duckrun 0.2.0__py3-none-any.whl → 0.2.2__py3-none-any.whl

duckrun/core.py

@@ -195,21 +195,22 @@ class Duckrun:
 
     def __init__(self, workspace: str, lakehouse_name: str, schema: str = "dbo", 
                  sql_folder: Optional[str] = None, compaction_threshold: int = 10,
-                 scan_all_schemas: bool = False):
+                 scan_all_schemas: bool = False, storage_account: str = "onelake"):
         self.workspace = workspace
         self.lakehouse_name = lakehouse_name
         self.schema = schema
         self.sql_folder = sql_folder.strip() if sql_folder else None
         self.compaction_threshold = compaction_threshold
         self.scan_all_schemas = scan_all_schemas
-        self.table_base_url = f'abfss://{workspace}@onelake.dfs.fabric.microsoft.com/{lakehouse_name}.Lakehouse/Tables/'
+        self.storage_account = storage_account
+        self.table_base_url = f'abfss://{workspace}@{storage_account}.dfs.fabric.microsoft.com/{lakehouse_name}.Lakehouse/Tables/'
         self.con = duckdb.connect()
         self.con.sql("SET preserve_insertion_order = false")
         self._attach_lakehouse()
 
     @classmethod
     def connect(cls, connection_string: str, sql_folder: Optional[str] = None,
-                compaction_threshold: int = 100):
+                compaction_threshold: int = 100, storage_account: str = "onelake"):
         """
         Create and connect to lakehouse.
         
@@ -219,11 +220,13 @@ class Duckrun:
             connection_string: OneLake path "ws/lh.lakehouse/schema" or "ws/lh.lakehouse"
             sql_folder: Optional path or URL to SQL files folder
             compaction_threshold: File count threshold for compaction
+            storage_account: Storage account name (default: "onelake")
         
         Examples:
             dr = Duckrun.connect("ws/lh.lakehouse/schema", sql_folder="./sql")
             dr = Duckrun.connect("ws/lh.lakehouse/schema")  # no SQL folder
             dr = Duckrun.connect("ws/lh.lakehouse")  # defaults to dbo schema
+            dr = Duckrun.connect("ws/lh.lakehouse", storage_account="xxx-onelake")  # custom storage
         """
         print("Connecting to Lakehouse...")
         
@@ -261,7 +264,7 @@ class Duckrun:
                 "  connect('workspace/lakehouse.lakehouse')  # defaults to dbo"
             )
         
-        return cls(workspace, lakehouse_name, schema, sql_folder, compaction_threshold, scan_all_schemas)
+        return cls(workspace, lakehouse_name, schema, sql_folder, compaction_threshold, scan_all_schemas, storage_account)
 
     def _get_storage_token(self):
         return os.environ.get("AZURE_STORAGE_TOKEN", "PLACEHOLDER_TOKEN_TOKEN_NOT_AVAILABLE")
@@ -295,7 +298,7 @@ class Duckrun:
             token = token_obj.token
             os.environ["AZURE_STORAGE_TOKEN"] = token
         
-        url = f"abfss://{self.workspace}@onelake.dfs.fabric.microsoft.com/"
+        url = f"abfss://{self.workspace}@{self.storage_account}.dfs.fabric.microsoft.com/"
         store = AzureStore.from_url(url, bearer_token=token)
         
         base_path = f"{self.lakehouse_name}.Lakehouse/Tables/"
@@ -354,7 +357,13 @@ class Duckrun:
             attached_count = 0
             for schema_name, table_name in tables:
                 try:
-                    view_name = f"{schema_name}_{table_name}" if self.scan_all_schemas else table_name
+                    if self.scan_all_schemas:
+                        # Create proper schema.table structure in DuckDB
+                        self.con.sql(f"CREATE SCHEMA IF NOT EXISTS {schema_name}")
+                        view_name = f"{schema_name}.{table_name}"
+                    else:
+                        # Single schema mode - use just table name
+                        view_name = table_name
                     
                     self.con.sql(f"""
                         CREATE OR REPLACE VIEW {view_name}
@@ -371,7 +380,7 @@ class Duckrun:
             print(f"{'='*60}\n")
             
