sql-glider 0.1.2__py3-none-any.whl → 0.1.3__py3-none-any.whl

{sql_glider-0.1.2.dist-info → sql_glider-0.1.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sql-glider
-Version: 0.1.2
+Version: 0.1.3
 Summary: SQL Utility Toolkit for better understanding, use, and governance of your queries in a native environment.
 Project-URL: Homepage, https://github.com/rycowhi/sql-glider/
 Project-URL: Repository, https://github.com/rycowhi/sql-glider/
@@ -26,6 +26,8 @@ Requires-Dist: rich>=13.0.0
 Requires-Dist: rustworkx>=0.15.0
 Requires-Dist: sqlglot[rs]>=25.0.0
 Requires-Dist: typer>=0.9.0
+Provides-Extra: databricks
+Requires-Dist: databricks-sdk>=0.20.0; extra == 'databricks'
 Description-Content-Type: text/markdown
 
 # SQL Glider
@@ -40,6 +42,7 @@ SQL Glider provides powerful column-level and table-level lineage analysis for S
 
 - **Forward Lineage:** Trace output columns back to their source tables and columns
 - **Reverse Lineage:** Impact analysis - find which output columns are affected by a source column
+- **Query Dissection:** Decompose SQL into components (CTEs, subqueries, UNION branches) for unit testing
 - **Table Extraction:** List all tables in SQL files with usage type (INPUT/OUTPUT) and object type (TABLE/VIEW/CTE)
 - **Multi-level Tracing:** Automatically handles CTEs, subqueries, and complex expressions
 - **Graph-Based Lineage:** Build and query lineage graphs across thousands of SQL files
@@ -171,15 +174,32 @@ List all tables involved in SQL files with usage and type information:
 
 ```bash
 # List all tables in a SQL file
-uv run sqlglider tables query.sql
+uv run sqlglider tables overview query.sql
 
 # JSON output with detailed table info
-uv run sqlglider tables query.sql --output-format json
+uv run sqlglider tables overview query.sql --output-format json
 
 # Export to CSV
-uv run sqlglider tables query.sql --output-format csv --output-file tables.csv
+uv run sqlglider tables overview query.sql --output-format csv --output-file tables.csv
 ```
 
+### Pull DDL from Remote Catalogs
+
+Fetch DDL definitions from remote data catalogs (e.g., Databricks Unity Catalog):
+
+```bash
+# Pull DDL for all tables used in a SQL file (outputs to stdout)
+uv run sqlglider tables pull query.sql --catalog-type databricks
+
+# Save DDL files to a folder (one file per table)
+uv run sqlglider tables pull query.sql -c databricks -o ./ddl/
+
+# List available catalog providers
+uv run sqlglider tables pull --list
+```
+
+**Note:** Requires optional dependencies. Install with: `pip install sql-glider[databricks]`
+
 **Example Output (JSON):**
 ```json
 {
@@ -204,6 +224,94 @@ uv run sqlglider tables query.sql --output-format csv --output-file tables.csv
