palo client for node.js

Note: this document is just a draft, actual coding is still in an early stage.

Node_palo is a database module for node.js that connects to a Palo database server. It's easy to implement in a project with middleware like express to create (single page) webapplications aimed at data analysis, budgetting or reporting.

Features include:

• Act as a proxy to a palo server, allowing access on any http port
• Conversion of CSV server results to JSON objects, with optional column mapping
• Custom authentication and authorization (ACL)
• Create custom methods to build complex datastructures serverside, then send to client

What is Palo?

Jedox Palo is an in memory online analytical processing (MOLAP) database that is available with both an open source and a commercial license. Node_palo is tested with a default compile of the latest open source version of Palo.

Usage

Test from a node console:

var palo = require('palo');
var paloUrl = 'http://localhost:7777/server/login?user=test&password=test';
palo.call(paloUrl, console.log);

Result: { success: true, data: [ [ 'HJvn', '3600' ] ], statusCode: 200, headers: { 'x-palo-sv': '1977569839' } }

TODO: Show usage with node htpp or express

TODO

• Create parameter bindings for the server results
• Options for palo.get
• Find caveats in palo server results. For some server methods the nr of columns returned seem to vary. Parsing of results must be adapted accordingly.

Eventually:

• Create a jQuery plugin (jquery.palo.js) for the client side with convenient ajax method, csv to json conversion and more
• Create another project called palo_explorer which offers a nice web UI for palo server administration; build on express, node_palo, ?
• Redis cache (palo-server has a cache token mechanism)
• Custom authentication and ACL (palo-acl can be cumbersome) to hide palo sid from user (and cache the sid)

Documentation