Node .JS Crawler to JSON output is empty - json

Question

So I am trying my hand at Node.js. I want to build a simple crawler which scans a page and then returns all links back in a json file. However, when I run the script it returns 0 links.
Here is my code in its entirety:
var request = require('request');
var cheerio = require('cheerio');
var fs = require("fs");
var url = 'https://stackoverflow.com/questions';
//Create the blank array to fill:
var obj = {
table: []
};
var i = 0;
request(url, function(err, resp, body){
$ = cheerio.load(body);
links = $('a'); //jquery get all hyperlinks
$(links).each(function(i, link){
var actualLink = $(link).attr('href');
obj.table.push({id: i, url:actualLink}); //add some data
i++;
});
});
var json = JSON.stringify(obj);
console.log(json);
The output in the terminal is so:
$ !!
node nodetest.js
{"table":[]}
Can anyone see why this is blank? Bonus points for writing the final json to a file :)

Answer 1

You must use obj inside the success callback of the request, that's where it gets populated:
request(url, function(err, resp, body) {
$ = cheerio.load(body);
links = $('a'); //jquery get all hyperlinks
$(links).each(function(i, link) {
var actualLink = $(link).attr('href');
obj.table.push({id: i, url:actualLink}); //add some data
});
// Only here you can be sure that the "obj" variable is properly
// populated because that's where the HTTP request completes
var json = JSON.stringify(obj);
console.log(json);
});
In your code you have placed the console.log outside the request success which is asynchronous and thus the obj variable is not yet populated.
Also notice that you don't need the i variable. It will be passed to the each callback automatically, you don't need to be explicitly declaring or incrementing it.
As far as writing the result to a file is concerned, you could use the fs.writeFile function:
fs.writeFile("/tmp/test", json, function(err) {
if(!err) {
console.log("File successfully saved");
}
});