import csv
from collections.abc import Generator
from datetime import datetime
from pathlib import Path


def filesize_csv(input_dir: Path, output_dir: Path) -> None:
    output_file = output_dir / "files.csv"
    with output_file.open("w") as fw:
        writer = csv.writer(fw)
        writer.writerow(["date", "filename", "mtime", "filesize"])

        for j in input_dir.glob("*.json"):
            p_date = j.stem
            p_fname = j.name
            stat = j.stat()
            p_mtime = stat.st_mtime
            p_size = stat.st_size
            writer.writerow([p_date, p_fname, p_mtime, p_size])


def newer_than_last_clean(
    input_glob: Generator[Path], output_glob: Generator[Path], desc: str | None = None
) -> bool:
    last_raw_update: datetime = datetime(1, 1, 1)
    oldest_clean: datetime = datetime.now()
    for file in input_glob:
        mdate = datetime.fromtimestamp(file.stat().st_mtime)
        if mdate > last_raw_update:
            last_raw_update = mdate
    for file in output_glob:
        mdate = datetime.fromtimestamp(file.stat().st_mtime)
        if mdate < oldest_clean:
            oldest_clean = mdate
    if oldest_clean > last_raw_update:
        print(
            "INFO: Skip creating cleaned data"
            + f"{f' for {desc}' if desc else ''}"
            + f", last processing {oldest_clean} newer than {last_raw_update}"
        )
        return False
    return True


def ensure_dirs(input_dir: Path, output_dir: Path):
    if not input_dir.is_dir():
        raise ValueError
    output_dir.mkdir(exist_ok=True, parents=True)


def main(input: str, output: str) -> None:
    input_dir = Path(input)
    output_dir = Path(output)
    ensure_dirs(input_dir, output_dir)

    filesize_csv(input_dir, output_dir)


if __name__ == "__main__":
    import sys

    if not len(sys.argv) == 3:
        print("Please provide exactly one input directory and one output directory.")
        sys.exit(1)
    inp = sys.argv[1]
    out = sys.argv[2]
    main(inp, out)