nodejs execution comes out of order?

Question

I'm adding node.js to an existing site for its "server side Java Script" capabilities, and have been following through several of the tutorials to get a handle on it. ...Of course, as an experienced programmer learning a new thing, I'm experimenting a little beyond what the tutorials might give. So, instead of replacing the effort with every iteration, I've been adding to what the code does in hopes that when I'm done I have one bit of code that exercises all the materials covered.

In doing this I have uncovered a clear example of out-of-order execution that seems to me to be completely unacceptible, but then maybe I'm missing something, which is why I'm posting (though it's also valuable to others who follow that if I'm right, all node.js programmers should know about it).

Here's a link to the tutorial that led me to this. This is the start of it.

When I got to the file-access section, since we were learning how to create and delete files shortly after learning how to populate variable values from the URL, I decided to use a conditional value to delete a newly created or updated file based on what the user had provided in the URL. This seems entirely rational to me!

But, it doesn't work out correctly; even though the code to delete the file occurs AFTER the code to create the file, in fact the deletion occurs before the creation and so, in the end, the file still exists! Perhaps more correctly, an instance of file creation happens after the potential file deletion, even though the order in the code is otherwise. (Note that this is a testing environment, so it's not as if some other user accessed the system causing a creation without the deletion condition being met.) This is easily repeatable.

Here's the code:

var http = require("http");
var dt = require("./firstmodeule.js");
var url = require("url");
var fs = require("fs");
var fn = "index.html";
var log = "nodejs.log";

http.createServer(function (req, res)
{
    res.writeHead(200, {"Content-Type": "text/html"});
    res.write("<p>New current Date & Time: "+dt.myDateTime()+"<br>");
    res.write(req.url+"<br>"); // This captures the right hand of the URL.
    var q = url.parse(req.url, true).query;
    var txt = q.year + " " + q.month;
    res.write(txt+"</p>");
    res.write("<p></p>");
    fs.appendFile(log, dt.myDateTime()+" "+txt+"\n", function (err)
    {
      if (err) throw err;
      console.log("saved?");
    });
    fs.readFile(fn, function(err, data)
    {
       fs.writeFile("delable.index", data, function(err)
       {
          if (err)
          {
             throw err;
          } else {
             console.log("wrote to delable.index");
          }
       });
       res.write(data);
       res.end();
    });
    if (q.month = "Jan")
    {
       fs.unlink("delable.index", function (err)
       {
          if (err)
          {
             throw err;
          } else {
             console.log("Deleted delable.index");
          }
       });
    }
    //res.end(); // End CANNOT be here if we"re doing a res.write(data);
}).listen(8080);

Of course you can get the tiny bit of code for firstmodule.js from the cited tutorial - or just delete it since it has no bearing on this.

Here's an example of the URL:

http://MySite:8080/?year=2020&month=Jan

Here's an example of the output I get:

$ node example.js
Deleted delable.index
saved?
wrote to delable.index
Deleted delable.index
saved?
wrote to delable.index

Contrary to the order of the code, we get the deleted part first!

OK, so, what gives? Does node.js just do things in whatever order it wants?

For anyone curious, this is all extremely current code as of this date. The OS is Fedora 28 downloaded in the last few days, and the node package is nodejs-8.11.4-1.fc28.x86_64.

Answer 1

Change fs.appendFile to fs.appendFileSync and change fs.readFile to fs.readFileSync and then the functions will be synchronous and hence in the order of execution to match the order of your code.