snowflake-data-validation 1.3.6__tar.gz → 1.3.7__tar.gz

{snowflake_data_validation-1.3.6 → snowflake_data_validation-1.3.7}/.gitignore

@@ -195,6 +195,12 @@ cython_debug/
 # Requirements file generated by pip-compile
 requirements.txt
 
+# Streamlit in Snowflake: baked metadata (generated at deploy; not committed)
+data-migration-orchestrator/src/data_migration_orchestrator/data_migration_cloud/dashboard/_streamlit_baked_metadata.py
+data-migration-orchestrator/src/data_migration_orchestrator/data_validation_cloud/dashboard/_streamlit_baked_metadata.py
+data-migration-orchestrator/src/data_migration_orchestrator/data_migration_cloud/dashboard/sis_streamlit_metadata.py
+data-migration-orchestrator/src/data_migration_orchestrator/data_validation_cloud/dashboard/sis_streamlit_metadata.py
+
 # Data Exchange Agent
 data-exchange-agent/src/mocked_api/*
 .vscode
@@ -208,3 +214,4 @@ e2e-tests/config.yaml
 
 # Temporary files
 temp/
+.worktrees/

{snowflake_data_validation-1.3.6 → snowflake_data_validation-1.3.7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: snowflake-data-validation
-Version: 1.3.6
+Version: 1.3.7
 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/
@@ -26,6 +26,7 @@ Classifier: Topic :: Software Development :: Libraries
 Classifier: Topic :: Software Development :: Libraries :: Application Frameworks
 Classifier: Topic :: Software Development :: Libraries :: Python Modules
 Requires-Python: >=3.10
+Requires-Dist: connectorx>=0.3.0
 Requires-Dist: cryptography>=3.4.0
 Requires-Dist: deepdiff>=8.0.0
 Requires-Dist: jinja2>=3.1.6

{snowflake_data_validation-1.3.6 → snowflake_data_validation-1.3.7}/pyproject.toml

@@ -26,6 +26,7 @@ classifiers = [
   "Topic :: Scientific/Engineering :: Information Analysis",
 ]
 dependencies = [
+  "connectorx>=0.3.0",
   "cryptography>=3.4.0",
   "deepdiff>=8.0.0",
   "snowflake-connector-python>=4.0.0",

{snowflake_data_validation-1.3.6 → snowflake_data_validation-1.3.7}/src/snowflake/snowflake_data_validation/__version__.py

@@ -13,4 +13,4 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-__version__ = "1.3.6"
+__version__ = "1.3.7"

{snowflake_data_validation-1.3.6 → snowflake_data_validation-1.3.7}/src/snowflake/snowflake_data_validation/configuration/model/configuration_model.py

@@ -84,7 +84,7 @@ class ConfigurationModel(BaseModel):
     def check_tables(self) -> None:
         table: TableConfiguration
         for table in self.tables:
-            if self.source_platform.lower() == Platform.TERADATA.value:
+            if self.source_platform.casefold() == Platform.TERADATA.value.casefold():
                 self._add_database(object=table)
             self._apply_mappings_to_table(table)
             self._check_max_failed_rows_number(table)
@@ -95,7 +95,7 @@ class ConfigurationModel(BaseModel):
     def check_views(self) -> None:
         view: ViewConfiguration
         for view in self.views:
-            if self.source_platform.lower() == Platform.TERADATA.value:
+            if self.source_platform.casefold() == Platform.TERADATA.value.casefold():
                 self._add_database(object=view)
             self._apply_mappings_to_view(view)
             self._check_max_failed_rows_number(view)
@@ -153,10 +153,9 @@ class ConfigurationModel(BaseModel):
             view._load_target_fully_qualified_name()
 
         # Set internal identifiers for SQL Server platform to handle special identifier formatting.
-        if self.source_platform.casefold() == Platform.SQLSERVER.value.casefold():
-            view._set_internal_identifiers(
-                source_platform=Platform.SQLSERVER,
-            )
+        view._set_internal_identifiers(
+            source_platform=Platform(self.source_platform.lower())
+        )
 
     def _check_max_failed_rows_number(self, table: TableConfiguration) -> None:
         if table.max_failed_rows_number is None:

