fix: don't use Path.rename if files are on different filesystems

openfoodfacts_exports/exports/csv.py

@@ -1,4 +1,6 @@
 import logging
+import shutil
+import tempfile
 from pathlib import Path
 
 import duckdb
@@ -34,11 +36,13 @@ def generate_mobile_app_dump(parquet_path: Path, output_path: Path) -> None:
     if not parquet_path.exists():
         raise FileNotFoundError(f"{str(parquet_path)} was not found.")
 
-    query = MOBILE_APP_DUMP_SQL_QUERY.replace(
-        "{dataset_path}", str(parquet_path)
-    ).replace("{output_path}", str(output_path))
-    try:
+    with tempfile.TemporaryDirectory() as tmp_dir:
+        tmp_file_path = Path(tmp_dir) / "mobile_dump.csv"
+        query = MOBILE_APP_DUMP_SQL_QUERY.replace(
+            "{dataset_path}", str(parquet_path)
+        ).replace("{output_path}", str(tmp_file_path))
         duckdb.sql(query)
-    except duckdb.Error as e:
-        logger.error(f"Error executing query: {query}\nError message: {e}")
-        raise
+        # Move dataset file to output_path
+        shutil.move(tmp_file_path, output_path)
+
+    logger.info("Mobile app dump generation done")

openfoodfacts_exports/exports/parquet.py

@@ -1,4 +1,5 @@
 import logging
+import shutil
 import tempfile
 from pathlib import Path
 
@@ -161,8 +162,8 @@ def export_parquet(dataset_path: Path, output_path: Path) -> None:
         convert_jsonl_to_parquet(
             output_file_path=tmp_file_path, dataset_path=dataset_path
         )
-        # Copy dataset file to datasets directory
-        tmp_file_path.rename(output_path)
+        # Move dataset file to output_path
+        shutil.move(tmp_file_path, output_path)
 
     if settings.ENABLE_HF_PUSH:
         push_parquet_file_to_hf(data_path=output_path)