snowflake-data-validation 1.22.0__tar.gz → 1.23.0__tar.gz

{snowflake_data_validation-1.22.0 → snowflake_data_validation-1.23.0}/.gitignore

@@ -93,6 +93,20 @@ docs/_build/
 # PyBuilder
 .pybuilder/
 target/
+# DV target-platform packages (not PyBuilder ``target/`` build output)
+!data-migration-orchestrator/src/data_migration_orchestrator/data_validation_cloud/platforms/target/
+!data-migration-orchestrator/src/data_migration_orchestrator/data_validation_cloud/platforms/target/**
+!data-migration-orchestrator/tests/platforms/target/
+!data-migration-orchestrator/tests/platforms/target/**
+!snowflake-data-validation/src/snowflake/snowflake_data_validation/target/
+!snowflake-data-validation/src/snowflake/snowflake_data_validation/target/**
+# Re-ignore bytecode under the un-ignored target/ trees (negation above overrides __pycache__/)
+data-migration-orchestrator/src/data_migration_orchestrator/data_validation_cloud/platforms/target/**/__pycache__/
+data-migration-orchestrator/tests/platforms/target/**/__pycache__/
+snowflake-data-validation/src/snowflake/snowflake_data_validation/target/**/__pycache__/
+data-migration-orchestrator/src/data_migration_orchestrator/data_validation_cloud/platforms/target/**/*.pyc
+data-migration-orchestrator/tests/platforms/target/**/*.pyc
+snowflake-data-validation/src/snowflake/snowflake_data_validation/target/**/*.pyc
 
 # Jupyter Notebook
 .ipynb_checkpoints

{snowflake_data_validation-1.22.0 → snowflake_data_validation-1.23.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: snowflake-data-validation
-Version: 1.22.0
+Version: 1.23.0
 Summary: Snowflake Data Validation
 Project-URL: Bug Tracker, https://github.com/snowflake-eng/migrations-data-validation/issues
 Project-URL: Source code, https://github.com/snowflake-eng/migrations-data-validation/
@@ -56,6 +56,7 @@ Requires-Dist: pytest-mock>=3.14.1; extra == 'all'
 Requires-Dist: pytest>=8.3.3; extra == 'all'
 Requires-Dist: redshift-connector>=2.1.8; extra == 'all'
 Requires-Dist: setuptools>=70.0.0; extra == 'all'
+Requires-Dist: sqlean-py>=3.45.0; extra == 'all'
 Requires-Dist: teradatasql>=17.20.0.0; extra == 'all'
 Requires-Dist: twine==6.0.1; extra == 'all'
 Provides-Extra: azure-synapse
@@ -74,6 +75,8 @@ Provides-Extra: postgresql
 Requires-Dist: psycopg[binary]>=3.2.0; extra == 'postgresql'
 Provides-Extra: redshift
 Requires-Dist: redshift-connector>=2.1.8; extra == 'redshift'
+Provides-Extra: sqlite
+Requires-Dist: sqlean-py>=3.45.0; extra == 'sqlite'
 Provides-Extra: sqlserver
 Requires-Dist: azure-identity>=1.15.0; extra == 'sqlserver'
 Requires-Dist: pyodbc>=5.2.0; extra == 'sqlserver'
@@ -118,6 +121,14 @@ For SQL Server support:
 pip install "snowflake-data-validation[sqlserver]"
 ```
 
+For SQLite support:
+
+```bash
+pip install "snowflake-data-validation[sqlite]"
+```
+
+> **SQLite prerequisite:** This extra installs [sqlean.py](https://github.com/nalgeon/sqlean), a drop-in `sqlite3` replacement that bundles the [sqlean-crypto](https://github.com/nalgeon/sqlean/blob/main/docs/crypto.md) extension. The `crypto_md5` function provided by this extension is required for row-level MD5 computation during SQLite data validation. Wheels are available for **Linux and macOS** only; Windows is not supported by `sqlean.py`.
+
 For development and testing:
 
 ```bash
@@ -205,6 +216,50 @@ For complete command reference, configuration options, and examples, see the [Da
 
