A Script to Download CSV Data from EPA ECHO

download_epa.js

#!/usr/bin/env casperjs
'use strict';

// The sledgehammer approach: Use a JavaScript testing framework to simulate
// web browsing. Slow as all hell, but gets the job done.
//
// Caveat Utilitor: This was hacked together on a musky Portland afternoon
// under the influence of several cans of excellent Oakshire "Watershed" IPA.
//
// Installation
// ============
//
// Ensure script is executable:
//
//     chmod +x download_epa.js
//
// On OS X:
//
//     brew install casperjs --devel
//
// `--devel` is required because of the PhantomJS version Homebrew ships. This
// will probably work on older versions of CasperJS.
//
//
// Usage:
// ======
//
//   ./download_epa.js [some] [permit] [ids] | xargs wget -O output.csv
//
// Released under the terms of WTFPL: http://www.wtfpl.net/about/

// Load up the CasperJS module and instantiate a driver.
var casper = require('casper').create();

// Pass desired station IDs on the command line.
if ( casper.cli.args.length == 0 ) {
  casper.die('You must pass at least one Permit ID to this script!');
}
var permitList = casper.cli.args

casper.start('http://echo.epa.gov/facility_search_results?permit=' + permitList.join(), function() {
  var queryID = this.evaluate(function() {
    return document.getElementById('QueryID').value;
  });

  var trackColumns = this.evaluate(function() {
    return document.getElementById('trackColumns').value;
  });

  // Can't figure out downloading at the moment, so just echo the CSV url to
  // stdout. It can then be piped to something like `xargs wget`.
  this.echo('https://ofmpub.epa.gov/echo/rest_services.get_download?qid=' + queryID + '&qcolumns=' + trackColumns);
});

casper.run();