rowcount`` stays at -1.
+    num_modified_rows: Optional[int] = None

{databricks_sql_connector-4.3.0 → databricks_sql_connector-4.4.0}/src/databricks/sql/client.py

@@ -340,9 +340,9 @@ class Connection:
                 http_path=http_path,
                 port=kwargs.get("_port", 443),
                 client_context=client_context,
-                user_agent=self.session.useragent_header
-                if hasattr(self, "session")
-                else None,
+                user_agent=(
+                    self.session.useragent_header if hasattr(self, "session") else None
+                ),
                 enable_telemetry=enable_telemetry,
             )
             raise e
@@ -390,9 +390,11 @@ class Connection:
 
         driver_connection_params = DriverConnectionParameters(
             http_path=http_path,
-            mode=DatabricksClientType.SEA
-            if self.session.use_sea
-            else DatabricksClientType.THRIFT,
+            mode=(
+                DatabricksClientType.SEA
+                if self.session.use_sea
+                else DatabricksClientType.THRIFT
+            ),
             host_info=HostDetails(host_url=server_hostname, port=self.session.port),
             auth_mech=TelemetryHelper.get_auth_mechanism(self.session.auth_provider),
             auth_flow=TelemetryHelper.get_auth_flow(self.session.auth_provider),
@@ -874,7 +876,10 @@ class Cursor:
 
         self.connection: Connection = connection
 
-        self.rowcount: int = -1  # Return -1 as this is not supported
+        # -1 until a statement runs. Set to the affected-row count after a DML
+        # statement (INSERT/UPDATE/DELETE/MERGE); stays -1 for SELECT and any
+        # statement the server does not report a modified-row count for.
+        self.rowcount: int = -1
         self.buffer_size_bytes: int = result_buffer_size_bytes
         self.active_result_set: Union[ResultSet, None] = None
         self.arraysize: int = arraysize
@@ -1039,6 +1044,9 @@ class Cursor:
                 self.active_result_set.close()
         finally:
             self.active_result_set = None
+            # Reset rowcount to its -1 default so a prior DML's count never
+            # leaks into a subsequent SELECT (or unreported) statement.
+            self.rowcount = -1
 
     def _check_not_closed(self):
         if not self.open:
@@ -1081,6 +1089,14 @@ class Cursor:
                 headers=headers,
             )
 
+        # REMOVE deletes a remote resource and never touches the local
+        # filesystem, so it does not require staging_allowed_local_path.
+        if row.operation == "REMOVE":
+            return self._handle_staging_remove(
+                presigned_url=row.presignedUrl,
+                headers=headers,
+            )
+
         # For non-streaming operations, validate staging_allowed_local_path
         if isinstance(staging_allowed_local_path, type(str())):
             _staging_allowed_local_paths = [staging_allowed_local_path]
@@ -1133,14 +1149,12 @@ class Cursor:
         )
 
         # TODO: Create a retry loop here to re-attempt if the request times out or fails
+        # REMOVE is handled above, before staging_allowed_local_path validation,
+        # since it does not touch the local filesystem.
         if row.operation == "GET":
             return self._handle_staging_get(**handler_args)
         elif row.operation == "PUT":
             return self._handle_staging_put(**handler_args)
-        elif row.operation == "REMOVE":
-            # Local file isn't needed to remove a remote resource
-            handler_args.pop("local_file")
-            return self._handle_staging_remove(**handler_args)
         else:
             raise ProgrammingError(
                 f"Operation {row.operation} is not supported. "
@@ -1367,6 +1381,14 @@ class Cursor:
             query_tags=query_tags,
         )
 
+        # Surface the affected-row count for DML (INSERT/UPDATE/DELETE/MERGE) as
+        # cursor.rowcount instead of the hardcoded -1. num_modified_rows is None
+        # for SELECT (and statements the server does not report a count for) →
+        # leave rowcount at its -1 default.
+        num_modified_rows = getattr(self.active_result_set, "num_modified_rows", None)
+        if num_modified_rows is not None:
+            self.rowcount = num_modified_rows
+
         if self.active_result_set and self.active_result_set.is_staging_operation:
             self._handle_staging_operation(
                 staging_allowed_local_path=self.connection.staging_allowed_local_path,
@@ -1505,8 +1527,22 @@ class Cursor:
 
         :returns self
         """
+        # Per PEP 249, rowcount after executemany reflects the total rows
+        # affected across all parameter sets (or -1 when undeterminable). Each
+        # execute() resets self.rowcount and sets it from its own statement, so
+        # we accumulate the reported counts here. If no statement reports a
+        # count (e.g. all SELECT, or the server does not report one), rowcount
+        # stays at its -1 default.
+        total_rowcount = -1
         for parameters in seq_of_parameters:
             self.execute(operation, parameters, query_tags=query_tags)
+            if self.rowcount >= 0:
+                total_rowcount = (
+                    self.rowcount
+                    if total_rowcount < 0
+                    else total_rowcount + self.rowcount
+                )
+        self.rowcount = total_rowcount
         return self
 
     @log_latency(StatementType.METADATA)

{databricks_sql_connector-4.3.0 → databricks_sql_connector-4.4.0}/src/databricks/sql/common/unified_http_client.py

@@ -148,11 +148,13 @@ class UnifiedHttpClient:
             "num_pools": self.config.pool_connections,
             "maxsize": self.config.pool_maxsize,
             "retries": self._retry_policy,
-            "timeout": urllib3.Timeout(
-                connect=self.config.socket_timeout, read=self.config.socket_timeout
-            )
-            if self.config.socket_timeout
-            else None,
+            "timeout": (
+                urllib3.Timeout(
+                    connect=self.config.socket_timeout, read=self.config.socket_timeout
+                )
+                if self.config.socket_timeout
+                else None
+            ),
             "ssl_context": ssl_context,
         }
 