 - `CTE`: Common Table Expression (WITH clause)
 - `UNKNOWN`: Cannot determine type from SQL alone
 
+### Query Dissection
+
+Decompose SQL queries into constituent parts for unit testing and analysis:
+
+```bash
+# Dissect a SQL file (text output)
+uv run sqlglider dissect query.sql
+
+# JSON output with full component details
+uv run sqlglider dissect query.sql --output-format json
+
+# CSV output for spreadsheet analysis
+uv run sqlglider dissect query.sql --output-format csv
+
+# Export to file
+uv run sqlglider dissect query.sql -f json -o dissected.json
+
+# With templating support
+uv run sqlglider dissect query.sql --templater jinja --var schema=analytics
+
+# From stdin
+echo "WITH cte AS (SELECT id FROM users) SELECT * FROM cte" | uv run sqlglider dissect
+```
+
+**Example Input:**
+```sql
+WITH order_totals AS (
+    SELECT customer_id, SUM(amount) AS total
+    FROM orders
+    GROUP BY customer_id
+)
+INSERT INTO analytics.summary
+SELECT * FROM order_totals WHERE total > 100
+```
+
+**Example Output (JSON):**
+```json
+{
+  "queries": [{
+    "query_index": 0,
+    "statement_type": "INSERT",
+    "total_components": 3,
+    "components": [
+      {
+        "component_type": "CTE",
+        "component_index": 0,
+        "name": "order_totals",
+        "sql": "SELECT customer_id, SUM(amount) AS total FROM orders GROUP BY customer_id",
+        "is_executable": true,
+        "dependencies": [],
+        "location": "WITH clause"
+      },
+      {
+        "component_type": "TARGET_TABLE",
+        "component_index": 1,
+        "name": "analytics.summary",
+        "sql": "analytics.summary",
+        "is_executable": false,
+        "location": "INSERT INTO target"
+      },
+      {
+        "component_type": "SOURCE_QUERY",
+        "component_index": 2,
+        "sql": "SELECT * FROM order_totals WHERE total > 100",
+        "is_executable": true,
+        "dependencies": ["order_totals"],
+        "location": "INSERT source SELECT"
+      }
+    ]
+  }]
+}
+```
+
+**Extracted Component Types:**
+- `CTE`: Common Table Expressions from WITH clause
+- `MAIN_QUERY`: The primary SELECT statement
+- `SUBQUERY`: Nested SELECT in FROM clause
+- `SCALAR_SUBQUERY`: Single-value subquery in SELECT list, WHERE, HAVING
+- `TARGET_TABLE`: Output table for INSERT/CREATE/MERGE (not executable)
+- `SOURCE_QUERY`: SELECT within DML/DDL statements
+- `UNION_BRANCH`: Individual SELECT in UNION/UNION ALL
+
+**Use Cases:**
+- Unit test CTEs and subqueries individually
+- Extract DQL from CTAS, CREATE VIEW, INSERT statements
+- Analyze query structure and component dependencies
+- Break apart complex queries for understanding
+
 ### Different SQL Dialects
 
