# -*- coding: utf-8 -*-
"""md2docx.py — 方案类文档 Markdown → Word（宋体 / 小四 / 首行缩进 2 字符 / 1.5 倍行距）

用法（Windows，务必先设 UTF-8）:
    set PYTHONUTF8=1
    python scripts/md2docx.py 输入.md 输出.docx

依赖:
    pip install python-docx

说明:
    只负责排版落地，不改动内容。
    支持: 标题(#~####)、正文段落、管道表格、无序/有序列表、- [ ] 复选框、**加粗**、代码块。
    复杂合并单元格、图片、公式请在 Word 中人工微调。
"""

import os
import re
import sys

from docx import Document
from docx.enum.table import WD_TABLE_ALIGNMENT
from docx.enum.text import WD_ALIGN_PARAGRAPH
from docx.oxml.ns import qn
from docx.shared import Cm, Pt

# ==================== 可按甲方模板调整的参数 ====================
FONT_NAME = '宋体'              # 中西文统一宋体
BODY_SIZE = Pt(12)              # 小四 = 12pt
FIRST_LINE_INDENT = Pt(24)      # 首行缩进 2 字符（2 x 12pt）
LINE_SPACING = 1.5              # 行间距 1.5 倍
HEADING_SIZES = {1: Pt(16), 2: Pt(15), 3: Pt(14), 4: Pt(12)}
HEADING_ALIGN_CENTER = {1}      # 一级标题居中
TABLE_FONT_SIZE = Pt(12)
TABLE_HEADER_BOLD = True
CODE_FONT_SIZE = Pt(10.5)
PAGE_WIDTH, PAGE_HEIGHT = Cm(21), Cm(29.7)   # A4
MARGIN_TB, MARGIN_LR = Cm(2.54), Cm(3.17)
# ================================================================

CHECKBOX = {False: '\u2610', True: '\u2611'}   # ☐ / ☑


def set_run_font(run, size=None, bold=None):
    """四处字体全设，避免 Word 回落到等线 / Calibri 造成字体不统一。"""
    run.font.name = FONT_NAME
    rpr = run._element.get_or_add_rPr()
    rfonts = rpr.find(qn('w:rFonts'))
    if rfonts is None:
        rfonts = rpr.makeelement(qn('w:rFonts'), {})
        rpr.append(rfonts)
    rfonts.set(qn('w:ascii'), FONT_NAME)
    rfonts.set(qn('w:hAnsi'), FONT_NAME)
    rfonts.set(qn('w:eastAsia'), FONT_NAME)
    rfonts.set(qn('w:cs'), FONT_NAME)
    if size is not None:
        run.font.size = size
    if bold is not None:
        run.font.bold = bold


def add_runs(paragraph, text, size, bold=False):
    """把 **加粗** 解析成多个 run。"""
    for piece in re.split(r'(\*\*.+?\*\*)', text):
        if not piece:
            continue
        if piece.startswith('**') and piece.endswith('**') and len(piece) > 4:
            set_run_font(paragraph.add_run(piece[2:-2]), size, True)
        else:
            set_run_font(paragraph.add_run(piece), size, bold)


def add_body(doc, text):
    p = doc.add_paragraph()
    pf = p.paragraph_format
    pf.first_line_indent = FIRST_LINE_INDENT
    pf.line_spacing = LINE_SPACING
    add_runs(p, text, BODY_SIZE)
    return p


def add_heading(doc, text, level):
    p = doc.add_paragraph()
    pf = p.paragraph_format
    pf.line_spacing = LINE_SPACING
    pf.space_before = Pt(6)
    pf.space_after = Pt(6)
    if level in HEADING_ALIGN_CENTER:
        pf.alignment = WD_ALIGN_PARAGRAPH.CENTER
    add_runs(p, text, HEADING_SIZES.get(level, BODY_SIZE), True)
    return p


def add_list_item(doc, text, prefix='\u2022 '):
    p = doc.add_paragraph()
    pf = p.paragraph_format
    pf.left_indent = FIRST_LINE_INDENT
    pf.line_spacing = LINE_SPACING
    add_runs(p, prefix + text, BODY_SIZE)
    return p


def is_table_sep(line):
    s = line.strip()
    return bool(s) and set(s) <= set('|-: ') and '-' in s


def split_row(line):
    s = line.strip()
    if s.startswith('|'):
        s = s[1:]
    if s.endswith('|'):
        s = s[:-1]
    return [c.strip() for c in s.split('|')]