 ## Changelog
 
+### v1.23.0
+
+**New features**
+
+- Added configurable text comparison mode (`logical` vs `raw`) for column-level validation.
+- Added SQLite as a supported source platform for data validation.
+
+**Improvements**
+
+- Implemented Oracle ISO 8601 timestamp normalization for Snowflake migration.
+- Enhanced timestamp normalization for Teradata with fractional-second precision.
+- Implemented timestamp timezone preservation for Azure Synapse validation.
+- Decoupled target platform for data validation, enabling non-Snowflake target comparisons.
+
+**Bug fixes**
+
+- Fixed row-hash early-stop so all failed rows are reported when early stopping is disabled.
+- Fixed empty-source handling to record an explicit validation failure instead of silently passing.
+
+### v1.22.1
+
+No user-facing changes in this release.
+
+### v1.22.0
+
+**New features**
+
+- Added accepted transformations (V1) for data validation.
+
+**Improvements**
+
+- Preserved ISO 8601 timestamp timezone for SQL Server validation.
+- Preserved ISO 8601 timestamp timezone for Teradata validation.
+- Replaced the pipe with the `AND` keyword in `WHERE` clause generation.
+- Aligned Postgres `BYTEA` normalization with Snowflake `BINARY` `0x` prefix.
+- Aligned Oracle `RAW`/`BLOB` normalization with Snowflake `BINARY` `0x` prefix.
+
+**Bug fixes**
+
+- Implemented a NULL hashing sentinel for SQL templates.
+- Fixed SQL Server datetime normalization for L2/L3 validation.
+- Fixed Oracle `DATE`/`TIMESTAMP` normalization to match Snowflake ISO FF9.
+- Replaced `TRIM` with `RTRIM` in L2/L3 validation query templates.
+
 ### v1.21.0
 
 **Improvements**

{snowflake_data_validation-1.22.0 → snowflake_data_validation-1.23.0}/README.md

@@ -35,6 +35,14 @@ For SQL Server support:
 pip install "snowflake-data-validation[sqlserver]"
 ```
 
+For SQLite support:
+
+```bash
+pip install "snowflake-data-validation[sqlite]"
+```
+
+> **SQLite prerequisite:** This extra installs [sqlean.py](https://github.com/nalgeon/sqlean), a drop-in `sqlite3` replacement that bundles the [sqlean-crypto](https://github.com/nalgeon/sqlean/blob/main/docs/crypto.md) extension. The `crypto_md5` function provided by this extension is required for row-level MD5 computation during SQLite data validation. Wheels are available for **Linux and macOS** only; Windows is not supported by `sqlean.py`.
+
 For development and testing:
 
 ```bash
@@ -122,6 +130,50 @@ For complete command reference, configuration options, and examples, see the [Da
 
 ## Changelog
 
+### v1.23.0
+
+**New features**
+
+- Added configurable text comparison mode (`logical` vs `raw`) for column-level validation.
+- Added SQLite as a supported source platform for data validation.
+
+**Improvements**
+
+- Implemented Oracle ISO 8601 timestamp normalization for Snowflake migration.
+- Enhanced timestamp normalization for Teradata with fractional-second precision.
+- Implemented timestamp timezone preservation for Azure Synapse validation.
+- Decoupled target platform for data validation, enabling non-Snowflake target comparisons.
+
+**Bug fixes**
+
+- Fixed row-hash early-stop so all failed rows are reported when early stopping is disabled.
+- Fixed empty-source handling to record an explicit validation failure instead of silently passing.
+
+### v1.22.1
+
+No user-facing changes in this release.
+
+### v1.22.0
+
+**New features**
+
+- Added accepted transformations (V1) for data validation.
+
+**Improvements**
+
+- Preserved ISO 8601 timestamp timezone for SQL Server validation.
+- Preserved ISO 8601 timestamp timezone for Teradata validation.
+- Replaced the pipe with the `AND` keyword in `WHERE` clause generation.
+- Aligned Postgres `BYTEA` normalization with Snowflake `BINARY` `0x` prefix.
+- Aligned Oracle `RAW`/`BLOB` normalization with Snowflake `BINARY` `0x` prefix.
+
+**Bug fixes**
+
+- Implemented a NULL hashing sentinel for SQL templates.
+- Fixed SQL Server datetime normalization for L2/L3 validation.
+- Fixed Oracle `DATE`/`TIMESTAMP` normalization to match Snowflake ISO FF9.
+- Replaced `TRIM` with `RTRIM` in L2/L3 validation query templates.
+
 ### v1.21.0
 
