test_scholar.py•8.72 kB
import asyncio
import os
import httpx
from mcp_scholar.scholar import (
search_scholar,
parse_profile,
extract_profile_id_from_url,
convert_inverted_index_to_text,
convert_google_scholar_to_openalex, # 导入新函数
get_paper_references, # 导入引用函数
)
async def test_search_scholar():
print("测试 search_scholar 函数...")
# 搜索关于人工智能的论文，获取前3篇
results = await search_scholar("artificial intelligence", 3)
print(f"找到 {len(results)} 篇论文:")
for i, paper in enumerate(results, 1):
print(f"\n论文 {i}:")
print(f"标题: {paper['title']}")
print(f"作者: {paper['authors']}")
print(f"摘要: {paper['abstract']}")
print(f"引用次数: {paper['citations']}")
async def test_parse_real_profile():
print("\n测试解析OpenAlex学者档案...")
# 使用一个OpenAlex作者ID作为测试
profile_id = "A2208157607" # OpenAlex上的一个作者ID
try:
# 解析个人主页
papers = await parse_profile(profile_id, top_n=5)
print(f"解析出 {len(papers)} 篇最高引用论文:")
for i, paper in enumerate(papers, 1):
print(f"\n论文 {i}:")
print(f"标题: {paper['title']}")
print(f"引用: {paper['citations']}")
print(
f"摘要: {paper['abstract'][:150]}..."
if len(paper["abstract"]) > 150
else paper["abstract"]
)
print(f"年份: {paper.get('year', 'N/A')}")
if "authors" in paper:
print(f"作者: {paper['authors']}")
if "venue" in paper:
print(f"发表于: {paper['venue']}")
except Exception as e:
print(f"解析时出错: {e}")
async def test_inverted_index_conversion():
print("\n测试倒排索引转换函数...")
# 示例倒排索引
inverted_index = {"这是": [0], "一个": [1], "测试": [2], "示例": [3]}
text = convert_inverted_index_to_text(inverted_index)
print(f"转换结果: {text}")
# 空索引测试
empty_text = convert_inverted_index_to_text({})
print(f"空索引转换结果: '{empty_text}'")
async def test_google_scholar_profile():
print("\n测试从谷歌学术主页解析学者信息...")
# 使用一个知名学者的谷歌学术页面URL作为示例
# 这里使用Andrew Ng的谷歌学术主页作为例子
google_scholar_url = "https://scholar.google.com/citations?user=mG4imMEAAAAJ"
# 从URL提取ID
profile_id = extract_profile_id_from_url(google_scholar_url)
if profile_id and profile_id.startswith("google:"):
try:
print(f"从URL中提取的谷歌学术ID: {profile_id}")
# 解析个人主页（内部会自动转换为OpenAlex ID）
papers = await parse_profile(profile_id, top_n=3)
print(f"解析出 {len(papers)} 篇最高引用论文:")
for i, paper in enumerate(papers, 1):
print(f"\n论文 {i}:")
print(f"标题: {paper['title']}")
print(f"引用: {paper['citations']}")
print(
f"摘要: {paper['abstract'][:]}..."
if len(paper.get("abstract", "")) > 150
else paper.get("abstract", "无摘要")
)
print(f"年份: {paper.get('year', 'N/A')}")
if "authors" in paper:
print(f"作者: {paper['authors']}")
if "venue" in paper:
print(f"发表于: {paper['venue']}")
except Exception as e:
print(f"解析谷歌学术主页时出错: {e}")
else:
print(f"无法从URL提取有效的谷歌学术ID: {google_scholar_url}")
async def test_papers_by_year():
print("\n测试按年份筛选论文...")
