|
Thread Rules
1. This is not a "do my homework for me" thread. If you have specific questions, ask, but don't post an assignment or homework problem and expect an exact solution.
2. No recruiting for your cockamamie projects (you won't replace facebook with 3 dudes you found on the internet and $20)
3. If you can't articulate why a language is bad, don't start slinging shit about it. Just remember that nothing is worse than making CSS IE6 compatible.
4. Use [code] tags to format code blocks. |
On December 01 2013 05:10 Warri wrote:Show nested quote +On December 01 2013 05:00 Manit0u wrote:On December 01 2013 04:42 Warri wrote:Woa java is driving me nuts today. I have this endless loop:
boolean pause=true;
while (true) {
if (!pause) {
update();
}
}
At the start of the program pause is set to true and is not changed in the class which contains the loop. It is only changed in another class, which contains a gui. And heres the weird part: pause gets set to false correctly, but the loop still doesnt run. However, if i add a System.out.println("foo"); inside the loop, it works. So my assumption is, java falsely "optimizes", aka removes, the loop, because the condition is never changed to true inside the class. How can i prevent that without adding dummy commands? Can you make it more like this? Class A
while (true) {
if (!ClassB->queryPause(pause)) {
update();
}
}
Class B
boolean pause = true;
... do stuff with pause ...
public boolean queryPause(boolean pause)
{
return pause;
}
I think it would be even better if you had 3 classes. 2 that actually do stuff and have all the data and another one acting as an interface for the other two to communicate. This way you can easily shuffle stuff around and change things later. Hm i dont understand what the parameter is for in queryPause, if pause is a member of Class B? I tried this: Class A:
while (true) {
if (!B.getPause()) {
update();
}
}
Class B:
boolean pause=true;
...
public boolean getPause() {
return pause;
}
This also doesnt work. But again, if i change the getPause() method to
public boolean getPause() {
System.out.println("qwe");
return pause;
}
it starts working... i dont get it 
Don't you have to pass the pause variable to getPause method first? I'm a bit confused here and I have no idea why it would suddenly work if you add the print message.
|
That's just weird. o.O
Does it work with other statements in place of the print statement(out of curiosity)? Maybe the return statement just got lonely 
|
|
|
int i;
public boolean getPause() {
i++;
return pause;
}
doesnt work.
int i;
public boolean getPause() {
System.out.println(i);
return pause;
}
works. As i said, i have absolutely no idea why it does that.
|
On December 01 2013 06:06 Warri wrote:
int i;
public boolean getPause() {
i++;
return pause;
}
doesnt work.
int i;
public boolean getPause() {
System.out.println(i);
return pause;
}
works. As i said, i have absolutely no idea why it does that.
I'm not experienced in java, but it sounds like a threading issue maybe? If that's the case then what you need is a memory barrier so that java knows to check the ram for the updated value rather than reusing what's in the cpu register, or set the variable to volatile if java has such a thing. The println function would act as memory barrier and that would be why that "fixes" it.
edit: yea java has volatile, this is what you need http://stackoverflow.com/questions/2423622/volatile-vs-static-in-java
|
On December 01 2013 06:12 Die4Ever wrote:Show nested quote +On December 01 2013 06:06 Warri wrote:
int i;
public boolean getPause() {
i++;
return pause;
}
doesnt work.
int i;
public boolean getPause() {
System.out.println(i);
return pause;
}
works. As i said, i have absolutely no idea why it does that. I'm not experienced in java, but it sounds like a threading issue maybe? If that's the case then what you need is a memory barrier so that java knows to check the ram for the updated value rather than reusing what's in the cpu register, or set the variable to volatile if java has such a thing. The println function would act as memory barrier and that would be why that "fixes" it. edit: yea java has volatile, this is what you need http://stackoverflow.com/questions/2423622/volatile-vs-static-in-java
Yeah that works, thank you!
|
On December 01 2013 06:28 Warri wrote:Show nested quote +On December 01 2013 06:12 Die4Ever wrote:On December 01 2013 06:06 Warri wrote:
int i;
public boolean getPause() {
i++;
return pause;
}
doesnt work.
int i;
public boolean getPause() {
System.out.println(i);
return pause;
}
works. As i said, i have absolutely no idea why it does that. I'm not experienced in java, but it sounds like a threading issue maybe? If that's the case then what you need is a memory barrier so that java knows to check the ram for the updated value rather than reusing what's in the cpu register, or set the variable to volatile if java has such a thing. The println function would act as memory barrier and that would be why that "fixes" it. edit: yea java has volatile, this is what you need http://stackoverflow.com/questions/2423622/volatile-vs-static-in-java Yeah that works, thank you!
