Created
April 25, 2012 21:35
-
-
Save Fabryz/2493656 to your computer and use it in GitHub Desktop.
Save all images contained in a website domain/paths
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
{ | |
"name": "rakeup", | |
"version": "0.0.1", | |
"author": "Fabrizio Codello", | |
"engines": { | |
"node": "0.6.x" | |
}, | |
"private": true, | |
"dependencies": { | |
"http-agent": "0.1.x", | |
"jsdom": "0.2.x" | |
} | |
} |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
var httpAgent = require('http-agent'), | |
jsdom = require('jsdom'), | |
http = require('http'), | |
url = require('url'), | |
fs = require('fs'), | |
exec = require('child_process').exec, | |
child; | |
// check if directory exists | |
function dirExistsSync(dir) { | |
try { | |
return fs.statSync(dir).isDirectory(); | |
} catch (err) { | |
return false | |
} | |
} | |
// retrieve a resource and save it | |
function saveImage(options, name) { | |
http.get(options, function(res) { | |
var imageData = ''; | |
res.setEncoding('binary'); | |
res.on('data', function(chunk){ | |
imageData += chunk; | |
}) | |
res.on('end', function() { | |
fs.writeFile(savePath + name, imageData, 'binary', function(err) { | |
if (err) throw err; | |
}); | |
}); | |
}).on('error', function(e) { | |
console.log("Got error: " + e.message); | |
}); | |
} | |
var domain = 'boards.4chan.org', | |
domainPaths = ['v', 'sci'], | |
savePath = './images/', | |
count = 0; | |
var agent = httpAgent.create(domain, domainPaths); | |
agent.addListener('next', function (err, agent) { | |
jsdom.env({ | |
html: agent.body, | |
scripts: [ | |
'http://code.jquery.com/jquery-1.7.2.min.js' | |
] | |
}, function (err, window) { | |
var $ = window.jQuery; | |
// grab all image URLs and save them on savePath | |
$('img').each(function(index) { | |
var imgSrc = $(this).attr('src'); | |
// add http: as by default the path is //path | |
var imgPath = (~imgSrc.indexOf('http:') ? imgSrc : 'http:'+ imgSrc); | |
var fullURL = url.parse(imgPath); | |
var options = { | |
host: fullURL.host, | |
port: 80, | |
path: fullURL.path | |
}; | |
var paths = fullURL.path.split('/'), | |
fileName = paths[paths.length - 1]; | |
saveImage(options, fileName); | |
count++; | |
}); | |
agent.next(); | |
}); | |
}); | |
agent.addListener('stop', function(agent) { | |
console.log('* '+ count +' images saved.'); | |
console.log('* Done.'); | |
}); | |
// create savePath if does not exists | |
if (!dirExistsSync(savePath)) { | |
fs.mkdir(savePath, function(err) { | |
if (err) throw err; | |
}); | |
} else { | |
// if exists remove all directory contents | |
child = exec('rm -rf '+ savePath +'*'); | |
} | |
console.log('* Grabbing all images from '+ domain +'.'); | |
console.log('* Paths: /'+ domainPaths.join(' /') +'.'); | |
console.log('* Download dir: '+ savePath); | |
agent.start(); |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment