Overview

This article describes Harlinn.Windows, a Visual Studio 2022 solution.

Before you can build the solution, please read the build instructions and configure your environment accordingly.

Introduction

Harlinn.Windows is a collection of libraries that I have put together that serves as a big part of the runtime we use at work. A few are my own; but most are; more or less, well-known open source libraries.

There are currently 630 projects in the solution, and this article provides a high-level overview of what's inside.

Why did I not use cmake and vcpkg?

This is a good question, and initially I tried to do just that, but there were some snags:

The size of the release build, using vcpkg, for grpc.lib was 325 MB; while the size of the release build of Harlinn.grpc.dll is just 3.7 MB. Currently the combined size of all the native dlls created for a release build of the solution is 118 MB.
Some packages relied on different versions of the same package, including their own private copy of a package.
Some packages were compiled with /sdl on, while others were not. I prefer to turn /sdl and /GS off for release builds, and on for debug builds. This may not be as significant as it used to, but you still get better performance when turning /sdl and /GS off.
Many packages will not compile with the /std:c++latest option, which enables all currently implemented compiler and standard library features proposed for the next draft standard, as well as some in-progress and experimental features.
Many packages creates static libraries, not dlls.

Don't get me wrong; I think CMake and vcpkg are great tools, but putting everything into a single solution makes it easier, at least for me, to understand how everything is tied together.

Why put Harlinn in front of every dll name?

To avoid naming conflicts. I own the harlinn.com domain, so it is unlikely that anybody would, by accident, put Harlinn in front of the name of their dll. These dlls are not drop-in replacements for the dlls that vcpkg and cmake generate, so I think it makes sense to ensure that they are uniquely named.

What about the unit tests?

A few of those are included, but still just a fraction of those available for the original packages.

In particular, I've included many of the unit tests, that can be expected to compile on Windows, for:

Abseil - 4430 tests pass, 26 fails, and 12 tests are disabled
boringssl
- crypto_test - 910 tests pass, 0 fails, and 0 tests are disabled
- ssl_test - 330 tests pass, 0 fails, and 0 tests are disabled
gRPC - the tests are implemented as 94 separate executables, using multiple approaches to testing. I think that the tests pass; but I am not certain, since much of the validation was performed manually.
bullet - 31 tests pass, 0 fails, and 0 tests are disabled
gdal - 84 tests pass, 0 fails, and 0 tests are disabled
skia - A fair number of the tests works, about 1500 of them, but then the test program terminates. As far as I have been able to determine, this seems to be related to the test code.

I've also created a fair number of tests for my own projects:

Harlinn.Common.Core.Tests: 3192
Harlinn.Windows.Tests: 271
Harlinn.OCI.Tests: 49
Harlinn.ODBC.Tests: 305
Harlinn.Timeseries.Tests: 10
Harlinn.Julia.Tests: 11

All tests can be found under the Tests folder.

What about the examples?

So far I've included 202 example projects.

All Examples can be found under the Examples and DotNet\Examples folders.

How stable is this?

This is a snapshot of the solution, and while most things seems to work, there are bound to be some snags. The most common problem will be a function or variable that is not exported from a dll, and there are probably other bugs too. Some, probably, created by yours truly.

Still, with this solution you can now:

Create nice 2D effects:
Create animated 3D visualizations:
Learn a lot about some of the most widely used open source packages available, using Visual Studio 2022 and C++.

Harlinn.Common.Core

Harlinn.Common.Core is a library packaged as a dll that provide functionality that is useful for most Windows C++ projects.

Dependencies

The boost C++ libraries

Other 3rd party code included in Harlinn.Common.Core

The Harlinn.Common.Core library contains code from a number of libraries:

The xxHash library, version 0.8.1
The CRC-32C (Castagnoli) for C++ library
The xxhash_cx library

About `char`, `wchar_t`, `signed char` and `unsigned char`

The library tries to treat char and wchar_t as types used for the sole purpose of representing character data, while signed char and unsigned char are expected to represent signed and unsigned 8-bit integers, respectively.

Normally this is not much of a problem, but consider:

constexpr char buffer1[] = { 1,2,3 };
constexpr unsigned char buffer2[] = { 1,2,3 };
constexpr char buffer3[] = "hello";
constexpr int buffer4[] = { 1,2,3 };
constexpr wchar_t_ buffer5[] = { 1,2,3 };
constexpr wchar_t_ buffer6[] = L"hello";
constexpr auto length1 = LengthOf( buffer1 );
constexpr auto length2 = LengthOf( buffer2 );
constexpr auto length3 = LengthOf( buffer3 );
constexpr auto length4 = LengthOf( buffer4 );
constexpr auto length5 = LengthOf( buffer5 );
constexpr auto length6 = LengthOf( buffer6 );

LengthOf expects buffer1 and buffer5 to be zero-terminated; but since they are not, the compiler will not be able to evaluate LengthOf( buffer1 ) and LengthOf( buffer5 ). LengthOf( buffer3 ) and LengthOf( buffer6 ) evaluate to 5, while LengthOf( buffer2 ) and LengthOf( buffer4 ) evaluate to 3.

The code in HCCPersistent.h, which is used by the code in HCCLogger.h, relies on this, eliminating a significant number of calls to strlen and wcslen at runtime.

Main header files:

HCC7BitEncoding.h: provides constexpr support for 7-bit encoded values.
HCCActiveObject.h: provides:
- template<typename DerivedT> class ActiveObject:
  
  An active object implementation based on ConcurrentQueue<>.

HCCApplication.h: provides:

class Application:

Every program needs to create one, and only one, instance of this class, or a class derived from this class.

int main( int argc, char* argv[], char* envp[] )
{
    Harlinn::Common::Core::ApplicationOptions options;
    options.Load( );
    Harlinn::Common::Core::Application application( options );
    application.Start( );

    // your code goes here, perhaps:
    int result = foo( );

    application.Stop( );
    return result;
}

Currently this configures, starts and stops the logging back-end on behalf of the program.

HCCArray.h: provides:
- template<typename T, size_t N> ArrayEx:
  
  A template derived from std::array that supports constexpr concatenation using operator +.
- template<typename T, size_t N> class Array:
  
  An alternative to std::array that supports constexpr concatenation using operator +.
- template<size_t N> class ByteArray:
  
  A template that supports constexpr serialization of binary data.
- template<size_t N, bool networkByteOrder = false, bool use7BitEncodedSize = true> class ArrayReader:
  
  A template that implements constexpr deserialization of data.
- template<size_t N, bool networkByteOrder = false, bool use7BitEncodedSize = true> class ArrayWriter:
  
  A template that implements constexpr serialization of data.

HCCBinaryReader.h: provides:

BinaryReaderBase:

template<typename Derived, 
        bool networkByteOrder, 
        bool use7BitEncodedSize>
class BinaryReaderBase

A template that can be used to deserialize binary data.

BufferedBinaryReader:

template<typename StreamT, 
        size_t BufferSizeValue, 
        bool networkByteOrder = false, 
        bool use7BitEncodedSize = true> 
class BufferedBinaryReader

A template that can be used to deserialize binary data from a stream using an internal buffer.

BinaryReader:

template< IO::StreamReader StreamT, 
        bool networkByteOrder = false, 
        bool use7BitEncodedSize = true >
class BinaryReader

A template that can be used to deserialize binary data from a stream.

HCCBinaryWriter.h: provides:

BinaryWriterBase:

template<typename Derived, 
        bool networkByteOrder, 
        bool use7BitEncodedSize>
class BinaryWriterBase

A template that can be used to serialize binary data.