 **Improvements**

{snowflake_data_validation-1.22.0 → snowflake_data_validation-1.23.0}/pyproject.toml

@@ -65,7 +65,7 @@ sdv = "snowflake.snowflake_data_validation.main_cli:data_validation_app"
 snowflake-data-validation = "snowflake.snowflake_data_validation.main_cli:data_validation_app"
 
 [project.optional-dependencies]
-all = ["snowflake-data-validation[development,sqlserver,teradata,redshift,postgresql,azure-synapse]"]
+all = ["snowflake-data-validation[development,sqlserver,teradata,redshift,postgresql,azure-synapse,sqlite]"]
 development = [
   "pytest>=8.3.3",
   "pytest-cov>=6.0.0",
@@ -84,6 +84,9 @@ sqlserver = [
 teradata = [
   "teradatasql>=17.20.0.0",
 ]
+sqlite = [
+  "sqlean.py>=3.45.0",
+]
 redshift = [
   "redshift-connector>=2.1.8",
 ]

{snowflake_data_validation-1.22.0 → snowflake_data_validation-1.23.0}/src/snowflake/snowflake_data_validation/azure_synapse/azure_synapse_arguments_manager.py

@@ -38,9 +38,9 @@ from snowflake.snowflake_data_validation.utils.constants import (
 class AzureSynapseArgumentsManager(ArgumentsManagerBase):
     """Arguments manager for Azure Synapse Analytics connections and configuration."""
 
-    def __init__(self) -> None:
+    def __init__(self, target_platform: Platform = Platform.SNOWFLAKE) -> None:
         """Initialize the Azure Synapse arguments manager."""
-        super().__init__(Platform.AZURE_SYNAPSE, Platform.SNOWFLAKE)
+        super().__init__(Platform.AZURE_SYNAPSE, target_platform)
 
     @property
     def is_snowflake_to_snowflake(self) -> bool:

{snowflake_data_validation-1.22.0 → snowflake_data_validation-1.23.0}/src/snowflake/snowflake_data_validation/azure_synapse/extractor/templates/azure_synapse_column_metrics_templates.yaml

@@ -424,9 +424,11 @@ DATETIME:
   min:
     metric_query: MIN("{{ col_name }}")
     metric_return_datatype: DATETIME
+    metric_normalization: FORMAT({{ metric_query }}, 'yyyy-MM-dd''T''HH:mm:ss.fffffff') + '00'
   max:
     metric_query: MAX("{{ col_name }}")
     metric_return_datatype: DATETIME
+    metric_normalization: FORMAT({{ metric_query }}, 'yyyy-MM-dd''T''HH:mm:ss.fffffff') + '00'
   count_null:
     metric_query: SUM(CASE WHEN "{{ col_name }}" IS NULL THEN 1 ELSE 0 END)
     metric_return_datatype: INT
@@ -435,9 +437,11 @@ DATETIME2:
   min:
     metric_query: MIN("{{ col_name }}")
     metric_return_datatype: DATETIME2
+    metric_normalization: FORMAT({{ metric_query }}, 'yyyy-MM-dd''T''HH:mm:ss.fffffff') + '00'
   max:
     metric_query: MAX("{{ col_name }}")
     metric_return_datatype: DATETIME2
+    metric_normalization: FORMAT({{ metric_query }}, 'yyyy-MM-dd''T''HH:mm:ss.fffffff') + '00'
   count_null:
     metric_query: SUM(CASE WHEN "{{ col_name }}" IS NULL THEN 1 ELSE 0 END)
     metric_return_datatype: INT
@@ -446,9 +450,11 @@ SMALLDATETIME:
   min:
     metric_query: MIN("{{ col_name }}")
     metric_return_datatype: SMALLDATETIME
+    metric_normalization: FORMAT({{ metric_query }}, 'yyyy-MM-dd''T''HH:mm:ss.000000000')
   max:
     metric_query: MAX("{{ col_name }}")
     metric_return_datatype: SMALLDATETIME
+    metric_normalization: FORMAT({{ metric_query }}, 'yyyy-MM-dd''T''HH:mm:ss.000000000')
   count_null:
     metric_query: SUM(CASE WHEN "{{ col_name }}" IS NULL THEN 1 ELSE 0 END)
     metric_return_datatype: INT
@@ -457,9 +463,11 @@ DATETIMEOFFSET:
   min:
     metric_query: MIN("{{ col_name }}")
     metric_return_datatype: DATETIMEOFFSET
+    metric_normalization: FORMAT({{ metric_query }}, 'yyyy-MM-dd''T''HH:mm:ss.fffffff') + '00' + FORMAT({{ metric_query }}, 'zzz')
   max:
     metric_query: MAX("{{ col_name }}")
     metric_return_datatype: DATETIMEOFFSET
+    metric_normalization: FORMAT({{ metric_query }}, 'yyyy-MM-dd''T''HH:mm:ss.fffffff') + '00' + FORMAT({{ metric_query }}, 'zzz')
   count_null:
     metric_query: SUM(CASE WHEN "{{ col_name }}" IS NULL THEN 1 ELSE 0 END)
     metric_return_datatype: INT

