Last active
August 29, 2015 14:06
-
-
Save hakatashi/7584c64e643a48084587 to your computer and use it in GitHub Desktop.
Twitpic download script. Create 'twitpic' directory and make a list of ids as 'twitpics.json' like "["/6jbgg1","/5ryufc","/5hpuoq"]". Copyleft.
This file contains hidden or 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 cheerio = require('cheerio'); | |
var request = require('request'); | |
var async = require('async'); | |
var fs = require('fs'); | |
var querystring = require('querystring'); | |
var url = require('url'); | |
var pics = JSON.parse(fs.readFileSync('twitpics.json')); | |
var skipped = []; | |
var twitpic = function (method, resource, params, callback) { | |
var paramString = querystring.stringify(params); | |
request({ | |
url: 'http://api.twitpic.com/2/' + resource + '.json' + '?' + paramString, | |
json: true, | |
method: method, | |
}, callback); | |
}; | |
async.eachLimit(pics, 3, function (pic, done) { | |
var id = pic.slice(1); | |
var viewPage = 'http://twitpic.com' + pic; | |
var fullPage = viewPage + '/full'; | |
async.parallel([ | |
function (done) { | |
twitpic('GET', 'media/show', { | |
id: id | |
}, function (error, response, data) { | |
if (error) return done(error); | |
console.log(id + ': info downloaded, ratelimit remains ' + response.headers['x-ratelimit-remaining']); | |
// validate json file | |
if (typeof data === 'undefined' || !data.id || !data.user || !data.timestamp) { | |
return done(new Error(id + ': Invalid JSON')); | |
} | |
fs.writeFile('twitpic/' + id + '.json', JSON.stringify(data), done); | |
}); | |
}, | |
function (done) { | |
request(fullPage, function (error, response, body) { | |
if (error) return done(error); | |
try { | |
var $ = cheerio.load(body); | |
var imageURL = $('#media-full > img').attr('src'); | |
var resolvedURL = url.resolve(fullPage, imageURL); | |
var parsedURL = url.parse(resolvedURL); | |
var extension = parsedURL.pathname.match('\\.[^.]+$')[0]; | |
var ws = fs.createWriteStream('twitpic/' + id + extension); | |
ws.on('error', done); | |
ws.on('finish', function () { | |
console.log(id + ': image downloaded'); | |
return done(); | |
}); | |
request(imageURL).pipe(ws); | |
} catch (error) { | |
return done(error); | |
} | |
}); | |
} | |
], function (error) { | |
if (error) { | |
console.error(id + ': ' + error); | |
skipped.push(pic); | |
} | |
return done(); | |
}); | |
}, function (error) { | |
if (error) { | |
console.error(error); | |
} else { | |
console.log('Successful. Skipped: ' + JSON.stringify(skipped)); | |
} | |
}); |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment