# -*- coding: utf-8 -*-
r"""B118: chuyển _agy_sx41_excel.md -> DOCX (mode bw: bỏ emoji) + verify + thống kê."""
import os, re, subprocess, sys, zipfile
from docx import Document

DIR = r'C:\Claw 2\Bao cao - Hanh chinh\Bao-cao-sx'
MD = os.path.join(DIR, '_agy_sx41_excel.md')
DOCX = os.path.join(DIR, 'BAO_CAO_SX_TUAN41_PHAN_VAN_BAN_AGY.docx')
SCRIPT = r'C:\Users\admin\AppData\Local\hermes\skills\productivity\markdown-to-docx\scripts\md2docx.py'

r = subprocess.run([sys.executable, SCRIPT, MD, DOCX, 'bw'], capture_output=True, text=True,
                   encoding='utf-8', errors='replace', timeout=600)
print('  md2docx exit=%s' % r.returncode, ((r.stdout or '') + (r.stderr or '')).strip()[-160:])
d = Document(DOCX)
toan = '\n'.join(p.text for p in d.paragraphs) + '\n'.join(c.text for t in d.tables for row in t.rows for c in row.cells)
emj = re.findall(r'[\U0001F000-\U0001FAFF\u2600-\u27BF\uFE0F]', toan)
print('  DOCX: %d B | đoạn %d | bảng %d | ảnh %d | dấu ** còn %d | emoji còn %d' % (
    os.path.getsize(DOCX), len(d.paragraphs), len(d.tables),
    len([n for n in zipfile.ZipFile(DOCX).namelist() if n.startswith('word/media/')]), toan.count('**'), len(emj)))
for k in ('62 nhà máy', '530', 'HOAN TAT'):
    print('   %-14s %s' % (k, 'CÓ' if k in toan else 'KHÔNG'))
print('  ký tự văn bản:', len(toan))
print('  ->', DOCX)