AiDOPWarehouse/doc/_adjust_front_matter.py

# -*- coding: utf-8 -*-
"""Adjust front matter: drop 文档控制, rename 更改记录, page breaks."""
from __future__ import annotations

import shutil
import tempfile
from pathlib import Path

from docx import Document
from docx.enum.text import WD_BREAK
from docx.oxml import OxmlElement
from docx.text.paragraph import Paragraph

BASE = Path(
    r"C:\Users\skygu\OneDrive\Projects\AIDOP\项目\项目管理\产互联项目管理\交付文档"
)


def remove_paragraph(p: Paragraph) -> None:
    p._element.getparent().remove(p._element)


def clear_page_breaks(p: Paragraph) -> None:
    for br in p._element.findall(".//{http://schemas.openxmlformats.org/wordprocessingml/2006/main}br"):
        br.getparent().remove(br)


def ensure_page_break_before(p: Paragraph) -> None:
    clear_page_breaks(p)
    text = p.text
    p.text = ""
    br_run = p.add_run()
    br_run.add_break(WD_BREAK.PAGE)
    if text:
        p.add_run(text)


def adjust_file(path: Path) -> None:
    work = path
    tmp_path = None
    try:
        doc = Document(path)
    except PermissionError:
        tmp_path = Path(tempfile.gettempdir()) / f"docfix_{path.name}"
        shutil.copy2(path, tmp_path)
        work = tmp_path
        doc = Document(work)

    removed_control = False
    renamed = False
    version_break = False
    toc_break = False

    for p in list(doc.paragraphs):
        if p.text.strip() == "文档控制":
            remove_paragraph(p)
            removed_control = True

    for p in doc.paragraphs:
        if p.text.strip() == "更改记录":
            p.text = "版本记录"
            renamed = True
            break

    for p in doc.paragraphs:
        if p.text.strip() == "版本记录":
            ensure_page_break_before(p)
            version_break = True
            break

    for p in doc.paragraphs:
        if p.text.strip() == "目录":
            ensure_page_break_before(p)
            toc_break = True
            break

    if not renamed and not any(p.text.strip() == "版本记录" for p in doc.paragraphs):
        raise RuntimeError(f"未找到「版本记录」: {path.name}")
    if not version_break:
        raise RuntimeError(f"未找到「版本记录」: {path.name}")
    if not toc_break:
        raise RuntimeError(f"未找到「目录」: {path.name}")

    doc.save(work)
    if tmp_path is not None:
        try:
            shutil.copy2(work, path)
            tmp_path.unlink(missing_ok=True)
        except PermissionError:
            print(f"  !! {path.name} 被占用，已写入: {work}")
            return
    print(
        f"{path.name}: 删文档控制={'Y' if removed_control else 'N'} "
        f"版本记录分页=Y 目录分页=Y"
    )


def iter_delivery_docx():
    for module in ["S0", "S1", "S2", "S3", "S4"]:
        mod_dir = BASE / module
        if not mod_dir.is_dir():
            continue
        for path in sorted(mod_dir.rglob("*.docx")):
            if not path.name.startswith("~$"):
                yield path


def main() -> None:
    for path in iter_delivery_docx():
        adjust_file(path)


if __name__ == "__main__":
    main()