C++ :: When Are Rvalue References To Primitive Integers Short-lived Or Long-lived
Sep 4, 2012
I have done some experimentation on rvalue references with the TDM-GCC 4.6.1 compiler and made some interesting observations that I cannot explain away with theories. I have a very simple program that does not deal with objects but int primitives and that has defined 2 functions:
foo1 (returning a local variable by rvalue reference) and
foo2 (returning a local variable by value)
#include <iostream>
using namespace std;
int &&foo1();
int foo2();
int main() {
int&& variable1 = foo1();
[Code] .....
It seems the value returned by foo1 and received by variable1 dies out after some time - perhaps, a brief period of some milliseconds. Notice that I have prevented cout from printing "My name is softwarelover" by commenting it out. If I allow that statement to run, the result is different. Instead of printing 5, 0 it prints 0, 0. Seems like it is because of the time-delay introduced by "cout << "My name is softwarelover." that 5 turns into 0.
Is the above how an rvalue reference is supposed to behave when referring to a primitive integer which a function returned by reference as opposed to return-by-value? By the way, why is it 0, why not garbage?
Notice also that variable2 never seems to die out, no matter how many times I print it with cout! variable2 refers to a primitive integer which a function returned by value, not return-by-reference.
I am trying to understand RValue-references as return values of functions. First let's consider a simple function, that transforms a string into upper case letters.
It compiles, but I get the output 0 . Here I am wondering why the code above does not move the substr correctly while the code below does (prints out 1):
In both cases abc is a temporary object inside of the function and gets deleted after the function is left. But why does the second version work and the first one does not?
My understanding is that the rhs of line 1 construct only a temporary object. getMe() then return the reference of this temp object and bind it to thg (as a lvalue reference). After line one, the temp object is really destroyed (hence the first output line). At this point thg is really binding to a destroyed, invalid object. But somehow the 2nd line still prints the correct value of 10 is because the memory storage is not yet corrupted (still holding the previous value). Is this correct?
I think I may have found a new way of checking for 3d polygon collisions, but I'm not sure. The method involves...
1. finding the planes that the primitives lie on 2. finding the line where the planes intersect 3. if both polys have points on both sides of the line AND have points that overlap on the 1d space of the line, then they intersect.
I have some half done code testing this, and so far it seems to be sound and fairly fast. These are some average time-tests done on my machine for each part:
So, in my platform, conversion from an unsigend integer primitive data type to any bigger integer primitive data type never extends the most significant bit of the former integer and conversion from an signed integer primitive data type to any bigger integer primitive data type always extends the most significant bit of the former integer. This is convenient to mantain the same value when converting between integer primitive data types of the same signedness (i.e, signed integers or unsigned integers).
I have defined my own class, Queue, which inherits from my own class, LinkedList. I have been using templates to allow Queues to be of int, string, etc types.
But now I want to be able to store objects in my Queue type. And so the problem I have is that in my LinkedList class, I have two instances where I initialize an instance of my generic type T to 0.
For instance, the removeFirst() method starts like this:
template <typename T> T LinkedList<T>::removeFirst() { T a = 0;
And so the compiler complains that it can't convert from int to [in this case] Command&.
I have a double variable and depending on certain conditions I need to set certain bits of an unsigned short Variable. For example, if double var is odd I need to set the 15th bit of the unsigned short variable.
I am using print/sprintf with a "%i" format string. Works fine if the input is indeed a 32bit integer. But how can i put in 8/16 bit (ie short & byte) 'integers'? If i just throw them in, they are always taken as unsigned, as the topmost bit/s is/are casted to zero... [URL] ....
I'm having trouble reading a block of bytes into a vector of short ints. My code is as follows:
Code: FileStream.seekg(2821); vector<short> g_Data; int iter = 0; g_Data.reserve(maxNumOfInts);
[Code] ....
The relevant block of data starts at offset 2821 in the file. Every two bytes are a signed short integer. What's odd is that it's giving me the correct results only part of the time. At offset 1052421 and 1052422 there are two bytes 40 and 1F that are correctly read in as 8000, but at offset 1052415 and 1052416 bytes 88 and 13 are read in as -120 instead of 5000.
I don't see anything wrong with my code, though, unless I'm misunderstanding completely how to convert an array of two bytes into a single float. Is my method correct? Better still, is there some way to just convert en mass an array of bytes into a vector of signed short ints?
I've been working on this project which inserts data for different types of books lately, and I'm trying to utilize inheritance and polymorphism. The issue I've been having is after I create an object with my MediaFactory, I go to insert data into that type of object, but it won't reach the correct child class. The parent class in this case is MediaData. The class I'm trying to reach is Childrens, and the class that falls in between is called Book. The function I'm calling to insert the information is setData(ifstream&). This function simply places the information from a txt file into an object with the insertion operator.
Right now the program runs into the setData function of Book instead of Childrens. I need Childrens so I can enter all the required attributes (title,author,year). The call to this function is in MediaManager through the function called buildMedia, and my test is running into the case if (type == 'Y'). From there a pointer of type MediaData is created and pointed to a new object returned of type Childrens. I'm using my debugger in Xcode which does show that the correct object type (Childrens) was created.
I've tried a couple things so far. First I created another function called getData. From there I passed in *media and infile (txt input file) into that function, which then passed it right into setData with a pointer to the object and infile in the parameter. That didn't work, so I also tried going back into Childrens and removing MediaData from all my virtual functions and replacing it with Book. I got the same result when that happened; It morphed into Book class instead.
I have a portion of my UML as a reference. Childrens is a Book; Book is a MediaData. I also included all code for MediaManager, MediaHash, MediaFactory, MediaData, Book, and Childrens.
I did not include the operator overloads in the .cpp files to eliminate some redundancy.
//----------------------------------------------------------------------------- // MediaManager.h // Manager class for MediaData type objects //-----------------------------------------------------------------------------
#ifndef MediaManager_H #define MediaManager_H #include "MediaHash.h" #include "MediaFactory.h" #include "MediaData.h" using namespace std; class MediaManager {
I am trying to make a utility program for work that will update multiple projects with local dll references. Basically I work with two solutions (for talk sake solutIon1 and solutIon2). Generally solutIon1 will reference the dll's built In solutIon2 which reside on a server. However for debugging proposes I sometimes need to D/L the solutIon2 projects and build them local-ally, so that I can reference the solutIon2 dll's local-ally (this Is so that I can easily attach the dll and step Into the code). However this require changing the reference paths, so that I am pointing to the local-ally built dll's, which Is quite a laborious task.
So the question is how would I update references in solution1 from the program that I am making. I don't really know what to start reading about as I have never done anything like this before.
I'm a beginner at c++ and I need to write a program that reads a set of integers and then finds and prints the sum of the even and odd integers. The program cannot tell the user how many integers to enter. I need to have separate totals for the even and odd numbers. what would I need to use so that I can read whatever number of values the user inputs and get the sum of even and odd?
I am attempting to combine two vectors into a vector of pairs. I want to be able to alter the first and second of each pair and have those alterations reflected in the original vectors. I thought the following code might work but get compilation errors about a lack of viable overload for "=" for the line with the call to std::transform:
void f() { std::vector<int> a = {1,2,3,4,5}; std::vector<int> b = {6,7,8,9,0};
I am checking to see if two references are bound to the same object. My instincts tell Me, "Check their addresses. If they match, they are bound to the same." At the same time, I have not found anything in the C++ standard which would support this approach. Am I missing something? Is there wording which backs up My instincts? Is there a standard function to do this?