{snowflake_data_validation-1.22.0 → snowflake_data_validation-1.23.0}/src/snowflake/snowflake_data_validation/azure_synapse/extractor/templates/azure_synapse_datatypes_normalization_templates.yaml

@@ -32,12 +32,15 @@ NCHAR: TRIM("{{ col_name }}")
 NVARCHAR: TRIM("{{ col_name }}")
 
 # ---- Date / Time types ----
+# Append '00' to FORMAT('.fffffff') so the output reaches FF9 (Snowflake TIMESTAMP_NTZ format).
+# Use '+' (not CONCAT) so a NULL input propagates to NULL — CONCAT would coerce NULL to ''.
+# FORMAT(..., 'zzz') reliably emits '+HH:MM'/'-HH:MM' for every offset, including UTC.
 DATE: FORMAT("{{ col_name }}", 'yyyy-MM-dd')
 TIME: CONVERT(VARCHAR(8), "{{ col_name }}", 108)
-DATETIME: FORMAT("{{ col_name }}", 'yyyy-MM-dd-HH:mm:ss')
-DATETIME2: FORMAT("{{ col_name }}", 'yyyy-MM-dd-HH:mm:ss')
-SMALLDATETIME: FORMAT("{{ col_name }}", 'yyyy-MM-dd-HH:mm:ss')
-DATETIMEOFFSET: FORMAT("{{ col_name }}" AT TIME ZONE 'UTC', 'yyyy-MM-dd-HH:mm:ss')
+DATETIME: FORMAT("{{ col_name }}", 'yyyy-MM-dd''T''HH:mm:ss.fffffff') + '00'
+DATETIME2: FORMAT("{{ col_name }}", 'yyyy-MM-dd''T''HH:mm:ss.fffffff') + '00'
+SMALLDATETIME: FORMAT("{{ col_name }}", 'yyyy-MM-dd''T''HH:mm:ss.000000000')
+DATETIMEOFFSET: FORMAT("{{ col_name }}", 'yyyy-MM-dd''T''HH:mm:ss.fffffff') + '00' + FORMAT("{{ col_name }}", 'zzz')
 
 # ---- Binary types ----
 BINARY: CONVERT(VARCHAR(MAX), "{{ col_name }}", 2)

{snowflake_data_validation-1.22.0 → snowflake_data_validation-1.23.0}/src/snowflake/snowflake_data_validation/azure_synapse/query/query_generator_azure_synapse.py

@@ -83,15 +83,22 @@ class QueryGeneratorAzureSynapse(QueryGeneratorBase):
         table_context: TableContext,
         *,
         order_by: bool = False,
+        column_codec_map=None,  # accepted for interface compatibility; not applied on this platform
     ) -> str:
         """Generate an inline SELECT returning ``(index_cols, ROW_MD5)``.
 
         Uses ``HASHBYTES('MD5', ...)`` and ``CONVERT(VARCHAR(32), ..., 2)``
         syntax via the ``azure_synapse_row_md5_select.sql.j2`` template.
         """
-        return self._build_row_md5_select(table_context, order_by=order_by)
+        return self._build_row_md5_select(table_context, order_by=order_by, column_codec_map=column_codec_map)
 
-    def _build_row_md5_select(self, table_context: TableContext, *, order_by: bool = False) -> str:
+    def _build_row_md5_select(
+        self,
+        table_context: TableContext,
+        *,
+        order_by: bool = False,
+        column_codec_map=None,  # accepted for interface compatibility; not applied on this platform
+    ) -> str:
         """Synapse-specific row MD5 SELECT (case-insensitive type template lookup).
 
