gitea-admin/gh-christianlouis-docuelevate
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix(api): use processed_file_path from DB in upload retry path resolution

Browse Source 
The upload retry logic now checks file_record.processed_file_path first
(the GPT-suggested filename stored during finalization), before falling
back to legacy hash-based and original-filename-based path patterns.

This fixes the case where the processed file has a different name than
the original (e.g., '2023-10-01_Unknown.pdf' vs 'cable_graphic.pdf')
and the retry couldn't find the file on disk.

Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>
This commit is contained in:
copilot-swe-agent[bot]
2026-02-28 12:56:01 +00:00
parent 5fe0f6f864
commit bb98afac82
2 changed files with 108 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files
app/api/files.py
+33 -17
View File
@@ -763,28 +763,44 @@ def retry_subtask(
logger.info(
f"Retrying upload task {subtask_name} for file {file_id}: "
f"original_filename={file_record.original_filename!r}, "
f"filehash={file_record.filehash!r}"
f"filehash={file_record.filehash!r}, "
f"processed_file_path={file_record.processed_file_path!r}"
)
workdir = settings.workdir
processed_dir = os.path.join(workdir, "processed")
# Try to find the processed file
base_filename = os.path.splitext(file_record.original_filename)[0]
potential_paths = [
os.path.join(processed_dir, f"{file_record.filehash}.pdf"),
os.path.join(processed_dir, f"{base_filename}_processed.pdf"),
os.path.join(processed_dir, file_record.original_filename),
]
# Try to find the processed file using multiple strategies:
# 1. DB-stored processed_file_path (most reliable - set during finalization)
# 2. Hash-based path in processed dir
# 3. Original filename with _processed suffix
# 4. Original filename in processed dir
file_path = None
checked_paths = []
for path in potential_paths:
exists = os.path.exists(path)
checked_paths.append(f"{path!r} (exists={exists})")
logger.info(f"Checking processed file path: {path!r}, exists={exists}")
# Check 1: processed_file_path from DB
if file_record.processed_file_path:
exists = os.path.exists(file_record.processed_file_path)
checked_paths.append(f"processed_file_path={file_record.processed_file_path!r} (exists={exists})")
logger.info(f"Checking processed_file_path: {file_record.processed_file_path!r}, exists={exists}")
if exists:
file_path = path
break
file_path = file_record.processed_file_path
# Check 2-4: Legacy path patterns in processed directory
if not file_path:
workdir = settings.workdir
processed_dir = os.path.join(workdir, "processed")
base_filename = os.path.splitext(file_record.original_filename)[0]
legacy_paths = [
os.path.join(processed_dir, f"{file_record.filehash}.pdf"),
os.path.join(processed_dir, f"{base_filename}_processed.pdf"),
os.path.join(processed_dir, file_record.original_filename),
]
for path in legacy_paths:
exists = os.path.exists(path)
checked_paths.append(f"{path!r} (exists={exists})")
logger.info(f"Checking processed file path: {path!r}, exists={exists}")
if exists:
file_path = path
break
if not file_path:
paths_detail = "; ".join(checked_paths)