Function _normalize_heading_line

skills/pdf-parser/scripts/parse_pdf.py:170–176

(line: str)

def _normalize_heading_line(line: str) -> str:
    normalized = re.sub(
        r"^\s*(?:(?:\d+(?:\.\d+)*[.)]?)|(?:[IVXivx]+[.)]?)|(?:[一二三四五六七八九十]+[、.)]?))?\s*",
        "",
        (line or "").strip(),
    )
    return normalized.strip().lower().rstrip(":：")


def _get_rapidocr_engine():