VGO Part Merger

merge.py

'''
Utility for merging orchestra parts in a printer-friendly manner

Motivation: print.mit.edu has an unfortunate quirk where only one document can
be uploaded at a time. This is quite inconvenient when printing a bunch of
orchestral parts because there are many instruments.

Only dependencies are python and the pypdf package. Its generic usage is of the
form

    python merge.py -i INPUT_FOLDER -o OUTPUT_FILE

Further options are documented under the -h option.

List of implemented features:
    - merges all top-level files in the specified folder
    - can specify the number of copies for each part

List of desired features:
    - config file for repeated operations
    - better control over how to align each part's pages
    - download directly from sharepoint url?
'''

import argparse
import os

from pypdf import PdfReader, PdfWriter

def input_int(prompt):
    while True:
        try:
            result = int(input(prompt))
            if result < 0:
                raise ValueError
            return result
        except ValueError:
            print('Please enter a nonnegative integer')

# fancy cli thing
parser = argparse.ArgumentParser(
        prog='parts-merger',
        description='merges parts')

parser.add_argument('-i', '--input',
                    required=True,
                    help='directory containing parts to merge')
parser.add_argument('-o', '--output',
                    required=True,
                    help='file to write merged pdf to')
parser.add_argument('-c', '--config',
                    help='load the specified configuration file')
parser.add_argument('-n', '--no-blank',
                    help='don\'t insert a blank page in front of a part if needed',
                    action='store_true')

args = parser.parse_args()
if not os.path.isdir(args.input):
    raise ValueError('must supply a directory as input!')
# for simplicity, only iterate through top-level files
parts = [f for f in os.listdir(args.input)
         if os.path.isfile(os.path.join(args.input, f))]
parts.sort()
outFile = PdfWriter()
# TODO: check for config file's existence
should_save_config = False
for part in parts:
    # TODO: use config file to determine copies
    copies = input_int(f'How many copies for {part}? ')
    if copies == 0:
        continue
    partFile = PdfReader(os.path.join(args.input, part))
    # get page dimensions
    pages = partFile.pages
    if len(pages) == 0:
        print('Empty file, skipping...')
        continue
    box = partFile.pages[0].mediabox
    num_pages = partFile.get_num_pages()
    for _ in range(copies):
        # we use a naive approach: if the number of pages is even, do
        # [BLANK] [PART] [BLANK]
        # and if the number of pages is odd, do
        # [PART] [BLANK]
        if num_pages % 2 == 0:
            box = partFile.pages[0].mediabox
            outFile.add_blank_page(box.width, box.height)
        outFile.append(partFile)
        box = partFile.pages[-1].mediabox
        outFile.add_blank_page(box.width, box.height)
outFile.write(args.output)
outFile.close()
if should_save_config:
    # TODO: write copy counts to config file
    pass