{databricks_sql_connector-4.3.0 → databricks_sql_connector-4.4.0}/src/databricks/sql/exc.py

@@ -3,6 +3,7 @@ import logging
 
 logger = logging.getLogger(__name__)
 
+
 ### PEP-249 Mandated ###
 # https://peps.python.org/pep-0249/#exceptions
 class Error(Exception):
@@ -155,7 +156,8 @@ class CursorAlreadyClosedError(RequestError):
 
 class TelemetryRateLimitError(Exception):
     """Raised when telemetry endpoint returns 429 or 503, indicating rate limiting or service unavailable.
-    This exception is used exclusively by the circuit breaker to track telemetry rate limiting events."""
+    This exception is used exclusively by the circuit breaker to track telemetry rate limiting events.
+    """
 
 
 class TelemetryNonRateLimitError(Exception):

{databricks_sql_connector-4.3.0 → databricks_sql_connector-4.4.0}/src/databricks/sql/result_set.py

@@ -50,6 +50,7 @@ class ResultSet(ABC):
         is_staging_operation: bool = False,
         lz4_compressed: bool = False,
         arrow_schema_bytes: Optional[bytes] = None,
+        num_modified_rows: Optional[int] = None,
     ):
         """
         A ResultSet manages the results of a single command.
@@ -82,6 +83,8 @@ class ResultSet(ABC):
         self._is_staging_operation = is_staging_operation
         self.lz4_compressed = lz4_compressed
         self._arrow_schema_bytes = arrow_schema_bytes
+        # Affected-row count for DML; None for SELECT / unreported.
+        self.num_modified_rows = num_modified_rows
 
     def __iter__(self):
         while True:
@@ -264,6 +267,7 @@ class ThriftResultSet(ResultSet):
             is_staging_operation=execute_response.is_staging_operation,
             lz4_compressed=execute_response.lz4_compressed,
             arrow_schema_bytes=execute_response.arrow_schema_bytes,
+            num_modified_rows=execute_response.num_modified_rows,
         )
 
         # Initialize results queue if not provided

{databricks_sql_connector-4.3.0 → databricks_sql_connector-4.4.0}/src/databricks/sql/telemetry/telemetry_client.py

@@ -490,9 +490,9 @@ class TelemetryClientFactory:
     on the telemetry endpoint.
     """
 
-    _clients: Dict[
-        str, _TelemetryClientHolder
-    ] = {}  # Map of host_url -> TelemetryClientHolder
+    _clients: Dict[str, _TelemetryClientHolder] = (
+        {}
+    )  # Map of host_url -> TelemetryClientHolder
     _executor: Optional[ThreadPoolExecutor] = None
     _initialized: bool = False
     _lock = threading.RLock()  # Thread safety for factory operations
@@ -639,13 +639,13 @@ class TelemetryClientFactory:
                             client_context=client_context,
                             extra_headers=extra_headers,
                         )
-                        TelemetryClientFactory._clients[
-                            host_url
-                        ] = _TelemetryClientHolder(client)
+                        TelemetryClientFactory._clients[host_url] = (
+                            _TelemetryClientHolder(client)
+                        )
                     else:
-                        TelemetryClientFactory._clients[
-                            host_url
-                        ] = _TelemetryClientHolder(NoopTelemetryClient())
+                        TelemetryClientFactory._clients[host_url] = (
+                            _TelemetryClientHolder(NoopTelemetryClient())
+                        )
         except Exception as e:
             logger.debug("Failed to initialize telemetry client: %s", e)
             # Fallback to NoopTelemetryClient to ensure connection doesn't fail

{databricks_sql_connector-4.3.0 → databricks_sql_connector-4.4.0}/src/databricks/sql/utils.py

@@ -876,7 +876,7 @@ def _create_python_tuple(t_col_value_wrapper):
 
 
 def concat_table_chunks(
-    table_chunks: List[Union["pyarrow.Table", ColumnTable]]
+    table_chunks: List[Union["pyarrow.Table", ColumnTable]],
 ) -> Union["pyarrow.Table", ColumnTable]:
     if len(table_chunks) == 0:
         return table_chunks
@@ -894,11 +894,27 @@ def concat_table_chunks(
                 result_table[j].extend(table_chunks[i].column_table[j])
         return ColumnTable(result_table, table_chunks[0].column_names)
     else:
+        return _concat_arrow_tables(table_chunks)
+
+
+def _concat_arrow_tables(table_chunks: List["pyarrow.Table"]) -> "pyarrow.Table":
+    """Concatenate Arrow tables, tolerant of the installed pyarrow version.
+
+    ``promote_options`` was added in pyarrow 14.0.0, replacing the older
+    ``promote`` boolean. The connector's declared floor is pyarrow>=14, but a
+    base install (no ``[pyarrow]`` extra) can run against a runtime's older
+    bundled pyarrow -- e.g. DBR 13.3 (12.x) / 14.3 -- where ``promote_options``
+    raises ``TypeError: unexpected keyword argument``. Fall back to the legacy
+    ``promote=True`` there (equivalent to ``promote_options="default"``).
+    """
+    try:
         return pyarrow.concat_tables(table_chunks, promote_options="default")
+    except TypeError:
+        return pyarrow.concat_tables(table_chunks, promote=True)
 
 
 def serialize_query_tags(
-    query_tags: Optional[Dict[str, Optional[str]]]
+    query_tags: Optional[Dict[str, Optional[str]]],
 ) -> Optional[str]:
     """
     Serialize query_tags dictionary to a string format.