 ```bash
@@ -475,7 +583,7 @@ Options:
 ### Tables Command
 
 ```
-sqlglider tables <sql_file> [OPTIONS]
+sqlglider tables overview <sql_file> [OPTIONS]
 
 Arguments:
   sql_file                    Path to SQL file to analyze [required]
@@ -491,6 +599,66 @@ Options:
   --help                      Show help message and exit
 ```
 
+```
+sqlglider tables pull <sql_file> [OPTIONS]
+
+Arguments:
+  sql_file                    Path to SQL file to analyze [optional, reads from stdin if omitted]
+
+Options:
+  --catalog-type, -c          Catalog provider (e.g., 'databricks') [required if not in config]
+  --ddl-folder, -o            Output folder for DDL files [optional, outputs to stdout if omitted]
+  --dialect, -d               SQL dialect (spark, postgres, snowflake, etc.) [default: spark]
+  --templater, -t             Templater for SQL preprocessing (e.g., 'jinja', 'none') [optional]
+  --var, -v                   Template variable in key=value format (repeatable) [optional]
+  --vars-file                 Path to variables file (JSON or YAML) [optional]
+  --list, -l                  List available catalog providers and exit
+  --help                      Show help message and exit
+```
+
+**Databricks Setup:**
+
+Install the optional Databricks dependency:
+```bash
+pip install sql-glider[databricks]
+```
+
+Configure authentication (via environment variables or `sqlglider.toml`):
+```bash
+export DATABRICKS_HOST="https://your-workspace.cloud.databricks.com"
+export DATABRICKS_TOKEN="dapi..."
+export DATABRICKS_WAREHOUSE_ID="abc123..."
+```
+
+### Dissect Command
+
+```
+sqlglider dissect [sql_file] [OPTIONS]
+
+Arguments:
+  sql_file                    Path to SQL file to analyze [optional, reads from stdin if omitted]
+
+Options:
+  --dialect, -d               SQL dialect (spark, postgres, snowflake, etc.) [default: spark]
+  --output-format, -f         Output format: 'text', 'json', or 'csv' [default: text]
+  --output-file, -o           Write output to file instead of stdout [optional]
+  --templater, -t             Templater for SQL preprocessing (e.g., 'jinja', 'none') [optional]
+  --var, -v                   Template variable in key=value format (repeatable) [optional]
+  --vars-file                 Path to variables file (JSON or YAML) [optional]
+  --help                      Show help message and exit
+```
+
+**Output Fields:**
+- `component_type`: Type of component (CTE, MAIN_QUERY, SUBQUERY, etc.)
+- `component_index`: Sequential order within the query (0-based)
+- `name`: CTE name, subquery alias, or target table name
+- `sql`: The extracted SQL for this component
+- `is_executable`: Whether the component can run standalone (TARGET_TABLE is false)
+- `dependencies`: List of CTE names this component references
+- `location`: Human-readable context (e.g., "WITH clause", "FROM clause")
+- `depth`: Nesting level (0 = top-level)
+- `parent_index`: Index of parent component for nested components
+
 ### Graph Commands
 
 ```
@@ -612,6 +780,10 @@ See [ARCHITECTURE.md](ARCHITECTURE.md) for detailed technical documentation.
 ```
 src/sqlglider/
 ├── cli.py                    # Typer CLI entry point
+├── dissection/
+│   ├── analyzer.py           # DissectionAnalyzer for query decomposition
+│   ├── formatters.py         # Output formatters (text, JSON, CSV)
+│   └── models.py             # ComponentType, SQLComponent, QueryDissectionResult
 ├── graph/
 │   ├── builder.py            # Build graphs from SQL files
 │   ├── merge.py              # Merge multiple graphs

