Created
May 18, 2015 21:18
-
-
Save G-Ray/b31d5f5950dac5e96613 to your computer and use it in GitHub Desktop.
cpasbien scraper
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
socket.on('msg', function(msg){ | |
if(msg['cpb_request']) { | |
var url = msg['cpb_request']; //url de cpasbien avec recherche sur les films en x264 | |
console.log(url); | |
request(url, function(error, response, html) { | |
var cheerio = require('cheerio'); | |
var $ = cheerio.load(html); | |
var json = {"MoviesList": []} | |
var MovieList = []; | |
var order = []; | |
var movies = []; | |
var i = 0; | |
var j =0; | |
$('.ligne0 a, .ligne1 a').each(function(el, link) { | |
var title = $(link).text(); | |
order.push(title); | |
j++; | |
request($(link).attr('href'), function(error, response, html) { | |
var movie = {"imdb":"", "title":"", "synopsis":"", "torrent_url":"", "poster":"", "description":""}; | |
var html = cheerio.load(html); | |
var poster = html('#bigcover img').attr('src'); | |
var torrent = html('#telecharger').attr('href'); | |
movie.imdb = torrent.substring(torrent.lastIndexOf("/") + 1, torrent.length); | |
movie.imdb = movie.imdb.substring(0, movie.imdb.lastIndexOf(".")); | |
var synopsis = html('#textefiche p').eq(1).text(); | |
movie.synopsis = synopsis; | |
movie.description = synopsis; | |
movie.poster = poster; | |
movie.title = title; | |
movie.torrent_url = 'http://cpasbien.pe' + torrent; | |
movies.push(movie); | |
i++ | |
//Replace movies in the correct order | |
if(i === j) { | |
for(var k=0; k<order.length; k++) { | |
for(var l=movies.length-1; l>=0; l--) { | |
if(movies[l].title === order[k]) { | |
MovieList.push(movies[l]); | |
movies.splice(l, 1); | |
} | |
} | |
} | |
var data = {"MovieList":MovieList}; | |
socket.emit('cpb_request', data); | |
} | |
}); | |
}) | |
}); | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
G-ray je comprends pas où est déclarer l'url de recherche cpasbien "cpb_request" ??
Tu l'as déclaré dans un autre fichier ?? Je débute en même temps avec node et le scrapping.
Thx