exporters.py

"""导出模块 - RIS / BibTeX / CSV"""

import csv
import io
import re


def export_ris(papers: list) -> str:
    """导出 RIS 格式（EndNote 直接导入）"""
    if not papers:
        return ""
    lines = []
    for p in papers:
        try:
            lines.append("TY  - JOUR")
            if getattr(p, "title", None):
                lines.append(f"TI  - {p.title}")
            for author in getattr(p, "authors", None) or []:
                lines.append(f"AU  - {author}")
            if getattr(p, "year", None):
                lines.append(f"PY  - {p.year}")
                lines.append(f"DA  - {p.year}///")
            if getattr(p, "journal", None):
                lines.append(f"JO  - {p.journal}")
                lines.append(f"JA  - {p.journal}")
                lines.append(f"T2  - {p.journal}")
            # 尝试提取卷号、期号、页码（如果有）
            if getattr(p, "volume", None):
                lines.append(f"VL  - {p.volume}")
            if getattr(p, "issue", None):
                lines.append(f"IS  - {p.issue}")
            if getattr(p, "pages", None):
                lines.append(f"SP  - {p.pages}")
            if getattr(p, "issn", None):
                lines.append(f"SN  - {p.issn}")
            if getattr(p, "doi", None):
                lines.append(f"DO  - {p.doi}")
            if getattr(p, "pmid", None):
                lines.append(f"AN  - PMID:{p.pmid}")
            if getattr(p, "abstract", None):
                abs_text = p.abstract[:2000]
                lines.append(f"AB  - {abs_text}")
            for kw in getattr(p, "keywords", None) or []:
                lines.append(f"KW  - {kw}")
            if getattr(p, "oa_url", None):
                lines.append(f"UR  - {p.oa_url}")
            elif getattr(p, "doi", None):
                lines.append(f"UR  - https://doi.org/{p.doi}")
            # Phase 5a: 学术元数据
            if getattr(p, "orcid", None):
                lines.append(f"OI  - {p.orcid}")
            if getattr(p, "article_type", None):
                lines.append(f"M3  - {p.article_type}")
            if getattr(p, "conference", None):
                lines.append(f"CT  - {p.conference}")
            for fund in getattr(p, "funding", None) or []:
                lines.append(f"RG  - {fund}")
            lines.append("ER  - ")
            lines.append("")
        except Exception as e:
            print(f"[WARN] Skipping paper in RIS export: {e}")
            continue

    return "\n".join(lines)


def _bibtex_escape(text: str) -> str:
    """BibTeX 特殊字符转义（LaTeX 保留字符）"""
    if not text:
        return ""
    text = text[:2000]
    # 注意：\ { } 是 BibTeX/LaTeX 结构字符（命令前缀、分组符），不应按字面转义
    # 只转义会被 LaTeX 解释的特殊字面字符
    text = text.replace("%", "\\%")
    text = text.replace("#", "\\#")
    text = text.replace("&", "\\&")
    text = text.replace("_", "\\_")
    text = text.replace("$", "\\$")
    text = text.replace("~", "\\~{}")
    text = text.replace("^", "\\^{}")
    return text


def export_bibtex(papers: list) -> str:
    """导出 BibTeX 格式"""
    if not papers:
        return ""
    lines = []
    for i, p in enumerate(papers, 1):
        try:
            key = _make_bibtex_key(p, i)
            entry_type = _infer_entry_type(p)
            lines.append(f"@{entry_type}{{{key},")
            if getattr(p, "title", None):
                lines.append(f"  title = {{{_bibtex_escape(p.title)}}},")
            if getattr(p, "authors", None):
                lines.append(
                    f"  author = {{{' and '.join(_bibtex_escape(a) for a in p.authors)}}},"
                )
            if getattr(p, "journal", None):
                lines.append(f"  journal = {{{_bibtex_escape(p.journal)}}},")
            if getattr(p, "year", None):
                lines.append(f"  year = {{{p.year}}},")
            # 补充学术字段
            if getattr(p, "volume", None):
                lines.append(f"  volume = {{{_bibtex_escape(p.volume)}}},")
            if getattr(p, "issue", None):
                lines.append(f"  number = {{{_bibtex_escape(p.issue)}}},")
            if getattr(p, "pages", None):
                lines.append(f"  pages = {{{_bibtex_escape(p.pages)}}},")
            if getattr(p, "issn", None):
                lines.append(f"  issn = {{{_bibtex_escape(p.issn)}}},")
            if getattr(p, "doi", None):
                lines.append(f"  doi = {{{_bibtex_escape(p.doi)}}},")
            if getattr(p, "pmid", None):
                lines.append(f"  pmid = {{{p.pmid}}},")
            if getattr(p, "abstract", None):
                lines.append(f"  abstract = {{{_bibtex_escape(p.abstract)}}},")
            if getattr(p, "keywords", None):
                lines.append(
                    f"  keywords = {{{', '.join(_bibtex_escape(kw) for kw in p.keywords)}}},"
                )
            if getattr(p, "oa_url", None):
                lines.append(f"  url = {{{_bibtex_escape(p.oa_url)}}},")
            elif getattr(p, "doi", None):
                lines.append(
                    f"  url = {{{_bibtex_escape(f'https://doi.org/{p.doi}')}}},"
                )
            # Phase 5a: 学术元数据
            if getattr(p, "orcid", None):
                lines.append(f"  orcid = {{{_bibtex_escape(p.orcid)}}},")
            if getattr(p, "article_type", None):
                lines.append(f"  type = {{{_bibtex_escape(p.article_type)}}},")
            if getattr(p, "conference", None):
                lines.append(f"  booktitle = {{{_bibtex_escape(p.conference)}}},")
            if getattr(p, "funding", None):
                lines.append(f"  note = {{{_bibtex_escape('; '.join(p.funding))}}},")
            lines.append("}")
            lines.append("")
        except Exception as e:
            print(f"[WARN] Skipping paper in BibTeX export: {e}")
            continue

