dbt-cube-sync 0.1.0a8__py3-none-any.whl → 0.1.0a10__py3-none-any.whl

dbt_cube_sync/connectors/superset.py

@@ -60,13 +60,19 @@ class SupersetConnector(BaseConnector):
         """Authenticate and get JWT token"""
         login_url = f"{self.base_url}/api/v1/security/login"
         payload = {
-            "username": self.config['username'],
             "password": self.config['password'],
             "provider": "db",
-            "refresh": True
+            "refresh": "true",
+            "username": self.config['username']
         }
-        
+
         response = self.session.post(login_url, json=payload)
+        if response.status_code == 401:
+            raise Exception(
+                f"Superset authentication failed (401). "
+                f"Check username/password and ensure provider='{payload['provider']}' is correct. "
+                f"Response: {response.text}"
+            )
         response.raise_for_status()
         
         data = response.json()

dbt_cube_sync/core/db_inspector.py

@@ -1,17 +1,16 @@
 """
-Database inspector - fetches column types using SQLAlchemy MetaData reflection.
+Database inspector - fetches column types using SQLAlchemy or direct SQL.
 
-Uses SQLAlchemy's Table(..., autoload_with=engine) for portable, database-agnostic
-column type extraction. This approach works consistently across PostgreSQL, MySQL,
-Snowflake, BigQuery, Redshift, and other databases.
+Uses Redshift-specific queries for Redshift databases (which don't support
+standard PostgreSQL reflection), and SQLAlchemy reflection for other databases.
 """
 from typing import Dict, Optional
-from sqlalchemy import create_engine, MetaData, Table
+from sqlalchemy import create_engine, MetaData, Table, text
 from sqlalchemy.engine import Engine
 
 
 class DatabaseInspector:
-    """Inspects database schema to extract column type information using SQLAlchemy reflection."""
+    """Inspects database schema to extract column type information."""
 
     def __init__(self, sqlalchemy_uri: str):
         """
@@ -20,8 +19,9 @@ class DatabaseInspector:
         Args:
             sqlalchemy_uri: SQLAlchemy connection URI (e.g., postgresql://user:pass@host:port/db)
         """
-        # Add connect_args for Redshift compatibility
-        if 'redshift' in sqlalchemy_uri.lower():
+        self.is_redshift = 'redshift' in sqlalchemy_uri.lower()
+
+        if self.is_redshift:
             self.engine: Engine = create_engine(
                 sqlalchemy_uri,
                 connect_args={'sslmode': 'prefer'}
@@ -30,13 +30,62 @@ class DatabaseInspector:
             self.engine: Engine = create_engine(sqlalchemy_uri)
 
         self.metadata = MetaData()
-        self._table_cache: Dict[str, Table] = {}
+        self._table_cache: Dict[str, Dict[str, str]] = {}
+
+    def _get_redshift_columns(self, schema: str, table_name: str) -> Dict[str, str]:
+        """
+        Get column types from Redshift using LIMIT 0 query (fastest method).
+
+        Executes SELECT * FROM table LIMIT 0 and reads column types from cursor description.
+        This is very fast because it doesn't scan any data - just returns metadata.
+        """
+        columns = {}
+
+        # LIMIT 0 query - returns no rows but gives us column metadata
+        query = text(f'SELECT * FROM "{schema}"."{table_name}" LIMIT 0')
+
+        # Redshift type OID to name mapping (common types)
+        redshift_type_map = {
+            16: 'boolean',
+            20: 'bigint',
+            21: 'smallint',
+            23: 'integer',
+            25: 'text',
+            700: 'real',
+            701: 'double precision',
+            1042: 'char',
+            1043: 'varchar',
+            1082: 'date',
+            1083: 'time',
+            1114: 'timestamp',
+            1184: 'timestamptz',
+            1700: 'numeric',
+            2950: 'uuid',
+        }
+
+        try:
+            with self.engine.connect() as conn:
+                result = conn.execute(query)
+                # Get column info from cursor description
+                # Format: (name, type_code, display_size, internal_size, precision, scale, null_ok)
+                if result.cursor and result.cursor.description:
+                    for col_desc in result.cursor.description:
+                        col_name = col_desc[0]
+                        type_code = col_desc[1]
+                        # Map type code to type name, fallback to 'varchar' if unknown
+                        col_type = redshift_type_map.get(type_code, 'varchar')
+                        columns[col_name] = col_type
+
+        except Exception as e:
+            print(f"Warning: Could not inspect Redshift table {schema}.{table_name}: {e}")
+
+        return columns
 
     def get_table_columns(self, schema: str, table_name: str) -> Dict[str, str]:
         """
         Get column names and their data types for a specific table.
 
-        Uses SQLAlchemy MetaData reflection for portable column extraction.
+        Uses Redshift-specific queries for Redshift, SQLAlchemy reflection for others.
 
