-
Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.
C++ Help?
- Thread starter AcciDante
- Start date
- Status
Yes, you can assign pointers to each other. A pointer's value is a location in memory. When you do an assignment of `ptr1 = ptr2` the address stored in ptr2 overwrites the address stored in ptr1.Zoramon089 said:
Oh... oh my.Linkhero1 said:
Code:
if ((_start < 'A' || _start > 'H') || (_start < 'a' || _start > 'h')) {
std::cout << "Fail... ";
exit(1);
}Right, but I highly doubt Link's code is meant to be maintained for longdeadbeef said:
Anyhow, in that case you'd set a lowercaseLowerBound char, lowercaseUpperBound, uppercaseLowerBound, and uppercaseUpperBound, and compare to those. Then you're nice and extensible.
Code:
char lowercaseLowerBound = 'a';
char lowercaseUpperBound = 'h';
char uppercaseLowerBound = 'A';
char uppercaseUpperBound = 'H';
if ((_start < lowercaseLowerBound || _start > lowercaseUpperBound) ||
(_start < uppercaseLowerBound || _start > uppercaseUpperBound)) {
std::cout << "Fail... ";
exit(1);
}Zoramon089 said:
As other people have said, yes, you can do assignment with pointers - pointers just contain a memory address. That's where the fun/trouble comes into it - in your example above, what are you wanting to be assigned? Because you're just sticking the memory address of sdfsdf into the ptr2's location. No change was made to either sdfsdf or hrtrh. Essentially you've got two pointers to the first object now.
Zoramon089
Banned
Having some pointer problems that I don't really understand. This is just a small snippet of my code, shortened for the sake of this post, but including everything relevant.
So as you can see I have a class I created named myClass which takes in an int as a parameter and simply saves it into an int field in the class. It also has a function called getId() which simply returns that field. The problem here is that when I print out the int I get
Which I don't understand
}
Code:
int id=0;
int i;
vector<*myClass> theList;
while(true) // My program is looping but I don't feel like typing out everything else
{
if(theList.empty())
theList.push_back(&myClass(id));
else
{
printf("id %i", theList[0]->getId())
theList[0] = &myClass(id));
}So as you can see I have a class I created named myClass which takes in an int as a parameter and simply saves it into an int field in the class. It also has a function called getId() which simply returns that field. The problem here is that when I print out the int I get
Code:
id -858993460Which I don't understand
}
I'm not understanding &myClass(id).
What exactly are you doing there?
If that's your instantiation of the class, it's running out of scope and getting destroyed so you're accessing invalid memory, thus why you're getting the random integer.
Edit: If what I said is correct then:
myClass * whatever = new myClass(id);
theList.push_back(whatever);
should work for you.
just remember you need to manually delete all instances of myClass if you're creating them on the heap.
What exactly are you doing there?
If that's your instantiation of the class, it's running out of scope and getting destroyed so you're accessing invalid memory, thus why you're getting the random integer.
Edit: If what I said is correct then:
myClass * whatever = new myClass(id);
theList.push_back(whatever);
should work for you.
just remember you need to manually delete all instances of myClass if you're creating them on the heap.
Zoramon089
Banned
josephdebono said:
Yeah, reading more about it, I kind of thought that was the issue but I did this as well and got the same problem.
Code:
int id=0;
int i;
vector<*myClass> theList;
while(true) // My program is looping but I don't feel like typing out everything else
{
if(theList.empty())
{
myClass m = myClass(id);
theList.push_back(&m);
}
else
{
printf("id %i", theList[0]->getId())
myClass m = myClass(id);
theList[0] = &m);
}}
Zoramon089 said:
You need to new it.
myClass * whatever = new myClass(id);
theList.push_back(whatever);
And as stated above, you have to manually delete the instances when clearing the vector.
Remember, unless you're returning or storing the object itself (which you aren't since you're only storing a pointer to the object) it will get destroyed when you leave scope.
Zoramon089
Banned
THANKS
That was the problem. Man...switching from java to C++...syntactically it's not too different but all the small differences with pointers and references make a HUGE difference in what works and what doesn't
That was the problem. Man...switching from java to C++...syntactically it's not too different but all the small differences with pointers and references make a HUGE difference in what works and what doesn't
Zoramon089 said:
Code:
int id=0;
int i;
vector<myClass*> theList; // the star goes afterward to indicate a pointer of type myClass
while(true) // My program is looping but I don't feel like typing out everything else
{
if(theList.empty())
theList.push_back(new myClass(id)); // this allocates memory for and initiates an instance of myClass, and returns a pointer to it
else
{
printf("id %i", theList[0]->getId())
// theList[0] = &myClass(id)); // commented this line, since it doesn't make sense to me what you're even trying to do here.
