PDF from images. Script downscales all jpg images and then merges them into pdfs with pdftk.

stitch_pdf.py

#!/usr/bin/python3.5
import os
import glob
import pathlib
from pathlib import Path

# How to use:
# Run in directory with folders named from 1 and up to 99. Each folder should contain *.JPG files.
# Script will downscale all files that are not yet downscaled to "downscaled" subdirectory and stitch to multiple pdfs: pdf per folder + single full pdf.
# Example of input directories:
#├── 1
#│   ├── DSCF2853.JPG
#│   ├── DSCF2854.JPG
#│   ├── DSCF2855.JPG
#│   ├── DSCF2856.JPG
#│   └── DSCF2857.JPG
#├── 2
#│   ├── DSCF2858.JPG
#│   ├── DSCF2859.JPG
#│   ├── DSCF2860.JPG
#│   └── DSCF2861.JPG
#├── 3
#│   ├── DSCF2862.JPG
#│   ├── DSCF2863.JPG
#│   ├── DSCF2864.JPG
#│   ├── DSCF2865.JPG
#│   └── DSCF2866.JPG
#└── stitch_pdf.py
#
# Example of output:
#├── 1
#│   ├── DSCF2853.JPG
#│   ├── DSCF2854.JPG
#│   ├── DSCF2855.JPG
#│   ├── DSCF2856.JPG
#│   └── DSCF2857.JPG
#├── 2
#│   ├── DSCF2858.JPG
#│   ├── DSCF2859.JPG
#│   ├── DSCF2860.JPG
#│   └── DSCF2861.JPG
#├── 3
#│   ├── DSCF2862.JPG
#│   ├── DSCF2863.JPG
#│   ├── DSCF2864.JPG
#│   ├── DSCF2865.JPG
#│   └── DSCF2866.JPG
#├── downscaled
#│   ├── 1
#│   │   ├── DSCF2853.JPG
#│   │   ├── DSCF2854.JPG
#│   │   ├── DSCF2855.JPG
#│   │   ├── DSCF2856.JPG
#│   │   └── DSCF2857.JPG
#│   ├── 2
#│   │   ├── DSCF2858.JPG
#│   │   ├── DSCF2859.JPG
#│   │   ├── DSCF2860.JPG
#│   │   └── DSCF2861.JPG
#│   └── 3
#│       ├── DSCF2862.JPG
#│       ├── DSCF2863.JPG
#│       ├── DSCF2864.JPG
#│       ├── DSCF2865.JPG
#│       └── DSCF2866.JPG
#├── stitch_pdf.py
#└── pdfs
#    ├── theory01.pdf
#    ├── theory02.pdf
#    ├── theory03.pdf
#    └── theory.pdf



filepaths = glob.glob("./*")
day_paths = []

for filepath in filepaths:
    filename = os.path.basename(filepath)
    try:
        day_number = int(filename)
        day_paths.append(filepath)
    except ValueError:
        pass

day_paths = sorted(day_paths)
print("Directories to be processed: ")
for day_path in day_paths:
    print(day_path)

pdf_dir = "./pdfs"
pathlib.Path(pdf_dir).mkdir(parents=True, exist_ok=True)

pdf_files = []
for day_path in day_paths:
    dir_name = os.path.basename(day_path)
    day_number = int(dir_name)
    
    print("Processing day {}...".format(day_number))
    day_images = sorted(glob.glob(day_path + "/*.JPG"))
    
    downscaled_dir = "./downscaled/" + str(day_number)
    pathlib.Path(downscaled_dir).mkdir(parents=True, exist_ok=True)
    
    skipped = True
    
    downscaled_images = []
    for day_image in day_images:
        image_name = os.path.basename(day_image)
        downscaled_path = downscaled_dir + "/" + image_name
        downscaled_images.append(downscaled_path)

        if Path(downscaled_path).is_file():
            continue
        else:
            skipped = False
            os.system("convert -resize 25% -quality 90% {} {}".format(day_image, downscaled_path))
    
    if skipped:
        print("Images already downscaled!")
    else:
        print("Images downscaled!")
    
    pdf_file = pdf_dir + "/theory{:02}.pdf".format(day_number)
    os.system("convert {} {}".format(" ".join(downscaled_images), pdf_file))
    
    pdf_files.append(pdf_file)
    
pdf_files = sorted(pdf_files)
os.system("pdftk {} cat output {}".format(" ".join(pdf_files), pdf_dir + "/theory.pdf"))