cloudinary-backup

#!/usr/bin/python3
#=======================================================================================================================
# Program to back up content from a Cloudinary cloud instance
# Requirements:
#     cloudinary (PIP module)
#     CLOUDINARY_URL environment variable is set to "cloudinary://<API_key>:<API_secret>@<cloud_name>
# Synopsis:
#     cloudinary-backup [options]
# Options:
#     -v  Verbose output
#     -d  Delete local files that do not exist in the cloud
# More info: see https://yktoo.com/en/blog/post/2019/12/13-cloudinary-content-backup/
#=======================================================================================================================

import sys
import logging
import os
import pathlib
import urllib.request
import cloudinary


MAX_RESOURCE_RESULTS = 1000
BACKUP_DIR = os.path.join(str(pathlib.Path.home()), 'Archive/Backup/cloudinary')


class CloudinaryBackup:
    """Synchronises the local copy of all resources with the Cloudinary library."""

    def __init__(self, resource_type: str, del_redundant=False) -> None:
        self.resource_type = resource_type
        self.delete_redundant = del_redundant
        self.resources = []
        self.files = {}
        self.root_file_dir = os.path.join(BACKUP_DIR, self.resource_type)

    def query_resources(self):
        """Load the list of resources from the server into self.resources."""
        self.resources = []
        self.files = {}

        # Iterate until there's no next_cursor
        next_cursor = None
        req_num = 0
        while True:
            req_num += 1

            # Fetch the data
            result = cloudinary.api.resources(
                resource_type=self.resource_type,
                max_results=MAX_RESOURCE_RESULTS,
                next_cursor=next_cursor)

            # Accumulate the resource in the list
            self.resources += result['resources']

            # Fetch the cursor for the next set, if any
            if 'next_cursor' not in result:
                break
            next_cursor = result['next_cursor']

        logging.info('Fetched %d %s resources with %d API calls', len(self.resources), self.resource_type, req_num)

    def list_files(self):
        """List all backed up files into self.files."""
        total_size = 0
        for root, dirs, files in os.walk(self.root_file_dir):
            for file in files:
                # The full path to the resource file
                full_path = os.path.join(root, file)

                # Fetch file size
                size = os.path.getsize(full_path)

                # Determine the 'id' of the resource by stripping off the base path
                id = os.path.relpath(full_path, self.root_file_dir)

                # Add the file to the dictionary
                self.files[id] = size
                total_size += size
                logging.debug('Found file %s ⇒ %s (%d bytes)', id, full_path, size)

        logging.info('Found %d files, total size %d bytes', len(self.files), total_size)

    def fetch_missing_files(self):
        """Download files that are missing or have a different size."""
        remote_size = 0
        local_size = 0
        download_size = 0
        download_count = 0
        for resource in self.resources:
            # Fetch the remote resource's properties
            rsrc_id = resource['public_id'] + '.' + resource['format']
            rsrc_size = resource['bytes']
            remote_size += rsrc_size

            # First check for the presence of the local file
            if rsrc_id not in self.files:
                logging.debug('Resource %s (%d bytes) is not found, downloading', rsrc_id, rsrc_size)
                need_download = True

            # Then verify its size
            else:
                file_size = self.files[rsrc_id]
                if rsrc_size != file_size:
                    logging.debug(
                        'Resource %s has a different size (local: %d, remote: %d), downloading',
                        rsrc_id, file_size, rsrc_size)
                    need_download = True

                # Resource is unchanged
                else:
                    need_download = False
                    local_size += file_size

            # Download the file
            if need_download:
                file_path = os.path.join(self.root_file_dir, rsrc_id)

                # Make sure target directory exists
                os.makedirs(os.path.dirname(file_path), exist_ok=True)

                # Download the file into the final location
                urllib.request.urlretrieve(resource['secure_url'], file_path)
                download_size += rsrc_size
                download_count += 1
                logging.info('Downloaded %s (%d bytes)', rsrc_id, rsrc_size)

        logging.info(
            'Finished fetching missing files. Total bytes: remote %d, local %d, downloaded %d (%d files)',
            remote_size, local_size, download_size, download_count)

    def find_redundant_files(self):
        """Find local files that no longer exist on the server."""
        # Make a set of existing resource IDs
        resource_ids = {r['public_id'] + '.' + r['format'] for r in self.resources}

        # Iterate local files
        del_count = 0
        for file_id in self.files.keys():
            if file_id not in resource_ids:
                del_count += 1
                logging.debug('File %s is not present in the cloud, deleting', file_id)
                os.remove(os.path.join(self.root_file_dir, file_id))

        logging.info('Finished searching for redundant files: %d files have been removed', del_count)

    def run(self):
        """Perform the backup."""
        # Load the list of resources
        self.query_resources()

        # List all files backed up so far
        self.list_files()

        # Download missing files
        self.fetch_missing_files()

        # Find redundant files, if needed
        if self.delete_redundant:
            self.find_redundant_files()


# Main routine
if __name__ == '__main__':

    # Parse command line / setup logging
    logging.basicConfig(level=logging.DEBUG if '-v' in sys.argv else logging.INFO, format='%(levelname)-8s %(message)s')
    delete_redundant = '-d' in sys.argv

    # Run the backup
    CloudinaryBackup('image', del_redundant=delete_redundant).run()
    CloudinaryBackup('video', del_redundant=delete_redundant).run()