}I'm editing after I post each reply so you may have missed some of what I'm saying. Now, I understand what you're doing but I don't know if you're doing this to learn or for an actual 'useful' task. In reality, I see no reason for you to storing the pointers there, so:
vector<myClass> theList;
theList.push_back(myClass(id));
theList[0].getId();
would work just as well and remove the unnecessary pointer headaches.
vector<myClass> theList;
theList.push_back(myClass(id));
theList[0].getId();
would work just as well and remove the unnecessary pointer headaches.
Zoramon089
Banned
No, they need to be pointers for other functionality in the program. Mainly because myClass is a derived class of another class and sometimes I need to cast it as that based class. I can only use the dynamic_cast if it's a pointer
cpp_is_king
Member
josephdebono said:
I tend to agree with this most of the time, but you need to understand the difference. Sometimes this can lead to significantly better performance (not having your accesses go through a pointer is significantly better for cache-locality), sometimes it can lead to significantly worse performance (vector operations often result in lots of copies being made, if you have a non-trivial copy-constructor, or even if your objects are just large, this can be a big performance hit), sometimes it's flat out broken (e.g. if your destructors free external resources like file handles, sockets, etc), and sometimes it isn't even possible (i.e. if you need to rely on polymorphic behavior).
For future reference, if you're using dynamic_cast<> chances are you're doing somethign wrong. Keep going with it while you're still learning the language, but just keep it in the back of your mind that there's almost certainly a better way.Zoramon089 said:
Zoramon089
Banned
cpp_is_king said:
Yeah, i've tried a number of ways already. I even consulted this topic earlier and that ended up being the method that worked best. I basically have a vector of a base class and I need to add a bunch of derived classes to it, so I can loop through all of them and call a fucntion in the base class on all of them.
That works just fine but sometimes I need to call a function specific to a base class (which i can determine accurately based on the index in the vector). When i tried a simple cast of (derivedClass)list[index] it kept giving me errors about there being no way to cast to the derived class.
Zoramon089
Banned
Yeah, and the dynamic cast was what ended up working!fenners said:
Hey people, I'm back again. I'm trying to find a good way to read in file, but have it exclude all whitespace and punctuation. I'm looking at istream right now, but I don't see a way that I can make it exclude punctuation. Do you know of something else I could use, or am I just not looking in the right place?
If you're just looking for alpha characters (a-zA-Z) you could just use istream and check the retrieved characters to see if they're <45 ('A') or >122 ('z')rexor0717 said:
edit: to be more clear:
Code:
char c;
while(whatever) {
instream.get(c);
if (c >= 'A' || c <= 'z') {
// do what you will with c
}
}Nice, thanks. Did the first two (really digging the referencesLathentar said:
) but I'm a little stumped by how to approach the 3rd.I know that for all order bits, there must be an even number of said bit (or 0) to keep the kid from crying. As for finding out the most lucrative combination for the smart kid, I'm guessing I just need to come up with a search tree that finds the most lucrative combination of numbers that contains an odd count for all order bits (that aren't 0)
But the way you made it sound, perhaps I'm overthinking it?
So I have 2 functions with the following prototypes:
(functionB uses functionA).
Is there any way I can directly pass the ellipsis arguments from functionB to functionA?
Thank you very much, I appreciate the help.
Code:
void functionA(const char* str, ...);
bool functionB(int,int,const char*,...);
bool functionB(int a,int b, const char* string,...){
//code
//call of functionA();
//return true;
}(functionB uses functionA).
Is there any way I can directly pass the ellipsis arguments from functionB to functionA?
Thank you very much, I appreciate the help.
cpp_is_king
Member
poweld said:Nice, thanks. Did the first two (really digging the references
I know that for all order bits, there must be an even number of said bit (or 0) to keep the kid from crying. As for finding out the most lucrative combination for the smart kid, I'm guessing I just need to come up with a search tree that finds the most lucrative combination of numbers that contains an odd count for all order bits (that aren't 0)
But the way you made it sound, perhaps I'm overthinking it?
