Home » Blog » Tags » advanced

advanced

"Modern C++ Template Programming" with Nicolai Josuttis

By Meeting C++ | May 2, 2018 02:19 AM | Tags: workshop training templates meetingcpp intermediate c++17 c++14 c++11 advanced

Meeting C++ 2018 offers also a workshop with Nicolai Josuttis:

Modern C++ Template Programming

by Jens Weller

From the article:

Each and every C++ programmer uses templates. Containers such as vector<> or array<>, strings, algorithms such as sort(), iterators, and I/O streams are all implemented as generic code. Modern C++ adds type traits, smart pointers, and template member functions such as emplace(), and generic lambdas as a tricky form of generic code.

Nevertheless the knowledge and understanding of how to implement and use templates is very limited and each and every programmer is sooner or later getting uncertain.

This workshop therefore discusses templates for a whole day to make clear what it means to use templates and how to use them in practice. As a result the general understanding of templates will be improved and generic code might become more helpful and less surprising.

Quick Q: int a[] = {1,2,}; Weird comma allowed. Any particular reason?

By Adrien Hamelin | Apr 20, 2018 12:29 PM | Tags: advanced

Quick A: For convenience.

Recently on SO:

int a[] = {1,2,}; Weird comma allowed. Any particular reason?

It makes it easier to generate source code, and also to write code which can be easily extended at a later date. Consider what's required to add an extra entry to:

int a[] = {
   1,
   2,
   3
};

... you have to add the comma to the existing line and add a new line. Compare that with the case where the three already has a comma after it, where you just have to add a line. Likewise if you want to remove a line you can do so without worrying about whether it's the last line or not, and you can reorder lines without fiddling about with commas. Basically it means there's a uniformity in how you treat the lines.

Now think about generating code. Something like (pseudo-code):

output("int a[] = {");
for (int i = 0; i < items.length; i++) {
    output("%s, ", items[i]);
}
output("};");

No need to worry about whether the current item you're writing out is the first or the last. Much simpler.

uninitialized_tag in C++--Marius Elvert

By Adrien Hamelin | Apr 17, 2018 12:17 PM | Tags: performance advanced

Optimise or not?

uninitialized_tag in C++

by Marius Elvert

From the article:

No doubt, C++ is one of those languages you can use to squeeze out every last drop of your CPU’s processing power. On the other hand, it also allows a high amount of abstraction. However, micro-optimization seldom works well with nice abstractions...

C++17: The two line visitor explained--Marius Elvert

By Adrien Hamelin | Apr 16, 2018 01:00 PM | Tags: c++17 advanced

The power of C++17.

C++17: The two line visitor explained

by Marius Elvert

From the article:

If you have ever used an “idiomatic” C++ variant datatype like Boost.Variant or the new C++17 std::variant, you probably wished you could assemble a visitor to dispatch on the type by assembling a couple of lambda expressions like this:

auto my_visitor = visitor{
  [&](int value) { /* ... */ },
  [&](std::string const& value) { /* ... */ },
};

Quick Q: With “-fno-exceptions”, what happens with “new T”?

By Adrien Hamelin | Apr 16, 2018 12:46 PM | Tags: advanced

Quick A: The behaviour will likely stay the same.

Recently on SO:

With “-fno-exceptions”, what happens with “new T”?

The way I understand it, operator new is defined by libstdc++. If you now compile your own code with -fno-exceptions, you cannot catch any exceptions, but you will still be linking against the normal version of libstdc++, which does throw an exception.

So yes, new T will throw an exception, even with -fno-exception.

However, if you compiled libstdc++ with -fno-exception as well, things become different. Now, new T cannot throw an exception but, if I read the libstdc++ manual right it will call abort() instead.

It seems that, if you want your new T to return NULL on failure, the only way is to explicitely specify nothrow...

Quick Q: Does access control matter for deleted constructors?

By Adrien Hamelin | Apr 9, 2018 02:58 PM | Tags: c++11 advanced

Quick A: No

Recently on SO:

Does access control matter for deleted constructors?

Since it's overload resolution that makes the program ill-formed in this case, and not access specifiers (which are checked later), there is no difference in outcome. The compiler will always complain that a deleted function was picked.

But since the idiom before C++11 was "declare but not define a private copy c'tor to disable copying", I would consider it going along with the same idiom, and therefore favorable. You are using the "old slang" with some new language to describe the same thing, except better.

Guidelines For Rvalue References In APIs--Jonathan Müller

By Adrien Hamelin | Mar 29, 2018 03:39 PM | Tags: c++11 advanced

Everything you need to know.

Guidelines For Rvalue References In APIs

by Jonathan Müller

From the article:

I’ll be giving a talk at ACCU about when to use which pointer types and why.

While working on that I made some guidelines for rvalue references in interfaces which didn’t quite fit the talk, so I’m writing about them here.

When should you use rvalue references as function parameters?

When as return types?

What are ref-qualified member functions and when and how should you use them?

Let’s tackle it one by one...

The “unsigned for value range” antipattern--Arthur O’Dwyer

By Adrien Hamelin | Mar 19, 2018 01:12 PM | Tags: experimental advanced

What do you think?

The “unsigned for value range” antipattern

by Arthur O’Dwyer

From the article:

Background: Signed Integers Are (Not Yet) Two’s Complement
At the WG21 committee meeting which is currently underway in Jacksonville, JF Bastien will be presenting a proposal to make C++’s int data type wrap around on overflow. That is, where today the expression INT_MAX + 1 has undefined behavior, JF would like to see that expression formally defined to come out equal to INT_MIN...

Selective argument application -- Krzysztof Ostrowski

By fj | Mar 6, 2018 09:04 AM | Tags: c++17 advanced

Presentation of a technique that enables transparent interception of the selected arguments passed to any function by means of a user-provided predicate metafunction.

Selective argument application

by Krzysztof Ostrowski

From the article:

Presentation of a design of an interceptor abstraction built over C++17 utilities that records values of the selected arguments passed to any action transparently, then applies that action to the captured arguments, and eventually passes the result value to the next action.