Could you also check one other way to do it that could possibly fix the issue without the need for volatile types? What happens if you prototype functions that use pause variable? To do that you have to declare the function before you define it, like so:
boolean getPause();
int main()
{
getPause();
return 0;
}
boolean getPause();
{
do_stuff();
}
That is, of course, if Java supports prototypes...
Edit: Scratch that. I read up on that and it seems that Java has method signatures but not method prototypes.
|
On December 02 2013 02:18 Manit0u wrote:Show nested quote +On December 01 2013 06:28 Warri wrote:On December 01 2013 06:12 Die4Ever wrote:On December 01 2013 06:06 Warri wrote:
int i;
public boolean getPause() {
i++;
return pause;
}
doesnt work.
int i;
public boolean getPause() {
System.out.println(i);
return pause;
}
works. As i said, i have absolutely no idea why it does that. I'm not experienced in java, but it sounds like a threading issue maybe? If that's the case then what you need is a memory barrier so that java knows to check the ram for the updated value rather than reusing what's in the cpu register, or set the variable to volatile if java has such a thing. The println function would act as memory barrier and that would be why that "fixes" it. edit: yea java has volatile, this is what you need http://stackoverflow.com/questions/2423622/volatile-vs-static-in-java Yeah that works, thank you! Could you also check one other way to do it that could possibly fix the issue without the need for volatile types? What happens if you prototype functions that use pause variable? To do that you have to declare the function before you define it, like so:
boolean getPause();
int main()
{
getPause();
return 0;
}
boolean getPause();
{
do_stuff();
}
That is, of course, if Java supports prototypes... Edit: Scratch that. I read up on that and it seems that Java has method signatures but not method prototypes.
As far as i understand your post, that seems to be what an interface does in java. I dont think that would change anything though :p
http://en.wikipedia.org/wiki/Interface_(Java)#Overview
Specifically "Interface methods are public, abstract and never final. Think of them as prototypes only; no implementations are allowed."
|
Hey I just had a quick question about interfaces - why is it useful to use them when you're simply reiterating the definition of a method? Like so:
+ Show Spoiler +void Update();
void Update()
{
stuff
}
It doesn't seem like you're adding anything to the method you're using by implementing an interface, in fact it seems like you're only adding more code typing to do.
|
On December 02 2013 08:44 Roe wrote:Hey I just had a quick question about interfaces - why is it useful to use them when you're simply reiterating the definition of a method? Like so: + Show Spoiler +void Update();
void Update()
{
stuff
} It doesn't seem like you're adding anything to the method you're using by implementing an interface, in fact it seems like you're only adding more code typing to do.
Sorry, I assumed you mean Java's interfaces. I think I've not come across such a keyword in Objective-C and C, but it's still possible that other languages use the keyword.
In my opinion, interface is like what a contract is. If you inherit the class, then you have to play by the house's rules. Every further implementation of the interface also binds you to that "contract". 
I'm not too experienced with Java, so take this with a grain of salt.
Edit #1: As far as I remember, when I was doing Java RMI for university, the client only knows about the interface but not about implementation, so if I'm not mistaken, you can also hide implementation from the user.
Edit #2: Oh, yeah, I just remembered what I've recently studied in Software Engineering.
Basically, if you have an interface:
interface Encryption {
public String encrypt(String);
}
and then:
class LowerLevelEncryption implements Encryption {
// constructor
LowerLevelEncryption() {
}
public String encrypt(String blahblah) {
...
}
}
class HigherLevelEncryption implements Encryption {
// constructor
HigherLevelEncryption() {
}
public String encrypt(String blahblah) {
...
}
}
Now you can take advantage of polymorphism by choosing which encryption level to use at runtime.
E.g.