BufferedBinaryWriter:

template<IO::StreamWriter StreamWriterT, 
        size_t BufferSizeValue, 
        bool networkByteOrder = false, 
        bool use7BitEncodedSize = true >
class BufferedBinaryWriter

A template that can be used to serialize binary data to a stream using an internal buffer.

BinaryWriter:

template<IO::StreamWriter StreamT, 
        bool networkByteOrder = false, 
        bool use7BitEncodedSize = true>
class BinaryWriter

A template that can be used to serialize binary data to a stream.

HCCBuffer.h: provides Blocks::Stream a block oriented memory stream.
HCCCom.h: provides:
- class Unknown: A wrapper for the IUnknown interface.
- template<typename T> class UnknownPtr: A generic smart pointer for COM interfaces.
- template< class InterfaceType > class IUnknownImplementation: An implementation of the IUnknown interface.

HCCConverters.h: provides various overloads of:

  template <typename ResultType, typename ArgumentType>
  inline ResultType ConvertTo( ArgumentType arg )

implementing efficient and 'safe' conversion between basic data types.

HCCCRC.h: provides a few ways of calculating the CRC value for data.
HCCCrypto.h: provides functionality for working with certificates.
HCCCurrency.h: declares the Currency class.
HCCDateTime.h: provides functionality for handling date and time, declares:
- class TimeSpan
- class DateTime
- class Stopwatch
HCCEnvironment.h: provides easy access to system information
HCCEse.h: provides C++ classes for working with the Extensible Storage Engine
HCCException.h: provides a wide range of exception classes.
HCCGuid.h: functionality for working with UUIDs/GUIDs, declares class Guid.
HCCHandle.h: declares template<typename DerivedT, typename HandleT> class Handle
HCCHTTP.h: provides C++ classes for working with the HTTP Server API.
HCCIO.h: provides file, directory, and stream I/O related C++ classes.
HCCIOContext.h: declares:
- class Context: uses a thread-pool to process asynchronous I/O for an I/O completion port
- class ContextHandler: An instance of the Context class delegates the work for a particular OS handle to the Process override implemented for a class derived from ContextHandler.
- template<typename DerivedT> FileHandler: A template derived from ContextHandler used to perform file related asynchronous I/O.
- class FileRequest: Base class for file related asynchronous I/O requests.
- class ReadFileRequest: Represents an asynchronous file read request.
- class WriteFileRequest: Represents an asynchronous file write request.
- class FileDispatcher: Derived from FileHandler, broadcasts I/O completion events to callbacks attached to boost::signals2::signal<> handlers.
HCCLib.h: provides a number of useful funtions:
- ByteSwap:
```
template<typename T>
    requires IsInteger<std::remove_cvref_t<T>>
inline constexpr T ByteSwap( T value )
```
  Swaps the bytes of the argument, which must be some type of integer. Performs this operation, if possible, at compile-time. If the operation must be performed at runtime, the implementation will call _byteswap_ushort, _byteswap_ulong, _byteswap_uint64, or just return the argument; based on the byte size of the argument type. ByteSwap for float, double and enums are implemented as:
```
template<typename T>
    requires (IsFloatingPoint<std::remove_cvref_t<T>> || std::is_enum_v<std::remove_cvref_t<T>> )
inline constexpr T ByteSwap( const T value ) noexcept
{
    using Type = std::remove_cvref_t<T>;
    using UIntType = MakeUnsigned<Type>;
    std::bit_cast<Type>( ByteSwap( std::bit_cast<UIntType>( value ) ) );
}
```
- BitMask:
```
template<size_t N>
struct BitMask
{
    using type = ...
    static constexpr type value = ...;
};
```
  BitMask::type: is the type of the smallest unsigned integer capable of holding a bit-mask of N bits, where N <= 64.
  
