Geospatial applications created using java script(and nosql)

© COPYRIGHT 2016 MARKLOGIC CORPORATION. ALL RIGHTS RESERVED.
Geospatial Applications created using
JavaScript (and NoSQL)
Tamas Piros (@tpiros | me.tamas.io)

SLIDE: 2 © COPYRIGHT 2016 MARKLOGIC CORPORATION. ALL RIGHTS RESERVED.
Agenda
▪ What is full stack JavaScript and why should you care?
▪ What is geospatial data?
▪ How is this related to NoSQL and databases?
▪ JSON data
▪ Server Side JavaScript
▪ Node.js
▪ Demo

JavaScript is eating the world

And if you need more convincing

A bit more on JavaScript
▪ JavaScript has been around, it is here now and will be here for a while (whether
you like it / use it or not)
▪ Ruling the browser since ‘95

<html>
<body>
<script src="jquery.js"></script>
<script>
$.ajax({ url: "/items" }).done(function(html) {
$("#results").append(html);
});
</script>
</body>
</html>
AJAX & jQuery era
● Asynchronous requests to websites
● Page updates & data loads, without the need to reload the site
● First real step towards interactive websites

The evolution of JS frameworks / libraries
▪ And a ‘few’ frameworks and libraries followed
▪ AngularJS
▪ Ember
▪ Knockout
▪ React
▪ Backbone
▪ Dojo
▪ ExtJS
▪ D3.js
▪ Velocity
▪ etc etc etc

var http = require('http');
var server = http.createServer(function(request, response) {
response.writeHead(200, {
'Content-type': 'text/plain'
});
response.write('hello node');
response.end();
});
server.listen(8080);
JavaScript at the server-side
● Ubiquity in the browser had trickled down the stack
● You can run JavaScript at the server-side
● It’s fast - asynchronous & non-blocking

The future of JavaScript
▪ JavaScript has a prosperous future ES6/2015 and ES7/2016
▪ classes
▪ arrow functions
▪ string templates
▪ promises
▪ object destructuring
▪ array includes
▪ async functions
▪ rest parameters

Geospatial data
- data represents geographical information which can include site locations (as
well as vectors and raster data)
- (site) locations are made up of (geospatial) points
- (geospatial) points can be placed on a coordinate system and they consist of a
pair of latitude and longitude values

Geospatial data
Munich: (+) 48.1351° N, (+) 11.5820° E
Sydney: (-) 33.8600° S, (+) 151.2094° E

Exif data
Exchangeable image file
format (standard)
- metadata about
images (and sounds)
- make, model,
orientation, exposure
time, creation date,
focal length, …
- with location services
on: GPS data

Application Architecture
User Interface
● Data views
● User workflow
● Browser
Middle-tier
● Business rules
● Application logic
JSON over HTTP
Pros
● Same language throughout the stack
● Lightweight data format
● Data format ‘natively’ understood by
JavaScript
Con(s)
● Missing persistent data storage

User Interface
● Data views
● User workflow
● Browser
Middle-tier
● Business rules
JSON over HTTP
Wouldn’t it be nice to add a
database to this architecture that
can:
● store JSON documents natively (along
with XML, binary and RDF)?
● allow you to construct queries using
JavaScript?
● have ACID properties instead of
eventual consistency?
● Give you all the indexes you need and
allow you to execute search out of the
box?
● Apply role based, document level
security?
● Execute SPARQL queries?
● Manage the database via REST API
calls?

User Interface
● Data views
● User workflow
● Browser
Middle-tier
● Business rules
JSON over HTTP
MarkLogic can:
● store JSON documents natively (along
with XML, binary and RDF)
● allow you to construct queries using
JavaScript
● have ACID properties instead of
eventual consistency
● Give you all the indexes you need and
allow you to execute search out of the
box
● Apply role based, document level
security
● Execute SPARQL queries
● Manage the database via REST API
calls
JSON/XML over HTTP
Database-tier
● Persistent storage

- support for WGS84 coordinate system as well as raw coordinate system
- support for geospatial queries, indexes and shapes
- points, (complex) polygons, circles, boxes
- support for Well-Known Text (WKT) and Well-Known Binary (WKB)
representation of geospatial data
- point, linestring, triangle, multipoint, multilinestring, multipolygon,
geometrycollection
Geospatial data in MarkLogic

var latitude = 10.3910;
var longitude = -75.4794;
var miles = function(distance) { return distance * 1.60934 };
cts.search(cts.jsonPropertyChildGeospatialQuery('location',
'coordinates',
cts.circle(miles(5), cts.point(latitude, longitude))
));
Server-side JavaScript in MarkLogic
● Runs on Google’s V8 engine (JavaScript compiler)
● Allows you to execute JavaScript code close to your data (“stored procedures”)
○ both native JavaScript (including some ES2015) and proprietary JavaScript

var marklogic = require('marklogic');
var db = marklogic.createDatabaseClient
(connection);
var qb = marklogic.queryBuilder;
db.documents.query(
qb.where(
qb.geospatial(
qb.geoProperty(
qb.property('location'),
qb.property('coordinates')),
qb.circle(10, 10.3910, -75.4794)
)
)
).result().then(function(response) {
console.log(response);
});
Node.js Client API
● Registered npm package
● Focus on application
features rather than
plumbing:
○ read/write (bulk), patch
(updates), queries,
projections, extensions,
alerting, semantics …
● Supports all geospatial
query types

Resources
- Get the database for free!
- GeoPhoto (GitHub)
- Samplestack (GitHub)
- Character Search v1 (GitHub)
- Character Search v2 (GitHub)
- MarkLogic Java API (GitHub)
- MarkLogic Node.js API (GitHub)
- How is MarkLogic different from MongoDB? (Article)
- Free Training
- More on MarkLogic’s Node.js API (free training)

Danke schön!

Geospatial applications created using java script(and nosql)

More Related Content

What's hot (20)

Viewers also liked (15)

Similar to Geospatial applications created using java script(and nosql) (20)

More from Comsysto Reply GmbH (12)

Recently uploaded (20)

Geospatial applications created using java script(and nosql)