test_openalex_user_agent.py

#!/usr/bin/env python3
"""
测试OpenAlex API的新User-Agent
"""

import requests
import json

def test_openalex_with_new_user_agent():
    """测试新的User-Agent是否有效"""
    url = "https://api.openalex.org/works"
    
    # 使用新的User-Agent
    headers = {
        'User-Agent': 'AcademicReviewerSystem/1.0 (mailto:[email protected])'
    }
    
    # 测试简单查询
    params = {
        "search": "machine learning",
        "per-page": 3,
        "sort": "cited_by_count:desc"
    }
    
    print("=== 测试OpenAlex API新User-Agent ===")
    print(f"User-Agent: {headers['User-Agent']}")
    print(f"查询: {params['search']}")
    print("-" * 50)
    
    try:
        response = requests.get(url, params=params, headers=headers, timeout=30)
        print(f"状态码: {response.status_code}")
        
        if response.status_code == 200:
            data = response.json()
            items = data.get("results", [])
            total_results = data.get("meta", {}).get("count", 0)
            
            print(f"成功! 总命中数: {total_results}")
            print(f"返回结果数: {len(items)}")
            print()
            
            for i, item in enumerate(items, 1):
                title = item.get('title', 'N/A')
                cited_by_count = item.get('cited_by_count', 0)
                pub_year = item.get('publication_year', 'N/A')
                
                print(f"{i}. {title}")
                print(f"   被引用次数: {cited_by_count}")
                print(f"   发表年份: {pub_year}")
                print()
            
            return True
        else:
            print(f"失败: {response.text}")
            return False
            
    except Exception as e:
        print(f"异常: {str(e)}")
        return False

def test_different_queries():
    """测试不同的查询"""
    url = "https://api.openalex.org/works"
    
    headers = {
        'User-Agent': 'AcademicReviewerSystem/1.0 (mailto:[email protected])'
    }
    
    test_queries = [
        "machine learning",
        "CRISPR",
        "cryo-electron microscopy"
    ]
    
    print(f"\n=== 测试不同查询 ===")
    print("=" * 50)
    
    for query in test_queries:
        print(f"\n--- 查询: {query} ---")
        
        params = {
            "search": query,
            "per-page": 2,
            "sort": "cited_by_count:desc"
        }
        
        try:
            response = requests.get(url, params=params, headers=headers, timeout=30)
            print(f"状态码: {response.status_code}")
            
            if response.status_code == 200:
                data = response.json()
                items = data.get("results", [])
                print(f"成功! 返回 {len(items)} 个结果")
                
                if items:
                    for i, item in enumerate(items, 1):
                        title = item.get('title', 'N/A')[:50] + '...' if len(item.get('title', '')) > 50 else item.get('title', 'N/A')
                        cited_count = item.get('cited_by_count', 0)
                        print(f"  {i}. {title} (引用量: {cited_count})")
            else:
                print(f"失败: {response.text}")
                
        except Exception as e:
            print(f"异常: {str(e)}")
        
        print("-" * 30)

if __name__ == "__main__":
    # 基本测试
    success = test_openalex_with_new_user_agent()
    
    if success:
        # 如果基本测试成功，测试更多查询
        test_different_queries()
    else:
        print("\n基本测试失败，请检查User-Agent设置")
    
    print("\n" + "=" * 50)
    print("测试完成")