Create Optimized DVD/CD Images of Folder for Backup

The Problem:
I have music, videos, pictures, and files that all need backups. I have them organized how I want, but they don't always fit perfectly on to a DVD. I want a simple program that I point to a directory and it will create optimized image files for burning to a dvd. I would also like to do a few quick 'dry runs' to be sure that I like which files go in each image and that I don't have an image that is only 10% full!

The Solution:
Well, if you can't find one... make one! I decided to write a script that linked up with PowerISO's commandline tool (piso) to create images of files. You will need several Gigs of space to store the images... This is nice because you don't have to go through all your files and figure out how to divide them among images... You just do a few dry runs and then when you are happy, let it go all night!

The Script
This script uses the command line utility of PowerISO to create disc images of a folder and tries to maximize the disc usage while still preserving the file structure of the provided folder. You will need to add the PowerISO install path to your path environment variable
file_dvd_backup.pyfrom optparse import OptionParser<br />import time, logging, os, sys, string, shutil, tempfile<br />from os import sep<br /><br />class ImageTooLargeException(Exception):<br /> def __init__(self, value):<br /> self.value = value<br /> def __str__(self):<br /> return repr(self.value)<br /><br />class DiscImage:<br /> def __init__(self, file_path, base_dir, max = 4.37, password = None):<br /> self.file_name = '.'.join(file_path.split(sep)[-1].split('.')[:-1])<br /> self.logger = logging.getLogger('DiscImage:%s'%self.file_name)<br /> self.temp = None<br /> self.files = []<br /> self.dirs = []<br /> self.command = 'piso'<br /> if base_dir[-1] != sep:<br /> base_dir += sep<br /> self.base_dir = base_dir<br /> self.file_path = file_path<br /> self.max = max *1000*1000*1000 # Size in bytes<br /> self.password = password<br /> self.size = 0<br /> <br /> def setCommand(self, command):<br /> self.command = command<br /> <br /> def add(self, file):<br /> file_size = os.stat(file).st_size<br /> if self.size + file_size > self.max:<br /> raise ImageTooLargeException('File could not be added because it was too large: %s' %file)<br /> self.size += file_size<br /> self.files.append(file)<br /> self.logger.log(5, 'Adding File to Image: %s' %file)<br /> rel_dir = sep.join(file.split(sep)[:-1]).replace(self.base_dir, '')<br /> if rel_dir not in self.dirs:<br /> self.dirs.append(rel_dir)<br /> <br /> def append(self, file):<br /> self.add(file)<br /> <br /> def percent_full(self):<br /> return self.size/self.max*100<br /> <br /> def create(self):<br /> # I copy all to a temp directory because Power ISO does not have an add/append command and will not take more than a few -add command line arguments<br /> self.copyFiles()<br /> piso = self.command<br /> if self.command != 'piso':<br /> cmd = '"%s"' %self.command<br /> else:<br /> cmd = [self.command]<br /> cmd = [self.command]<br /> cmd.append('create')<br /> cmd.append('-o %s' % self.file_path)<br /> <br /> if self.password is not None:<br /> cmd.append('-setpassword %s' %self.password)<br /> <br /> cmd.append('-add \"%s\" /' %self.temp)<br /> <br /> cmd_str = ' '.join(cmd)<br /> if self.password is not None:<br /> cmd_str = cmd_str.replace(self.password, '*'*len(self.password))<br /> <br /> self.logger.debug('Creating Image with Command: \n%s '% (cmd_str))<br /> os.system(' '.join(cmd))<br /> <br /> self.cleanTemp()<br /> <br /> def copyFiles(self):<br /> if self.temp is None:<br /> self.temp = tempfile.mkdtemp()<br /> for file in self.files:<br /> file_dir = os.path.join(self.temp, sep.join(file.replace(self.base_dir, '').split(sep)[:-1]))<br /> if not os.access(file_dir, os.F_OK):<br /> os.makedirs(file_dir)<br /> self.logger.log(3, "Creating Temp File: %s" %os.path.join(file_dir, file.split(sep)[-1]))<br /> shutil.copy(file, os.path.join(file_dir, file.split(sep)[-1]))<br /> <br /> <br /> def cleanTemp(self):<br /> for root, dirs, files in os.walk(self.temp, topdown=False):<br /> for name in files:<br /> try:<br /> os.remove(os.path.join(root, name))<br /> except:<br /> self.logger.error('Could not remove temp file: %s' %os.path.join(root, name))<br /> for name in dirs:<br /> try:<br /> os.rmdir(os.path.join(root, name))<br /> except:<br /> self.logger.error('Could not remove temp dir: %s' %os.path.join(root, name))<br /> try:<br /> os.rmdir(self.temp)<br /> except:<br /> self.logger.error('Could not remove temp dir: %s' %os.path.join(root, name))<br /> <br /> self.temp = None<br /><br /><br />class Dir2Img:<br /> def __init__(self, src, dest, name_base, img_type, max = 4.37, password = None):<br /> self.logger = logging.getLogger('Dir2Img')<br /> self.command = None<br /> self.src = src<br /> self.dest = dest<br /> self.name_base = name_base<br /> self.img_type = img_type<br /> self.max = max<br /> self.password = password<br /> self.images = []<br /> self.parseImages()<br /> <br /> def setCommand(self, command):<br /> self.command = command<br /> for image in self.images:<br /> image.setCommand(command)<br /> <br /> def parseImages(self):<br /> img = self.getNextImage()<br /> img2 = None<br /> for root, dirs, files in os.walk(self.src, topdown=True):<br /> for name in files:<br /> full_path = os.path.join(root, name)<br /> img, img2 = self.addToImage(full_path, img, img2)<br /> <br /> def addToImage(self, full_path, img, img2 = None):<br /> # Try and add it to the current image<br /> try:<br /> img.append(full_path)<br /> except ImageTooLargeException:<br /> # See if the current image is less than 75 percent full<br /> if img.percent_full()< file_size =" os.stat(full_path).st_size/(1000*1000*1000)"> self.max:<br /> raise ImageTooLargeException('File Too Large! File: %s, Size: %fMB' % (full_path, file_size*1000))<br /> <br /> if img2 is not None:<br /> # If we are already dealing with a second image, add the file to that and make it the primary image<br /> return self.addToImage(full_path, img2, None)<br /> else:<br /> # If not, then create a secondary image and add the large file to it<br /> img2 = self.getNextImage()<br /> img2.append(full_path)<br /> else:<br /> # Image is pretty much full<br /> if img2 is not None:<br /> # If dealing with a secondary image, start adding smaller files to it<br /> return self.addToImage(full_path, img2, None)<br /> else:<br /> # Create a new image<br /> img = self.getNextImage()<br /> img.append(full_path)<br /> return img, img2<br /> <br /> def getNextImage(self):<br /> i = ''<br /> if len(self.images) > 0:<br /> i = str(len(self.images))<br /> file_path = os.path.join(self.dest, '%s%s.%s' %(self.name_base, i, self.img_type))<br /> img = DiscImage(file_path, self.src, self.max, self.password)<br /> self.images.append(img)<br /> self.logger.log(1, 'New Image: %s' %file_path)<br /> if self.command is not None:<br /> img.setCommand(self.command)<br /> return img<br /> <br /> def create(self):<br /> for image in self.images:<br /> image.create()<br /> <br /> def usageReport(self):<br /> s = ''<br /> for image in self.images:<br /> s += '%s: %f%% Full\n' %(image.file_name, image.percent_full())<br /> return s<br /> <br /> def fullReport(self):<br /> s = ''<br /> for image in self.images:<br /> s += '\n%s\n'%(image.file_name)<br /> s += '\tUsage: %f%% Full\n' %(image.percent_full())<br /> s += '\tDirectories:\n'<br /> for dir in image.dirs:<br /> s += '\t\t%s\n' %dir<br /> s += '\n'*2<br /> return s<br /> <br /> <br />def setupLogging(level):<br /> if not os.access('log', os.F_OK):<br /> os.mkdir('log')<br /> logging.basicConfig(level=level,<br /> format='%(asctime)s [%(levelname)s] %(message)s',<br /> filename=os.path.abspath('log/file_dvd_backup.log'),<br /> filemode='w')<br /> # define a Handler which writes INFO messages or higher to the sys.stderr<br /> console = logging.StreamHandler()<br /> console.setLevel(level)<br /> # set a format which is simpler for console use<br /> formatter = logging.Formatter('%(name)-5s: [%(levelname)-5s] %(message)s')<br /> # tell the handler to use this format<br /> console.setFormatter(formatter)<br /> # add the handler to the root logger<br /> logging.getLogger('').addHandler(console)<br /> logger = logging.getLogger('DVDBackup')<br /> return logger<br /> <br />def main():<br /> # This method parses the options set at the command line and initializes the logger.