Fin Ref Paper

from scripts.ref_paper import ReferenceExtractor, ExtractedReference

extractor = ReferenceExtractor()

# 从多个来源提取
extracted = extractor.extract_from_multiple(
    sources=[
        "output/fin-literature/LIT_REVIEW.md",
        "output/fin-ideas/IDEA_REPORT.md",
        "output/fin-novelty/NOVELTY_REPORT.md",
    ]
)

print(f"提取到 {len(extracted)} 条参考文献")
for ref in extracted[:3]:
    print(f"  - {ref.author} ({ref.year}). {ref.title}. {ref.journal}.")

import requests

def get_crossref_metadata(doi: str) -> dict:
    """通过DOI获取完整的文献元数据"""
    resp = requests.get(
        f"https://api.crossref.org/works/{doi}",
        headers={"User-Agent": "fin-research/1.0 (mailto:research@example.com)"}
    )
    if resp.status_code == 200:
        return resp.json()["message"]
    return {}

def enrich_reference(ref: ExtractedReference) -> ExtractedReference:
    """补全缺失字段"""
    if ref.doi:
        meta = get_crossref_metadata(ref.doi)
        ref.volume = meta.get("volume", ref.volume)
        ref.issue = meta.get("issue", ref.issue)
        ref.pages = meta.get("page", ref.pages)
        ref.publisher = meta.get("publisher", "")
        # 规范化作者列表
        authors_raw = meta.get("author", [])
        ref.author = "; ".join([
            f"{a['family']}, {a['given'][0]}." 
            for a in authors_raw
        ])
    return ref

GET https://api.crossref.org/works/10.1111/jofi.12345

Response:
{
  "message": {
    "title": ["The Effect of Carbon Trading on Innovation"],
    "author": [
      {"family": "Zhang", "given": "Yi"},
      {"family": "Li", "given": "Xiao"}
    ],
    "container-title": ["Journal of Finance"],
    "volume": "78",
    "issue": "3",
    "page": "1234-1289",
    "DOI": "10.1111/jofi.12345",
    "published": {"date-parts": [[2023, 6]]}
  }
}

def to_bibtex(ref: ExtractedReference, style: str = "jf") -> str:
    """将参考文献转换为BibTeX格式"""
    
    # 生成唯一引用键
    first_author = ref.author.split(";")[0].split(",")[0].strip()
    key = f"{first_author}{ref.year}"
    
    # 期刊缩写映射 (常见金融期刊)
    journal_abbr = {
        "Journal of Finance": "J. Finance",
        "Journal of Financial Economics": "J. Financ. Econ.",
        "Review of Financial Studies": "Rev. Financ. Stud.",
        "Quarterly Journal of Economics": "Q. J. Econ.",
        "American Economic Review": "Am. Econ. Rev.",
        "经济研究": "J. Econ. (China)",
        "金融研究": "J. Finance (China)",
    }
    
    journal = journal_abbr.get(ref.journal, ref.journal)
    
    bib_entry = f"""@article{{{key},
  author = {{{'; '.join([a.strip() for a in ref.author.split(';')])}}},
  title = {{{ref.title}}},
  journal = {{{journal}}},
  year = {{{ref.year}}},
  volume = {{{ref.volume or ''}}},
  number = {{{ref.issue or ''}}},
  pages = {{{ref.pages or ''}}},
  doi = {{{ref.doi or ''}}},
}}"""
    
    return bib_entry

@article{Zhang2023,
  author = {Zhang, Y. and Li, X.},
  title = {The Effect of Carbon Trading on Innovation: Evidence from China},
  journal = {J. Finance},
  year = {2023},
  volume = {78},
  number = {3},
  pages = {1234--1289},
  doi = {10.1111/jofi.12345},
}

@article{Li2022,
  author = {Li, H. and Wang, J.},
  title = {Digital Finance and Corporate Innovation: A Panel Study},
  journal = {J. Financ. Econ.},
  year = {2022},
  volume = {145},
  number = {2},
  pages = {567--589},
  doi = {10.1016/j.jfineco.2022.01.001},
}

from scripts.ref_paper import ReferenceFormatter

formatter = ReferenceFormatter()

# JF/JFE/RFS格式 (AEA/biblio style)
jf_refs = formatter.format(refs, style="jf")
# 输出: Author (Year). Title. Journal Volume: Pages.

