synchronized test

[Shrinath M Aithal]

source: katherine sierra and bert bates. (self test of chapter 9)

Given:

public static synchronized void main(String[] args) throws
InterruptedException {
Thread t = new Thread();
t.start();
System.out.print("X");
t.wait(10000);
System.out.print("Y");
}

What is the result of this code?
(skipping irrelevant parts of choices given, )

D. It prints XY with a 10-second delay between X and Y.

G. An exception is thrown at runtime.

Answer in book says :
" G is correct. The code does not acquire a lock on t before calling t.wait(), so it throws
an IllegalThreadStateException.
If the wait were placed inside a synchronized(t)
block, then the answer would have been D. "

My problem is :
I actually tried to get D as answer by wrapping t.wait(10000); with synchronized(t){ t.wait(10000); } but didn't get the required result.. how should i've done it?

[avi sinha]

nothing is wrong with the code if you wrap the wait method invokation in a synchronised block.
running fine on my system.

   class a{
	
	public static synchronized void main(String[] args) throws
InterruptedException {
Thread t = new Thread();
t.start();
System.out.print("X");

synchronized(t){

t.wait(10000);
}

System.out.print("Y");
}
}


--------------------Configuration: <Default>--------------------
XY
Process completed.
 

avi sinha

[Shrinath M Aithal]

No dude, I too tried the same code, and its not "waiting" 10 seconds before printing y..
I am getting the output as "xy" fine, but not with 10 second gap..
did you get 10 second gap?

[avi sinha]

Shrinath M Aithal wrote:No dude, I too tried the same code, and its not "waiting" 10 seconds before printing y..
I am getting the output as "xy" fine, but not with 10 second gap..
did you get 10 second gap?

ya i am getting the lag between x & y

[Shrinath M Aithal]

super confusion now
why isn't it waiting in my system ? I tried running it on another system too, both not giving lag between x and y..

and I tried other programs with "wait()" command, no program is waiting on this command.. ???
why isn't wait working with my system?

[avi sinha]

Shrinath M Aithal wrote:super confusion now
why isn't it waiting in my system ? I tried running it on another system too, both not giving lag between x and y..

and I tried other programs with "wait()" command, no program is waiting on this command.. ???
why isn't wait working with my system?

i got the point actually when i executed it again it gave XY instantly.

reason:

wait() method will force the thread to stop executing until

1> the thread on which it has been called becomes dead or specified time

since you haven't overridden run method here the thread t becomed dead as soon as it is started hence there is no lag.

avi sinha

[avi sinha]

   class a extends Thread{
	
	public static synchronized void main(String[] args) throws
InterruptedException {
Thread t = new Thread(new a());
t.start();
System.out.print("X");

synchronized(t){
	
t.wait(10000);
}

System.out.print("Y");
}


public void run(){
	int i=0;
	while(i<10000);
		
}
}
 

this will give you lag for sure

[Shrinath M Aithal]

avi sinha wrote:
since you haven't overridden run method here the thread t becomed dead as soon as it is started hence there is no lag.

so you mean to say, "t" is a dead thread when we are calling t.wait(10000); ???
and you can acquire synchronization on a dead thread?

and one more thing .. just to bug avi sinha ---> how did you get lag the first time? and why???

and the code you posted above, is infinitely waiting even after printing x and y and is inside your run method forever..

[Shrinath M Aithal]

I think i found a proper way..

watch this :

   public class test{   
       
public static synchronized void main(String[] args) throws Exception {   
Thread t=new Thread(new Runnable(){
 public void run(){
  System.out.print("x");
  synchronized(this){
  try{
   wait(10000);
   }catch(Exception e){}
  }
  System.out.print("y");
 }});
t.start();

}
}

 

that does the job exactly as required.. but is that mature way?

[avi sinha]

Shrinath M Aithal wrote:

so you mean to say, "t" is a dead thread when we are calling t.wait(10000); ???
and you can acquire synchronization on a dead thread?

first point to notify you is that you can synchronize a method or a block on objects not a thread.
and even if a thread is dead by isAlive() method it's still a valid object irrespective of its status

and one more thing .. just to bug avi sinha ---> how did you get lag the first time? and why???

since you are reading k & B so i can say you know that a very little is guaranteed in multithreading.

adding to the above statement threads in java are system dependent. os decides the thread everytime to be executed. so you can't be sure that at a particular instant a thread will be executed.

so everytime even if you are not overridding the run method thread t can take different time intervals for completion.

i think this much is sufficient for you.

and the code you posted above, is infinitely waiting even after printing x and y and is inside your run method forever..

i posted the code to show the case when the wait() method will again allow the main thread to go to runnable state before the completion of run method for thread t DEAD STATE

also if the exception is not being thrown there is no any difference between declaring or handling the exception.there will be no any effect of it on code's behaviour.

avi sinha

[avi sinha]

Shrinath M Aithal wrote:I think i found a proper way..

watch this :

   public class test{   
       
public static synchronized void main(String[] args) throws Exception {   
Thread t=new Thread(new Runnable(){
 public void run(){
  System.out.print("x");
  synchronized(this){
  try{
   wait(10000);
   }catch(Exception e){}
  }
  System.out.print("y");
 }});
t.start();

}
}

 

that does the job exactly as required.. but is that mature way?

the original code you posted in you first post & this code is entirely different.

wait method in the first code was being called by the main thread & here it is being called by the thread you ve created. & it is not allowing any other thread to execute because there is no any thread hence it is just waiting for 10 secs

avi sinha

[Shrinath M Aithal]

so, does it suggest that the answer given in book is wrong? because it says

If the wait were placed inside a synchronized(t)
block, then the answer would have been D.

that didn't happen.. so it doesn't work as said?

>> about bugging Mr avi sinha :
I meant to ask you how you got the "10 second " lag the first time I agree with whatever you said about the thread taking different times to complete printing x and y on different executions.. But usually, its instantaneous.. and you getting proper 10 second lag, is that co incidence ?? ;)

lets drop it.. that'll not help with threads lets talk about the priority of this discussion " is the answer specified in k&b about option "D" wrong? "

[Shrinath M Aithal]

avi sinha wrote:
the original code you posted in you first post & this code is entirely different.

wait method in the first code was being called by the main thread & here it is being called by the thread you ve created. & it is not allowing any other thread to execute because there is no any thread hence it is just waiting for 10 secs

avi sinha

ya you are right on that one.. I was on the aim of creating a lag between x & y, so forgot the big picture about what we are talking..

so the code you posted would be right on that case
but please do something to stop the infinite running thread.. add t.stop() or something to main before it ends..

[avi sinha]

Shrinath M Aithal wrote:so, does it suggest that the answer given in book is wrong? because it says

If the wait were placed inside a synchronized(t)
block, then the answer would have been D.

that didn't happen.. so it doesn't work as said?

>> about bugging Mr avi sinha :
I meant to ask you how you got the "10 second " lag the first time I agree with whatever you said about the thread taking different times to complete printing x and y on different executions.. But usually, its instantaneous.. and you getting proper 10 second lag, is that co incidence ?? ;)

lets drop it.. that'll not help with threads lets talk about the priority of this discussion " is the answer specified in k&b about option "D" wrong? "

well i think the book is saying about the possiblity . ya its possible but not always