The Big Programming Thread - Page 799

On November 17 2016 18:32 Manit0u wrote:
Does anyone have any insight on possible RoR developer interview questions?

I'm applying for a new job, trying to move from PHP to RoR but my RoR knowledge is pretty basic (but I have plenty of general experience in web dev).

On November 17 2016 19:09 Khalum wrote:


Am I imagining this or do you seriously apply for new jobs every 2-3 months?

On November 17 2016 19:22 Manit0u wrote:


Last time I've changed a job was 9 months ago. This one isn't moving me forward in any way so why should I stay stagnant?

Besides, there's nothing wrong in checking out various possibilities, even if I don't get the job or won't want it.

On November 17 2016 19:47 Morfildur wrote:
The company I've started working for has previously done all development through service providers. It's now my Job to build up an inhouse developer team.

Trouble is, a lot of the structure is... not fit for software development. Things like the Firewall spoofing all ssl requests and cerificates, which breaks pretty much all package Management tools that verify cerificates when downloading packages. Also, developers don't have local admin, because only admins were previously allowed to have local admin, which means we have to call an admin every time we need to installieren something or change an entry in the hosts file.

I've pushed for a meeting with the admins next week to change the situation, but I could use a few more arguments for a less restrictive or netter completely open Firewall and local admins for developers. Dev environment will be local VMs, PHP and Java for the current Magento and soon Hybris shop and we want to use openvpn to Connect to partners.

On November 17 2016 19:47 Morfildur wrote:
The company I've started working for has previously done all development through service providers. It's now my Job to build up an inhouse developer team.

Trouble is, a lot of the structure is... not fit for software development. Things like the Firewall spoofing all ssl requests and cerificates, which breaks pretty much all package Management tools that verify cerificates when downloading packages. Also, developers don't have local admin, because only admins were previously allowed to have local admin, which means we have to call an admin every time we need to installieren something or change an entry in the hosts file.

I've pushed for a meeting with the admins next week to change the situation, but I could use a few more arguments for a less restrictive or netter completely open Firewall and local admins for developers. Dev environment will be local VMs, PHP and Java for the current Magento and soon Hybris shop and we want to use openvpn to Connect to partners.

On November 17 2016 19:25 Khalum wrote:


Wasn't meant as criticism, I just had the impression that I read about you applying for some new job every 10 pages here. No idea how that crossed my mind tbh.

function mystery(input) {
	var secret = 5;
  
  function mystery2(modifier) {
  	modifier *= input;
    
    return secret * modifier;
  }
  
  return mystery2;
}

var hidden = mystery(4);
var result = hidden(2);

// What is the result?

On November 18 2016 23:21 Manit0u wrote:


Well, I've had the interview today and got the job starting next month (they've decided to hire me on the spot during the interview). No more PHP for me as I enter RoR development in an inernational company

I really enjoyed the JS question:

function mystery(input) {
	var secret = 5;
  
  function mystery2(modifier) {
  	modifier *= input;
    
    return secret * modifier;
  }
  
  return mystery2;
}

var hidden = mystery(4);
var result = hidden(2);

// What is the result?

On November 19 2016 00:52 Acrofales wrote:
Why did you think that question was particularly interesting?

On November 19 2016 05:56 travis wrote:
We started talking about threading in my class now. The instructor ran an a couple amusing programs. The first created 100 threads, and each thread ran a loop that displayed it's thread name plus "i", with i being a variable incremented from 0 to 4 in the loop. And then "all done" when all the threads had been made. The amount of time it took each thread to go through its loop differed, and the "all done" came out at varying places in the output. I guess this was to show that threads are competing for processor resources and that which thread has priority is often outside the scope of your control?


The second program created the same 100 threads but this time each thread shared a static int x, which started at 0. Each thread ran a loop a million times, and in the loop it did x++. When everything was said and done x was much lower than 100 million, and each time it varied. This is because the threads were interrupting each other during the x++ operation. While one was about to assign x to the new value (say... x+1 = 6), another may have just said that x will become (x+1 = 6) as well. Then the first thread assigns x = 6, and then the next says x = 6. and really the static x only got incremented once instead of twice. Or maybe instead of 3 our 4 or 10 times even. I thought this was pretty cool!

So what controls threads then anyways? Is it solely the operating system? I mean obviously it depends on your processor but is all the controlling done by the operating system?

