PlantAI-rag-new/sdk/app/services/docgen/md2docx.py

"""services/docgen/md2docx.py —— Markdown → Word（行级状态机）。

- 行级状态机 → Word 内建样式：
  Normal / Heading 1-3 / List Bullet / List Number / Table Grid / Quote
- 中文：正文宋体 / 标题黑体；font.name 与 w:eastAsia 同时设置
- 表格表头加粗灰底
"""
from __future__ import annotations

import re
from pathlib import Path

from docx import Document
from docx.enum.text import WD_ALIGN_PARAGRAPH
from docx.oxml import OxmlElement
from docx.oxml.ns import qn
from docx.shared import Pt, RGBColor

# 字体：正文宋体 / 标题黑体（Windows 中文环境标准；wps-office 渲染兼容）
_FONT_BODY = "宋体"
_FONT_HEADING = "黑体"

_HEADING_RE = re.compile(r"^(#{1,6})\s+(.*)$")
_LIST_BULLET_RE = re.compile(r"^\s*[-*+]\s+(.*)$")
_LIST_NUMBER_RE = re.compile(r"^\s*(\d+)[.)、]\s*(.*)$")
_QUOTE_RE = re.compile(r"^\s*>\s?(.*)$")
_TABLE_ROW_RE = re.compile(r"^\s*\|(.+)\|\s*$")


def _set_run_font(run, font_name: str, size_pt: float | None = None,
                  bold: bool | None = None) -> None:
    """设置西文 + 中文字体（font.name 与 w:eastAsia 同时设置）。"""
    run.font.name = font_name
    rpr = run._element.get_or_add_rPr()
    rfonts = rpr.find(qn("w:rFonts"))
    if rfonts is None:
        rfonts = OxmlElement("w:rFonts")
        rpr.append(rfonts)
    rfonts.set(qn("w:eastAsia"), font_name)
    if size_pt is not None:
        run.font.size = Pt(size_pt)
    if bold is not None:
        run.font.bold = bold


def _shade_cell(cell, fill: str) -> None:
    """单元格底色（表头灰底）。"""
    tc_pr = cell._tc.get_or_add_tcPr()
    shd = OxmlElement("w:shd")
    shd.set(qn("w:val"), "clear")
    shd.set(qn("w:color"), "auto")
    shd.set(qn("w:fill"), fill)
    tc_pr.append(shd)


def _parse_table_row(line: str) -> list[str]:
    """| a | b | → [a, b]（去首尾空格）。"""
    cells = line.strip().strip("|").split("|")
    return [c.strip() for c in cells]


def _is_separator_row(cells: list[str]) -> bool:
    """|---|---| 分隔行。"""
    return all(re.fullmatch(r":?-{2,}:?", c) for c in cells if c.strip() != "") and any(c.strip() for c in cells)


def docx_from_markdown(markdown: str, out_path: str | Path) -> Path:
    """Markdown → docx（行级状态机）。返回写入路径。

    - 标题：#/##/### → Heading 1/2/3（更深的折叠到 3）
    - 列表：- / * / + → List Bullet；1. / 1) → List Number
    - 引用：> → Quote
    - 表格：| 行 → Table Grid；首行为表头（加粗灰底）
    - 其余 → Normal
    """
    doc = Document()
    lines = (markdown or "").split("\n")

    # 表格缓冲（连续 | 行）
    table_lines: list[str] = []

    def _flush_table() -> None:
        if not table_lines:
            return
        rows = [_parse_table_row(line) for line in table_lines]
        rows = [r for r in rows if not _is_separator_row(r)]
        if rows:
            width = max(len(r) for r in rows)
            rows = [r + [""] * (width - len(r)) for r in rows]
            table = doc.add_table(rows=len(rows), cols=width)
            table.style = "Table Grid"
            for r_idx, row in enumerate(rows):
                for c_idx, cell_text in enumerate(row):
                    cell = table.cell(r_idx, c_idx)
                    cell.text = ""
                    para = cell.paragraphs[0]
                    run = para.add_run(cell_text)
                    _set_run_font(run, _FONT_BODY, 10.5)
                    if r_idx == 0:  # 表头：加粗灰底
                        _set_run_font(run, _FONT_BODY, 10.5, bold=True)
                        _shade_cell(cell, "D9D9D9")
        table_lines.clear()

    def _flush_and(block: list[str]) -> None:
        if block:
            _flush_table()
            block.clear()

    for raw in lines:
        line = raw.rstrip()
        # 表格行缓冲
        if _TABLE_ROW_RE.match(line):
            table_lines.append(line)
            continue
        _flush_table()

        if not line.strip():
            continue
        m = _HEADING_RE.match(line)
        if m:
            level = min(len(m.group(1)), 3)  # 更深标题折叠到 Heading 3
            para = doc.add_heading("", level=level)
            para.style = doc.styles[f"Heading {level}"]
            run = para.add_run(m.group(2).strip())
            _set_run_font(run, _FONT_HEADING, {1: 16, 2: 14, 3: 12}[level], bold=True)
            para.paragraph_format.space_before = Pt(10)
            continue
        m = _LIST_BULLET_RE.match(line)
        if m:
            para = doc.add_paragraph(style="List Bullet")
            run = para.add_run(m.group(1).strip())
            _set_run_font(run, _FONT_BODY, 12)
            continue
        m = _LIST_NUMBER_RE.match(line)
        if m:
            para = doc.add_paragraph(style="List Number")
            run = para.add_run(m.group(2).strip())
            _set_run_font(run, _FONT_BODY, 12)
            continue
        m = _QUOTE_RE.match(line)
        if m:
            para = doc.add_paragraph(style="Quote")
            run = para.add_run(m.group(1).strip())
            _set_run_font(run, _FONT_BODY, 12)
            continue
        # 普通段落
        para = doc.add_paragraph()
        run = para.add_run(line.strip())
        _set_run_font(run, _FONT_BODY, 12)

    _flush_table()
    out = Path(out_path)
    out.parent.mkdir(parents=True, exist_ok=True)
    doc.save(out)
    return out