collective/mediawiki_dump_tools
17
0
Fork 0
Code Issues Pull Requests Activity

improve print debugging.

Browse Source
This commit is contained in:
Nathan TeBlunthuis 2025-12-10 19:50:47 -08:00
parent c6e96c2f54
commit f4a9491ff2
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
src/wikiq/resume.py
View File

@ -241,14 +241,14 @@ def merge_parquet_files(original_path, temp_path, merged_path):
original_pq = pq.ParquetFile(original_path) original_pq = pq.ParquetFile(original_path)
except Exception: except Exception:
# Original file is invalid (empty or corrupted) # Original file is invalid (empty or corrupted)
print(f"Note: No data in original file (namespace had no records or file was not properly written)", file=sys.stderr) print(f"Note: No data in original file {original_path} (namespace had no records or file was not properly written)", file=sys.stderr)
return None return None
try: try:
temp_pq = pq.ParquetFile(temp_path) temp_pq = pq.ParquetFile(temp_path)
except Exception: except Exception:
# Temp file is invalid (empty or corrupted) - no new data was written # Temp file is invalid (empty or corrupted) - no new data was written
print(f"Note: No new data in temp file (namespace had no records after resume point)", file=sys.stderr) print(f"Note: No new data in temp file {temp_path} (namespace had no records after resume point)", file=sys.stderr)
return None return None
merged_writer = None merged_writer = None