sql_glider-0.1.3.dist-info/RECORD

@@ -0,0 +1,34 @@
+sqlglider/__init__.py,sha256=gDf7s52dMcX7JuCZ1SLawcB1vb3U0yJCohu9RQAATBY,125
+sqlglider/_version.py,sha256=q5nF98G8SoVeJqaknL0xdyxtv0egsqb0fK06_84Izu8,704
+sqlglider/cli.py,sha256=9sweHRVLk2iBSzCzT2Gcj8y1g1XKzq26iApQsMaFbx4,51786
+sqlglider/global_models.py,sha256=2vyJXAuXOsXQpE-D3F0ejj7eR9z0nDWFjTkielhzM8k,356
+sqlglider/catalog/__init__.py,sha256=2PqFPyzFXJ14FpSUcBmVK2L-a_ypWQHAbHFHxLDk_LE,814
+sqlglider/catalog/base.py,sha256=R7htHC43InpH4uRjYk33dMYYji6oylHns7Ye_mgfjJE,3116
+sqlglider/catalog/databricks.py,sha256=Ho1crIKv1bw-fXkWUhQhcKfiYQEGinGSxBS2zoVLB3o,9504
+sqlglider/catalog/registry.py,sha256=KD1XrvK46xSrK5IikzbdbTSk_-wwRTXvyBxXn3m-Rx0,3391
+sqlglider/dissection/__init__.py,sha256=ObXM7AXTJZvheIg36ps9KuFsXPV7WmWamaA4xPfxP4s,396
+sqlglider/dissection/analyzer.py,sha256=-GD3-lTbfBthq1BW6HiDjvJx2y4LDmnUVHIVIb0HqrU,27085
+sqlglider/dissection/formatters.py,sha256=M7gsmTNljRIeLIRv4D0vHvqJVrTqWSpsg7vem83zSzY,7302
+sqlglider/dissection/models.py,sha256=RRD3RIteqbUBY6e-74skKDvMH3qeAUaqA2sFcrjP5GQ,3618
+sqlglider/graph/__init__.py,sha256=4DDdrPM75CmeQWt7wHdBsjCm1s70BHGLYdijIbaUEKY,871
+sqlglider/graph/builder.py,sha256=WdMUwKlB6UGtr7CA-J5Lj7D2GMQJZzteDetzr7Pe4Kk,11916
+sqlglider/graph/merge.py,sha256=uUZlm4BN3S9gRL66Cc2mzhbtuh4SVAv2n4cN4eUEQBU,4077
+sqlglider/graph/models.py,sha256=EYmjv_WzDSNp_WfhJ6H-qBIOkAcoNKS7GRUryfKrHuY,9330
+sqlglider/graph/query.py,sha256=LHU8Cvn7ZPPSEnqdDn2pF8f1_LQjIvNIrZqs8cFlb6U,9433
+sqlglider/graph/serialization.py,sha256=7JJo31rwSlxnDhdqdTJdK4Dr_ZcSYetXfx3_CmndSac,2662
+sqlglider/lineage/__init__.py,sha256=llXMeI5_PIZaiBo8tKk3-wOubF4m_6QBHbn1FtWxT7k,256
+sqlglider/lineage/analyzer.py,sha256=HyyjGMP7VvEmvt-V-qT48C-41Usj2OmT5FPYYKdJsSs,48218
+sqlglider/lineage/formatters.py,sha256=_Y9wcTX4JXn1vVnZ1xI656g1FF2rMjcAVc-GHjbd9QA,10389
+sqlglider/templating/__init__.py,sha256=g3_wb6rSDI0usq2UUMDpn-J5kVwlAw3NtLdwbxL6UHs,1435
+sqlglider/templating/base.py,sha256=y5bWAW7qXl_4pPyo5KycfHwNVvt1-7slZ63DAsvTE1s,2902
+sqlglider/templating/jinja.py,sha256=o01UG72N4G1-tOT5LKK1Wkccv4nJH2VN4VFaMi5c1-g,5220
+sqlglider/templating/registry.py,sha256=BJU3N2qNVMTUtkgbibyqo8Wme_acXQRw5XI-6ZVgyac,3476
+sqlglider/templating/variables.py,sha256=5593PtLBcOxsnMCSRm2pGAD5I0Y9f__VV3_J_HfXVlQ,8010
+sqlglider/utils/__init__.py,sha256=KGp9-UzKz_OFBOTFoSy-g-NXDZsvyWXG_9-1zcC6ePE,276
+sqlglider/utils/config.py,sha256=iNJgSXFw3pmL2MCdvW3SJp4X2T3AQP2QyQuXIXT-6H0,4761
+sqlglider/utils/file_utils.py,sha256=5_ff28E0r1R7emZzsOnRuHd-7zIX6873eyr1SuPEr4E,1093
+sql_glider-0.1.3.dist-info/METADATA,sha256=D83HzMc1l3AHbnR3Y9aOzwTbp6yMKR9tQMXWTWkH1Sw,28445
+sql_glider-0.1.3.dist-info/WHEEL,sha256=WLgqFyCfm_KASv4WHyYy0P3pM_m7J5L9k2skdKLirC8,87
+sql_glider-0.1.3.dist-info/entry_points.txt,sha256=HDuakHqHS5C0HFKsMIxMYmDU7-BLBGrnIJcYaVRu-s0,251
+sql_glider-0.1.3.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+sql_glider-0.1.3.dist-info/RECORD,,

