google-pso-data-validator 8.9.2__tar.gz → 8.9.3__tar.gz

{google_pso_data_validator-8.9.2 → google_pso_data_validator-8.9.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: google-pso-data-validator
-Version: 8.9.2
+Version: 8.9.3
 Summary: A package to enable easy data validation
 Author: PSO DVT Engineering team
 Author-email: data-validator-eng@google.com
@@ -10,6 +10,7 @@ Classifier: Programming Language :: Python
 Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
 Classifier: Operating System :: OS Independent
 Requires-Python: >=3.9
 Description-Content-Type: text/markdown

{google_pso_data_validator-8.9.2 → google_pso_data_validator-8.9.3}/google_pso_data_validator.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: google-pso-data-validator
-Version: 8.9.2
+Version: 8.9.3
 Summary: A package to enable easy data validation
 Author: PSO DVT Engineering team
 Author-email: data-validator-eng@google.com
@@ -10,6 +10,7 @@ Classifier: Programming Language :: Python
 Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
 Classifier: Operating System :: OS Independent
 Requires-Python: >=3.9
 Description-Content-Type: text/markdown

{google_pso_data_validator-8.9.2 → google_pso_data_validator-8.9.3}/setup.py

@@ -18,7 +18,7 @@ import setuptools
 
 name = "google-pso-data-validator"
 description = "A package to enable easy data validation"
-version = "8.9.2"
+version = "8.9.3"
 release_status = "Development Status :: 4 - Beta"
 
 with open("README.md", "r") as fh:
@@ -90,6 +90,7 @@ setuptools.setup(
         "Programming Language :: Python :: 3.9",
         "Programming Language :: Python :: 3.10",
         "Programming Language :: Python :: 3.11",
+        "Programming Language :: Python :: 3.12",
         "Operating System :: OS Independent",
     ],
     python_requires=">=3.9",

{google_pso_data_validator-8.9.2 → google_pso_data_validator-8.9.3}/third_party/ibis/ibis_oracle/__init__.py

@@ -12,6 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+import re
 from typing import Iterable, Literal, Optional, Tuple, Dict, Any
 
 import sqlalchemy as sa
@@ -20,7 +21,6 @@ from sqlalchemy.dialects.oracle.base import (
     RESERVED_WORDS as ORACLE_RESERVED_WORDS,
 )
 from sqlalchemy.dialects.oracle.oracledb import OracleDialect_oracledb
-import re
 
 import ibis.expr.datatypes as dt
 from ibis.backends.base.sql.alchemy import BaseAlchemyBackend
@@ -37,6 +37,55 @@ EXTRA_RESERVED_WORDS = set(
 )
 
 
+# We use a hardcoded list instead of checking the `oracle_maintained` column
+# in `all_users` because we want more granular control over which schemas to ignore.
+# For example, some customers use SYSTEM and DBSNMP for their own monitoring tables.
+ORACLE_SYSTEM_SCHEMAS = {
+    "ANONYMOUS",
+    "APPQOSSYS",
+    "AUDSYS",
+    "CTXSYS",
+    "DBSFWUSER",
+    "DGPDB_INT",
+    "DIP",
+    "DVF",
+    "DVSYS",
+    "EXFSYS",
+    "FLOWS_FILES",
+    "GGSYS",
+    "GSMADMIN_INTERNAL",
+    "GSMCATUSER",
+    "GSMUSER",
+    "LBACSYS",
+    "MDDATA",
+    "MDSYS",
+    "MGMT_VIEW",
+    "OJVMSYS",
+    "OLAPSYS",
+    "ORACLE_OCM",
+    "ORDDATA",
+    "ORDPLUGINS",
+    "ORDSYS",
+    "OUTLN",
+    "PDBADMIN",
+    "REMOTE_SCHEDULER_AGENT",
+    "SI_INFORMTN_SCHEMA",
+    "SPATIAL_CSW_ADMIN_USR",
+    "SPATIAL_WFS_ADMIN_USR",
+    "SYS",
+    "SYS$UMF",
+    "SYSBACKUP",
+    "SYSDG",
+    "SYSKM",
+    "SYSMAN",
+    "SYSRAC",
+    "TSMSYS",
+    "WMSYS",
+    "XDB",
+    "XS$NULL",
+}
+
+
 def _ora_denormalize_name(self, name):
     """Oracle specific version of sqlalchemy/engine/default.py.denormalize_name()
 
@@ -252,8 +301,25 @@ class Backend(BaseAlchemyBackend):
         padded values, which DVT may want to trim"""
         return char_type[0] in ["CHAR", "NCHAR"]
 
-    def list_databases(self) -> list:
-        return sa.inspect(self.con).get_schema_names()
+    def list_databases(self, like: Optional[str] = None) -> list[str]:
+        """Return a list of user schemas in the Oracle database.
+
+        Excludes Oracle-maintained system schemas to avoid performance bottlenecks
+        and permission errors during table discovery.
+
+        Args:
+            like: Optional pattern string to filter returned schema names.
+
+        Returns:
+            list[str]: Filtered list of schema names.
+        """
+        all_schemas = sa.inspect(self.con).get_schema_names()
+        schemas = [
+            schema
+            for schema in all_schemas
+            if schema.upper() not in ORACLE_SYSTEM_SCHEMAS
+        ]
+        return self._filter_with_like(schemas, like)
 
     def dvt_tuple_in_supported(self) -> bool:
         """Return True if backend client supports native SQL tuple/struct IN expressions."""

{google_pso_data_validator-8.9.2 → google_pso_data_validator-8.9.3}/third_party/ibis/ibis_teradata/__init__.py

@@ -38,23 +38,25 @@ class Backend(BaseSQLBackend):
 
     def do_connect(
         self,
-        host: str = "localhost",
+        host: str = None,
         user_name: str = None,
         password: str = None,
-        port: int = 1025,
-        logmech: str = "TD2",
+        port: int = None,
+        logmech: str = None,
         use_no_lock_tables: str = "False",
         json_params: Mapping[str, Any] = None,
     ) -> None:
-        self.teradata_config = {
+        self.teradata_config = dict(json_params) if json_params else {}
+        explicit_config = {
             "host": host,
             "user": user_name,
             "password": password,
             "dbs_port": port,
             "logmech": logmech,
         }
-        if json_params:
-            self.teradata_config.update(json_params)
+        self.teradata_config.update(
+            {key: value for key, value in explicit_config.items() if value is not None}
+        )
 
         self.client = teradatasql.connect(**self.teradata_config)
         self.con = self.client.cursor()

{google_pso_data_validator-8.9.2 → google_pso_data_validator-8.9.3}/third_party/ibis/ibis_teradata/api.py

@@ -18,11 +18,11 @@ from data_validation.util import dvt_config_string_to_dict
 
 
 def teradata_connect(
-    host: str = "localhost",
+    host: str = None,
     user_name: str = None,
     password: str = None,
-    port: int = 1025,
-    logmech: str = "TD2",
+    port: int = None,
+    logmech: str = None,
     use_no_lock_tables: str = "False",
     json_params: str = None,
 ):