         L1 ``columns_metadata`` stores source types in lowercase (e.g. int,
@@ -121,6 +128,7 @@ class QueryGeneratorAzureSynapse(QueryGeneratorBase):
         self,
         table_context: TableContext,
         order_by_columns: list[str] | None = None,
+        column_codec_map=None,  # accepted for interface compatibility; not applied on this platform
     ) -> str:
         """SELECT with per-column normalization matching L3 MD5 templates.
 

{snowflake_data_validation-1.22.0 → snowflake_data_validation-1.23.0}/src/snowflake/snowflake_data_validation/configuration/model/configuration_model.py

@@ -82,24 +82,33 @@ class ConfigurationModel(BaseModel):
         return self
 
     def check_tables(self) -> None:
+        if not self.tables:
+            return
+        target_platform = Platform(self.target_platform.lower())
         table: TableConfiguration
         for table in self.tables:
             if self.target_database and table.target_database is None:
                 self._add_database(object=table)
-            self._apply_mappings_to_table(table)
+            self._apply_mappings_to_table(table, target_platform=target_platform)
             self._normalize_fully_qualified_name_based_on_source(table)
+            table._load_target_fully_qualified_name(target_platform=target_platform)
+            table._set_internal_identifiers()
             self._check_max_failed_rows_number(table)
             self._check_chunk_number(table)
             self.set_exclude_metrics(table)
             self.set_apply_metric_column_modifier(table)
 
     def check_views(self) -> None:
+        if not self.views:
+            return
+        target_platform = Platform(self.target_platform.lower())
         view: ViewConfiguration
         for view in self.views:
             if self.target_database and view.target_database is None:
                 self._add_database(object=view)
-            self._apply_mappings_to_view(view)
+            self._apply_mappings_to_view(view, target_platform=target_platform)
             self._normalize_fully_qualified_name_based_on_source(view)
+            view._load_target_fully_qualified_name(target_platform=target_platform)
             self._check_max_failed_rows_number(view)
             self._check_chunk_number(view)
             self.set_exclude_metrics(view)
@@ -121,7 +130,9 @@ class ConfigurationModel(BaseModel):
                     "Schema mappings cannot have empty string as a target schema."
                 )
 
-    def _apply_mappings_to_table(self, table: TableConfiguration) -> None:
+    def _apply_mappings_to_table(
+        self, table: TableConfiguration, *, target_platform: Platform
+    ) -> None:
         needs_reload = False
 
         mapped_database = self.database_mappings.get(table.source_database)
@@ -135,7 +146,7 @@ class ConfigurationModel(BaseModel):
             needs_reload = True
 
         if needs_reload:
-            table._load_target_fully_qualified_name()
+            table._load_target_fully_qualified_name(target_platform=target_platform)
             table._set_internal_identifiers()
 
     def _normalize_fully_qualified_name_based_on_source(
@@ -147,7 +158,9 @@ class ConfigurationModel(BaseModel):
         if is_source_oracle:
             table.fully_qualified_name = f"{table.source_schema}.{table.source_table}"
 
-    def _apply_mappings_to_view(self, view: ViewConfiguration) -> None:
+    def _apply_mappings_to_view(
+        self, view: ViewConfiguration, *, target_platform: Platform
+    ) -> None:
         needs_reload = False
 
         mapped_database = self.database_mappings.get(view.source_database)
@@ -161,11 +174,12 @@ class ConfigurationModel(BaseModel):
             needs_reload = True
 
         if needs_reload:
-            view._load_target_fully_qualified_name()
+            view._load_target_fully_qualified_name(target_platform=target_platform)
 