{sql_glider-0.1.2.dist-info → sql_glider-0.1.3.dist-info}/entry_points.txt

@@ -1,6 +1,9 @@
 [console_scripts]
 sqlglider = sqlglider.cli:app
 
+[sqlglider.catalogs]
+databricks = sqlglider.catalog.databricks:DatabricksCatalog
+
 [sqlglider.templaters]
 jinja = sqlglider.templating.jinja:JinjaTemplater
 none = sqlglider.templating.base:NoOpTemplater

sqlglider/_version.py

@@ -28,7 +28,7 @@ version_tuple: VERSION_TUPLE
 commit_id: COMMIT_ID
 __commit_id__: COMMIT_ID
 
-__version__ = version = '0.1.2'
-__version_tuple__ = version_tuple = (0, 1, 2)
+__version__ = version = '0.1.3'
+__version_tuple__ = version_tuple = (0, 1, 3)
 
 __commit_id__ = commit_id = None

sqlglider/catalog/__init__.py

@@ -0,0 +1,30 @@
+"""Catalog module for fetching DDL from remote data catalogs.
+
+This module provides a plugin system for connecting to various data catalogs
+(e.g., Databricks Unity Catalog) and fetching table DDL definitions.
+
+Example:
+    >>> from sqlglider.catalog import get_catalog, list_catalogs
+    >>> print(list_catalogs())
+    ['databricks']
+    >>> catalog = get_catalog("databricks")
+    >>> catalog.configure({"warehouse_id": "abc123"})
+    >>> ddl = catalog.get_ddl("my_catalog.my_schema.my_table")
+"""
+
+from sqlglider.catalog.base import Catalog, CatalogError
+from sqlglider.catalog.registry import (
+    clear_registry,
+    get_catalog,
+    list_catalogs,
+    register_catalog,
+)
+
+__all__ = [
+    "Catalog",
+    "CatalogError",
+    "get_catalog",
+    "list_catalogs",
+    "register_catalog",
+    "clear_registry",
+]

sqlglider/catalog/base.py

@@ -0,0 +1,99 @@
+"""Base classes for catalog system.
+
+This module defines the abstract interface for catalog providers and provides
+the exception class for catalog-related errors.
+"""
+
+from abc import ABC, abstractmethod
+from typing import Any, Dict, List, Optional
+
+
+class CatalogError(Exception):
+    """Exception raised when catalog operations fail."""
+
+    pass
+
+
+class Catalog(ABC):
+    """Abstract base class for catalog providers.
+
+    All catalog implementations must inherit from this class and implement
+    the required methods. Catalogs are discovered via entry points and
+    can be used to fetch DDL definitions from remote data catalogs.
+
+    Example:
+        >>> class MyCatalog(Catalog):
+        ...     @property
+        ...     def name(self) -> str:
+        ...         return "my-catalog"
+        ...
+        ...     def get_ddl(self, table_name: str) -> str:
+        ...         # Fetch DDL from remote catalog
+        ...         return "CREATE TABLE ..."
+        ...
+        ...     def get_ddl_batch(self, table_names: List[str]) -> Dict[str, str]:
+        ...         return {name: self.get_ddl(name) for name in table_names}
+    """
+
+    @property
+    @abstractmethod
+    def name(self) -> str:
+        """Return the catalog provider name.
+
+        This name is used to identify the catalog in configuration
+        and CLI options.
+
+        Returns:
+            The unique name of this catalog provider.
+        """
+        pass
+
+    @abstractmethod
+    def get_ddl(self, table_name: str) -> str:
+        """Fetch DDL for a single table from the remote catalog.
+
+        Args:
+            table_name: The fully qualified table name (e.g., "catalog.schema.table").
+
+        Returns:
+            The DDL statement for creating the table.
+
+        Raises:
+            CatalogError: If the DDL cannot be fetched (table not found,
+                         authentication failure, network error, etc.).
+        """
+        pass
+
+    @abstractmethod
+    def get_ddl_batch(self, table_names: List[str]) -> Dict[str, str]:
+        """Fetch DDL for multiple tables from the remote catalog.
+
+        This method may be more efficient than calling get_ddl() multiple
+        times, as implementations can batch requests where supported.
+
+        Args:
+            table_names: List of fully qualified table names.
+
+        Returns:
+            Dictionary mapping table names to their DDL statements.
+            Tables that couldn't be found will have None as their value.
+
+        Raises:
+            CatalogError: If the batch operation fails entirely.
+        """
+        pass
+
+    def configure(self, config: Optional[Dict[str, Any]] = None) -> None:
+        """Configure the catalog with provider-specific settings.
+
+        This method is called after instantiation to pass configuration
+        from sqlglider.toml or environment variables.
+
+        Args:
+            config: Provider-specific configuration dictionary.
+                   Keys and values depend on the catalog implementation.
+
+        Raises:
+            CatalogError: If required configuration is missing or invalid.
+        """
+        pass

