#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
图文匹配检测服务
完整的Web服务器实现
"""
import http.server
import socketserver
import os
import json
import urllib.parse
from pathlib import Path
import time
import requests
import logging
# 配置日志
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)
class ImageTextMatchingHandler(http.server.SimpleHTTPRequestHandler):
def __init__(self, *args, **kwargs):
# 设置静态文件目录
super().__init__(*args, directory="web", **kwargs)
def end_headers(self):
# 添加CORS头
self.send_header('Access-Control-Allow-Origin', '*')
self.send_header('Access-Control-Allow-Methods', 'GET, POST, OPTIONS')
self.send_header('Access-Control-Allow-Headers', 'Content-Type')
super().end_headers()
def do_OPTIONS(self):
# 处理预检请求
self.send_response(200)
self.end_headers()
def do_POST(self):
"""处理POST请求"""
if self.path == '/upload':
self.handle_upload()
elif self.path == '/api/image_text_matching':
self.handle_matching_api()
else:
self.send_error(404)
def do_GET(self):
"""处理GET请求"""
if self.path == '/api/health':
# 健康检查端点
self.send_response(200)
self.send_header('Content-Type', 'application/json')
self.end_headers()
response = {"status": "ok", "service": "image_text_matching"}
self.wfile.write(json.dumps(response).encode())
else:
# 静态文件处理
super().do_GET()
def handle_upload(self):
"""处理文件上传"""
try:
content_length = int(self.headers['Content-Length'])
post_data = self.rfile.read(content_length)
# 解析multipart/form-data
content_type = self.headers['Content-Type']
if 'multipart/form-data' in content_type:
boundary = content_type.split('boundary=')[1].encode()
parts = post_data.split(b'--' + boundary)
for part in parts:
if b'Content-Disposition: form-data' in part and b'filename=' in part:
# 提取文件名
lines = part.split(b'\r\n')
filename = "unknown.jpg"
for line in lines:
if b'Content-Disposition' in line:
try:
filename_start = line.find(b'filename="') + 10
filename_end = line.find(b'"', filename_start)
filename = line[filename_start:filename_end].decode()
except:
pass
break
# 提取文件内容
content_start = part.find(b'\r\n\r\n') + 4
content_end = part.rfind(b'\r\n')
if content_end == -1:
content_end = len(part)
file_content = part[content_start:content_end]
if len(file_content) > 0:
# 保存文件
upload_dir = Path("uploads")
upload_dir.mkdir(exist_ok=True)
timestamp = str(int(time.time() * 1000))
file_extension = filename.split('.')[-1] if '.' in filename else 'jpg'
save_filename = f"upload_{timestamp}.{file_extension}"
file_path = upload_dir / save_filename
with open(file_path, 'wb') as f:
f.write(file_content)
# 返回文件URL
file_url = f"http://localhost:8080/uploads/{save_filename}"
response = {"success": True, "url": file_url}
self.send_response(200)
self.send_header('Content-Type', 'application/json')
self.end_headers()
self.wfile.write(json.dumps(response).encode())
return
self.send_error(400, "Bad Request - No valid file found")
except Exception as e:
logger.error(f"Upload error: {e}")
self.send_error(500, f"Internal Server Error: {str(e)}")
def handle_matching_api(self):
"""处理图文匹配API请求"""
try:
content_length = int(self.headers['Content-Length'])
post_data = self.rfile.read(content_length)
# 解析JSON数据
data = json.loads(post_data.decode('utf-8'))
# 调用图文匹配检测服务
result = self.call_matching_service(data)
# 返回结果
self.send_response(200)
self.send_header('Content-Type', 'application/json')
self.end_headers()
self.wfile.write(json.dumps(result, ensure_ascii=False).encode('utf-8'))
except json.JSONDecodeError as e:
logger.error(f"JSON decode error: {e}")
self.send_error(400, "Invalid JSON data")
except Exception as e:
logger.error(f"Matching API error: {e}")
error_response = {
"code": 500,
"message": f"服务器内部错误: {str(e)}",
"result": []
}
self.send_response(500)
self.send_header('Content-Type', 'application/json')
self.end_headers()
self.wfile.write(json.dumps(error_response, ensure_ascii=False).encode('utf-8'))
def call_matching_service(self, data):
"""调用外部图文匹配检测服务"""
api_url = "http://localhost:29500/v1/image_text_matching"
try:
# 准备请求数据
request_data = {
"illustration_url": data.get("illustration_url", ""),
"caption_text": data.get("caption_text", ""),
"context_info": data.get("context_info", "")
}
logger.info(f"Calling matching service with data: {request_data}")
# 调用外部服务
response = requests.post(api_url, json=request_data, timeout=30)
if response.status_code == 200:
result = response.json()
logger.info(f"Matching service response: {result}")
return result
else:
logger.error(f"Matching service error: {response.status_code} - {response.text}")
return {
"code": response.status_code,
"message": f"外部服务错误: {response.text}",
"result": []
}
except requests.exceptions.RequestException as e:
logger.error(f"Request to matching service failed: {e}")
return {
"code": 503,
"message": f"无法连接到图文匹配检测服务 (localhost:29500): {str(e)}",
"result": []
}
except Exception as e:
logger.error(f"Unexpected error in matching service call: {e}")
return {
"code": 500,
"message": f"服务调用失败: {str(e)}",
"result": []
}
def setup_directories():
"""创建必要的目录结构"""
directories = ["web", "uploads"]
for directory in directories:
os.makedirs(directory, exist_ok=True)
print(f"✓ 创建目录: {directory}")
def create_index_html():
"""创建前端HTML文件"""
html_content = '''
图文匹配检测服务
'''
with open("web/index.html", "w", encoding="utf-8") as f:
f.write(html_content)
print("✓ 创建前端文件: web/index.html")
def start_server(port=8080):
"""启动Web服务器"""
print("正在初始化图文匹配检测服务...")
# 设置目录结构
setup_directories()
# 创建前端文件
create_index_html()
print("\n" + "="*50)
print("🚀 图文匹配检测服务启动中...")
print("="*50)
# 启动服务器
with socketserver.TCPServer(("", port), ImageTextMatchingHandler) as httpd:
server_address = ("", port)
print(f"✓ 服务器绑定地址: http://localhost:{port}")
print(f"✓ 静态文件目录: web/")
print(f"✓ 上传文件目录: uploads/")
print(f"✓ 外部API地址: http://localhost:29500")
try:
print(f"\n🌐 服务器启动成功！")
print(f"📱 请在浏览器中访问: http://localhost:{port}")
print(f"🔧 管理界面: http://localhost:{port}/api/health")
print("\n" + "="*50)
print("按 Ctrl+C 停止服务器")
print("="*50)
httpd.serve_forever()
except KeyboardInterrupt:
print("\n\n⏹️ 正在停止服务器...")
httpd.shutdown()
print("✓ 服务器已停止")
if __name__ == "__main__":
import sys
# 解析命令行参数
port = 8080
if len(sys.argv) > 1:
try:
port = int(sys.argv[1])
except ValueError:
print("❌ 端口号必须是数字")
sys.exit(1)
# 启动服务器
start_server(port)