         # Set internal identifiers for SQL Server platform to handle special identifier formatting.
         view._set_internal_identifiers(
-            source_platform=Platform(self.source_platform.lower())
+            source_platform=Platform(self.source_platform.lower()),
+            target_platform=target_platform,
         )
 
     def _check_max_failed_rows_number(self, table: TableConfiguration) -> None:
@@ -190,4 +204,6 @@ class ConfigurationModel(BaseModel):
 
     def _add_database(self, object: TableConfiguration) -> None:
         object.target_database = self.target_database
-        object._load_target_fully_qualified_name()
+        object._load_target_fully_qualified_name(
+            target_platform=Platform(self.target_platform.lower())
+        )

{snowflake_data_validation-1.22.0 → snowflake_data_validation-1.23.0}/src/snowflake/snowflake_data_validation/configuration/model/connection_types.py

@@ -27,6 +27,7 @@ from snowflake.snowflake_data_validation.configuration.model.connections import
     SnowflakeCredentialsConnection,
     SnowflakeDefaultConnection,
     SnowflakeNamedConnection,
+    SqliteConnection,
     SqlServerCredentialsConnection,
     TeradataCredentialsConnection,
 )
@@ -49,4 +50,6 @@ Connection = Union[
     RedshiftCredentialsConnection,
     # Azure Synapse connections
     AzureSynapseCredentialsConnection,
+    # SQLite connections
+    SqliteConnection,
 ]

{snowflake_data_validation-1.22.0 → snowflake_data_validation-1.23.0}/src/snowflake/snowflake_data_validation/configuration/model/connections/__init__.py

@@ -45,6 +45,11 @@ from snowflake.snowflake_data_validation.snowflake.model.snowflake_credentials_c
     SnowflakeCredentialsConnection,
 )
 
+# Import SQLite connection models
+from snowflake.snowflake_data_validation.sqlite.model.sqlite_connection import (
+    SqliteConnection,
+)
+
 __all__ = [
     "AzureSynapseCredentialsConnection",
     "SqlServerCredentialsConnection",
@@ -54,4 +59,5 @@ __all__ = [
     "SnowflakeCredentialsConnection",
     "PostgresqlCredentialsConnection",
     "RedshiftCredentialsConnection",
+    "SqliteConnection",
 ]

{snowflake_data_validation-1.22.0 → snowflake_data_validation-1.23.0}/src/snowflake/snowflake_data_validation/configuration/model/table_configuration.py

@@ -9,7 +9,10 @@ from typing_extensions import Self
 from snowflake.snowflake_data_validation.configuration.model.validation_configuration import (
     ValidationConfiguration,
 )
-from snowflake.snowflake_data_validation.utils.constants import NEWLINE, ObjectType
+from snowflake.snowflake_data_validation.target.target_platform_factory import (
+    TargetPlatformFactory,
+)
+from snowflake.snowflake_data_validation.utils.constants import NEWLINE, ObjectType, Platform
 from snowflake.snowflake_data_validation.utils.helper import Helper
 from snowflake.snowflake_data_validation.utils.helpers.helper_database import (
     HelperDatabase,
@@ -123,16 +126,16 @@ class TableConfiguration(BaseModel):
         self.source_schema = decomposed_tuple[1]
         self.source_table = decomposed_tuple[2]
 
-    def _load_target_fully_qualified_name(self) -> None:
-        self.target_database = HelperDatabase.normalize_to_snowflake_identifier(
-            self.target_database or self.source_database
-        )
-        self.target_schema = HelperDatabase.normalize_to_snowflake_identifier(
-            self.target_schema or self.source_schema
-        )
-        self.target_name = HelperDatabase.normalize_to_snowflake_identifier(
-            self.target_name or self.source_table
+    def _load_target_fully_qualified_name(
+        self, target_platform: Platform | None = None
+    ) -> None:
+        target_capability = TargetPlatformFactory.create(
+            target_platform or Platform.SNOWFLAKE
         )
+        normalize = target_capability.normalize_target_identifier
+        self.target_database = normalize(self.target_database or self.source_database)
+        self.target_schema = normalize(self.target_schema or self.source_schema)
+        self.target_name = normalize(self.target_name or self.source_table)
         self.target_fully_qualified_name = (
             f"{self.target_database}.{self.target_schema}.{self.target_name}"
         )

{snowflake_data_validation-1.22.0 → snowflake_data_validation-1.23.0}/src/snowflake/snowflake_data_validation/executer/executor_factory.py

