docparse-scraper-status-socket

Setups up an http express app for an single scraper. This app allows external services to get the status of the scraper and start new scrapes

Scraper Status Socket

Allow node.js based scrapers to connect to a status socket so that external services can query the current status of the scraper and check if it is running or stopped.

When a scraper starts up, it registers a new service with seaport using the role scraperStatusSupplierCode. For example the NGrid Electric scraper has the supplierCode NGE so it would register with the role scraperStatusNGE

Then this module will connect to an axon rep socket on the port returned by seaport.

Querying

To query the status of a scraper, get the port of the status socket from seaport and send a status request over http

var inspect = require('eyespect').inspector();
var request = require('request')
var config = require('nconf')
var supplierCode = 'NGE'
var seaConfig = config.get('seaport')
var seaHost = seaConfig.host
var seaPort = seaConfig.port
var ports = seaport.connect(seaHost, seaPort)
var role = 'scraperStatus' + supplierCode // eg "scraperStatusNGE" 
var scraperServices = ports.query(role) // an array of services 
var scraperService = scraperServices[0] // use the first one as an example 
var statusHost = scraperService.host
var statusPort = scraperService.port
 
var url = 'http://' + statusHost + ':' + statusPort + '/status'
var opts = {
  url: url,
  json: true
}
request(opts, function (errresreply) {
  if (err) {
    inspect(err, 'error getting status of scraper')
    return
  }
  inspect(reply, 'scraper status reply')
})

The reply is json and should look like

reply = {
  "supplierCode": "NGE",
  "status": "running",
  "running": true
}