In coatless/dropcli: Interface with Linux Dropbox CLI

Overview

This document contains a brief overview of the features of dropcli and how to use these features. First, we discuss software requirements. Secondly, we address the need to install the software. Third, we address how to use it. Fourth, we provide a teaching example.

Package Requirements

In order to use dropcli, you must have:

1. Linux based operating system
2. R
3. Dropbox account

Unfortunately, the linux OS requirement is not negotiable. Principally, the CLI provided by Dropbox's dropbox.py interface is only available on Linux. For details, please see No support for CLI exists on OS X or Windows.

If you need a package for multiple platforms, please see the rdrop2 by Karthik Ram. This package uses Dropbox's WebAPI via OAuth instead of an operating system specific install. The downside to this approach is you must initiate the push instead of having a background process take care of it.

Installation

To install Dropbox's dropbox.py interface, there are two options:

1. Try using dropcli's dropcli::install_dropbox_cli()
2. Follow the guide given at: Dropbox Headless Install.

Option 1: Install via R

This process is somewhat automated within R via:

dropcli::install_dropbox_cli()

You will need to copy and paste the URL in your browser to link the dropbox account with your dropbox.

Then, skip to the Turn on autostart subsection. This immediately follows after Option 2.

Option 2: Install via Terminal

To install in terminal, we'll need to perform a headless or text-only install.

Dropbox provides a guide for it at: Dropbox Headless Install. However, you may need to resort to using Google's cache to access an up-to-date copy of the instructions as Dropbox attempt to automatically detect and serve the appropriate client to you based on your current operating system.

With that being said, the setup required is as follows:

For 32-bit:

cd ~ && wget -O - "https://www.dropbox.com/download?plat=lnx.x86" | tar xzf -

For 64-bit:

cd ~ && wget -O - "https://www.dropbox.com/download?plat=lnx.x86_64" | tar xzf -

Then, we launch from shell the Dropbox daemon out of the newly created .dropbox-dist folder.

~/.dropbox-dist/dropboxd

Here, you will be prompted to follow a URL. Please do so! Copy and paste it in your browser.

Next, we'll download the dropbox.py interface via:

wget -O dropbox.py https://www.dropbox.com/download?dl=packages/dropbox.py

And voila! We're done with the install.

Turn on autostart!

After the package has been installed, please turn on the autostart option. This makes sure that once the system user is logged in then the dropbox daemon is automatically running. In turn, this allows for the use of the CLI within R as the state of the system is destroyed after each call.

To turn on autostart use either:

1. R via dropcli::dropbox_autostart()
2. Shell via ~/dropbox.py autostart y

Playing with Commands

There are a few helpful commands

| Command | Description | dropcli Functions | |------------|--------------------------------------------------------------|------------------------| | status | get current status of the dropboxd | dropbox_status() | | throttle | set bandwidth limits for Dropbox | MIA | | help | provide help |dropbox_help(topic) | | puburl | get public url of a file in your dropbox's public folder |dropbox_puburl(fp) | | stop | stop dropboxd |dropbox_stop() | | running | return whether dropbox is running |dropbox_running() | | start | start dropboxd |dropbox_start() | | filestatus | get current sync status of one or more files |dropbox_filestatus() | | ls | list directory contents with current sync status | MIA | | autostart | automatically start dropbox at login |dropbox_autostart() | | exclude | ignores/excludes a directory from syncing | MIA | | lansync | enables or disables LAN sync | MIA | | sharelink | get a shared link for a file in your dropbox | dropbox_sharelink(fp)| | proxy | set proxy settings for Dropbox | MIA |

Teaching Example

The seemingly rapid development of this interface was inspired by Michael Levy's UseR2016 Talk (See the part at 7:20) that details live code piping from the instructor's computer.

This alleviated my concerns for students not being able to:

1. See the code being typed during STAT 385 @ UIUC;
2. Type code sufficiently fast;
3. Avoid the sitting five feet from me while checking Pokemon Go

The idea presented was very straightforward: Provide students with a direct link to the code file that updates upon every save done by the file editor.

To do that, there are a few things we need:

1. A Linux machine (No support for CLI exists on OS X or Windows)
2. RStudio Server (Pro preferred)
3. An obscure Dropbox knowledge base article entitled How do I force a file or folder to download or render on dropbox.com?
4. dropcli

Alternatively, you can use a local Dropbox install and manually go through the process of enabling a Public document shared URL. Though, I found this to be the easier approach as I run the STATS@UIUC Online Analytical Environment.

With that being said, to generate such a link either:

1. pick a file in the Public directory and use:

dropcli::dropbox_puburl_live("path/to/Public/file.R")`

1. pick a file in the not in the Public directory and use:

dropcli::dropbox_sharelink_live("path/to/file.R")`

Thus, you now have a shareable link. It doesn't look pretty, but it is shareable.

At a later time, support may be added for using a URL shortener within R.

coatless/dropcli documentation built on May 13, 2019, 8:46 p.m.