ocr speed improved

test_ocr_batch.py

@@ -0,0 +1,52 @@
+#!/usr/bin/env python3
+"""
+Test OCR batch processing and initialization improvements.
+"""
+import sys
+import os
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), 'LightRAG-main'))
+
+from lightrag.optimized_ocr_processor import OptimizedOCRProcessor
+from lightrag.document_processor import DocumentProcessor
+import asyncio
+import time
+
+def test_ocr_initialization():
+    print("Testing OCR initialization improvements...")
+    processor = OptimizedOCRProcessor(use_gpu=True, batch_size=4, max_workers=2)
+    
+    # Wait for initialization (should be quick if pre-warmed)
+    start = time.time()
+    ready = processor._ensure_ocr_initialized(timeout=30.0)
+    elapsed = time.time() - start
+    print(f"OCR ready: {ready}, elapsed: {elapsed:.2f}s")
+    
+    if ready:
+        print("OCR metrics:", processor.get_metrics())
+    else:
+        print("OCR not available")
+    
+    processor.close()
+
+async def test_document_processor():
+    print("\nTesting DocumentProcessor with batch OCR...")
+    processor = DocumentProcessor()
+    # Use a dummy PDF file (if exists) or just test initialization
+    test_pdf = "test_meaningful.pdf"
+    if os.path.exists(test_pdf):
+        print(f"Processing {test_pdf}...")
+        start = time.time()
+        result = await processor.process_document(test_pdf)
+        elapsed = time.time() - start
+        print(f"Processing completed in {elapsed:.2f}s")
+        print(f"Success: {result.success}")
+        print(f"Pages: {result.metadata.get('pages', 'N/A')}")
+        print(f"Processed with OCR: {result.metadata.get('processed_with_ocr', False)}")
+        print(f"Content length: {len(result.content)}")
+    else:
+        print(f"Test PDF not found at {test_pdf}, skipping processing test.")
+
+if __name__ == "__main__":
+    test_ocr_initialization()
+    asyncio.run(test_document_processor())
+    print("\nAll tests completed.")