{snowflake_data_validation-1.3.6 → snowflake_data_validation-1.3.7}/src/snowflake/snowflake_data_validation/configuration/model/view_configuration.py

@@ -49,6 +49,10 @@ class ViewConfiguration(TableConfiguration):
         self.internal_source_table = self.source_table
         self.internal_target_name = self.target_name
 
+        # Skip internal identifiers generation for Teradata views, because temporary table is not necessary.
+        if source_platform == Platform.TERADATA:
+            return
+
         # Apply quoting if original identifier was non-regular
         source_needs_quotes = HelperDatabase.is_delimited_identifier(self.source_table)
         target_needs_quotes = HelperDatabase.is_delimited_identifier(self.target_name)
@@ -89,6 +93,7 @@ class ViewConfiguration(TableConfiguration):
         self.internal_fully_qualified_name = (
             f"{self.source_database}.{self.source_schema}.{self.internal_source_table}"
         )
+
         self.internal_target_fully_qualified_name = (
             f"{self.target_database}.{self.target_schema}.{self.internal_target_name}"
         )

{snowflake_data_validation-1.3.6 → snowflake_data_validation-1.3.7}/src/snowflake/snowflake_data_validation/connector/connector_base.py

@@ -42,6 +42,7 @@ class ConnectorBase(ABC):
         self.output_path = output_path
         self.logger = logging.getLogger(__name__)
         self._failure_tracker = ConnectionFailureTracker()
+        self._direct_read_uri: str | None = None
 
     @abstractmethod
     def connect(

{snowflake_data_validation-1.3.6 → snowflake_data_validation-1.3.7}/src/snowflake/snowflake_data_validation/executer/sync_validation_executor.py

@@ -47,6 +47,8 @@ from snowflake.snowflake_data_validation.utils.base_output_handler import (
 from snowflake.snowflake_data_validation.utils.constants import (
     CHUNK_ID_COLUMN_KEY,
     DEFAULT_TOLERANCE,
+    ObjectType,
+    Platform,
     Result,
     ValidationLevel,
 )
@@ -122,6 +124,16 @@ class SyncValidationExecutor(BaseValidationExecutor):
             result_processor=self.source_extractor.process_schema_query_result,
         )
 
+        # Teradata view validation does not need view materialization.
+        if (
+            source_table_context.object_type == ObjectType.VIEW
+            and self.context.source_platform == Platform.TERADATA
+        ):
+            source_result = data_validation_api.normalize_table_metadata(
+                platform=self.context.source_platform,
+                metadata=source_result,
+            )
+
         self._emit_extraction_message(target_table_context)
         target_result = data_validation_api.execute_query(
             table_context=target_table_context,
@@ -130,13 +142,25 @@ class SyncValidationExecutor(BaseValidationExecutor):
             result_processor=self.target_extractor.process_schema_query_result,
         )
 
-        validation_result = data_validation_api.validate_schema(
-            source_query_result=source_result,
-            target_query_result=target_result,
-            table_context=source_table_context,
-            column_mappings=column_mappings,
-            datatypes_mappings=self.context.datatypes_mappings,
-        )
+        if (
+            source_table_context.object_type == ObjectType.VIEW
+            and self.context.source_platform == Platform.TERADATA
+        ):
+            validation_result = data_validation_api.validate_schema_simple(
+                source_query_result=source_result,
+                target_query_result=target_result,
+                table_context=source_table_context,
+                column_mappings=column_mappings,
+                datatypes_mappings=self.context.datatypes_mappings,
+            )
+        else:
+            validation_result = data_validation_api.validate_schema(
+                source_query_result=source_result,
+                target_query_result=target_result,
+                table_context=source_table_context,
+                column_mappings=column_mappings,
+                datatypes_mappings=self.context.datatypes_mappings,
+            )
 
         self._emit_validation_result_message(
             validation_type=ValidationLevel.SCHEMA_VALIDATION.value,

