#!/usr/bin/env python3
import urllib.request
import urllib.parse
import json
import os

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

def extract_abstract(w):
    abstract_inv = w.get("abstract_inverted_index", {})
    if not abstract_inv:
        return ""
    pairs = []
    for word, positions in abstract_inv.items():
        for pos in positions:
            pairs.append((pos, word))
    pairs.sort()
    return " ".join(pw for _, pw in pairs)

def search_openalex(query, per_page=15):
    url = f"https://api.openalex.org/works?search={urllib.parse.quote(query)}&per_page={per_page}&sort=cited_by_count:desc"
    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())
    results = []
    for w in data.get("results", []):
        results.append({
            "title": w.get("title", ""),
            "doi": w.get("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": extract_abstract(w),
            "language": w.get("language",""),
            "query": query,
        })
    return results, data.get("meta",{}).get("count", 0)

# English queries
en_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",
]

# Chinese queries
cn_queries = [
    "共享服务中心 价值创造",
    "全球商业服务 跨国企业",
    "财务共享 跨境 合规",
    "GBS 能力中心 中国",
    "共享服务中心 数字化转型",
    "共享服务中心 成本效率",
]

all_papers = {}

print("=" * 60)
print("搜索英文文献 (OpenAlex)")
print("=" * 60)

for q in en_queries:
    try:
        results, total = search_openalex(q)
        added = 0
        for p in results:
            key = p["doi"] or p["title"]
            if key and key not in all_papers:
                all_papers[key] = p
                added += 1
        print(f"✓ EN '{q}': {total} total, {added} new")
    except Exception as e:
        print(f"✗ EN '{q}': {e}")

print(f"\n英文文献去重后: {len(all_papers)} 篇")

print("\n" + "=" * 60)
print("搜索中文文献 (OpenAlex)")
print("=" * 60)

for q in cn_queries:
    try:
        results, total = search_openalex(q)
        added = 0
        for p in results:
            key = p["doi"] or p["title"]
            if key and key not in all_papers:
                all_papers[key] = p
                added += 1
        print(f"✓ CN '{q}': {total} total, {added} new")
    except Exception as e:
        print(f"✗ CN '{q}': {e}")

print(f"\n总去重论文数: {len(all_papers)}")

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

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

# Save English summary
en_papers = [p for p in papers if p.get("language","") == "en" or not p.get("language")]
cn_paper_list = [p for p in papers if p.get("language","") == "zh" or p.get("language","") == "zh-Hant"]

# Full markdown
lines = ["# GBS/共享服务中心 学术论文汇总\n"]
lines.append(f"**搜索日期**: 2026-06-27\n")
lines.append(f"**数据来源**: OpenAlex API\n")
lines.append(f"**总论文数**: {len(papers)} (英文: {len(en_papers)}, 中文: {len(cn_paper_list)})\n")

lines.append("\n## 按被引次数排序的论文列表\n")
lines.append("| # | 标题 | 年份 | 被引 | 语言 | 期刊 |")
lines.append("|---|------|------|------|------|------|")
for i, p in enumerate(papers, 1):
    title = p["title"][:90] if p["title"] else "N/A"
    lines.append(f"| {i} | {title} | {p['year']} | {p['cited_by_count']} | {p['language']} | {p['source'][:40]} |")

lines.append("\n---\n## 详细论文信息与摘要\n")
for i, p in enumerate(papers, 1):
    lines.append(f"### {i}. {p['title']}")
    lines.append(f"- **年份**: {p['year']} | **被引次数**: {p['cited_by_count']} | **语言**: {p['language']}")
    lines.append(f"- **期刊/来源**: {p['source']}")
    lines.append(f"- **作者**: {', '.join(p['authors'][:6])}")
    if p['doi']:
        lines.append(f"- **DOI**: [{p['doi']}]({p['doi']})")
    lines.append(f"- **搜索关键词**: {p['query']}")
    if p['abstract']:
        lines.append(f"- **摘要**: {p['abstract'][:600]}")
    else:
        lines.append(f"- **摘要**: 无")
    lines.append("")

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

# Top cited dedicated file
lines2 = ["# 高被引论文TOP30\n\n"]
for i, p in enumerate(papers[:30], 1):
    lines2.append(f"## {i}. {p['title']}\n")
    lines2.append(f"- **被引**: {p['cited_by_count']} | **年份**: {p['year']} | **语言**: {p['language']}")
    lines2.append(f"- **期刊**: {p['source']}")
    lines2.append(f"- **作者**: {', '.join(p['authors'][:5])}")
    lines2.append(f"- **DOI**: {p['doi']}")
    lines2.append(f"- **摘要**: {p['abstract'][:800]}\n")

with open(f"{out_dir}/高被引论文TOP30.md", "w", encoding="utf-8") as f:
    f.write("\n".join(lines2))

print(f"\n文件已保存到 {out_dir}/")
print(f"  - openalex_all_papers.json ({len(papers)} papers)")
print(f"  - 学术论文汇总.md")
print(f"  - 高被引论文TOP30.md")
print(f"\nTop 5 by citations:")
for p in papers[:5]:
    print(f"  [{p['cited_by_count']}] {p['title'][:70]}")