  BitMask::value: is the value of the bit-mask with the N lowest bits set to 1.
- LengthOf: returns the number of elements in an array, length of a char or wchar_t zero-terminated string, or the number of elements in a container.
- template<typename T> inline int Compare( T v1, T v2 ) noexcept: compares two arguments of the same basic type, returning < 0 if the first value is considered to be less than the second; 0 if they are considered equal, and otherwise > 0. For char or wchar_t zero-terminated strings, nullptr is considered equal to a string with 0 length.
- AreNearlyEqual:
```
template<typename T>
    requires std::is_floating_point_v<T>
constexpr bool AreNearlyEqual( T a, T b, T smallNumber = static_cast<T>( 0.0001 ) ) noexcept
```
  returns true if the absolute value of the difference between a and b is less than smallNumber.
- class XXH64Hasher: use this class to incrementally generate a hash value from multiple inputs.
- IsPowerOfTwo: returns true if an integer is a power of 2.
- NextPowerOfTwo: returns the next higher value, that is a power of 2, greater than the argument.
HCCLMDB.h: declares a set of C++ classes for working with LMDB based on 3rdParty\Harlinn.LMDB.
HCCLogger.h: logger implementation.
HCCLogging.h: preprocessor macros used for logging.

HCCMath.h: constexpr math, namespace: Harlinn::Common::Core::Math

IsSameValue:

constexpr bool IsSameValue( double x, double y ) noexcept;
constexpr bool IsSameValue( float x, float y ) noexcept;

Returns true if x and y holds the same value, even if they hold NaN values.

IsZero:

constexpr bool IsZero( double x ) noexcept;
constexpr bool IsZero( float x ) noexcept;

Returns true if x is 0.0 or -0.0.

signum:

template< typename T>
    requires IsUnsignedInteger<std::remove_cvref_t<T>> || IsBoolean<std::remove_cvref_t<T>>
inline constexpr T signum( T val ) noexcept;

template< typename T>
    requires IsSignedInteger<std::remove_cvref_t<T>>
inline constexpr T signum( T val ) noexcept;

template< typename T>
    requires IsFloatingPoint<std::remove_cvref_t<T>>
inline constexpr T signum( T val ) noexcept;

Indicates the sign of val, returning -1, 0, or +1.

Deg2Rad:

template< typename T>
    requires IsFloatingPoint<std::remove_cvref_t<T>>
inline constexpr std::remove_cvref_t<T> Deg2Rad( T angleInDegrees ) noexcept;

Convert angleInDegrees from degrees to radians.

Rad2Deg:

template< typename T>
    requires IsFloatingPoint<std::remove_cvref_t<T>>
inline constexpr std::remove_cvref_t<T> Rad2Deg( T angleInRadians ) noexcept;

Convert angleInRadians from radians to degrees.

NextAfter:

inline constexpr double NextAfter( double x, double y ) noexcept;
inline constexpr double NextAfter( float x, float y ) noexcept;

Returns the next representable value of x in the direction of y.

NextDown:

template<typename T>
    requires IsFloatingPoint<T>
inline constexpr std::remove_cvref_t<T> NextDown( T x ) noexcept;

Returns the next representable value of x in the direction of negative infinity.

NextUp:

template<typename T>
    requires IsFloatingPoint<T>
inline constexpr std::remove_cvref_t<T> NextUp( T x ) noexcept;

Returns the next representable value of x in the direction of positive infinity.

IsNaN:

template<typename T>
    requires IsInteger<T>
constexpr inline bool IsNaN( T val ) noexcept;

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline bool IsNaN( T val ) noexcept;

Returns true if val is NaN.

IsInf:

template<typename T>
    requires IsInteger<T>
constexpr inline bool IsInf( T val ) noexcept;

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline bool IsInf( T val ) noexcept;

Returns true if val is infinite.

Abs:

template<typename T>
    requires IsSignedInteger<T>
constexpr inline std::remove_cvref_t<T> Abs( T val ) noexcept;

template<typename T>
    requires IsUnsignedInteger<T>
constexpr inline std::remove_cvref_t<T> Abs( T val ) noexcept;

template<typename T>
    requires IsFloatingPoint<T> 
constexpr inline std::remove_cvref_t<T> Abs( T val ) noexcept;

Returns the absolute value of val.