I didn't try that one, but the Goro Sort problem was pretty cool. I did it by first proving a mathematical formula for the optimal strategy and from there the actual algorithm is trivial. But convincing yourself that the optimal strategy is in fact optimal, and being able to compute the expected number of trials assuming you do use the optimal strategy is not immediately obvious.
I wonder if other people solved it a different way than i did though.
I'm writing a code that transverses through a directory you choose and searches for a file you specify. I'm trying to write something to display the directory the file is in. Say for example someone selects C:\ as the root directory and then readme.TXT as the file they're searching for they code will transverse through folders until it finds it. I'm having trouble returning the path to display.
items is multimap i declared earlier and item is just the name of the file the user is searching for. I am able to display how many folders and items are in the directory but I'm unable to produce a path to display.
Edit: nvm it's because of a function I'm using not working properly. I'm not sure why though but I guess I'll spend some time debugging it.
Code:
item_iterator m1;
pair<item_iterator, item_iterator> range = items.equal_range(item);
for (m1 = range.first; m1 != range.second; ++m1)
{
std::cout << "Directory Location: " << (*m1).second << endl;
}Edit: nvm it's because of a function I'm using not working properly. I'm not sure why though but I guess I'll spend some time debugging it.
Complex Shadow
Cudi Lame
Can someone explain me the diffrence between references and pointers.
Complex Shadow said:
They're kind of intertwined, and as someone who's never made a fantastic teacher, I'll defer to an article I once stumbled on. It's a pretty thorough run-down of the two and does a decent job of detailing specific usage scenarios.
http://www.eetimes.com/discussion/programming-pointers/4023307/References-vs-Pointers
A reference can never be reseated, but a pointer can.Complex Shadow said:
Code:
int& someNumRef = someNum;
someNumRef = 5; // someNum is now equal to 5
int* someNumPtr = &someNum;
*someNumPtr = 10; // someNum is now equal to 10
someNumPtr = &someOtherNum;
*someNumPtr = 20; // someOtherNum is now equal to 20cpp_is_king
Member
poweld said:
Also a reference cannot be NULL, but a pointer can.
Put more in terms of *why* they're different rather than the specific details of how, a reference acts like a value. References also imply a lack of ownership over the underlying resource.
So basically, if you just want to avoid copying something when you pass it into a function (because it's heavy or has non-copying semantics), and intend to treat it as a value, you should use a reference.
So basically, if you just want to avoid copying something when you pass it into a function (because it's heavy or has non-copying semantics), and intend to treat it as a value, you should use a reference.
Zombie Lyle
Member
Hey, I'm looking over past papers for my C++ exam next week and can't find the answer for this question.
In C++ how may bytes of memory are occupied by an array of 150 integers and an array of 150 character pointers.
Thanks in advance.
In C++ how may bytes of memory are occupied by an array of 150 integers and an array of 150 character pointers.
Thanks in advance.
I'm getting std:ut_of_range errors concerning a string check function in my program. For everyone's sanity I'm not going to post the actual application here but what usually causes this? I have a ton of loops going through entire strings in the function in question - should I be looking for places where it tries to look at what character there is at a position that isn't defined in the string?
ut_of_range errors concerning a string check function in my program. For everyone's sanity I'm not going to post the actual application here but what usually causes this? I have a ton of loops going through entire strings in the function in question - should I be looking for places where it tries to look at what character there is at a position that isn't defined in the string?You should be looking for places where you are looking at index positions in your string, for example when you use the bracket operator (e.g. myString[10]) or at (e.g. myString.at(10)). The error is being thrown because you're trying to access an index in the container that does not exist.Rapstah said:I'm getting std:
Yeah, I guess I just figured if it's asking for the size of an int they may be assuming 32bit. Also, good eye, I didn't see that it was char pointers, not char's.Tigel said:
In that case an int and a pointer to anything will always scale to the architecture, be it 32 or 64 bit.
Zombie Lyle
Member
Is a pointer always 4 bytes for an int, char or float?
Also guys, I hate to ask loads of questions but I'm useless at C++. I don't know how to use a vector as an array. The question is below.
Write the C++ code to dynamically create a vector of 100 long integers. Explain how you could use this block of memory as a two dimensional array of 10 by 10 long integers. Also give the code to release the vector.
Also guys, I hate to ask loads of questions but I'm useless at C++. I don't know how to use a vector as an array. The question is below.
Write the C++ code to dynamically create a vector of 100 long integers. Explain how you could use this block of memory as a two dimensional array of 10 by 10 long integers. Also give the code to release the vector.
- Status