On November 19 2016 05:56 travis wrote:
The second program created the same 100 threads but this time each thread shared a static int x, which started at 0. Each thread ran a loop a million times, and in the loop it did x++. When everything was said and done x was much lower than 100 million, and each time it varied. This is because the threads were interrupting each other during the x++ operation. While one was about to assign x to the new value (say... x+1 = 6), another may have just said that x will become (x+1 = 6) as well. Then the first thread assigns x = 6, and then the next says x = 6. and really the static x only got incremented once instead of twice. Or maybe instead of 3 our 4 or 10 times even. I thought this was pretty cool!

On November 19 2016 05:56 travis wrote:
So what controls threads then anyways? Is it solely the operating system? I mean obviously it depends on your processor but is all the controlling done by the operating system?

On November 20 2016 03:29 travis wrote:
I have an arraylist of vertex objects called "vertices"

My breadth first search method is passed a parameter "start", which is where I start my breadth first search from. The "start" parameter is a data variable that is a member of each vertex object.

For my breadth first search I am using a queue, and adding neighbors to the queue as I visit each vertex.

Is there an easy way to find the vertex in my arraylist that has the data that matches the parameter? Do I seriously have to do a for each loop just to find the vertex I am starting at? I guess any equivalent method in arraylist would just be doing the same thing anyways, though..

On November 20 2016 03:39 RoomOfMush wrote:

An ArrayList is not exactly a good data structure to look up values from. You will need that loop over the entire list.
Is that software design forced upon you or did you come up with that?

On November 20 2016 03:29 travis wrote:
I have an arraylist of vertex objects called "vertices"

My breadth first search method is passed a parameter "start", which is where I start my breadth first search from. The "start" parameter is a data variable that is a member of each vertex object.

For my breadth first search I am using a queue, and adding neighbors to the queue as I visit each vertex.

Is there an easy way to find the vertex in my arraylist that has the data that matches the parameter? Do I seriously have to do a for each loop just to find the vertex I am starting at? I guess any equivalent method in arraylist would just be doing the same thing anyways, though..

public class Graph{
   
   //use adjacency list here
 
   public Graph(int V);
   public void addEdge(int v, int w);
   public Iterable<Integer> adj(int v);
}

//s : the start vertex
public void bfs(int s, Graph g) {
   Queue<Integer> q = new Queue<Integer>
   q.enqueue(s);
   while (!q.isEmpty()){
      int v = q.dequeue()
      foreach(Integer w in g.adj(v)){
          //rest of your algorithm using some kind of visted structure
      }
   }
}

On November 20 2016 12:17 Nesserev wrote:

Please, use the code markup tags to format your code :/
This should become a rule, lol. It happens way too much.

[ code]...insert code here[ /code]
(remove the spaces in front of code and /code)


ex.

function meow() {
    console.log("meow");
}

On November 14 2016 09:30 Nesserev wrote:
I really hate that sentence... "Primitive types are passed by value while complex types are passed by reference." It's a lazy rule that sadly is accepted as truth by so many, because it seems to fit at first sight, but like the new POTUS said, WRONG.

Java ALWAYS uses pass by value.

Variables can only hold 2 types of values in Java:
- primitive values (numbers and bools)
- object references

Object references can hold the location of an object (objects are always allocated on the heap), and can be used to access the object that they're referring to. So, when you pass arguments to a function, you either pass a copy of a primitive value, or a copy of an object reference.

This has 3 important consequences, explained with a simple call: swap(a,b)
- if a and b hold a primitive type, the method cannot modify a and b
- if a and b hold an object reference, a method can change the state of a and b
- if a and b hold an object reference, a method cannot make a and b refer to a new object

public static void swap(X a, X b) 
{
    X t = a;
    a = b;
    b = t;
}

public static void main(String[] args) {
    X a = ...;
    X b = ...;
    swap(a,b) // does nothing
}

If Java really used pass by reference for complex types, the third consequence would not be true.
For example, C++ has real 'pass by reference':

void swap(int& a, int& b)
{
    int t = b;
    b = a;
    a = t;
}


int main() 
{
    int a = 1;
    int b = 2;
    swap(a,b); // swaps the values of a and b
    std::cout << a << ", " << b << std::endl; // prints "2, 1"
    return 0;
}

TL;DR: Don't say that Java passes primitive types by value and complex types by reference.

EDIT: fixed some typos in Java code