sqllocks-spindle 2.2.2__tar.gz → 2.2.4__tar.gz

{sqllocks_spindle-2.2.2/sqllocks_spindle.egg-info → sqllocks_spindle-2.2.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sqllocks-spindle
-Version: 2.2.2
+Version: 2.2.4
 Summary: Multi-domain, schema-aware synthetic data generator for Microsoft Fabric
 Author-email: Jonathan Stewart <jonathan@sqllocks.com>
 License: MIT

{sqllocks_spindle-2.2.2 → sqllocks_spindle-2.2.4}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "sqllocks-spindle"
-version = "2.2.2"
+version = "2.2.4"
 description = "Multi-domain, schema-aware synthetic data generator for Microsoft Fabric"
 readme = "README.md"
 license = { text = "MIT" }

{sqllocks_spindle-2.2.2 → sqllocks_spindle-2.2.4}/sqllocks_spindle/__init__.py

@@ -1,6 +1,6 @@
 """Spindle by SQLLocks — Multi-domain synthetic data generator for Microsoft Fabric."""
 
-__version__ = "2.2.2"
+__version__ = "2.2.3"
 
 from sqllocks_spindle.engine.generator import Spindle
 from sqllocks_spindle.domains.retail import RetailDomain

{sqllocks_spindle-2.2.2 → sqllocks_spindle-2.2.4}/sqllocks_spindle/fabric/warehouse_bulk_writer.py

@@ -240,11 +240,20 @@ class WarehouseBulkWriter:
             if pd.api.types.is_datetime64_ns_dtype(chunk_df[col]):
                 chunk_df[col] = chunk_df[col].astype("datetime64[us]")
 
-        # Inside Fabric Notebook — write via notebookutils (no upload needed)
+        # Inside Fabric Notebook — write to local temp then copy to OneLake.
+        # notebookutils.fs.put() is UTF-8 text only (per Microsoft Learn docs);
+        # binary Parquet data must go through fs.cp() from a local file.
         try:
             import notebookutils  # type: ignore[import-not-found]
-            parquet_bytes = chunk_df.to_parquet(index=False, engine="pyarrow")
-            notebookutils.fs.put(remote_path, parquet_bytes, overwrite=True)
+            import os as _os
+
+            local_tmp = tempfile.mktemp(suffix=".parquet")
+            try:
+                chunk_df.to_parquet(local_tmp, index=False, engine="pyarrow")
+                notebookutils.fs.cp(f"file://{local_tmp}", remote_path, recurse=False)
+            finally:
+                if _os.path.exists(local_tmp):
+                    _os.unlink(local_tmp)
             logger.info(
                 "Staged chunk %d for %s on OneLake via notebookutils (%d rows)",
                 chunk_idx, table_name, len(chunk_df),

{sqllocks_spindle-2.2.2 → sqllocks_spindle-2.2.4}/sqllocks_spindle/packs/runner.py

@@ -144,7 +144,7 @@ class PackRunner:
 
         # Record outputs in manifest
         for table_name, df in gen_result.tables.items():
-            table_paths = [p for p in files_written if table_name in p]
+            table_paths = [p for p in files_written if table_name in str(p)]
             builder.record_output(
                 table_name=table_name,
                 rows=len(df),

{sqllocks_spindle-2.2.2 → sqllocks_spindle-2.2.4/sqllocks_spindle.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sqllocks-spindle
-Version: 2.2.2
+Version: 2.2.4
 Summary: Multi-domain, schema-aware synthetic data generator for Microsoft Fabric
 Author-email: Jonathan Stewart <jonathan@sqllocks.com>
 License: MIT