             if self.scan_all_schemas:
-                print(f"\n💡 Note: Tables are prefixed with schema (e.g., dbo_tablename)")
+                print(f"\n💡 Note: Tables use schema.table format (e.g., aemo.calendar, dbo.results)")
                 print(f"   Default schema for operations: {self.schema}\n")
                 
         except Exception as e:
@@ -412,7 +421,8 @@ class Duckrun:
         full_params = {
             'ws': self.workspace,
             'lh': self.lakehouse_name,
-            'schema': self.schema
+            'schema': self.schema,
+            'storage_account': self.storage_account
         }
         if params:
             full_params.update(params)
@@ -661,7 +671,7 @@ class Duckrun:
             os.environ["AZURE_STORAGE_TOKEN"] = token
         
         # Setup OneLake Files URL (not Tables)
-        files_base_url = f'abfss://{self.workspace}@onelake.dfs.fabric.microsoft.com/{self.lakehouse_name}.Lakehouse/Files/'
+        files_base_url = f'abfss://{self.workspace}@{self.storage_account}.dfs.fabric.microsoft.com/{self.lakehouse_name}.Lakehouse/Files/'
         store = AzureStore.from_url(files_base_url, bearer_token=token)
         
         # Collect files to upload
@@ -768,7 +778,7 @@ class Duckrun:
             os.environ["AZURE_STORAGE_TOKEN"] = token
         
         # Setup OneLake Files URL (not Tables)
-        files_base_url = f'abfss://{self.workspace}@onelake.dfs.fabric.microsoft.com/{self.lakehouse_name}.Lakehouse/Files/'
+        files_base_url = f'abfss://{self.workspace}@{self.storage_account}.dfs.fabric.microsoft.com/{self.lakehouse_name}.Lakehouse/Files/'
         store = AzureStore.from_url(files_base_url, bearer_token=token)
         
         # Create local directory

{duckrun-0.2.0.dist-info → duckrun-0.2.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: duckrun
-Version: 0.2.0
+Version: 0.2.2
 Summary: Lakehouse task runner powered by DuckDB for Microsoft Fabric
 Author: mim
 License: MIT

duckrun-0.2.2.dist-info/RECORD

@@ -0,0 +1,7 @@
+duckrun/__init__.py,sha256=L0jRtD9Ld8Ti4e6GRvPDdHvkQCFAPHM43GSP7ARh6EM,241
+duckrun/core.py,sha256=VqfTL4fFE-XUXXsDy9VRFEPSQ21dfrkCGH_06C9CLNg,39416
+duckrun-0.2.2.dist-info/licenses/LICENSE,sha256=-DeQQwdbCbkB4507ZF3QbocysB-EIjDtaLexvqRkGZc,1083
+duckrun-0.2.2.dist-info/METADATA,sha256=J_Vw7Ps5afPRkofvyo-r7wufizjS431XgXpHdwaKwyo,18339
+duckrun-0.2.2.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+duckrun-0.2.2.dist-info/top_level.txt,sha256=BknMEwebbUHrVAp3SC92ps8MPhK7XSYsaogTvi_DmEU,8
+duckrun-0.2.2.dist-info/RECORD,,

duckrun-0.2.0.dist-info/RECORD

@@ -1,7 +0,0 @@
-duckrun/__init__.py,sha256=L0jRtD9Ld8Ti4e6GRvPDdHvkQCFAPHM43GSP7ARh6EM,241
-duckrun/core.py,sha256=XGw8MaXZdlnrZXHWlMjXDof4AfRC9RYrU_rZQTNOnOc,38677
-duckrun-0.2.0.dist-info/licenses/LICENSE,sha256=-DeQQwdbCbkB4507ZF3QbocysB-EIjDtaLexvqRkGZc,1083
-duckrun-0.2.0.dist-info/METADATA,sha256=AjMXnfXohPk1SChQULq8CCBjkFH7oTzqimxnLNjbJuc,18339
-duckrun-0.2.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-duckrun-0.2.0.dist-info/top_level.txt,sha256=BknMEwebbUHrVAp3SC92ps8MPhK7XSYsaogTvi_DmEU,8
-duckrun-0.2.0.dist-info/RECORD,,