了解Node.js应用程序中的控制流

Question

I am trying to understand control flow in Node.js applications. Specifically does control returns to the original function once callback method completes (like a callback stack in recursive calls). I wrote a simple program that make a GET call and return the data. Here is the program:

Code:

var async = require('async');
var http = require('http');

function getGoogleData(url, callback) {
    http.get(url, function(response) {
        if (response.statusCode == 200) {
            var googleInfo = '';
            response.on('data', function(chunk) {
                console.log("receiving data... ");
                googleInfo += chunk;
                return;
            });
            response.on('end', function() {
                console.log("End of data receive... ");
                response.setEncoding('utf8');
                return callback(null, googleInfo);
            });
        }
        console.log("I am here but why!");
        //callback(new Error("GET called failed status_code=" + response.statusCode));
    });
    console.log("Return from get google data");
}

async.waterfall([
    function(callback) {
        console.log("In func 1");
        getGoogleData("http://www.google.com", callback);
    },
    function(data, callback) {
        console.log("In func 2");
        callback(data);
    }],
    function (err, res) {
        console.log("In err fn");
    });

Here is output of the program: Output:

In func 1
Return from get google data
I am here but why!
receiving data...
receiving data...
End of data receive...
In func 2
In err fn

Can someone help me understand why 'I am here but why!' line gets printed as the second output line in console log even after returning from 'data' event emitter? What is the overall control flow here?

Answer 1

The reason you're seeing that message logged first is that all that the code inside the if block is doing is adding event handlers. Those events are emitted some time in the future, after your console.log has already executed.

It's a similar reason why "Return from get google data" gets printed before the request finishes, because the http request is asynchronous.