@@ -92,6 +92,15 @@ from snowflake.snowflake_data_validation.snowflake.query.query_generator_snowfla
 from snowflake.snowflake_data_validation.snowflake.script_writer.script_writer_snowflake import (
     ScriptWriterSnowflake,
 )
+from snowflake.snowflake_data_validation.sqlite.extractor.metadata_extractor_sqlite import (
+    MetadataExtractorSqlite,
+)
+from snowflake.snowflake_data_validation.sqlite.query.query_generator_sqlite import (
+    QueryGeneratorSqlite,
+)
+from snowflake.snowflake_data_validation.sqlite.script_writer.script_writer_sqlite import (
+    ScriptWriterSqlite,
+)
 from snowflake.snowflake_data_validation.sqlserver.connector.connector_sql_server import (
     ConnectorSqlServer,
 )
@@ -194,6 +203,11 @@ class ExecutorFactory:
                 metadata_extractor=MetadataExtractorPostgresql,
                 script_writer=ScriptWriterPostgresql,
             ),
+            Platform.SQLITE: PlatformConfig(
+                query_generator=QueryGeneratorSqlite,
+                metadata_extractor=MetadataExtractorSqlite,
+                script_writer=ScriptWriterSqlite,
+            ),
             Platform.AZURE_SYNAPSE: PlatformConfig(
                 query_generator=QueryGeneratorAzureSynapse,
                 metadata_extractor=MetadataExtractorAzureSynapse,

{snowflake_data_validation-1.22.0 → snowflake_data_validation-1.23.0}/src/snowflake/snowflake_data_validation/executer/sync_validation_executor.py

@@ -26,6 +26,7 @@ public workflow API module:
 3. Validation - Uses data_validation_api.validate_schema / validate_metrics
 """
 
+import hashlib
 import logging
 
 from concurrent.futures import ThreadPoolExecutor
@@ -374,6 +375,15 @@ class SyncValidationExecutor(BaseValidationExecutor):
             columns=lambda x: x.upper() + TARGET_SUFFIX
         )
 
+        # SQLite has no native MD5: the row-md5-select query returns CONCAT_TEXT
+        # instead of ROW_MD5. Compute hashlib MD5 here so the comparison columns match.
+        _concat_col = "CONCAT_TEXT" + SOURCE_SUFFIX
+        if _concat_col in source_md5_df.columns:
+            source_md5_df["ROW_MD5" + SOURCE_SUFFIX] = source_md5_df[_concat_col].apply(
+                lambda v: hashlib.md5((v if v is not None else "").encode("utf-8")).hexdigest().upper()
+            )
+            source_md5_df = source_md5_df.drop(columns=[_concat_col])
+
         source_transformed = QueryExecutionResult(dataframe=source_md5_df)
         target_transformed = QueryExecutionResult(dataframe=target_md5_df)
 

{snowflake_data_validation-1.22.0 → snowflake_data_validation-1.23.0}/src/snowflake/snowflake_data_validation/extractor/sql_queries_template_generator.py

@@ -71,8 +71,9 @@ class SQLQueriesTemplateGenerator:
         fully_qualified_name: str,
         where_clause: str,
         has_where_clause: bool,
-        column_selection_list: list[ColumnMetadata],
+        columns: list[str],
         is_temporary_table: bool,
+        use_as_exclude_list: bool = False,
     ):
         """
         Generate SQL query for table metadata based on the provided table configuration and platform.
@@ -85,8 +86,10 @@ class SQLQueriesTemplateGenerator:
             fully_qualified_name (str): The fully qualified name of the table.
             where_clause (str): Optional WHERE clause to filter results.
             has_where_clause (bool): Indicates if a WHERE clause is present.
-            column_selection_list (list[ColumnMetadata]): List of ColumnMetadata objects to include in the query.
+            columns (list[str]): Column names to include (``use_as_exclude_list=False``)
+                or exclude (``use_as_exclude_list=True``) from the catalog query.
             is_temporary_table (bool): Indicates if the table is temporary.
+            use_as_exclude_list (bool): When True, ``columns`` is treated as a NOT IN list.
 
