split_into_files.py

#!/usr/bin/env python
"""
Parse a listing like the one in stdin and write each file to disk.
"""

import re
import sys
from pathlib import Path

_PATH_LINE = re.compile(r"^/.*:$")
_DASH_LINE = re.compile(r"^[\-\u2500\u2501]+$")  # ascii or box-drawing dashes
_NUM_LINE = re.compile(r"^\s*\d+\s*\|\s?(.*)$")


def _is_path(line: str) -> bool:
    return bool(_PATH_LINE.match(line.strip()))


def _is_dash(line: str) -> bool:
    return bool(_DASH_LINE.match(line.strip()))


def _strip_prefix(line: str) -> str:
    m = _NUM_LINE.match(line)
    return m.group(1) if m else line


def parse(text: str) -> dict[str, str]:
    """Return {path: file_content} from the input text."""
    files: dict[str, list[str]] = {}
    lines = text.splitlines()
    i = 0
    while i < len(lines):
        if not _is_path(lines[i]):
            i += 1
            continue

        path = lines[i].strip()[:-1]  # drop trailing ':'
        i += 1

        if i < len(lines) and _is_dash(lines[i]):
            i += 1  # skip separator line

        content: list[str] = []
        while i < len(lines) and not _is_path(lines[i]):
            if _is_dash(lines[i]) and i + 1 < len(lines) and _is_path(lines[i + 1]):
                break  # end of this file block
            content.append(_strip_prefix(lines[i]))
            i += 1

        files[path] = "\n".join(content).rstrip("\n") + "\n"

    return files


def write_files(files: dict[str, str], root: Path = Path(".")) -> None:
    """Create directories and write file contents."""
    for rel_path, body in files.items():
        dst = root / rel_path.lstrip("/")
        dst.parent.mkdir(parents=True, exist_ok=True)
        dst.write_text(body, encoding="utf-8")


if __name__ == "__main__":
    data = sys.stdin.read()
    write_files(parse(data))