scripts/fix-faq-inline.py - master - ArtKAtMan/arbat

scripts/fix-faq-inline.py
76 строк2 KB
Mavis (M3)
fix: article markdown — bold/FAQ rendering, wider container, related grid 4-col
10 июл 2026, 18:41
10 июл 2026, 18:41a416f50
"""Second pass: split a giant H2 that runs into **Q1?** A1. **Q2?** A2. ... content.

Targets patterns like:
  ## FAQ**Вопрос 1?** Ответ 1.** Вопрос 2?** Ответ 2.

And turns them into:
  ## FAQ
  ** Вопрос 1? **
  Ответ 1.

  ** Вопрос 2? **
  Ответ 2.
"""
import re
from pathlib import Path

CONTENT_DIR = Path('src/content/articles')

# Match `## FAQ` followed immediately by `**...?**`
# Then we split the rest by `**...?**` markers.
# Be permissive: the H2 line can end with anything.
# We assume the H2 occupies the first line that starts with `##` and contains FAQ.
FAQ_INLINE = re.compile(
    r'^(##\s*FAQ[^\n]*?)\*\*([^*]+?\?)\*\*([\s\S]*?)(?=^##\s|\Z)',
    re.MULTILINE,
)


def fix_inline_faq(body: str) -> str:
    def repl(m: re.Match) -> str:
        h2 = m.group(1).rstrip()  # `## FAQ` (might have a trailing space)
        # normalize: drop everything after the heading keyword
        if not h2.rstrip().endswith('FAQ'):
            h2 = '## FAQ'
        first_q_inner = m.group(2).strip()
        rest = m.group(3)

        out = [h2, '', f'** {first_q_inner} **', '']
        # rest is `A1.** Q2?** A2.** Q3?** A3....`
        # Split on the bold markers
        parts = re.split(r'\*\*([^*]+?)\*\*', rest)
        # parts: [a1, q2, a2, q3, a3, ...]
        i = 1
        a = parts[0].strip()
        if a:
            out.append(a)
            out.append('')
        while i < len(parts) - 1:
            q = parts[i].strip()
            a_next = parts[i + 1].strip() if i + 1 < len(parts) else ''
            out.append(f'** {q} **')
            if a_next:
                out.append('')
                out.append(a_next)
            out.append('')
            i += 2
        return '\n'.join(out).rstrip() + '\n\n'

    return FAQ_INLINE.sub(repl, body)


def main():
    for f in sorted(CONTENT_DIR.glob('*.md')):
        text = f.read_text(encoding='utf-8')
        if '## FAQ' not in text:
            continue
        new = fix_inline_faq(text)
        if new != text:
            f.write_text(new, encoding='utf-8')
            print(f'faq-split: {f.name}')
        else:
            print(f'no change: {f.name}')


if __name__ == '__main__':
    main()