YogeLiu/doc_processer
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: update port

Browse Source
This commit is contained in:
liuyuanchuang
2026-02-05 22:07:39 +08:00
parent a3ca04856f
commit 36172ba4ff
3 changed files with 1 additions and 116 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
app/services/ocr_service.py
View File

@@ -489,7 +489,7 @@ class MineruOCRService(OCRServiceBase):
api_url: str = "http://127.0.0.1:8000/file_parse",
image_processor: Optional[ImageProcessor] = None,
converter: Optional[Converter] = None,
paddleocr_vl_url: str = "http://localhost:8001/v1",
paddleocr_vl_url: str = "http://localhost:8000/v1",
):
"""Initialize Local API service.

53
test_paddleocr_vl_integration.py
View File

@@ -1,53 +0,0 @@
"""Test script for PaddleOCR-VL integration in MineruOCRService."""
import cv2
import numpy as np
from app.services.ocr_service import MineruOCRService
from app.services.converter import Converter
from app.services.image_processor import ImageProcessor
def test_paddleocr_vl_integration():
"""Test that PaddleOCR-VL is called when image references are found."""
# Create a simple test image (white background with black text)
test_image = np.ones((100, 300, 3), dtype=np.uint8) * 255
cv2.putText(test_image, "x^2 + y^2 = 1", (50, 50),
cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 0, 0), 2)
# Initialize service
service = MineruOCRService(
api_url="http://127.0.0.1:8000/file_parse",
converter=Converter(),
image_processor=ImageProcessor(),
paddleocr_vl_url="http://localhost:8000/v1" # Your PaddleOCR-VL server
)
# Simulate markdown with image reference (this is what Mineru returns)
test_markdown = "![](images/af7f211f671f16f57d346e8e17611e68e0f4671bd1ae52ed59013c10eecef589.jpg)"
print("Testing formula extraction...")
result = service._extract_and_recognize_formulas(test_markdown, test_image)
print(f"\nOriginal markdown: {test_markdown}")
print(f"Processed markdown: {result}")
# Check if the image reference was replaced
if "![](images/" in result:
print("\n❌ FAILED: Image reference was not replaced")
else:
print("\n✅ SUCCESS: Image reference was replaced with formula")
if __name__ == "__main__":
print("=" * 60)
print("PaddleOCR-VL Integration Test")
print("=" * 60)
print("\nMake sure your PaddleOCR-VL server is running at:")
print("http://localhost:8000/v1")
print("\n" + "=" * 60 + "\n")
try:
test_paddleocr_vl_integration()
except Exception as e:
print(f"\n❌ Test failed with error: {e}")
import traceback
traceback.print_exc()

62
test_vllm_connection.py
View File

@@ -1,62 +0,0 @@
"""Quick test to verify PaddleOCR-VL connection."""
from openai import OpenAI
import base64
import cv2
import numpy as np
# Create test image
test_image = np.ones((100, 300, 3), dtype=np.uint8) * 255
cv2.putText(test_image, "x^2 = 4", (50, 50), cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 0, 0), 2)
# Encode to base64
success, encoded_image = cv2.imencode(".png", test_image)
if not success:
print("Failed to encode image")
exit(1)
image_base64 = base64.b64encode(encoded_image.tobytes()).decode("utf-8")
image_url = f"data:image/png;base64,{image_base64}"
# Test connection
client = OpenAI(
api_key="EMPTY",
base_url="http://100.115.184.74:8001/v1",
timeout=3600
)
print("Testing PaddleOCR-VL connection...")
print(f"Server: http://100.115.184.74:8001/v1")
print(f"Model: PaddleOCR-VL-0.9B")
print("-" * 60)
try:
messages = [
{
"role": "user",
"content": [
{
"type": "image_url",
"image_url": {"url": image_url}
},
{
"type": "text",
"text": "Formula Recognition:"
}
]
}
]
response = client.chat.completions.create(
model="PaddleOCR-VL-0.9B",
messages=messages,
temperature=0.0,
)
print("✅ SUCCESS!")
print(f"Response: {response.choices[0].message.content}")
except Exception as e:
print(f"❌ FAILED: {e}")
import traceback
traceback.print_exc()