Bundle, rectify and OCR (sometimes poorly) scanned multi-page PDFs in current directory

scanner-pdf-prep.py

import os, subprocess
from glob import  glob
from pprint import pprint

_, directories, files = next(os.walk('.'))

def raad_taal_en_titel(directory):
    split = directory.split(' ', 1)

    if len(split) is 2 and split[0] in ['eng', 'deu', 'nld']:
        return split

    return ('nld', directory)

# Dit zijn files die uit 1 stuk bestaan, dat is geen eigen dir waard
# dus pak ik die rechtstreeks op. Eindigt op PDF, niet op pdf!
inputfiles = [file[:-4] for file in files if file.endswith('.PDF') or file.endswith('.pdf')]

for file in inputfiles:
    taal, titel = raad_taal_en_titel(file)
    print(titel)
    ocrmypdf = ['ocrmypdf', '-l', taal, '--skip-text', '--deskew', f'{file}.PDF', f'{titel}.pdf']
    subprocess.run(ocrmypdf)

for directory in directories:
    files = glob(directory + '/*.PDF') + glob(directory + '/*.pdf')

    if not any(files):
        print(f"De directory {directory} bevat geen PDFs")
        continue

    taal, titel = raad_taal_en_titel(directory)
    print(titel)

    pdfunite = ['pdfunite'] + files + [f"{titel}.pdf"]
    ocrmypdf = ['ocrmypdf', '-l', taal, '--skip-text', '--deskew', f'{titel}.pdf', f'{titel}.pdf']

    subprocess.run(pdfunite)
    subprocess.run(ocrmypdf)

    "pdfunite {directory}/*.{pdf,PDF} {directory}.pdf"
    "exec ocrmypdf -l nld --skip-text --deskew {directory}.pdf --sidecar {directory}.txt"