Popular C++ recipes

Convert from Html To Pdf in ASP.NET MVC C# with SelectPdf Free Community Edition (C++)

2016-11-17T15:01:12-08:00

C++ recipe 580719 by SelectPdf (aspnet, mvc, pdf, selectpdf).

It’s very easy to use SelectPdf SDK for .NET in ASP.NET MVC applications. Take a look at the simple code below.

Hopfield Artificial Neural Network (C++)

2016-07-03T21:58:49-07:00

C++ recipe 580688 by Jay baller (ai, algorithm, algorithms, artificial_intelligence, neural_network).

(Discrete (Binary)) Hopfield Artificial Neural Network (ANN).

For more info see:

http://en.wikipedia.org/wiki/Hopfield_net

http://www.scholarpedia.org/article/Hopfield_network

MicroXml: Stand-alone library for basic XML features (C++)

2016-02-18T19:33:07-08:00

C++ recipe 580612 by Jack Trainor (parsing, xml).

MicroXml provides stand-alone support for the basic, most-used features of XML -- tags, attributes, and element values. It produces a DOM tree of XML nodes. MicroXml does not support DTDs, CDATAs and other advanced XML features.

MicroXml is easy to use and provides easy access to view/navigate its nodes in a debugger.

Boost::Python callback triggered from Non-Python created threads (C++)

2014-11-20T20:00:27-08:00

C++ recipe 578967 by Tomáš Rampas (boost, python).

Consider we have virtual/abstract C++ class that's fully implemented in Python. And for some sake of necessity we have a callback method (e.g. as some sort of event) that is being triggered from different thread on C++ side. In such case corresponding callback methods have to manage GIL state via PyGILState_STATE. So the resulting C++ callback class definition will look like follows (notice that Python method calls are wrapped up with GIL state handling code).

Numbers as Ranges - iterable integers (C++)

2014-10-02T12:13:52-07:00

C++ recipe 578945 by elazar (iterable, range). Revision 2.

This header file makes simple integers iterable. Note that it works only on C++11 or above.

Usage:

#include "num_range.h"
for (int i : 3) cout << i << endl;

Output: 0 1 2

Implementation note: This code is far too generic. We only need DerefableInt. The templates are there for no practical reason.

Cons: pollutes namespace std; nonstandard idiom;

Tri-nary Tree: insertion and deletion (C++)

2014-02-07T04:21:50-08:00

C++ recipe 578825 by Chen (tree).

This contains the insertion and deletion implement for the tri-nary tree data structure.

Elementary regex without third party libraries (C++)

2013-07-13T12:58:14-07:00

C++ recipe 578602 by greg zakharov (regex, wininet).

This is a dirty trick how to access primitive regular expression on Windows.

A-star Shortest Path Algorithm (C++)

2010-12-25T23:36:35-08:00

C++ recipe 577457 by FB36 (algorithm, algorithms, game, graph, graphs, routes). Revision 4.

A-star (A*) is a shortest path algorithm widely used for RTS games, GPS navigation etc.

Automatic Python PyObject ref-count management in C++ using a smart ptr (C++)

2011-12-19T15:24:03-08:00

C++ recipe 577985 by samh (extending).

Managing ref-counting is a complex and error prone business. If you choose C++ to extend or embed Python, you can simply use a modification on std::auto_ptr. Instead of calling delete on the managed pointer, it will decref it.

So now you can do:

auto_pyptr pyHelloStr(PyStr_FromString("Hello World!"));

and forget about having to decref it altogether! Just like auto_ptr you can get the PyObject * with get(), release it from managed control with release(), and rebind it with reset(new_ptr). You can also incref it by calling inc(), but be cautious as you can easily create a leak by increfing once to much.

Even faster prime generator (C++)

2011-11-27T21:48:25-08:00

C++ recipe 577966 by Sumudu Fernando (primes, prime_generator).

A very quick (segmented) sieve of Eratosthenes.

Takes ~6s on a midrange machine to hit all 50 847 534 primes less than 1 billion, ending with 999999937

If you want to actually do anything with every prime (beyond counting them), there are three places to add a statement doing whatever is necessary with "lastP" -- one at the top (handles the special case '2'), one in the middle (handles "small" primes which are actually used to sieve), and one at the bottom (handles "large" primes which merely survive sieving).

In principle one can use a function object as parameter to allow generic operations on the primes, so add that if you want more general-purpose code (perhaps I'll do that later)

For higher limits you need to switch to wider types and follow the commented guidelines for the constants. For a fixed limit, changing B_SIZE may affect performance so if needed tune it (profile as you go, of course!). But this will get quite slow if you go to much higher numbers.