    return "\n".join(lines)


def _infer_entry_type(paper) -> str:
    """根据论文字段推断 BibTeX 条目类型"""
    if getattr(paper, "journal", None):
        return "article"
    if hasattr(paper, "booktitle") and paper.booktitle:
        return "inproceedings"
    return "misc"


def export_csv(papers: list) -> str:
    """导出 CSV 格式"""
    if not papers:
        return ""
    output = io.StringIO()
    writer = csv.writer(output)

    # 表头（包含学术字段）
    writer.writerow(
        [
            "Title",
            "Authors",
            "Journal",
            "Year",
            "Volume",
            "Issue",
            "Pages",
            "DOI",
            "PMID",
            "ISSN",
            "Citations",
            "OA_URL",
            "Keywords",
            "Abstract",
            "ORCID",
            "ArticleType",
            "Conference",
            "Funding",
        ]
    )

    for p in papers:
        try:
            writer.writerow(
                [
                    getattr(p, "title", ""),
                    "; ".join(getattr(p, "authors", None) or []),
                    getattr(p, "journal", ""),
                    getattr(p, "year", ""),
                    getattr(p, "volume", ""),
                    getattr(p, "issue", ""),
                    getattr(p, "pages", ""),
                    getattr(p, "doi", ""),
                    getattr(p, "pmid", ""),
                    getattr(p, "issn", ""),
                    getattr(p, "citation_count", 0),
                    getattr(p, "oa_url", ""),
                    "; ".join(getattr(p, "keywords", None) or []),
                    (getattr(p, "abstract", "") or "")[:2000],
                    getattr(p, "orcid", ""),
                    getattr(p, "article_type", ""),
                    getattr(p, "conference", ""),
                    "; ".join(getattr(p, "funding", None) or []),
                ]
            )
        except Exception as e:
            print(f"[WARN] Skipping paper in CSV export: {e}")
            continue

    return output.getvalue()


def _make_bibtex_key(paper, index: int) -> str:
    """生成 BibTeX 引用键"""
    # 第一作者姓氏
    first_author = ""
    authors = getattr(paper, "authors", None) or []
    if authors:
        name = authors[0]
        if "," in name:
            # "Smith, John" -> "Smith"
            first_author = name.split(",")[0].strip()
        else:
            # "John Smith" -> "Smith"
            parts = name.strip().split()
            first_author = parts[-1] if parts else ""
        # 保留字母（含非 ASCII 如中文/日文/韩文/带重音的欧洲文字）
        first_author = re.sub(r"[^\w]", "", first_author)

    year = str(getattr(paper, "year", "") or "xxxx")
    author_part = first_author[:8] if first_author else "unknown"

    return f"{author_part}{year}_{index}"


# EndNote ref-type 映射：article_type 值 -> (display name, numeric ID)
_ENDNOTE_REF_TYPE_MAP = {
    "journal-article": ("Journal Article", 17),
    "review-article": ("Journal Article", 17),
    "review": ("Journal Article", 17),
    "editorial": ("Journal Article", 17),
    "letter": ("Journal Article", 17),
    "erratum": ("Journal Article", 17),
    "book-chapter": ("Book Section", 3),
    "book": ("Book", 1),
    "proceedings-article": ("Conference Paper", 5),
    "conference-paper": ("Conference Paper", 5),
    "dissertation": ("Thesis", 7),
    "thesis": ("Thesis", 7),
    "patent": ("Patent", 11),
    "report": ("Report", 6),
    "standard": ("Report", 6),
    "dataset": ("Report", 6),
}


def _endnote_ref_type(article_type: str):
    """将 article_type 映射为 EndNote ref-type (name, id)"""
    key = article_type.lower().strip().replace(" ", "-")
    return _ENDNOTE_REF_TYPE_MAP.get(key, ("Journal Article", 17))


def export_endnote_xml(papers: list) -> str:
    """导出 EndNote XML 格式（Mendeley / EndNote 直接导入）"""
    lines = ['<?xml version="1.0" encoding="UTF-8"?>']
    lines.append("<xml><records>")

