import urllib.request
import json
import os

out_dir = "/home/admin/GBS_Research/学术论文"
os.makedirs(out_dir, exist_ok=True)

queries = [
    "shared services value creation",
    "shared service center multinational",
    "global business services GBS",
    "financial shared services China",
    "shared services digital transformation",
    "shared services compliance cross-border",
    "shared services cost efficiency",
    "business process reengineering shared services",
]

all_papers = {}

for q in queries:
    url = f"https://api.openalex.org/works?search={urllib.parse.quote(q)}&per_page=15&sort=cited_by_count:desc"
    try:
        req = urllib.request.Request(url, headers={"User-Agent": "Mozilla/5.0", "Accept": "application/json"})
        with urllib.request.urlopen(req, timeout=30) as resp:
            data = json.loads(resp.read().decode())
            for w in data.get("results", []):
                doi = w.get("doi", "")
                if doi in all_papers:
                    continue
                abstract_inv = w.get("abstract_inverted_index", {})
                if abstract_inv:
                    pairs = []
                    for word, positions in abstract_inv.items():
                        for pos in positions:
                            pairs.append((pos, word))
                    pairs.sort()
                    abstract = " ".join(pw for _, pw in pairs)
                else:
                    abstract = ""
                all_papers[doi] = {
                    "title": w.get("title", ""),
                    "doi": doi,
                    "year": w.get("publication_year", ""),
                    "cited_by_count": w.get("cited_by_count", 0),
                    "authors": [a.get("author",{}).get("display_name","") for a in w.get("authorships",[])],
                    "source": w.get("primary_location",{}).get("source",{}).get("display_name","") if w.get("primary_location") else "",
                    "abstract": abstract,
                    "query": q,
                }
        print(f"✓ {q}: {data.get('meta',{}).get('count',0)} total, got {len(data.get('results',[]))}")
    except Exception as e:
        print(f"✗ {q}: {e}")

# Sort by citations
papers = sorted(all_papers.values(), key=lambda x: x.get("cited_by_count",0), reverse=True)

# Save JSON
with open(f"{out_dir}/openalex_papers.json", "w", encoding="utf-8") as f:
    json.dump(papers, f, ensure_ascii=False, indent=2)

# Save markdown summary
lines = ["# GBS/共享服务中心 学术论文汇总 (OpenAlex)\n"]
lines.append(f"共找到 {len(papers)} 篇不重复论文\n")
lines.append("| # | 标题 | 年份 | 被引 | DOI |")
lines.append("|---|------|------|------|-----|")
for i, p in enumerate(papers[:80], 1):
    title = p["title"][:80] if p["title"] else "N/A"
    doi_link = f"https://doi.org/{p['doi'].replace('https://doi.org/','')}" if p["doi"] else ""
    lines.append(f"| {i} | {title} | {p['year']} | {p['cited_by_count']} | {doi_link} |")

lines.append("\n---\n## 详细摘要\n")
for i, p in enumerate(papers[:80], 1):
    lines.append(f"### {i}. {p['title']}")
    lines.append(f"- **年份**: {p['year']} | **被引**: {p['cited_by_count']}")
    lines.append(f"- **期刊**: {p['source']}")
    lines.append(f"- **作者**: {', '.join(p['authors'][:5])}")
    lines.append(f"- **DOI**: {p['doi']}")
    lines.append(f"- **摘要**: {p['abstract'][:500]}")
    lines.append("")

with open(f"{out_dir}/openalex_summary.md", "w", encoding="utf-8") as f:
    f.write("\n".join(lines))

print(f"\n总论文数: {len(papers)}, 有摘要: {sum(1 for p in papers if p['abstract'])}")
print(f"Top cited: {papers[0]['title'][:60]} ({papers[0]['cited_by_count']} cites)" if papers else "")