def add_table(doc, rows):
    cols = max(len(r) for r in rows)
    table = doc.add_table(rows=0, cols=cols)
    table.style = 'Table Grid'
    table.alignment = WD_TABLE_ALIGNMENT.CENTER
    for i, row in enumerate(rows):
        cells = table.add_row().cells
        for j in range(cols):
            text = row[j] if j < len(row) else ''
            para = cells[j].paragraphs[0]
            para.paragraph_format.line_spacing = LINE_SPACING
            add_runs(para, text, TABLE_FONT_SIZE, bold=(i == 0 and TABLE_HEADER_BOLD))
    return table


def normalize_links(text):
    text = re.sub(r'!\[([^\]]*)\]\([^)]*\)', r'\1', text)      # 图片 -> alt 文本
    text = re.sub(r'\[([^\]]+)\]\(([^)]+)\)', r'\1（\2）', text)  # 链接 -> 文本（URL）
    return text


def setup_document():
    doc = Document()
    style = doc.styles['Normal']
    style.font.name = FONT_NAME
    style.font.size = BODY_SIZE
    try:
        style.element.get_or_add_rPr().get_or_add_rFonts().set(qn('w:eastAsia'), FONT_NAME)
    except Exception:
        pass

    sec = doc.sections[0]
    sec.page_width, sec.page_height = PAGE_WIDTH, PAGE_HEIGHT
    sec.top_margin = sec.bottom_margin = MARGIN_TB
    sec.left_margin = sec.right_margin = MARGIN_LR
    return doc


def convert(md_path, docx_path):
    with open(md_path, encoding='utf-8') as f:
        lines = f.read().split('\n')

    # 跳过 YAML frontmatter（技能元数据不进正文）
    start = 0
    if lines and lines[0].strip() == '---':
        for k in range(1, len(lines)):
            if lines[k].strip() == '---':
                start = k + 1
                break

    doc = setup_document()
    i, in_code = start, False

    while i < len(lines):
        line = lines[i].rstrip()
        stripped = line.strip()

        # 代码块
        if stripped.startswith('```'):
            in_code = not in_code
            i += 1
            continue
        if in_code:
            p = doc.add_paragraph()
            p.paragraph_format.line_spacing = 1.0
            p.paragraph_format.left_indent = FIRST_LINE_INDENT
            add_runs(p, line, CODE_FONT_SIZE)
            i += 1
            continue

        if not stripped:
            i += 1
            continue

        # 标题
        m = re.match(r'^(#{1,6})\s+(.*)$', stripped)
        if m:
            add_heading(doc, m.group(2).strip(), len(m.group(1)))
            i += 1
            continue

        # 表格
        if '|' in stripped and i + 1 < len(lines) and is_table_sep(lines[i + 1]):
            rows = []
            while i < len(lines) and '|' in lines[i]:
                if not is_table_sep(lines[i]):
                    rows.append(split_row(lines[i]))
                i += 1
            if rows:
                add_table(doc, rows)
            continue

        # 分隔线
        if re.match(r'^([-*_]\s*){3,}$', stripped):
            i += 1
            continue

        # 复选框（必须早于普通列表判断）
        m = re.match(r'^[-*+]\s+\[([ xX])\]\s*(.*)$', stripped)
        if m:
            add_list_item(doc, normalize_links(m.group(2)),
                          prefix=CHECKBOX[m.group(1).lower() == 'x'] + ' ')
            i += 1
            continue

        # 无序列表
        m = re.match(r'^[-*+]\s+(.*)$', stripped)
        if m:
            add_list_item(doc, normalize_links(m.group(1)))
            i += 1
            continue

        # 有序列表
        m = re.match(r'^(\d+)[.)]\s+(.*)$', stripped)
        if m:
            add_list_item(doc, normalize_links(m.group(2)), prefix=m.group(1) + '. ')
            i += 1
            continue

        # 引用
        if stripped.startswith('>'):
            add_body(doc, normalize_links(stripped.lstrip('>').strip()))
            i += 1
            continue

        # 普通正文
        add_body(doc, normalize_links(stripped))
        i += 1

    doc.save(docx_path)


def main():
    if len(sys.argv) < 3:
        print(__doc__)
        sys.exit(1)

    md_path, docx_path = sys.argv[1], sys.argv[2]
    if not os.path.isfile(md_path):
        print('输入文件不存在: %s' % md_path)
        sys.exit(1)

    out_dir = os.path.dirname(os.path.abspath(docx_path))
    if out_dir and not os.path.isdir(out_dir):
        os.makedirs(out_dir)

    convert(md_path, docx_path)
    print('已生成: %s' % docx_path)
    print('字体=%s 正文=%s 首行缩进=%s 行距=%s' %
          (FONT_NAME, BODY_SIZE.pt, FIRST_LINE_INDENT.pt, LINE_SPACING))


if __name__ == '__main__':
    main()