Home » Blog » Tags » advanced

advanced

Compose and curry as folds--Nick Athanasiou

By Adrien Hamelin | Nov 28, 2016 12:33 PM | Tags: experimental advanced

With the next version of C++.

Compose and curry as folds

by Nick Athanasiou

From the article:

In a previous post we introduced C++17 fold expressions and described a way to extend them for arbitrary callables. Implementation details don’t matter for what we’re elaborating on here but it should be clear that (given the tools we developed) the following is possible:

(Op<F>(args) + ...)
(... + Op<F>(args))

Infographics: Operation Costs in CPU Clock Cycles--“No Bugs” Hare

By Adrien Hamelin | Nov 11, 2016 12:54 PM | Tags: performance advanced

A very interesting article about the cost of our basic operations.

Infographics: Operation Costs in CPU Clock Cycles

by “No Bugs” Hare

From the article:

Whenever we need to optimise the code, we should profile it, plain and simple. However, sometimes it makes sense just to know ballpark numbers for relative costs of some popular operations, so you won’t do grossly inefficient things from the very beginning (and hopefully won’t need to profile the program later �� )...

Terminators--Adi Shavit

By Adrien Hamelin | Nov 7, 2016 12:02 PM | Tags: advanced

Do you know how a program ends?

Terminators

by Adi Shavit

From the article:

A GraphViz diagram that shows both normal and unexpected program termination flows in C++.

There are multiple ways a C++ program may terminate. These include both normal and unexpected termination.
This GraphViz diagram shows the program termination flows as defined by the standard...

checking expression validity in-place with C++17--Vittorio Romeo

By Adrien Hamelin | Nov 4, 2016 01:09 PM | Tags: experimental advanced

An introduction to the world of C++17.

checking expression validity in-place with C++17

by Vittorio Romeo

From the article:

When writing generic code, it is sometimes useful to check whether or not a particular SFINAE-friendly expression is valid (e.g. to branch at compile-time). Let's assume that we have the following class declarations...

struct Cat
{
    void meow() const { cout << "meow\n"; }
};

struct Dog
{
    void bark() const { cout << "bark\n"; }
};

...and that we would like to write a template function make_noise(x) that calls x.meow() and/or  x.bark() if they are well-formed expressions:

template <typename T>
void make_noise(const T& x)
{
    // Pseudocode:
    /*
        if(`x.meow()` is well-formed)
        {
            execute `x.meow();`
        }
        else if(`x.bark()` is well-formed)
        {
            execute `x.bark();`
        }
        else
        {
            compile-time error
        }
    */
}

In this article I'll show how to implement the pseudocode in:

C++11: using std::void_t and std::enable_if.

C++14: using boost::hana::is_valid and vrm::core::static_if.

C++17: using if constexpr(...), constexpr lambdas, and std::is_callable. This version will allow expression validity to be checked in-place (i.e. directly in the if constexpr predicate). Variadic preprocessor macros will also be used to make the user code easier to read and maintain...

Quick Q: Is there a case where ellipsis(vararg) should be preffered over variadic templates?

By Adrien Hamelin | Nov 1, 2016 12:04 PM | Tags: c++11 advanced

Quick A: There are several special cases where you may want that, but in general no.

Recently on SO:

Is there a case where ellipsis(vararg) should be preffered over variadic templates?

  1. If you provide a C API with C++ implementation, then templates are not an option for the API. Varargs are.
  2. If you need to support a compiler that doesn't support C++11 or newer standard, then variadic templates are not available. Varargs are.
  3. If you need a compilation firewall. I.e. you need to hide the implementation of the function from the header, then variadic template is not an option. Varargs are.
  4. On memory constrained systems (embedded), the different functions generated by the template may introduce too much bloat. That said, such systems are typically also real time, in which case varargs might also unacceptable due to branching and stack usage.

void foo(T& out) - How to fix output parameters--Jonathan Müller

By Adrien Hamelin | Oct 27, 2016 11:47 AM | Tags: c++11 advanced

Or how to improve readability and reduce errors.

void foo(T& out) - How to fix output parameters

by Jonathan Müller

From the article:

There are some cases where you need to return a value from a function but cannot use the return value. It happens, for example, in functions where you want to return multiple values at once. While you can pass multiple inputs to a function - the parameters, you cannot pass multiple return values in the same way.

C++ programmers tend to use a good old (lvalue) reference for that. You take a non-const reference as parameter and assign the output to that reference. The caller will pass a variable and upon function completion find the value of the variable changed.

Yet this approach has some problems: For starters, it is not obvious when just looking at the call that the variable is going to be changed. This is the reason that C++ style guides such as the one used by Google recommend using a pointer for that. The caller then has to explicitly pass in the address of the variable, making it explicit.

But with a pointer you can now pass in nullptr, you have to check for that in the function: A pointer where you really mean “reference” does not follow the guidelines I’ve been advocating for.

So is there not a universal solution?

There is, but first we need to understand the full scope of the problem.

DLib 19.2 released

By Meeting C++ | Oct 11, 2016 05:57 AM | Tags: intermediate efficiency deep learning cuda basics advanced

Today a new version of DLib is available:

DLib 19.2 released

Release notes

by Davis King

From the article:

... So the obvious thing to do was to add an implementation of MMOD with the HOG feature extraction replaced with a convolutional neural network.  The new version of dlib, v19.2, contains just such a thing.  On this page you can see a short tutorial showing how to train a convolutional neural network using the MMOD loss function.  It uses dlib's new deep learning API to train the detector end-to-end on the very same 4 image dataset used in the HOG version of the example program.  Happily, and very much to the surprise of myself and my colleagues, it learns a working face detector from this tiny dataset.

Metashell 3.0.0 is available

By Ábel Sinkovics | Oct 4, 2016 05:37 AM | Tags: advanced

Metashell provides a compile-time debugger for debugging template instantiations and macro usage.

Metashell 3.0.0 is available

From the website:

Some of the major new features in 3.0.0:

  • Profiling template instantiations
  • Debugging template instantiations in expressions involving SFINAE
  • Displaying preprocessed expressions

Commands for exploring the available headers (similar to the which and ls commands).
The full list of changes can be found here.

An online demo can be found at http://metashell.org/about/demo/

Installers can be downloaded from http://metashell.org/getting_metashell/installers#version-300