molab-rhino-api

Allows a web interface in addition to twitter to control the operation of an LED array connected to a Raspberry Pi (or Arduino). This makes use of websockets and the twitter API.

The Met Office Informatics Lab Rhino Project

The main parts are separated into three directories:

• interface - directory that contains the main application and web pages.
• python-server - directory that contains the server to pass information to the LEDs.
• deploy - directory that contains an application for automatic git hub deployment.

To make full use of the twitter functionality the raspberry pi or local computer needs to have an internet connection. If no internet connection see Important! below.

Configuration

Interface

Contains the front and back end.

app.js is the main node application, change the following to match your project:

var display = new Display("urlOfPyhtonServer", LED-count);

• set urlOfPyhtonServer to match the python server.

• LED-count to the number of LEDs.

• Swap between display options depending on the use of arduino or raspberry pi.

const KEYWORD - the keyword used to search twitter.

Note: the Client facing website is made specifically for iPad.

The iPad interface has a time delay upon a user interaction. If a button is selected there will be a time delay before another button can be selected. Edit the following line in send-data.js to change this delay;

timeOut = setTimeout(buttonTimeOut, 5000);

Python Server

Contains a bash script ledserver.sh to run the python server ledserver.py.

Adjust LED_COUNT in the python server to match the number of LEDs used.

In the bash script export FLASK_APP=pathToCall adjust pathToCall to set were the bash script can be successfully called from.

Requires python and the Neopixel library to control the connected LEDs.

Deploy

Contains a node application which will run a list of commands upon receiving a git webhook.

Edit the commands executed in function deploy() to match your project.

Edit these variables used by localtunnel to match your project:

• port
• projectRoot
• subdomain

Important!

This project requires access to twitter.

For the project to work successfully the following environment variables need to be set:

TWITTER_CONSUMER_KEY
TWITTER_CONSUMER_SECRET
TWITTER_ACCESS_TOKEN_KEY
TWITTER_TOKEN_SECRET

If these cannot be set or no internet connection is present, please remove lines 44-74 in app.js. This will remove the twitter functionality.

Building and running

Install NeoPixels onto your Raspberry Pi, this allows you to control the LED array. Follow the link to get up an running

https://learn.adafruit.com/neopixels-on-raspberry-pi/overview

Once that is installed clone the repo

git clone https://github.com/met-office-lab/molab-rhino-api.git

Once downloaded run [sudo] npm install in the deploy and interface directories.

Start the python server and node application by running npm start in the interface directory.

Whilst these are still running start the node application for deployment by running npm start in the deploy directory.

Deploy works using localtunnel which needs installed globally - sudo npm install -g localtunnel.

This now lets the master branch on the raspberry pi update when there are any changes to the master branch on GitHub (still a bit buggy).

Mapping

To map an LED array to display images and video see led-mapper for more details.

Once completed an array of the LED location should be produced. Set const LED_MAPPING in led-mapping.js to this array.

Forever

The deployment node application runs using the forever npm package. This needs installed globally sudo npm install forever -g. Forever monitor also need to be installed sudo npm install -g forever-monitor.

Alternative build

The python server and node server can be run different machines, make sure to edit the urlOfPyhtonServer accordingly.

Use node bin/www to start the node server and sudo ./ledserver.sh to start the python server.

Adding themes

There are two methods to adding themes, they can either be 1. media based themes or 2. programmed themes.

1. Media themes

These are either photos, videos or gifs, add the desired files to interface/public/data

These files then need index in interface/src/themes/data/theme-index.json using the following format:

{"name":"star", "fileName":"starry.js", "type":"programmed"}

• name - the keyword used to activate the theme
• fileName - the name of the video or photo, include the file extension.
• type - either set to video, image, gif, or programmed based on the the theme type.

Note video based themes will not auto play on a mobile device due to OS restrictions. This is bypassed by using gif files for mobile OS.

Media themes can be slow to load onto the LED array if the file size is very large (>15mb). Due to the low resolution of the array compressing files is not an issue.

2. Programmed themes

When adding themes create a new themeName.js file in the custom directory, changing themeName.

Then index this new theme by adding it into the index.js file in the custom directory.

Finally add this new theme to themeServer.js.

Contributing

1. Fork it!
2. Create your feature branch: git checkout -b my-new-feature
3. Commit your changes: git commit -am 'Add some feature'
4. Push to the branch: git push origin my-new-feature
5. Submit a pull request :D

License

GNU General Public License (GPL) v3