{snowflake_data_validation-1.3.6 → snowflake_data_validation-1.3.7}/src/snowflake/snowflake_data_validation/extractor/sql_queries_template_generator.py

@@ -489,6 +489,35 @@ class SQLQueriesTemplateGenerator:
         except Exception as e:
             raise Exception(f"Error generating template: {e}") from e
 
+    def generate_cell_validation_data_select_query(
+        self,
+        platform: str,
+        column_collection: list,
+        datatypes_normalization_renderer_templates: dict[str, str],
+        fully_qualified_name: str,
+        has_where_clause: bool,
+        where_clause: str,
+        order_by_columns: list[str] | None,
+    ) -> str:
+        """Render a SELECT that applies the same per-type normalization as L3 MD5."""
+        template_file = f"{platform}_cell_validation_data_select.sql.j2"
+        try:
+            template = self.env.get_template(template_file)
+            return template.render(
+                column_collection=column_collection,
+                datatypes_normalization_renderer_templates=datatypes_normalization_renderer_templates,
+                fully_qualified_name=fully_qualified_name,
+                has_where_clause=has_where_clause,
+                where_clause=where_clause,
+                order_by_columns=order_by_columns or [],
+            )
+        except jinja2.exceptions.TemplateNotFound:
+            raise ValueError(
+                f"Template not found: {template_file}. Please create it in the {self.template_dir} directory."
+            ) from None
+        except Exception as e:
+            raise Exception(f"Error generating template: {e}") from e
+
     # SNOWFLAKE ONLY
     def generate_chunk_row_md5_template_query(
         self,

{snowflake_data_validation-1.3.6 → snowflake_data_validation-1.3.7}/src/snowflake/snowflake_data_validation/orchestration/parallel_execution_engine.py

@@ -306,7 +306,11 @@ class ParallelExecutionEngine:
                 db_manager=db_manager,
             )
 
-            if table_configuration.object_type == ObjectType.VIEW:
+            is_view = table_configuration.object_type == ObjectType.VIEW
+            is_dialect_teradata = context.source_platform == Platform.TERADATA
+
+            # Materialize source view if needed (Teradata views don't require materialization)
+            if is_view and not is_dialect_teradata:
                 self.view_pre_processor.materialize_source_view(
                     view_configuration=table_configuration,
                     source_extractor=source_extractor,
@@ -378,10 +382,17 @@ class ParallelExecutionEngine:
                 db_manager=db_manager,
             )
 
-            if table_configuration.object_type == ObjectType.VIEW:
-                self.view_pre_processor.create_temporary_table_from_view(
+            is_view = table_configuration.object_type == ObjectType.VIEW
+            is_dialect_teradata = context.source_platform == Platform.TERADATA
+
+            if is_view and not is_dialect_teradata:
+                self.view_pre_processor.materialize_source_view(
                     view_configuration=table_configuration,
                     source_extractor=source_extractor,
+                    context=context,
+                )
+                self.view_pre_processor.materialize_target_view(
+                    view_configuration=table_configuration,
                     target_extractor=target_extractor,
                     context=context,
                 )

{snowflake_data_validation-1.3.6 → snowflake_data_validation-1.3.7}/src/snowflake/snowflake_data_validation/orchestration/table_metadata_processor.py

@@ -177,6 +177,15 @@ class TableMetadataProcessor:
             )
         )
 
+        column_metadata_result = data_validation_api.normalize_table_column_metadata(
+            platform=context.source_platform,
+            database=table_configuration.source_database,
+            schema=table_configuration.source_schema,
+            table=table_configuration.source_table,
+            object_type=table_configuration.object_type,
+            metadata=column_metadata_result,
+        )
+
         row_count_query = data_validation_api.generate_row_count_query(
             fully_qualified_name=table_configuration.fully_qualified_name,
             where_clause=table_configuration.where_clause,

{snowflake_data_validation-1.3.6 → snowflake_data_validation-1.3.7}/src/snowflake/snowflake_data_validation/orchestration/view_preprocessor.py

@@ -28,32 +28,6 @@ class ViewPreProcessor:
     queries to script files for later reference.
     """
 
