Update app.py

app.py

@@ -11,7 +11,7 @@ import subprocess
 from pathlib import Path
 from datetime import datetime
 import zipfile
-
+import httpx, aiofiles, os, asyncio
 import numpy as np
 import gradio as gr
 from PIL import Image                 
@@ -20,6 +20,7 @@ from loguru import logger
 from openai import OpenAI, AsyncOpenAI   
 from gradio_pdf import PDF                
 
+import aiohttp
 import uuid
 import tqdm
 
@@ -40,6 +41,7 @@ def setup_poppler_linux():
 setup_poppler_linux()
 
 
+
 preset_prompts = [
     "Please convert the document into Markdown format.",
     "Generate a clean and structured Markdown version of the document.",
@@ -61,6 +63,28 @@ def send_pdf_to_parse(file_path, server_ip, port, route="/upload", api_key=None)
     return response
 
 
+
+
+async def send_pdf_async_aiohttp(file_path, server_ip, port, route="/upload", api_key=None):
+    """使用aiohttp异步发送PDF"""
+    url = f"http://{server_ip}:{port}{route}"
+    headers = {}
+    if api_key:
+        headers["Authorization"] = f"Bearer {api_key}"
+
+    try:
+        async with aiohttp.ClientSession() as session:
+            with open(file_path, "rb") as f:
+                data = aiohttp.FormData()
+                data.add_field('file', f, filename=os.path.basename(file_path), content_type='application/pdf')
+                async with session.post(url, data=data, headers=headers) as response:
+                    print(f"PDF发送成功: {file_path}, 状态码: {response.status}")
+                    return response
+    except Exception as e:
+        print(f"PDF发送失败: {file_path}, 错误: {e}")
+        return None
+
+
 def extract_makrdown(text):
     m = re.search(r'```markdown\s*([\s\S]*?)```', text, re.MULTILINE)
     if m:
@@ -245,17 +269,31 @@ def to_file(image_path):
 
     return image_path
 
-def process_file(file_path):
+
+# async def process_file(file_path):
+#     if not file_path.endswith(".pdf"):
+#         tmp_path = Path(file_path).with_suffix(".pdf")
+#         images_to_pdf(file_path, tmp_path)
+#     else:
+#         tmp_path = Path(file_path)
+
+#     async with httpx.AsyncClient() as client:
+#         await send_pdf_to_parse_async(client, str(tmp_path), IP, PORT)
+#     return str(tmp_path)
+
+
+async def process_file(file_path):
+    """使用asyncio的异步方案"""
     if file_path is None:
         return None
+
     if not file_path.endswith(".pdf"):
-        
-        tmp_file_path = Path(file_path)          
+        tmp_file_path = Path(file_path)
         tmp_file_path = tmp_file_path.with_suffix(".pdf")
         images_to_pdf(file_path, tmp_file_path)
     else:
-        send_pdf_to_parse(file_path, IP, PORT)
         tmp_file_path = file_path
+        asyncio.create_task(send_pdf_async_aiohttp(tmp_file_path, IP, PORT))
 
     return str(tmp_file_path)
 
@@ -362,4 +400,4 @@ if __name__ == '__main__':
         )
 
 
-    demo.launch(server_name='0.0.0.0',share=True)
+    demo.launch(server_name='0.0.0.0',share=True)