SignBit:

template<typename T>
    requires IsSignedInteger<T>
constexpr inline bool SignBit( T val ) noexcept;

template<typename T>
    requires IsUnsignedInteger<T>
constexpr inline bool SignBit( T val ) noexcept;

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline bool SignBit( T val ) noexcept;

Returns true if the value of the sign of val is negative, otherwise false.

FRExp:

template<typename T>
    requires IsInteger<T>
constexpr inline std::pair<double,int> FRExp( T val ) noexcept;

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::pair<double,int> FRExp( T val ) noexcept;

template<typename T>
    requires IsInteger<T>
constexpr inline double FRExp( T val, int* exp ) noexcept;

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> FRExp( T val, int* exp ) noexcept;

template<typename T>
    requires IsInteger<T>
constexpr inline double FRExp( T val, int& exp ) noexcept;

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> FRExp( T val, int& exp ) noexcept;

Return (result,exp) such that result has a magnitude in the interval [1/2, 1) or 0, and val is equal to result * 2^exp.

ModF:

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::pair<std::remove_cvref_t<T>, std::remove_cvref_t<T>> ModF( T val ) noexcept;

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> ModF( T val, T* integerPart ) noexcept;

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> ModF( T val, T& integerPart ) noexcept;

Return a std::pair<T,T> of the fractional and integral parts of a number. Both parts have the same sign as the argument.

Min:

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> Min( T first, T second ) noexcept;

template<typename T>
    requires IsInteger<T>
constexpr inline std::remove_cvref_t<T> Min( T first, T second ) noexcept;

Returns the smaller of the argument values.

Max:

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> Max( T first, T second ) noexcept;

template<typename T>
    requires IsInteger<T>
constexpr inline std::remove_cvref_t<T> Max( T first, T second ) noexcept;

Returns the greater of the argument values.

Trunc:

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> Trunc( T val ) noexcept;

template<typename T>
    requires IsInteger<T>
constexpr inline std::remove_cvref_t<T> Trunc( T val ) noexcept;

Returns the nearest integral value of the same type as val whose absolute value is less than or equal to the absolute value of val.

Floor:

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> Floor( T val ) noexcept;

template<typename T>
    requires IsInteger<T>
constexpr inline std::remove_cvref_t<T> Floor( T val ) noexcept;

Returns the nearest integral value of the same type as val that is less than or equal to val.

Ceil:

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> Ceil( T val ) noexcept;

template<typename T>
    requires IsInteger<T>
constexpr inline std::remove_cvref_t<T> Ceil( T val ) noexcept;

Returns the nearest integral value of the same type as val that is greater than or equal to val.

Round:

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> Round( T val ) noexcept;

template<typename T>
    requires IsInteger<T>
constexpr inline std::remove_cvref_t<T> Round( T val ) noexcept;

Returns val rounded to the nearest integral value.

Clamp:

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> Clamp( T val, T minimumValue, T maximumValue ) noexcept;

template<typename T>
    requires ( IsInteger<T> )
constexpr inline std::remove_cvref_t<T> Clamp( T val, T minimumValue, T maximumValue ) noexcept;

If val compares less than minimumValue, returns minimumValue; otherwise if maximumValue compares less than val, returns maximumValue; otherwise returns val.

Lerp:

template<typename T, typename U>
    requires ( ( IsInteger<T> || IsFloatingPoint<T> ) && ( IsInteger<U> || IsFloatingPoint<U> ) )
constexpr inline std::conditional_t<IsFloatingPoint<T>, T, std::conditional_t<IsFloatingPoint<U>, U, T>> Lerp( T a, T b, U t ) noexcept;

Computes the linear interpolation between a and b for the parameter t (or extrapolation, when t is outside the range [0,1]).