# 使用OpenAlex作者ID
profile_id = "A2208157607"
try:
# 只获取2018年之后的论文
papers = await parse_profile(profile_id, top_n=5, min_year=2018)
print(f"找到 {len(papers)} 篇2018年之后的论文:")
for i, paper in enumerate(papers, 1):
print(f"\n论文 {i}:")
print(f"标题: {paper['title']}")
print(f"引用: {paper['citations']}")
print(f"年份: {paper.get('year', 'N/A')}")
# 测试年份范围
papers = await parse_profile(profile_id, top_n=5, min_year=2015, max_year=2020)
print(f"\n找到 {len(papers)} 篇2015-2020年间的论文:")
for i, paper in enumerate(papers, 1):
print(f"\n论文 {i}:")
print(f"标题: {paper['title']}")
print(f"引用: {paper['citations']}")
print(f"年份: {paper.get('year', 'N/A')}")
except Exception as e:
print(f"按年份筛选论文时出错: {e}")
async def test_papers_without_citation_sort():
print("\n测试不按引用次数排序的论文...")
# 使用OpenAlex作者ID
profile_id = "A2208157607"
try:
# 按发表年份排序（降序）- 修正为新的排序参数
papers = await parse_profile(profile_id, top_n=5, sort_by="date")
print(f"按发表日期排序的 {len(papers)} 篇论文:")
for i, paper in enumerate(papers, 1):
print(f"\n论文 {i}:")
print(f"标题: {paper['title']}")
print(f"引用: {paper['citations']}")
print(f"年份: {paper.get('year', 'N/A')}")
# 按标题字母顺序排序
papers = await parse_profile(profile_id, top_n=5, sort_by="title")
print(f"\n按标题排序的 {len(papers)} 篇论文:")
for i, paper in enumerate(papers, 1):
print(f"\n论文 {i}:")
print(f"标题: {paper['title']}")
print(f"引用: {paper['citations']}")
print(f"年份: {paper.get('year', 'N/A')}")
except Exception as e:
print(f"使用非引用排序时出错: {e}")
async def test_all_sorting_methods():
"""测试所有可用的排序方式"""
print("\n===== 测试所有排序方式 =====")
# 1. 测试search_scholar的排序
print("\n1. 测试search_scholar的排序方式:")
query = "artificial intelligence"
for sort_by in ["relevance", "citations", "date", "title"]:
print(f"\n--- 使用 {sort_by} 排序搜索结果 ---")
results = await search_scholar(query, 3, sort_by=sort_by)
print(f"找到 {len(results)} 篇论文:")
for i, paper in enumerate(results, 1):
print(f"论文 {i}: {paper['title']}")
if sort_by == "citations":
print(f"引用量: {paper['citations']}")
elif sort_by == "date":
print(f"发表年份: {paper['year']}")
# 2. 测试get_paper_references的排序
print("\n2. 测试get_paper_references的排序方式:")
# 使用一个知名论文ID (GPT-3论文)
paper_id = "W3098397289"
for sort_by in ["relevance", "citations", "date", "title"]:
print(f"\n--- 使用 {sort_by} 排序引用论文 ---")
refs = await get_paper_references(paper_id, 3, sort_by=sort_by)
print(f"找到 {len(refs)} 篇引用论文:")
for i, ref in enumerate(refs, 1):
print(f"论文 {i}: {ref['title']}")
if sort_by == "citations":
print(f"引用量: {ref['citations']}")
elif sort_by == "date":
print(f"发表年份: {ref['year']}")
# 3. 测试parse_profile的排序
print("\n3. 测试parse_profile的排序方式:")
profile_id = "A2208157607" # 使用一个OpenAlex作者ID
for sort_by in ["relevance", "citations", "date", "title"]:
print(f"\n--- 使用 {sort_by} 排序作者论文 ---")
papers = await parse_profile(profile_id, 3, sort_by=sort_by)
print(f"找到 {len(papers)} 篇作者论文:")
for i, paper in enumerate(papers, 1):
print(f"论文 {i}: {paper['title']}")
if sort_by == "citations":
print(f"引用量: {paper['citations']}")
elif sort_by == "date":
print(f"发表年份: {paper['year']}")
async def main():
# await test_search_scholar()
# await test_inverted_index_conversion()
# await test_parse_real_profile()
# await test_google_scholar_profile()
# await test_papers_by_year()
# await test_papers_without_citation_sort()
# 新增的排序测试
await test_all_sorting_methods()
if __name__ == "__main__":
print("开始测试 mcp_scholar 模块...")
asyncio.run(main())
print("\n测试完成!")