-    def create_temporary_table_from_view(
-        self,
-        view_configuration: ViewConfiguration,
-        source_extractor: MetadataExtractorBase | ScriptWriterBase,
-        target_extractor: MetadataExtractorBase | ScriptWriterBase,
-        context: Context,
-    ) -> None:
-        """Create temporary tables from views for both source and target platforms.
-
-        Uses the public workflow API to generate and execute view materialization
-        queries in a decoupled manner. For async mode, also writes the queries to
-        script files.
-
-        Args:
-            view_configuration: Configuration object containing view details and settings.
-            source_extractor: Extractor or script writer for the source platform.
-            target_extractor: Extractor or script writer for the target platform.
-            context: Context object containing platform and execution details.
-
-        Returns:
-            None
-
-        """
-        self.materialize_source_view(view_configuration, source_extractor, context)
-        self.materialize_target_view(view_configuration, target_extractor, context)
-
     def materialize_source_view(
         self,
         view_configuration: ViewConfiguration,

{snowflake_data_validation-1.3.6 → snowflake_data_validation-1.3.7}/src/snowflake/snowflake_data_validation/public/DATA_VALIDATION_API_OVERVIEW.md

@@ -88,9 +88,21 @@ The cell validation provides a lightweight alternative to MD5-based row validati
 
 | Function | Purpose |
 |----------|---------|
-| `result_set_cell_level_validation()` | Compares two `ResultSet` objects and determines if they are equivalent |
+| `result_set_to_dataframe()` | Converts a `ResultSet` object to a Polars DataFrame |
+| `result_set_cell_level_validation()` | Compares two `ResultSet` objects at the cell level and determines if they are equivalent |
+| `result_set_metrics_level_validation()` | Compares two `ResultSet` objects at the metrics level (min, max, count, etc.) |
 
-This function compares the dataframes of two result sets, checking for equivalence in terms of data values (ignoring order and duplicates).
+These functions compare result sets for equivalence. Cell-level validation checks individual cell values while metrics-level validation compares aggregate statistics.
+
+### Step 3.2.2: Query Comparison
+
+| Function | Purpose |
+|----------|---------|
+| `query_to_dataframe()` | Executes a query and converts the result to a Polars DataFrame |
+| `query_cell_level_validation()` | Compares two Polars DataFrames at the cell level |
+| `query_metrics_level_validation()` | Compares two Polars DataFrames at the metrics level with configurable tolerance |
+
+These functions provide direct DataFrame comparison capabilities without requiring `ResultSet` objects. They are useful for comparing query results from different sources.
 
 ### Step 3.3: Snapshot Validation
 

{snowflake_data_validation-1.3.6 → snowflake_data_validation-1.3.7}/src/snowflake/snowflake_data_validation/public/__init__.py

@@ -48,6 +48,7 @@ from snowflake.snowflake_data_validation.public.data_validation_api import (
     execute_view_materialization,
     extract_row_count,
     generate_case_sensitive_columns_query,
+    generate_cell_validation_data_query,
     generate_column_metadata_query,
     generate_compute_md5_queries,
     generate_create_md5_table_statement,
@@ -63,8 +64,12 @@ from snowflake.snowflake_data_validation.public.data_validation_api import (
     load_validation_results,
     process_table_column_metadata,
     process_target_table_column_metadata_from_source_table,
+    query_cell_level_validation,
+    query_metrics_level_validation,
+    query_to_dataframe,
     result_set_cell_level_validation,
     result_set_metrics_level_validation,
+    result_set_to_dataframe,
     run_async_generation,
     run_async_validation,
     run_source_validation,
@@ -173,6 +178,7 @@ __all__ = [
     "execute_view_materialization",
     "extract_row_count",
     "generate_case_sensitive_columns_query",
+    "generate_cell_validation_data_query",
     "generate_column_metadata_query",
     "generate_compute_md5_queries",
     "generate_create_md5_table_statement",
@@ -188,8 +194,12 @@ __all__ = [
     "load_validation_results",
     "process_table_column_metadata",
     "process_target_table_column_metadata_from_source_table",
+    "query_cell_level_validation",
+    "query_metrics_level_validation",
+    "query_to_dataframe",
     "result_set_cell_level_validation",
     "result_set_metrics_level_validation",
+    "result_set_to_dataframe",
     "run_async_generation",
     "run_async_validation",
     "run_source_validation",

snowflake_data_validation-1.3.7/src/snowflake/snowflake_data_validation/public/_teradata_internal.py

