Fragments files in the directory so they can be pushed into Github.

prepare_for_github.py

# Recursively goes through the TARGET directory and zips files that are
# over SIZE_LIMIT bytes into PART_SIZE_7z parts, deleting the originals.
#
# Uses 7zip utility that needs to be ADDED INTO THE $PATH. 
#
# Results in a directory structure that can be pushed into 
# the GitHub without issues with oversized files.
import os, sys, subprocess
from time import sleep

# Path to the folder that should be fragmented.
TARGET = "X:\\_"

# Maximum allowed filesize in bytes.
SIZE_LIMIT = 95000000

# 7-zip size of the part file.
PART_SIZE_7z = "95m"

def obtain_all_filepaths(target_dir):
    files = []
    for r, d, f in os.walk(target_dir):
        for file in f:
            files.append(os.path.join(r, file))
    
    print("Discovered: %s files." % len(files))
    return files

def get_files_over_limit(files):
    large = []
    for f in files:
        size = os.stat(f).st_size
        if size >= SIZE_LIMIT:
            large.append(f)

    print("Discovered: %s files over the size limit." % len(large))
    return large

def zip_file_into_parts(path):
    print("Splitting: %s" % path)
    command = ["7z", "a", "%s_part" % path, path, "-v%" % PART_SIZE_7z]
    print(command)
    system = subprocess.Popen(command, shell=True)
    returncode = system.wait()
    sleep(3) # Just to be sure 7zip releases the file.
    print("File split. Deleting the original.")
    return True

files = obtain_all_filepaths(TARGET)
large = get_files_over_limit(files)
for file in large:
    zip_file_into_parts(file)
    try:
        os.remove(file)
    except PermissionError as e:  
        print("PermissionError: %s" % e)
        print("Deleting failed for: %s" % file)
        continue