125 lines
4.1 KiB
Python
125 lines
4.1 KiB
Python
import os
|
|
import sys
|
|
|
|
def force_paddleocr_config():
|
|
"""Force PaddleOCR configuration by modifying environment and config files"""
|
|
|
|
print("Forcing PaddleOCR configuration...")
|
|
|
|
# 1. Set environment variable permanently
|
|
os.environ['LIGHTRAG_OCR_ENGINE'] = 'paddleocr'
|
|
print("Set LIGHTRAG_OCR_ENGINE=paddleocr environment variable")
|
|
|
|
# 2. Check if PaddleOCR is available
|
|
try:
|
|
import paddleocr
|
|
print("PaddleOCR is installed")
|
|
except ImportError:
|
|
print("PaddleOCR is not installed. Installing...")
|
|
os.system("pip install paddleocr")
|
|
print("PaddleOCR installed")
|
|
|
|
# 3. Verify configuration in LightRAG
|
|
config_path = "LightRAG-main/config.ini"
|
|
if os.path.exists(config_path):
|
|
with open(config_path, 'r') as f:
|
|
content = f.read()
|
|
|
|
if 'ocr_engine = paddleocr' in content:
|
|
print("LightRAG config.ini already set to paddleocr")
|
|
else:
|
|
print("LightRAG config.ini not set to paddleocr")
|
|
# Update the config file
|
|
content = content.replace('ocr_engine = aquaforest', 'ocr_engine = paddleocr')
|
|
with open(config_path, 'w') as f:
|
|
f.write(content)
|
|
print("Updated LightRAG config.ini to use paddleocr")
|
|
|
|
# 4. Create a test script to verify OCR functionality
|
|
test_script = """
|
|
import paddleocr
|
|
import cv2
|
|
import os
|
|
|
|
def test_paddleocr():
|
|
print("Testing PaddleOCR installation...")
|
|
try:
|
|
# Initialize PaddleOCR
|
|
ocr = paddleocr.PaddleOCR(use_angle_cls=True, lang='en')
|
|
|
|
# Test with a simple image if available
|
|
test_image = "test_ocr_image.png"
|
|
if os.path.exists(test_image):
|
|
result = ocr.ocr(test_image, cls=True)
|
|
print(f"PaddleOCR test successful: {len(result)} results")
|
|
else:
|
|
print("PaddleOCR initialized successfully (no test image available)")
|
|
|
|
except Exception as e:
|
|
print(f"PaddleOCR test failed: {e}")
|
|
|
|
if __name__ == "__main__":
|
|
test_paddleocr()
|
|
"""
|
|
|
|
with open("test_paddleocr_installation.py", "w", encoding='utf-8') as f:
|
|
f.write(test_script)
|
|
|
|
print("Created PaddleOCR test script")
|
|
|
|
# 5. Clear existing failed documents
|
|
print("Clearing existing failed documents...")
|
|
|
|
clear_script = """
|
|
import requests
|
|
import json
|
|
|
|
def clear_failed_documents():
|
|
base_url = "http://localhost:3015"
|
|
api_key = "lightrag-test-key"
|
|
|
|
headers = {
|
|
"X-API-Key": api_key
|
|
}
|
|
|
|
try:
|
|
# Get current documents
|
|
docs_response = requests.get(f"{base_url}/documents", headers=headers)
|
|
if docs_response.status_code == 200:
|
|
documents = docs_response.json()
|
|
failed_docs = documents.get('statuses', {}).get('failed', [])
|
|
|
|
if failed_docs:
|
|
print(f"Found {len(failed_docs)} failed documents to clear")
|
|
|
|
# Clear all documents
|
|
clear_response = requests.delete(f"{base_url}/documents", headers=headers)
|
|
if clear_response.status_code == 200:
|
|
print("Successfully cleared all documents")
|
|
else:
|
|
print(f"Failed to clear documents: {clear_response.status_code}")
|
|
else:
|
|
print("No failed documents found")
|
|
else:
|
|
print(f"Could not fetch documents: {docs_response.status_code}")
|
|
|
|
except Exception as e:
|
|
print(f"Error clearing documents: {e}")
|
|
|
|
if __name__ == "__main__":
|
|
clear_failed_documents()
|
|
"""
|
|
|
|
with open("clear_failed_docs.py", "w", encoding='utf-8') as f:
|
|
f.write(clear_script)
|
|
|
|
print("Created document clearing script")
|
|
|
|
print("\\nNext steps:")
|
|
print("1. Run: python test_paddleocr_installation.py (to verify PaddleOCR)")
|
|
print("2. Run: python clear_failed_docs.py (to clear failed documents)")
|
|
print("3. Restart the LightRAG server")
|
|
print("4. Run: python test_ocr_upload_final.py (to test OCR PDF upload)")
|
|
|
|
if __name__ == "__main__":
|
|
force_paddleocr_config() |