fix: multi-worker translation status and OCR fallback handling

Translation status (multi-worker support):
- Add filesystem lock files (.translating) to track in-progress translations
- Check lock files in /status API when job_state not found in current worker
- Remove lock files on translation success or failure

OCR fallback fix:
- Fix empty pages when layout analysis fails but OCR succeeds
- Change 'enhanced_results' in ocr_results to ocr_results.get('enhanced_results')
- This ensures fallback to text_regions when enhanced_results is empty list

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This commit is contained in:
egg
2025-12-14 16:36:36 +08:00
parent 1c37585be2
commit 3ccbdb8394
4 changed files with 36 additions and 3 deletions

View File

@@ -343,9 +343,26 @@ async def get_translation_status(
job_state = translation_service.get_job_state(task_id) job_state = translation_service.get_job_state(task_id)
if not job_state: if not job_state:
# No active job - check if any completed translations exist # No active job in this worker - check filesystem for status
if task.result_json_path: if task.result_json_path:
result_dir = Path(task.result_json_path).parent result_dir = Path(task.result_json_path).parent
# Check for in-progress translation (lock file exists)
lock_files = list(result_dir.glob("*.translating"))
if lock_files:
# Translation is in progress (possibly in another worker)
latest_lock = max(lock_files, key=lambda f: f.stat().st_mtime)
# Extract language from lock filename: {filename}_translated_{lang}.translating
lock_stem = latest_lock.stem # e.g., "scan_translated_en"
lang = lock_stem.split("_translated_")[-1] if "_translated_" in lock_stem else "unknown"
return TranslationStatusResponse(
task_id=task_id,
status=TranslationStatusEnum.TRANSLATING,
target_lang=lang,
progress=TranslationProgress(percentage=50.0) # Approximate progress
)
# Check for completed translations
translated_files = list(result_dir.glob("*_translated_*.json")) translated_files = list(result_dir.glob("*_translated_*.json"))
if translated_files: if translated_files:
# Return completed status for the most recent translation # Return completed status for the most recent translation

View File

@@ -439,14 +439,15 @@ class OCRToUnifiedConverter:
ocr_dimensions = ocr_results.get('ocr_dimensions', {}) ocr_dimensions = ocr_results.get('ocr_dimensions', {})
# Check if we have enhanced results from PPStructureEnhanced # Check if we have enhanced results from PPStructureEnhanced
if 'enhanced_results' in ocr_results: # Note: Must check for non-empty list, not just key existence (key may exist with empty list)
if ocr_results.get('enhanced_results'):
pages = self._extract_from_enhanced_results( pages = self._extract_from_enhanced_results(
ocr_results['enhanced_results'], ocr_results['enhanced_results'],
raw_text_regions=raw_text_regions, raw_text_regions=raw_text_regions,
ocr_dimensions=ocr_dimensions ocr_dimensions=ocr_dimensions
) )
# Check for traditional OCR results with text_regions at top level (from process_file_traditional) # Check for traditional OCR results with text_regions at top level (from process_file_traditional)
elif 'text_regions' in ocr_results: elif ocr_results.get('text_regions'):
pages = self._extract_from_traditional_ocr(ocr_results) pages = self._extract_from_traditional_ocr(ocr_results)
# Check for traditional layout_data structure # Check for traditional layout_data structure
elif 'layout_data' in ocr_results: elif 'layout_data' in ocr_results:

View File

@@ -708,7 +708,14 @@ class TranslationService:
f"Starting translation: task_id={task_id}, target={target_lang}" f"Starting translation: task_id={task_id}, target={target_lang}"
) )
# Create lock file to indicate translation in progress (for multi-worker support)
output_filename = result_json_path.stem.replace('_result', '')
lock_file_path = result_json_path.parent / f"{output_filename}_translated_{target_lang}.translating"
try: try:
# Create lock file
lock_file_path.touch()
# Load source JSON # Load source JSON
with open(result_json_path, 'r', encoding='utf-8') as f: with open(result_json_path, 'r', encoding='utf-8') as f:
result_json = json.load(f) result_json = json.load(f)
@@ -786,12 +793,19 @@ class TranslationService:
f"saved to {output_path}" f"saved to {output_path}"
) )
# Remove lock file on success
if lock_file_path.exists():
lock_file_path.unlink()
return True, output_path, None return True, output_path, None
except Exception as e: except Exception as e:
logger.error(f"Translation failed: {e}") logger.error(f"Translation failed: {e}")
import traceback import traceback
traceback.print_exc() traceback.print_exc()
# Remove lock file on failure
if lock_file_path.exists():
lock_file_path.unlink()
return False, None, str(e) return False, None, str(e)
def get_job_state(self, task_id: str) -> Optional[TranslationJobState]: def get_job_state(self, task_id: str) -> Optional[TranslationJobState]:

View File

@@ -30,6 +30,7 @@ FRONTEND_PORT=${FRONTEND_PORT:-5173}
# Production settings # Production settings
PROD_MODE=false PROD_MODE=false
# Translation status uses filesystem lock files for multi-worker support
UVICORN_WORKERS=${UVICORN_WORKERS:-4} UVICORN_WORKERS=${UVICORN_WORKERS:-4}
# Create PID directory # Create PID directory