    for p in papers:
        lines.append("<record>")
        # ref-type 必须是 <record> 的第一个子元素
        if getattr(p, "article_type", None):
            ref_type_name, ref_type_id = _endnote_ref_type(p.article_type)
        else:
            ref_type_name, ref_type_id = "Journal Article", 17
        lines.append(f'<ref-type name="{ref_type_name}">{ref_type_id}</ref-type>')
        lines.append("<contributors>")
        if p.authors:
            lines.append("<authors>")
            for author in p.authors:
                # EndNote XML 格式：姓, 名
                if "," in author:
                    parts = author.split(",", 1)
                    lines.append(
                        f'<author><style face="normal" font="default" size="100">{_xml_escape(parts[0].strip())}, {_xml_escape(parts[1].strip())}</style></author>'
                    )
                else:
                    lines.append(
                        f'<author><style face="normal" font="default" size="100">{_xml_escape(author)}</style></author>'
                    )
            lines.append("</authors>")
        lines.append("</contributors>")
        # Build <titles> block: title + secondary-title (conference)
        titles_children = []
        if p.title:
            titles_children.append(
                f'<title><style face="normal" font="default" size="100">{_xml_escape(p.title)}</style></title>'
            )
        if getattr(p, "conference", None):
            titles_children.append(
                f'<secondary-title><style face="normal" font="default" size="100">{_xml_escape(p.conference)}</style></secondary-title>'
            )
        if titles_children:
            lines.append("<titles>")
            for tc in titles_children:
                lines.append(tc)
            lines.append("</titles>")
        # <periodical> is a sibling of <titles> at record level, not nested inside
        if p.journal:
            lines.append(
                f'<periodical><full-title><style face="normal" font="default" size="100">{_xml_escape(p.journal)}</style></full-title></periodical>'
            )
        if p.year:
            lines.append(
                f'<dates><year><style face="normal" font="default" size="100">{p.year}</style></year></dates>'
            )
        # 补充学术字段
        if getattr(p, "volume", None):
            lines.append(
                f'<volume><style face="normal" font="default" size="100">{_xml_escape(p.volume)}</style></volume>'
            )
        if getattr(p, "issue", None):
            lines.append(
                f'<number><style face="normal" font="default" size="100">{_xml_escape(p.issue)}</style></number>'
            )
        if getattr(p, "pages", None):
            lines.append(
                f'<pages><style face="normal" font="default" size="100">{_xml_escape(p.pages)}</style></pages>'
            )
        if getattr(p, "issn", None):
            lines.append(
                f'<isbn-issn><style face="normal" font="default" size="100">{_xml_escape(p.issn)}</style></isbn-issn>'
            )
        if p.doi:
            lines.append(
                f'<electronic-resource-num><style face="normal" font="default" size="100">{_xml_escape(p.doi)}</style></electronic-resource-num>'
            )
        if p.abstract:
            abs_text = _xml_escape(p.abstract[:2000])
            lines.append(
                f'<abstract><style face="normal" font="default" size="100">{abs_text}</style></abstract>'
            )
        if p.keywords:
            lines.append("<keywords>")
            for kw in p.keywords:
                lines.append(
                    f'<keyword><style face="normal" font="default" size="100">{_xml_escape(kw)}</style></keyword>'
                )
            lines.append("</keywords>")
        # Phase 5a: 学术元数据
        # ORCID 使用 <custom1> 存储，避免与 DOI 的 <electronic-resource-num> 冲突
        if getattr(p, "orcid", None):
            orcid_value = _xml_escape(p.orcid)
            # 如果 ORCID 没有 https:// 前缀，添加标准前缀
            if not orcid_value.startswith("https://orcid.org/"):
                orcid_value = f"https://orcid.org/{orcid_value}"
            lines.append(
                f'<custom1><style face="normal" font="default" size="100">{orcid_value}</style></custom1>'
            )
        # conference is already included in <titles> block above
        if getattr(p, "funding", None):
            for fund in p.funding:
                lines.append(
                    f'<custom3><style face="normal" font="default" size="100">{_xml_escape(fund)}</style></custom3>'
                )
        if p.pmid:
            lines.append(
                f'<accession-num><style face="normal" font="default" size="100">{_xml_escape(p.pmid)}</style></accession-num>'
            )
        url = p.oa_url or (f"https://doi.org/{p.doi}" if p.doi else "")
        if url:
            lines.append(
                f'<urls><related-urls><url><style face="normal" font="default" size="100">{_xml_escape(url)}</style></url></related-urls></urls>'
            )
        lines.append("</record>")

    lines.append("</records></xml>")
    return "\n".join(lines)


def _xml_escape(text: str) -> str:
    """XML 特殊字符转义"""
    if not text:
        return ""
    return (
        text.replace("&", "&amp;")
        .replace("<", "&lt;")
        .replace(">", "&gt;")
        .replace('"', "&quot;")
        .replace("'", "&apos;")
    )