         Args:
             schema: Database schema name
@@ -45,30 +94,33 @@ class DatabaseInspector:
         Returns:
             Dictionary mapping column names to data types
         """
-        columns = {}
         cache_key = f"{schema}.{table_name}"
 
-        try:
-            # Check cache first
-            if cache_key in self._table_cache:
-                table = self._table_cache[cache_key]
-            else:
-                # Reflect table using SQLAlchemy MetaData
+        # Check cache first
+        if cache_key in self._table_cache:
+            return self._table_cache[cache_key]
+
+        columns = {}
+
+        if self.is_redshift:
+            # Use Redshift-specific query
+            columns = self._get_redshift_columns(schema, table_name)
+        else:
+            # Use standard SQLAlchemy reflection
+            try:
                 table = Table(
                     table_name,
                     self.metadata,
                     autoload_with=self.engine,
                     schema=schema
                 )
-                self._table_cache[cache_key] = table
+                for column in table.columns:
+                    columns[column.name] = str(column.type)
 
-            # Extract column types
-            for column in table.columns:
-                columns[column.name] = str(column.type)
-
-        except Exception as e:
-            print(f"Warning: Could not inspect table {schema}.{table_name}: {e}")
+            except Exception as e:
+                print(f"Warning: Could not inspect table {schema}.{table_name}: {e}")
 
+        self._table_cache[cache_key] = columns
         return columns
 
     def reflect_multiple_tables(

{dbt_cube_sync-0.1.0a8.dist-info → dbt_cube_sync-0.1.0a10.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dbt-cube-sync
-Version: 0.1.0a8
+Version: 0.1.0a10
 Summary: Synchronization tool for dbt models to Cube.js schemas and BI tools
 Author: Ponder
 Requires-Python: >=3.9,<4.0

{dbt_cube_sync-0.1.0a8.dist-info → dbt_cube_sync-0.1.0a10.dist-info}/RECORD

@@ -4,15 +4,15 @@ dbt_cube_sync/config.py,sha256=qhGE7CxTmh0RhPizgd3x3Yj-3L2LoC00UQIDT0q9FlQ,3858
 dbt_cube_sync/connectors/__init__.py,sha256=NG6tYZ3CYD5bG_MfNLZrUM8YoBEKArG8-AOmJ8pwvQI,52
 dbt_cube_sync/connectors/base.py,sha256=JLzerxJdt34z0kWuyieL6UQhf5_dUYPGmwkiRWBuSPY,2802
 dbt_cube_sync/connectors/powerbi.py,sha256=2Y8fTfh_6Q_Myma1ymipPh1U3HsfQKcktVequXXnIXI,1275
-dbt_cube_sync/connectors/superset.py,sha256=5YEqadVZRPFAJkgvhqkse3JuGJkQHfyvT88jy52ow_0,21429
+dbt_cube_sync/connectors/superset.py,sha256=D_pWTN0F84mUosnsm-NG_v9IlVE2dviIDh08WxHEOIA,21709
 dbt_cube_sync/connectors/tableau.py,sha256=jKve1zErzTbgPOtmPB92ZwZl4I6uEySedM51JiwlGrE,1261
 dbt_cube_sync/core/__init__.py,sha256=kgsawtU5dqEvnHz6dU8qwJbH3rtIV7QlK2MhtYVDCaY,46
 dbt_cube_sync/core/cube_generator.py,sha256=DtmaA_dtWmBVJnSWHVoQi-3KEsRc0axHZpCUEcKeYAk,11061
-dbt_cube_sync/core/db_inspector.py,sha256=Ccd9ieGNlwHDHdgMVDEOfjs7R9Mjj904OW1P-mDSsyo,3155
+dbt_cube_sync/core/db_inspector.py,sha256=V_cd12FBXj-1gB2JZeYmkQluUO-UYufy_tvfYoJXCGI,5073
 dbt_cube_sync/core/dbt_parser.py,sha256=KbhDoB0ULP6JDUPZPDVbm9yCtRKrW17ptGoJvVLtueY,12763
 dbt_cube_sync/core/models.py,sha256=2s5iZ9MEBGfSzkB4HJB5vG0mZqNXNJSfAD3Byw1IVe4,3203
 dbt_cube_sync/core/state_manager.py,sha256=7uXJtlZBIWj6s6XgAhNlP6UHdfhH0y461iyQlfidqGI,7233
-dbt_cube_sync-0.1.0a8.dist-info/METADATA,sha256=fsb721DeeHXUeeeLIihijjIiM6x7Wl8fUTeBMzlyoZo,10680
-dbt_cube_sync-0.1.0a8.dist-info/WHEEL,sha256=3ny-bZhpXrU6vSQ1UPG34FoxZBp3lVcvK0LkgUz6VLk,88
-dbt_cube_sync-0.1.0a8.dist-info/entry_points.txt,sha256=iEAB_nZ1AoSeFwSHPY2tr02xmTHLVFKp5CJeFh0AfCw,56
-dbt_cube_sync-0.1.0a8.dist-info/RECORD,,
+dbt_cube_sync-0.1.0a10.dist-info/METADATA,sha256=Foy8KI7-ILdZwTvejfjukArtnGNJJg85vwIlQoMS31w,10681
+dbt_cube_sync-0.1.0a10.dist-info/WHEEL,sha256=3ny-bZhpXrU6vSQ1UPG34FoxZBp3lVcvK0LkgUz6VLk,88
+dbt_cube_sync-0.1.0a10.dist-info/entry_points.txt,sha256=iEAB_nZ1AoSeFwSHPY2tr02xmTHLVFKp5CJeFh0AfCw,56
+dbt_cube_sync-0.1.0a10.dist-info/RECORD,,