sqlglider/catalog/databricks.py

@@ -0,0 +1,255 @@
+"""Databricks catalog implementation.
+
+This module provides integration with Databricks Unity Catalog for fetching
+table DDL definitions using the Databricks SDK.
+
+Requires the optional 'databricks' dependency:
+    pip install sql-glider[databricks]
+"""
+
+import os
+from typing import Any, Dict, List, Optional
+
+from sqlglider.catalog.base import Catalog, CatalogError
+
+# Lazy import to avoid requiring databricks-sdk unless actually used
+_databricks_sdk_available: Optional[bool] = None
+
+
+def _check_databricks_sdk() -> None:
+    """Check if databricks-sdk is installed."""
+    global _databricks_sdk_available
+    if _databricks_sdk_available is None:
+        try:
+            import databricks.sdk  # noqa: F401
+
+            _databricks_sdk_available = True
+        except ImportError:
+            _databricks_sdk_available = False
+
+    if not _databricks_sdk_available:
+        raise CatalogError(
+            "The 'databricks-sdk' package is required for Databricks catalog support. "
+            "Install it with: pip install sql-glider[databricks]"
+        )
+
+
+class DatabricksCatalog(Catalog):
+    """Databricks Unity Catalog provider.
+
+    Fetches table DDL using the Databricks SDK's statement execution API.
+
+    Authentication:
+        Authentication is handled by the Databricks SDK's unified authentication,
+        which automatically tries multiple methods in order:
+
+        1. Direct configuration (host + token in sqlglider.toml)
+        2. Environment variables (DATABRICKS_HOST, DATABRICKS_TOKEN, etc.)
+        3. Databricks CLI profile (~/.databrickscfg) - use 'profile' config option
+        4. Azure CLI authentication (for Azure Databricks)
+        5. Google Cloud authentication (for GCP Databricks)
+        6. OAuth M2M (client credentials) via environment variables:
+           - DATABRICKS_CLIENT_ID
+           - DATABRICKS_CLIENT_SECRET
+
+        For OAuth M2M, set these environment variables:
+            export DATABRICKS_HOST=https://your-workspace.cloud.databricks.com
+            export DATABRICKS_CLIENT_ID=your-client-id
+            export DATABRICKS_CLIENT_SECRET=your-client-secret
+
+        For Databricks CLI profile, either:
+            - Configure DEFAULT profile in ~/.databrickscfg
+            - Set profile name in sqlglider.toml: profile = "my-profile"
+
+    Configuration:
+        - warehouse_id (required): SQL warehouse ID for statement execution
+        - profile (optional): Databricks CLI profile name from ~/.databrickscfg
+        - host (optional): Databricks workspace URL
+        - token (optional): Personal access token (legacy, prefer OAuth)
+
+    Example:
+        >>> # Using environment variables or CLI profile (recommended)
+        >>> catalog = DatabricksCatalog()
+        >>> catalog.configure({"warehouse_id": "abc123def456"})
+        >>> ddl = catalog.get_ddl("my_catalog.my_schema.my_table")
+
+        >>> # Using specific CLI profile
+        >>> catalog.configure({"warehouse_id": "abc123", "profile": "dev-workspace"})
+    """
+
+    def __init__(self) -> None:
+        """Initialize the Databricks catalog."""
+        self._warehouse_id: Optional[str] = None
+        self._profile: Optional[str] = None
+        self._host: Optional[str] = None
+        self._token: Optional[str] = None
+        self._client: Any = None
+
+    @property
+    def name(self) -> str:
+        """Return the catalog provider name."""
+        return "databricks"
+
+    def configure(self, config: Optional[Dict[str, Any]] = None) -> None:
+        """Configure the Databricks catalog.
+
+        Args:
+            config: Configuration dictionary with optional keys:
+                - warehouse_id: SQL warehouse ID (required, or set DATABRICKS_WAREHOUSE_ID)
+                - profile: Databricks CLI profile name from ~/.databrickscfg
+                - host: Databricks workspace URL (only needed if not using profile/env)
+                - token: Personal access token (legacy, prefer OAuth or profile)
+
+        Raises:
+            CatalogError: If warehouse_id is not provided and not in environment.
+        """
+        config = config or {}
+
+        # Get warehouse_id from config or environment
+        self._warehouse_id = config.get("warehouse_id") or os.environ.get(
+            "DATABRICKS_WAREHOUSE_ID"
+        )
+        if not self._warehouse_id:
+            raise CatalogError(
+                "Databricks warehouse_id is required. "
+                "Set it in sqlglider.toml under [sqlglider.catalog.databricks] "
+                "or via the DATABRICKS_WAREHOUSE_ID environment variable."
+            )
+
+        # Get optional profile for CLI profile-based auth
+        self._profile = config.get("profile")
+
+        # Get optional host and token - only from config, not env vars
