zipbench.py

#!/usr/bin/env python3
import sys
import subprocess
import datetime
import csv
import os

def create_csv(csv_file, *args):
    f = open(csv_file, 'w')
    try:
        writer = csv.writer(f)
        writer.writerow(*args)
    finally:
        f.close()


def append_csv(csv_file, *args):
    f = open(csv_file, 'a')
    try:
        writer = csv.writer(f)
        writer.writerow(*args)
    finally:
        f.close()
        
def get_folder_size(folder):
    '''
    Get the size fo all files in a folder. Recursive process.
    https://stackoverflow.com/a/1392549/2188572
    '''
    total_size = 0
    for dirpath, dirnames, filenames in os.walk(folder):
        for f in filenames:
            fp = os.path.join(dirpath, f)
            try:
                total_size += os.path.getsize(fp)
            except OSError:
                continue
    return total_size

source  = sys.argv[1]
destination = sys.argv[2]
folder_size = get_folder_size(source)
csv_report_filename = os.path.basename(source) + "zipbench_report"
csv_report = os.path.expanduser("~/Desktop/%s.csv") % csv_report_filename
create_csv(
    csv_report,
    (
        'compression_type',
        'compression_level',
        'duration_seconds',
        'source_folder_size',
        'zip_file_size',
        'compression_ratio',
        )
    )
pwd = os.getcwd()
compression_types = ['LZMA', 'Deflate', 'Copy']
compression_levels = ['3', '5']
os.chdir(os.path.dirname(source))

for compression_type in compression_types:
    for compression_level in compression_levels:
        zip_start = datetime.datetime.now()
        full_zip = os.path.join(destination, '%s_%s' % (compression_type, compression_level[-1]))
        compression_type_string = '-mm=%s' % compression_type
        compression_level_string = '-mx=%s' % compression_level
        zip_cmd = ['7za', 'a', '-tzip', compression_type_string, compression_level_string, full_zip, os.path.basename(source)]
        print(zip_cmd)
        subprocess.call(zip_cmd)
        zip_end = datetime.datetime.now()
        append_csv(
            csv_report,
            (
                compression_type,
                compression_level,
                zip_end - zip_start,
                folder_size,
                os.path.getsize(full_zip + '.zip'),
                os.path.getsize(full_zip + '.zip') / folder_size,
                )
            )