gistfile1.txt

import re
from PIL import Image
from natsort import natsorted
from pathlib import Path
import subprocess


# convert 1-2.png -crop 2x1@ out-%d.png
# convert etukansi.png -background white -page a5 etu.pdf

matcher = "Pohjolan Leijona (?P<season>\d+) vuosikerta nro.(?P<issue>\d+) (?P<year>\d+)"
cwd = Path.cwd()

for d in Path(".").glob("Pohjolan*"):
    matches = re.match(matcher, str(d))
    season = matches.group("season")
    issue = matches.group("issue")
    year = matches.group("year")
    new_dir = cwd / f"{year}.{season}" / f"{issue}"

    if not new_dir.exists():
        new_dir.mkdir(parents=True)

    tmp_pages = d.glob("*.png")
    # tmp_pages = filter(lambda x: "kansi" not in str(x), tmp_pages)
    pages = natsorted(tmp_pages, key=lambda i: str(i))
    print(new_dir)

    page = 1
    for x in pages:
        with Image.open(x) as im:
            width, height = im.size

        if width > 1200:
            cmd = ["convert", str(x), "-crop", "2x1@", str(new_dir / "out-%d.png")]
            subprocess.check_call(cmd)
            for c in [0,1]:
                tmp_file = new_dir / f"out-{c}.png"
                to_pdf = tmp_file.replace( new_dir / f"page_{page + c}.png")
                pdf_cmd = ["convert", str(to_pdf), "-background", "white", "-page", "a5", f"{new_dir}/{to_pdf.stem}.pdf"]
                subprocess.check_call(pdf_cmd)
                to_pdf.unlink()
            page += 2
        else:
            to_pdf = None
            if "etu" in str(x):
                to_pdf = new_dir / "page_0_fc.pdf"
            elif "taka" in str(x):
                to_pdf = new_dir / f"page_{page}_bc.pdf"
            else:
                to_pdf = new_dir / f"page_{page}_xxx.odf"
                page += 1
            pdf_cmd = ["convert", str(x), "-background", "white", "-page", "a5", f"{to_pdf}"]
            subprocess.check_call(pdf_cmd)

        print(f"\t {x} {width}")