# GB/T 7714-2015格式 (中文顶刊)
cn_refs = formatter.format(refs, style="gb7714")
# 输出: [序号] 作者. 题目[文献类型]. 期刊名, 年, 卷(期): 页.

# 经济研究格式
jjyj_refs = formatter.format(refs, style="经济研究")

from scripts.ref_paper import ReferenceAuditor

auditor = ReferenceAuditor()

# 检查项
issues = auditor.audit(
    bib_file="references.bib",
    tex_file="paper.tex"
)

# 问题类型
# - UNUSED_REF: .bib中有但.tex中未引用
# - MISSING_REF: .tex中引用但.bib中缺失
# - DUPLICATE: 重复条目
# - FORMAT_ERROR: 格式错误
# - CASE_MISMATCH: 作者姓名大小写不一致

for issue in issues:
    print(f"[{issue.severity}] {issue.type}: {issue.message}")
    print(f"  Location: {issue.location}")

检查项:
□ 所有 \cite{} 键存在于 references.bib
□ 所有 references.bib 条目在 .tex 中被引用
□ 无重复条目
□ 作者姓名一致 (检查大小写)
□ DOI格式正确
□ 年份格式一致
□ 期刊名称标准化
□ 页码格式正确 (无"1234-89"应为"1234-1289")

Author, A. B. and Author, C. D. (Year). Title of the article. Journal Name Volume(Number): Pages.

Author, A. B. and C. D. Author (Year). Title. Journal Name Volume: Pages.

Author, A. B., and C. D. Author (Year). "Title of the Article." Journal Name Volume(Number): Pages.

[1] 作者1, 作者2. 文章标题[J]. 期刊名称, 年, 卷(期): 页码.

[1] 作者1, 作者2. 文章标题[J]. 期刊名称, 年, 卷(期): 页码.

% 生成时间: 2026-06-13
% 目标期刊: 经济研究
% 参考文献数量: 45

@article{Zhang2023,
  author = {Zhang, Y. and Li, X.},
  title = {碳排放权交易对企业绿色创新的影响},
  journal = {经济研究},
  year = {2023},
  volume = {58},
  number = {3},
  pages = {1234--1289},
  doi = {10.1111/j.1245-2823.2023.01234},
}

# 参考文献索引

## 理论基础
- Zhang2023: 碳排放权交易理论
- Li2022: 绿色创新理论框架

## 实证方法
- Callaway2021: CS-DID方法
- Sun2021: 事件研究估计

## 核心实证
- Wang2023: 碳交易对企业创新的正向影响
- Chen2022: 政策效果异质性分析

# 参考文献一致性审计报告

## 审计时间
2026-06-13

## 问题汇总
| 问题类型 | 数量 | 严重程度 |
|----------|------|----------|
| UNUSED_REF | 3 | WARNING |
| MISSING_REF | 1 | ERROR |
| DUPLICATE | 0 | - |
| FORMAT_ERROR | 2 | WARNING |

## 详细问题

### ERROR: 引用缺失
- \cite{Huang2022} 在正文中引用但bib中不存在

### WARNING: 未使用文献
- Borusyak2024 在bib中但未在正文中引用

[CHECKPOINT] 参考文献管理完成

审计结果:
✅ 所有引用键匹配
⚠️ 3条文献未使用 (可保留供附录)
✅ 格式正确

请选择:
1. 生成/更新 references.bib
2. 查看参考文献索引
3. 查看审计报告
4. 添加新文献