docs/prod.md

In JohnCoene/coronavirus: Tracking the Coronavirus

Application

Deploying the application requires a Postgres database.

Setup

Create the required config file to run the crawler. You should only have to do this once.

library(coronavirus)

# create and fill config file
create_config()

Fill in the config file created with the credentials to a Postgres database and optionally a free newsapi.org token, then run the crawler. The config file should look like.

database:
  name: database-name
  host: 123.123.123.12
  user: me
  password: my-password
newsapi:
  key: xxXx6X43X12YXx4Xx0XxXx7y # from newsapi.org

Every time you want to update the data, re-run crawl_coronavirus, it collects fresh data and overwrites everything.

# crawl data
crawl_coronavirus()

Finally launch the app.

# launch the dashboard
run_app()

Deploy

You can deploy on whatever server you like, install R and the Shiny Community server, then install the package from the terminal. Note that This also works on shinyapps.io though you will need to host the database elsewhere. However, docker might be the easiest.

Docker

The easiest was is to use docker, pull the container.

docker pull jcoenep/corona

Then run it as follows to copy the config file to said container.

docker run -v "$(pwd)"/_coronavirus.yml:/_coronavirus.yml -p 3000:80 jcoenep/corona

Then visit localhost:3000.

With R

Deploy on a Shiny Community server.

sudo su - -c "R -e \"install.packages('remotes')\""
sudo su - -c "R -e \"remotes::install_github('JohnCoene/coronavirus')\""

Once this done create a directory under /srv/shiny-server/, where you can create the config file.

cd /srv/shiny-server
mkdir coronavirus
cd ./coronavirus
R -e "coronavirus::create_config()"
vi _coronavirus.yml

Fill in the config file and create the app.

echo "coronavirus::run_app()" > app.R 

You can then visit http://my.server.ip:3838/coronavirus, you can change the port in the /etc/shiny-server/shiny-server.conf file, change it to 80 to have your app at http://my.server.ip/coronavirus.

Note that it's always a good idea to recrawl the data with crawl_coronavirus after reinstalling the package, in the event it requires changes database-side.

That is the dashboard set up and running, go to the data section to see how to create a cronjob to automatically update the data.

Open an issue if you have problems deploying, I'm more than happy helping.

Go to the next section on embeds to host your own embedded charts.

JohnCoene/coronavirus documentation built on July 15, 2020, 4:58 a.m.