Cognibot


Package cognibot provides a simple and concurrent web crawler that follows the robots.txt policies and crawl delays.

Note

By default, cognibot only crawls URLs with the same host names as the seeds.

Cognibot is very much inspired by of fetchbot.

Installation

To install, simply run in a terminal:

go get github.com/cognifly/cognibot

The package has two dependencies, cognilog a simple but effective color logging package and collectlinks a package that collects all URLs from a webpage.

Usage

The following example shows how to create a Fetch, add seeds and start the crawl thread.

Before running cognibot, create a directory called docs.

package main

import (
	"github.com/cognifly/cognibot"
)

func main() {
	fetch := cognibot.New()
	fetch.Seed("http://localhost:2016/")

	fetch.Start(1)
}

Cognibot

Basically, Cognibot has the following functionality:

A Fetch module that uses the http protocal to retrieve the robot.txt or web page at a URL.

A built in Robot exclusion protocal parser that determines whether an extracted link passes the robot restrictions.

A Queue or URL Frontier containing URLs yet to be fetched by the crawl thread or crawl goroutine.

A Visited list containing URLs fetched by the fecth mobule.

A Parser module that extracts links from a fetched webpage and saves the webpage to disk.

A Duplication elimination module that determines whether an extracted link is already in the queue or url frontier

How Cognibot Works

Crawling is performed by the a crawl thread or crawl goroutine. Add the number to Start specifying the number of threads to begin with. If the length of the queue *minus the length of visited is less than 100, cognibot spins up a new crawl thread or crawl goroutine. see example below.

package main

import (
	"github.com/cognifly/cognibot"
)

func main() {
	fetch := cognibot.New()
	fecth.MaxPages = 50
	fetch.Seed("http://localhost:2016/", "http://localhost:2017/", "http://localhost:2018/")

	fetch.Start(3)
}

A crawl thread or crawl goroutine begins by taking a URL from the queue or frontier and fetching the web page at the URL. The URL is then added to the Visited list.

Links are extacted from the fetched page's response body and then the webpage body is saved to disk.

Each extracted link goes through a series of tests to determine whether it should be added to the queue or URL frontier. First the crawl thread or crawl goroutine tests whether the Host of the URL is Found in the *HostInfo []Robot. Secondly it tests whether the URL is already in the the queue or URL frontier and Finaly it tests whether the URL under consideration passes the robot restrictions.

Cognibot Options

Cognibot has a number of fields that provide further customization:

• MaxPages : By default, the max is 100.

• HttpClient : By default, the Fetcher uses the net/http default Client to make requests. A different client can be set on the Fetcher.HttpClient field.

• CrawlDelay : That value is used only if there is no delay specified by the robots.txt of a given host.

• UserAgent : Sets the user agent string to use for the requests and to validate against the robots.txt entries.

License

The [BSD-style license] found in the LICENSE file, the same as cognilog package. The collectlinks package source code is under the [MIT license] (details in the source file).

Name kampsy kampamba chanda Email [email protected] Github https://github.com/kampsy Social google.com/+kampambachanda