+        # Let the SDK handle env var discovery for better unified auth support
+        self._host = config.get("host")
+        self._token = config.get("token")
+
+        # Reset client so it gets recreated with new config
+        self._client = None
+
+    def _get_client(self) -> Any:
+        """Get or create the Databricks WorkspaceClient.
+
+        The SDK uses unified authentication, trying methods in this order:
+        1. Explicit host/token if provided in config
+        2. Profile from ~/.databrickscfg if specified
+        3. Environment variables (DATABRICKS_HOST, DATABRICKS_TOKEN, etc.)
+        4. OAuth M2M via DATABRICKS_CLIENT_ID/DATABRICKS_CLIENT_SECRET
+        5. Azure CLI / Google Cloud auth for cloud-hosted workspaces
+
+        Returns:
+            The WorkspaceClient instance.
+
+        Raises:
+            CatalogError: If the SDK is not installed or authentication fails.
+        """
+        _check_databricks_sdk()
+
+        if self._client is None:
+            from databricks.sdk import WorkspaceClient
+
+            # Build kwargs for WorkspaceClient
+            # Only pass values that are explicitly configured
+            # Let SDK handle env var discovery for unified auth
+            kwargs: Dict[str, Any] = {}
+            if self._profile:
+                kwargs["profile"] = self._profile
+            if self._host:
+                kwargs["host"] = self._host
+            if self._token:
+                kwargs["token"] = self._token
+
+            try:
+                self._client = WorkspaceClient(**kwargs)
+            except Exception as e:
+                raise CatalogError(
+                    f"Failed to authenticate with Databricks: {e}"
+                ) from e
+
+        return self._client
+
+    def get_ddl(self, table_name: str) -> str:
+        """Fetch DDL for a single table from Databricks.
+
+        Uses SHOW CREATE TABLE to get the full DDL statement.
+
+        Args:
+            table_name: The fully qualified table name (catalog.schema.table).
+
+        Returns:
+            The CREATE TABLE DDL statement.
+
+        Raises:
+            CatalogError: If the table is not found or the query fails.
+        """
+        if not self._warehouse_id:
+            raise CatalogError(
+                "Catalog not configured. Call configure() with warehouse_id first."
+            )
+
+        client = self._get_client()
+
+        try:
+            # Execute SHOW CREATE TABLE statement
+            response = client.statement_execution.execute_statement(
+                warehouse_id=self._warehouse_id,
+                statement=f"SHOW CREATE TABLE {table_name}",
+                wait_timeout="30s",
+            )
+
+            # Check for errors
+            if response.status and response.status.state:
+                state = response.status.state.value
+                if state == "FAILED":
+                    error_msg = (
+                        response.status.error.message
+                        if response.status.error
+                        else "Unknown error"
+                    )
+                    raise CatalogError(
+                        f"Failed to get DDL for '{table_name}': {error_msg}"
+                    )
+
+            # Extract DDL from result
+            if response.result and response.result.data_array:
+                # SHOW CREATE TABLE returns a single row with the DDL
+                ddl_parts = []
+                for row in response.result.data_array:
+                    if row:
+                        ddl_parts.append(str(row[0]))
+                return "\n".join(ddl_parts)
+
+            raise CatalogError(f"No DDL returned for table '{table_name}'")
+
+        except CatalogError:
+            raise
+        except Exception as e:
+            raise CatalogError(f"Failed to fetch DDL for '{table_name}': {e}") from e
+
+    def get_ddl_batch(self, table_names: List[str]) -> Dict[str, str]:
+        """Fetch DDL for multiple tables from Databricks.
+
+        Currently executes individual queries for each table.
+        Future optimization could use parallel execution.
+
+        Args:
+            table_names: List of fully qualified table names.
+
+        Returns:
+            Dictionary mapping table names to their DDL statements.
+            Tables that couldn't be found will have error messages as values
+            prefixed with "ERROR: ".
+
+        Raises:
+            CatalogError: If the batch operation fails entirely.
+        """
+        results: Dict[str, str] = {}
+
+        for table_name in table_names:
+            try:
+                results[table_name] = self.get_ddl(table_name)
+            except CatalogError as e:
+                # Store error message for this table but continue with others
+                results[table_name] = f"ERROR: {e}"
+
+        return results