@@ -0,0 +1,287 @@
+# Copyright 2026 Snowflake Inc.
+# SPDX-License-Identifier: Apache-2.0
+
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+
+# http://www.apache.org/licenses/LICENSE-2.0
+
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Teradata-specific internal helper functions for the public API.
+
+This module contains private utility functions that support the public API
+for Teradata-specific operations, including normalization of HELP COLUMN results.
+
+WARNING: Functions in this module are internal implementation details and
+may change without notice. Do not import from this module directly.
+"""
+
+from __future__ import annotations
+
+import pandas as pd
+
+from snowflake.snowflake_data_validation.utils.constants import (
+    CALCULATED_COLUMN_SIZE_IN_BYTES_KEY,
+    CHARACTER_LENGTH_KEY,
+    COLUMN_NAME_KEY,
+    DATA_TYPE_KEY,
+    DATABASE_NAME_KEY,
+    IS_PRIMARY_KEY_KEY,
+    NULLABLE_KEY,
+    PRECISION_KEY,
+    SCALE_KEY,
+    SCHEMA_NAME_KEY,
+    TABLE_NAME_KEY,
+)
+
+
+TERADATA_TYPE_CODE_MAPPING = {
+    "A1": "ARRAY",
+    "AN": "MULTI-DIMENSIONAL ARRAY",
+    "AT": "TIME",
+    "BF": "BYTE",
+    "BO": "BLOB",
+    "BV": "VARBYTE",
+    "CF": "CHAR",
+    "CO": "CLOB",
+    "CV": "VARCHAR",
+    "D": "DECIMAL",
+    "DA": "DATE",
+    "DH": "INTERVAL DAY TO HOUR",
+    "DM": "INTERVAL DAY TO MINUTE",
+    "DS": "INTERVAL DAY TO SECOND",
+    "DY": "INTERVAL DAY",
+    "F": "FLOAT",
+    "HM": "INTERVAL HOUR TO MINUTE",
+    "HS": "INTERVAL HOUR TO SECOND",
+    "HR": "INTERVAL HOUR",
+    "I": "INTEGER",
+    "I1": "BYTEINT",
+    "I2": "SMALLINT",
+    "I8": "BIGINT",
+    "JN": "JSON",
+    "MI": "INTERVAL MINUTE",
+    "MO": "INTERVAL MONTH",
+    "MS": "INTERVAL MINUTE TO SECOND",
+    "N": "NUMBER",
+    "PD": "PERIOD(DATE)",
+    "PM": "PERIOD(TIMESTAMP WITH TIME ZONE)",
+    "PS": "PERIOD(TIMESTAMP)",
+    "PT": "PERIOD(TIME)",
+    "PZ": "PERIOD(TIME WITH TIME ZONE)",
+    "SC": "INTERVAL SECOND",
+    "SZ": "TIMESTAMP WITH TIME ZONE",
+    "TS": "TIMESTAMP",
+    "TZ": "TIME WITH TIME ZONE",
+    "XM": "XML",
+    "YM": "INTERVAL YEAR TO MONTH",
+    "YR": "INTERVAL YEAR",
+}
+
+TERADATA_STRING_DATA_TYPES = {
+    "CHAR",
+    "VARCHAR",
+    "CLOB",
+    "BYTE",
+    "VARBYTE",
+    "INTERVAL DAY TO HOUR",
+    "INTERVAL DAY TO MINUTE",
+    "INTERVAL DAY TO SECOND",
+    "INTERVAL DAY",
+    "INTERVAL HOUR TO MINUTE",
+    "INTERVAL HOUR TO SECOND",
+    "INTERVAL HOUR",
+    "INTERVAL MINUTE",
+    "INTERVAL MINUTE TO SECOND",
+    "INTERVAL SECOND",
+    "INTERVAL YEAR TO MONTH",
+    "INTERVAL YEAR",
+    "INTERVAL MONTH",
+    "PERIOD(DATE)",
+    "PERIOD(TIME)",
+    "PERIOD(TIME WITH TIME ZONE)",
+    "PERIOD(TIMESTAMP)",
+    "PERIOD(TIMESTAMP WITH TIME ZONE)",
+}
+
+TERADATA_NUMERIC_DATA_TYPES = {
+    "DECIMAL",
+    "NUMBER",