         Returns:
             str: The generated SQL query as a string.
@@ -99,7 +102,6 @@ class SQLQueriesTemplateGenerator:
         template_file = TABLE_METADATA_QUERY.format(platform=platform)
         try:
             template = self.env.get_template(template_file)
-            column_names = [col.name for col in column_selection_list]
             sql = template.render(
                 object_name=table_name,
                 object_schema=schema_name,
@@ -107,7 +109,8 @@ class SQLQueriesTemplateGenerator:
                 fully_qualified_name=fully_qualified_name,
                 where_clause=where_clause,
                 has_where_clause=has_where_clause,
-                columns=column_names,
+                columns=columns,
+                use_as_exclude_list=use_as_exclude_list,
                 is_temporary_table=is_temporary_table,
             )
             return sql

{snowflake_data_validation-1.22.0 → snowflake_data_validation-1.23.0}/src/snowflake/snowflake_data_validation/oracle/extractor/templates/oracle_column_metrics_templates.yaml

@@ -487,11 +487,11 @@ TIMESTAMP WITH LOCAL TIME ZONE:
   min:
     metric_query: MIN("{{ col_name }}")
     metric_return_datatype: TIMESTAMP WITH LOCAL TIME ZONE
-    metric_normalization: TO_CHAR(SYS_EXTRACT_UTC(CAST({{ metric_query }} AS TIMESTAMP WITH TIME ZONE)), 'YYYY-MM-DD"T"HH24:MI:SS.FF9')
+    metric_normalization: TO_CHAR({{ metric_query }}, 'YYYY-MM-DD"T"HH24:MI:SS.FF9')
   max:
     metric_query: MAX("{{ col_name }}")
     metric_return_datatype: TIMESTAMP WITH LOCAL TIME ZONE
-    metric_normalization: TO_CHAR(SYS_EXTRACT_UTC(CAST({{ metric_query }} AS TIMESTAMP WITH TIME ZONE)), 'YYYY-MM-DD"T"HH24:MI:SS.FF9')
+    metric_normalization: TO_CHAR({{ metric_query }}, 'YYYY-MM-DD"T"HH24:MI:SS.FF9')
   count_null:
     metric_query: SUM(CASE WHEN "{{ col_name }}" IS NULL THEN 1 ELSE 0 END)
     metric_return_datatype: INTEGER

{snowflake_data_validation-1.22.0 → snowflake_data_validation-1.23.0}/src/snowflake/snowflake_data_validation/oracle/extractor/templates/oracle_datatypes_normalization_templates.yaml

@@ -65,11 +65,12 @@ SDO_GEOMETRY: NVL(SDO_UTIL.TO_WKTGEOMETRY({{ col_name }}), '')
 # ---- Date / Time types ----
 # Format masks aligned with Snowflake TIMESTAMP_NTZ / TIMESTAMP_TZ normalization:
 #   DATE / TIMESTAMP → ISO 8601 with 9-digit fractional seconds (FF9)
-#   *TIME ZONE      → UTC extract, ISO with offset (REGEXP_REPLACE Z → +00:00)
+#   TIMESTAMP WITH TIME ZONE → ISO with per-row offset (TZH:TZM)
+#   TIMESTAMP WITH LOCAL TIME ZONE → session wall-clock ISO (no UTC extract)
 DATE: TO_CHAR(CAST({{ col_name }} AS TIMESTAMP), 'YYYY-MM-DD"T"HH24:MI:SS.FF9')
 TIMESTAMP: TO_CHAR({{ col_name }}, 'YYYY-MM-DD"T"HH24:MI:SS.FF9')
 TIMESTAMP WITH TIME ZONE: REGEXP_REPLACE(TO_CHAR({{ col_name }}, 'YYYY-MM-DD"T"HH24:MI:SS.FF9TZH:TZM'), 'Z$', '+00:00')
-TIMESTAMP WITH LOCAL TIME ZONE: TO_CHAR(SYS_EXTRACT_UTC(CAST({{ col_name }} AS TIMESTAMP WITH TIME ZONE)), 'YYYY-MM-DD"T"HH24:MI:SS.FF9')
+TIMESTAMP WITH LOCAL TIME ZONE: TO_CHAR({{ col_name }}, 'YYYY-MM-DD"T"HH24:MI:SS.FF9')
 
 # ---- Interval types ----
 # TO_CHAR on INTERVAL prefixes a sign and zero-pads fields (e.g. "+01-06").