Dropbox Backup

A simple python script that downloads all your Dropbox files to a directory of your choice. Typically used to backup your files to an external HDD. Asyncio used to speed up download times.

Usage

./run.sh

Output can be viewed in the dropbox_backup.log file. The runtime for this can be quite long, so it is recommended to run in the background (with &), or to automate this via crontab or task scheduler to run in the background intermittently.

Credentials

cp .dropbox-backup.env.example .dropbox-backup.env

Obtain credentials using the instructions in Obtaining a Dropbox Refresh Token, and supply them in the corresponding fields in .dropbox-backup.env

Destination directory can be specified to choose which directory you want your backup to be stored in.

Configuration

Since Dropbox API is rate limited, and rate limits are not readily published, we have to enforce rate limits ourselves, otherwise we will be blocked from making further requests. We are using asycio since this is a heavily network and file IO bound script.

The 2 mechanisms we use for RateLimiting are Semaphores and request delays. You can set these in .dropbox-backup.env as MAX_CONCURRENT_REQUESTS for the number of semaphores, and REQUEST_DELAY for the delay between requests. I found that a value of 50 for semaphores and 0.1 for request delay works well with the script. You will hit rate limits, but not enough to put you in too long of a timeout. These values mean that 50 requests can be ongoing at the same time, and requests will manually wait for 0.1s between network calls.

From experience, with good internet connection, this means the script will take 3hrs for roughly 500GB. Unfortunately, this appears to be as fast as we can expect with this API.

Alternatives include CyberDuck and MountDuck, which you can experiment with to see if you prefer. I enjoy the ease of automation this script provides and the fact that it is free and open source.

Obtaining a Dropbox Refresh Token

To automate your Dropbox API interactions and avoid manual authentication each time, follow these steps to obtain a refresh token:

1. Create a Dropbox App:

   • Go to the Dropbox App Console.
   • Click on "Create App".
   • Select "Scoped access" and choose the appropriate permissions (e.g., files and folders access).
   • After filling out the required details, click "Create App".

2. Authorize Your App:

   • Construct a URL to authorize your app, replacing CLIENT_ID with your app's client ID:

   https://www.dropbox.com/oauth2/authorize?client_id=CLIENT_ID&response_type=code&token_access_type=offline
   

   • Open this URL in a web browser. This will prompt you to log in to Dropbox and authorize the app.
   • After authorization, you will be redirected to the redirect URI you specified in your app settings, along with an authorization code in the URL.

3. Exchange Authorization Code for Access and Refresh Tokens:

   • Use the following curl command to exchange the authorization code for an access token and refresh token:

   curl -X POST https://api.dropboxapi.com/oauth2/token \
       --header "Content-Type: application/x-www-form-urlencoded" \
       --data "code=AUTHORIZATION_CODE&grant_type=authorization_code&client_id=CLIENT_ID&client_secret=CLIENT_SECRET"

   • Replace AUTHORIZATION_CODE with the code you received in the redirect.
   • Replace CLIENT_ID and CLIENT_SECRET with your app's credentials.