Encryption encryptionMethod = new HigherLevelEncryption();
and you can change it anytime to: encryptionMethod = new LowerLevelEncryption();
|
On December 02 2013 08:44 Roe wrote:Hey I just had a quick question about interfaces - why is it useful to use them when you're simply reiterating the definition of a method? Like so: + Show Spoiler +void Update();
void Update()
{
stuff
} It doesn't seem like you're adding anything to the method you're using by implementing an interface, in fact it seems like you're only adding more code typing to do.
it's about future updates to the code. maybe there will be new classes to implement that interface later.
|
|
|
Hw help:
From a test header file like so:
...
char s[]={"2/3","2/-3"};
...
I need to convert that string into a fraction (int array). The C files utilizes this function (fraction is a struct and all has been initialized):
Fraction string_to_fraction(const char *S)
{
Fraction result = {0,1}; //to initialize
sscanf(S,"%d/%d",result); //to read in the data as int to place into the result array
return result;
}
Am I doing something wrong? This assignment is more of "fill in the blanks" type fyi (standard library, variables initialized etc...) I am not looking for someone to do my hw but simply to point me in the right direction as when I try to execute the program that tests my function it simply crashes with no warning.
|
Yeah you need to pass in pointers to each element of your fraction. Your fraction has two elements. Your sscanf call would look something more like:
const char *in = "1/2";
int out[2];
sscanf( in, "%d/%d", out, out + 1 );
Edit: Oh you said Fraction is a struct. Okay like this then:
typedef struct Fraction
{
int numerator;
int denominator;
} Fraction;
Fraction f;
const char *stringFraction = "1/2";
sscanf( stringFraction, "%d/%d", &f.numerator, &f.denominator );
Does this make sense?
|
On December 03 2013 12:30 CecilSunkure wrote:Yeah you need to pass in pointers to each element of your fraction. Your fraction has two elements. Your sscanf call would look something more like: const char *in = "1/2";
int out[2];
sscanf( in, "%d/%d", out, out + 1 );
Edit: Oh you said Fraction is a struct. Okay like this then: typedef struct Fraction
{
int numerator;
int denominator;
} Fraction;
Fraction f;
const char *stringFraction = "1/2";
sscanf( stringFraction, "%d/%d", &f.numerator, &f.denominator );
Does this make sense?
yep.
I literally just figured it out after I posted. Your response only solidifies it. Thanks for the quick reply and appreciate the help.
|
While you're at numerator/denominator, is the alternative String.split() in Java? Just wondering.
|
Ok! So your favorite noob is back with a question for c++.
I have 2 classes, both within the same namespace.
One of the classes calls upon the other with a function and one of the parameters is of the first class.
I try to send it with a "this" pointer and that can't be done because the compiler complains that i try to send namespacename::classname* instead of just the classname*.
I tried adding the namespace to the functions parameter but that didn't work.
Basically:
m_stateMachine.SetState(m_stateFactory->GetState(0), this, 0);
bool CStateMachine::SetState(CState* _state, CGame* _game, unsigned int _id)
|
What's the actual error message? What's the prototype/signature of m_stateFactory->GetState(0)? And I assume the type of "this" is CGame* ?
|
On December 03 2013 23:04 iaretehnoob wrote:
What's the actual error message? What's the prototype/signature of m_stateFactory->GetState(0)? And I assume the type of "this" is CGame* ?
error C2664: 'tpe::CStateMachine::SetState' : cannot convert parameter 2 from 'tpe::CGame *const ' to 'CGame *'
Types pointed to are unrelated; conversion requires reinterpret_cast, C-style cast or function-style cast
Yes, "this" is CGame*.
And they are both in the tpe namespace.
|
On December 03 2013 22:40 Nausea wrote:Ok! So your favorite noob is back with a question for c++. I have 2 classes, both within the same namespace. One of the classes calls upon the other with a function and one of the parameters is of the first class. I try to send it with a "this" pointer and that can't be done because the compiler complains that i try to send namespacename::classname* instead of just the classname*. I tried adding the namespace to the functions parameter but that didn't work. Basically: m_stateMachine.SetState(m_stateFactory->GetState(0), this, 0); bool CStateMachine::SetState(CState* _state, CGame* _game, unsigned int _id)
Putting this might work:
bool CStateMachine::SetState(CState* _state, CGame* const _game, unsigned int _id)
I believe 'this' is a const pointer that cannot have its address modified.
|
|
|
|
|
|
EPT