<br /> parser = OptionParser()<br /> parser.add_option("-l", "--logging", type='int', dest="debug", help="LOG level. See http://docs.python.org/library/logging.html. Default: INFO", metavar="LOG", default=logging.INFO)<br /> parser.add_option("-s", "--src", type='string', dest="src", help="Source drive or DIRECTORY. Script will create images from this directory/drive.", metavar="DIRECTORY", default=None)<br /> parser.add_option("-d", "--dest", type='string', dest="dest", help="Create images in destination DIRECTORY. ", metavar="DIRECTORY", default=None)<br /> parser.add_option("-f", "--file", type='string', dest="file", help="Create image files using base FILENAME. Must be an iso, daa, or bin. Example: Test.iso. NOTE: If not specified, script will use the source directory as a base and create iso images.", metavar="FILENAME", default=None)<br /> parser.add_option("-p", "--password", type='string', dest="password", help="Password for the image. This is optional.", metavar="DIRECTORY", default=None)<br /> parser.add_option("-m", "--max", type='float', dest="max", help="Max disc size. Default is a 4.7 gig disc (4.37 max size).", metavar="SIZE(GB)", default=4.37)<br /> parser.add_option("", "--piso", type='string', dest="piso", help="PowerISO command. Usually piso. If the PowerISO folder is not in your environment path, you will need to specify the full COMMAND to piso.exe here. Default: piso", metavar="COMMAND", default=None)<br /> parser.add_option("", "--dry", action="store_true", dest='dry', help="Dry run only.", default=False)<br /> <br /> opts, args = parser.parse_args()<br /> logger = setupLogging(opts.debug)<br /> return opts, args, logger<br /><br />if __name__ == "__main__":<br /> opts, args, logger = main()<br /> if opts.dest is None:<br /> logger.critical('Destination not provided. Please specify a destination via the --dest=DIR command line option')<br /> sys.exit(1)<br /> if opts.src is None:<br /> logger.critical('Source not provided. Please specify a source via the --src=DIR command line option')<br /> sys.exit(1)<br /> <br /> if opts.file is None:<br /> name_base = opts.src.split(sep)[-1].replace(' ','')<br /> img_type = 'iso'<br /> else:<br /> name_base = '.'.join(opts.file.split('.')[:-1])<br /> img_type = opts.file.split('.')[-1]<br /> <br /> images = Dir2Img(opts.src, opts.dest, name_base, img_type, opts.max, opts.password)<br /> if opts.piso:<br /> images.setCommand(opts.piso)<br /> report = images.fullReport()<br /> logger.info('Image Report: \n%s' %report)<br /> <br /> if opts.dry == False:<br /> images.create()<br />
Further Explanation:
OK, so as an example, I want to backup all of my pictures. I keep them (as seen in a previous post) in the structure of YYYY\MM month and when I burn a backup image of all of my pictures. I would prefer to keep as many pictures of the same year together as possible, however, if there is a large file (say a video clip) in one folder that doesn't fit on a dvd that is only 50% full, I would prefer it to be on a new dvd image, and then continue to fill first dvd, and then fill the new dvd image. Confusing I know, it creates dvd's that are usually about 99% full, even with a few large files mixed in.

I also hate searching for which dvd has what folder. The script creates a list of directories contained on each disc and add that file to each image. It makes it very easy for figuring out which disc has what pictures. The script could easily be altered to contain a list of all folders AND files, but I thought that was overkill.