Differences to the official spiderable package

The official spiderable package uses phantomjs and the stdout to browse the page and get the missing body. The used technologies create many problems on different environments and bugfixing is really complex. Spiderable2 uses zombie as the browser of choice which is better integrated with node and the package is reworked in general.

We are discussing some of the problems which are existing with the official spiderable package at: iron-meteor/iron-router#1192

Installation

meteor add lufrai:spiderable2

Settings

You can add the following settings in your Meteor Settings Config json File:

{
	'spiderable2': {
		'timeout': 1000,
		'port': 3000,
		'verbose': false
	}
}

Dependencies

make apt-get install build-essential

Notes

This is an early prototype so please be careful and try it out before you deploy it in production!

Known Problems

There seems to be a problem in regards to meteor-pages, if you use that package spiderable will maybe not work correctly
Style tags which were added dynamically onTimeout will eventually not being correctly added to the output! I had a problem where I dynamically added a style display: block to a div Element and the style was missing in the spiderable2 output
Sometimes the performance is slow

Planned Features

verbose logging
caching
precaching (reocurring process caches urls by timeplan)

Contribution

I always love contribution so don't be afraid to create pull requests!

Forum Thread

Some weeks ago I created a forum topic to discuss missing features of the spiderable package. If you wanna give feedback for this package you obviously can add an answer to the thread: https://forums.meteor.com/t/what-do-you-miss-in-your-spiderable-package/1971?u=katy_wings

andrefgneves / spiderable2