SnapSearch-Client-Node-Heroku-Demo

This repository shows a simple Express application deployed with SnapSearch-Client-Node as an intercetor middleware.

Heroku like other PAAS systems are behind a reverse proxy, which means certain flags needs to be turned true for SnapSearch to work. These flags allow SnapSearch-Client-Node to acquire the original hostnames, ports and protocol that was used between the client and the reverse proxy.

See the index.js for the example source code.

The below shell snippet shows how to deploy this on your own Heroku. Note the usage of environment variables to pass the SnapSearch API credentials.

# clone the project
git clone https://github.com/SnapSearch/SnapSearch-Client-Node-Heroku-Demo.git

# get into the project's directory
cd SnapSearch-Client-Node-Heroku-Demo

# optional, run it locally
SNAPEMAIL=XXXX SNAPKEY=XXXX node index.js

# get heroku CLI and login
heroku login

# create a unique snapsearch demo application, where X is your unique random string
heroku create snapsearch-demo-X

# set your SnapSearch API Email and Key (you need to sign up on heroku) for this app
heroku config:set SNAPEMAIL=XXXX SNAPKEY=XXXX

# deploy on heroku 
git push heroku master

# visit the app with a non-robot-like useragent, you will not see the fully evaluated JS content
curl -A "Meh" https://snapsearch-demo-X.herokuapp.com/

# visit the app with a robot-like useragent, triggering the scrape, you'll see the fully evaluated JS content
curl -A "Googlebot" https://snapsearch-demo-X.herokuapp.com/

# compare the difference between the output!

# check the logs
heroku logs

Special Features of this Demo:

• Shows how to make the compression middleware work with SnapSearch!
• Shows how to deal with non-html resources such as sitemaps!
• Shows how to make it work in Heroku!