+    "INTEGER",
+    "BYTEINT",
+    "SMALLINT",
+    "BIGINT",
+}
+
+TERADATA_HELP_COLUMN_NAME = "COLUMN NAME"
+TERADATA_HELP_COLUMN_TYPE = "TYPE"
+TERADATA_HELP_COLUMN_NULLABLE = "NULLABLE"
+TERADATA_HELP_COLUMN_MAX_LENGTH = "MAX LENGTH"
+TERADATA_HELP_COLUMN_DECIMAL_TOTAL = "DECIMAL TOTAL DIGITS"
+TERADATA_HELP_COLUMN_DECIMAL_FRAC = "DECIMAL FRACTIONAL DIGITS"
+
+COLUMN_VALIDATED_KEY = "COLUMN_VALIDATED"
+TERADATA_HELP_COLUMN_PRIMARY = "PRIMARY?"
+
+
+def teradata_normalize_table_column_metadata(
+    database: str,
+    table: str,
+    metadata_df: pd.DataFrame,
+) -> pd.DataFrame:
+    """Normalize Teradata HELP COLUMN result into a standardized column metadata DataFrame.
+
+    This function transforms the output of Teradata's HELP COLUMN command into a
+    standardized DataFrame format used by the data validation framework.
+    Uses vectorized pandas operations for better performance.
+
+    Args:
+        database: The database name containing the table.
+        table: The table name.
+        metadata_df: DataFrame containing HELP COLUMN results with columns:
+            - Column Name: Name of the column
+            - Type: 2-character Teradata type code (e.g., 'CV', 'I', 'D')
+            - Nullable: Nullable indicator (Y/N/?/1/-)
+            - Max Length: Maximum length for character types
+            - Decimal Total Digits: Precision for numeric types
+            - Decimal Fractional Digits: Scale for numeric types
+
+    Returns:
+        pd.DataFrame: Normalized DataFrame with columns:
+            DATABASE_NAME, SCHEMA_NAME, TABLE_NAME, COLUMN_NAME, DATA_TYPE,
+            CHARACTER_LENGTH, PRECISION, SCALE, NULLABLE, IS_PRIMARY_KEY,
+            CALCULATED_COLUMN_SIZE_IN_BYTES
+
+    """
+    output_columns = [
+        DATABASE_NAME_KEY,
+        SCHEMA_NAME_KEY,
+        TABLE_NAME_KEY,
+        COLUMN_NAME_KEY,
+        DATA_TYPE_KEY,
+        CHARACTER_LENGTH_KEY,
+        PRECISION_KEY,
+        SCALE_KEY,
+        NULLABLE_KEY,
+        IS_PRIMARY_KEY_KEY,
+        CALCULATED_COLUMN_SIZE_IN_BYTES_KEY,
+    ]
+
+    if metadata_df.empty:
+        return pd.DataFrame(columns=output_columns)
+
+    result_df = pd.DataFrame()
+
+    result_df[COLUMN_NAME_KEY] = (
+        metadata_df[TERADATA_HELP_COLUMN_NAME].fillna("").astype(str).str.strip()
+    )
+
+    type_codes = (
+        metadata_df[TERADATA_HELP_COLUMN_TYPE].fillna("").astype(str).str.strip()
+    )
+    result_df[DATA_TYPE_KEY] = type_codes.map(TERADATA_TYPE_CODE_MAPPING).fillna(
+        type_codes
+    )
+
+    is_string = result_df[DATA_TYPE_KEY].isin(TERADATA_STRING_DATA_TYPES)
+    is_numeric = result_df[DATA_TYPE_KEY].isin(TERADATA_NUMERIC_DATA_TYPES)
+
+    max_length = pd.to_numeric(
+        metadata_df[TERADATA_HELP_COLUMN_MAX_LENGTH], errors="coerce"
+    ).fillna(0)
