test_openalex_simple.py

#!/usr/bin/env python3
"""
OpenAlex API 简单测试 - 修复版本
"""

import requests
import json

def test_openalex_basic():
    """测试OpenAlex API基本功能"""
    url = "https://api.openalex.org/works"
    
    # 使用正确的User-Agent格式
    headers = {
        'User-Agent': 'AcademicReviewerSystem/1.0 (mailto:[email protected])'
    }
    
    # 最简单的查询
    params = {
        "search": "machine learning",
        "per-page": 3
    }
    
    print("OpenAlex API 基本测试")
    print("=" * 50)
    print(f"URL: {url}")
    print(f"参数: {params}")
    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(f"   OpenAlex ID: {item.get('id', 'N/A')}")
                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_openalex_with_sort():
    """测试OpenAlex API排序功能"""
    url = "https://api.openalex.org/works"
    
    headers = {
        'User-Agent': 'AcademicReviewerSystem/1.0 (mailto:[email protected])'
    }
    
    # 测试按引用量排序
    params = {
        "search": "machine learning",
        "per-page": 5,
        "sort": "cited_by_count:desc"
    }
    
    print(f"\n=== 测试按引用量排序 ===")
    print(f"参数: {params}")
    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", [])
            
            print(f"成功! 返回 {len(items)} 个结果")
            
            for i, item in enumerate(items, 1):
                title = item.get('title', 'N/A')
                if len(title) > 60:
                    title = title[:60] + "..."
                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()
            
            # 显示引用量统计
            citations = [item.get('cited_by_count', 0) for item in items]
            print(f"引用量统计: {citations}")
            if citations:
                print(f"平均引用量: {sum(citations) / len(citations):.2f}")
                print(f"最大引用量: {max(citations)}")
            
            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_openalex_work_details():
    """测试OpenAlex工作详情查找"""
    # 使用一个已知的OpenAlex ID
    work_id = "W2755950973"  # 一个示例ID
    
    url = f"https://api.openalex.org/works/{work_id}"
    
    headers = {
        'User-Agent': 'AcademicReviewerSystem/1.0 (mailto:[email protected])'
    }
    
    print(f"\n=== 测试工作详情查找 ===")
    print(f"Work ID: {work_id}")
    print("-" * 50)
    
    try:
        response = requests.get(url, headers=headers, timeout=30)
        print(f"状态码: {response.status_code}")
        
        if response.status_code == 200:
            data = response.json()
            
            title = data.get('title', 'N/A')
            cited_by_count = data.get('cited_by_count', 0)
            citation_count = data.get('citation_count', 0)
            
            print(f"标题: {title}")
            print(f"被引用次数: {cited_by_count}")
            print(f"引用次数: {citation_count}")
            print(f"OpenAlex ID: {data.get('id', 'N/A')}")
            
            return True
        else:
            print(f"失败: {response.text}")
            return False
            
    except Exception as e:
        print(f"异常: {str(e)}")
        return False

if __name__ == "__main__":
    print("OpenAlex API 修复测试")
    print("=" * 60)
    
    # 基本测试
    test_openalex_basic()
    
    # 排序测试
    test_openalex_with_sort()
    
    # 工作详情测试
    test_openalex_work_details()
    
    print("\n" + "=" * 60)
    print("测试完成")