+    result_df[CHARACTER_LENGTH_KEY] = "0"
+    result_df.loc[is_string, CHARACTER_LENGTH_KEY] = (
+        max_length[is_string].astype(int).astype(str)
+    )
+
+    decimal_total = pd.to_numeric(
+        metadata_df[TERADATA_HELP_COLUMN_DECIMAL_TOTAL], errors="coerce"
+    ).fillna(-128)
+    decimal_frac = pd.to_numeric(
+        metadata_df[TERADATA_HELP_COLUMN_DECIMAL_FRAC], errors="coerce"
+    ).fillna(-128)
+
+    result_df[PRECISION_KEY] = "-1"
+    valid_precision = is_numeric & (decimal_total != -128)
+    result_df.loc[valid_precision, PRECISION_KEY] = (
+        decimal_total[valid_precision].astype(int).astype(str)
+    )
+
+    result_df[SCALE_KEY] = "-1"
+    valid_scale = is_numeric & (decimal_frac != -128)
+    result_df.loc[valid_scale, SCALE_KEY] = (
+        decimal_frac[valid_scale].astype(int).astype(str)
+    )
+
+    nullable_values = (
+        metadata_df[TERADATA_HELP_COLUMN_NULLABLE]
+        .fillna("")
+        .astype(str)
+        .str.strip()
+        .str.upper()
+    )
+    result_df[NULLABLE_KEY] = nullable_values == "Y"
+
+    primary_values = (
+        metadata_df[TERADATA_HELP_COLUMN_PRIMARY]
+        .fillna("")
+        .astype(str)
+        .str.strip()
+        .str.upper()
+    )
+    result_df[IS_PRIMARY_KEY_KEY] = primary_values == "Y"
+
+    result_df[CALCULATED_COLUMN_SIZE_IN_BYTES_KEY] = "0"
+    result_df.loc[is_string, CALCULATED_COLUMN_SIZE_IN_BYTES_KEY] = (
+        max_length[is_string].astype(int).astype(str)
+    )
+
+    result_df[DATABASE_NAME_KEY] = database
+    result_df[SCHEMA_NAME_KEY] = database
+    result_df[TABLE_NAME_KEY] = table
+
+    return result_df[output_columns]
+
+
+def teradata_normalize_table_metadata(
+    metadata_df: pd.DataFrame,
+) -> pd.DataFrame:
+    """Normalize Teradata HELP COLUMN result into minimal table metadata format.
+
+    This function transforms the output of Teradata's HELP COLUMN command into a
+    minimal DataFrame format used for simple schema validation.
+    Uses vectorized pandas operations for better performance.
+
+    Args:
+        metadata_df: DataFrame containing HELP COLUMN results.
+
+    Returns:
+        pd.DataFrame: Normalized DataFrame with columns:
+            COLUMN_VALIDATED, DATA_TYPE
+
+    """
+    output_columns = [COLUMN_VALIDATED_KEY, DATA_TYPE_KEY]
+
+    if metadata_df.empty:
+        return pd.DataFrame(columns=output_columns)
+
+    result_df = pd.DataFrame()
+
+    result_df[COLUMN_VALIDATED_KEY] = (
+        metadata_df[TERADATA_HELP_COLUMN_NAME].fillna("").astype(str).str.strip()
+    )
+
+    type_codes = (
+        metadata_df[TERADATA_HELP_COLUMN_TYPE].fillna("").astype(str).str.strip()
+    )
+    result_df[DATA_TYPE_KEY] = type_codes.map(TERADATA_TYPE_CODE_MAPPING).fillna(
+        type_codes
+    )
+
+    return result_df[output_columns]