CopySign:

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> CopySign( T magnitude, T signValue ) noexcept;

Returns a value with the magnitude of magnitude and the sign of signValue.

ScaleByN:

inline constexpr double ScaleByN( double x, int n ) noexcept;
inline constexpr double ScaleByN( float x, int n ) noexcept;

Multiplies a floating point value x by FLT_RADIX raised to power n.

FMod:

inline constexpr double FMod( double x, double y ) noexcept;
inline constexpr float FMod( float x, float y ) noexcept;

Computes the floating-point remainder of the division operation x/y.

Exp:

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> Exp( T x ) noexcept;

Computes e (Euler's number, 2.7182818) raised to the given power of x.

Hypot:

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> Hypot( T x, T y ) noexcept;

Computes the hypotenuse.

Log:

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> Log( T x ) noexcept;

Computes the natural (base e) logarithm of x.

Log2:

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> Log2( T x ) noexcept;

Computes the base 2 logarithm of x.

Log10:

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> Log10( T x ) noexcept;

Computes the base 10 logarithm of x.

Sin:

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> Sin( T x ) noexcept;

Computes the sine of x, where x is in radians.

ASin:

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> ASin( T x ) noexcept;

Computes the inverse sine of x, in radians.

Cos:

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> Cos( T x ) noexcept;

Computes cosine of x, where x is in radians.

ACos:

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> ACos( T x ) noexcept;

Computes the inverse cosine of x, with the result in radians.

Tan:

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> Tan( T x ) noexcept;

Computes tangent of x, where x is in radians.

ATan, ATan2:

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> ATan( T x ) noexcept;

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> ATan( T x, T y ) noexcept

template<typename T>
    requires IsFloatingPoint<T>
constexpr inline std::remove_cvref_t<T> ATan2( T x, T y ) noexcept

Compute the inverse tangent of y or y/x, respectively.

SinCos:

template<typename T>
    requires IsFloatingPoint<T>
inline constexpr void SinCos( T x, T& sinResult, T& cosResult ) noexcept;

Simultaneously compute the sine and cosine of x, where x is in radians.

HCCObj.h: provides wrapper C++ classes for a wide range of Windows COM interfaces.
HCCObjBase.h: provides wrapper C++ classes for central Windows COM interfaces.
HCCObservable.h: declares template<typename DerivedT> class Observable and a couple macros that can be used to implement observable properties.
HCCPersistent.h: provides a flexible and efficient serialization and deserialization mechanism.
HCCPersistentDecoder.h: provides a generic decoder for data serialized using the mechanism provided in HCCPersistent.h.
HCCPipe.h: Synchronous and synchronous I/O using named pipes.
HCCProcess.h: classes for working with Windows processes.
HCCPropSys.h: classes for working with the (Windows Property System)[https://docs.microsoft.com/en-us/windows/win32/api/_properties/]
HCCReference.h: support for intrusive reference counted objects.
HCCSecurity.h: classes for working with Windows Security and identity
HCCShell.h: classes for working with the Windows Shell API.
HCCSocket.h: Synchronous and synchronous I/O using sockets.
HCCSync.h: classes for working with synchronization.
HCCThread.h: classes for working with threads.
HCCTraits.h: useful meta programming features.
HCCTuple.h: yet another tuple implementation.
HCCVariant.h: provides functionality for working with the VARIANT structure.
HCCVector.h: yet another vector implementation.
HCCXml.h: classes for working with the MSXML SDK.

Harlinn.Windows

Harlinn.Windows provides functionality for developing user interfaces in C++.

It is used to implement 73 of the examples.

Main header files:

HWDXGI.h: provides support for the DirectX Graphics Infrastructure API.
HWGraphicsD3D12.h: provides support for the Direct3D 12 API.
HWGraphics.h: provides support for the Direct2D API.
HWImaging.h: provides support for the Windows Imaging Component API
HWHandles.h: provides support for the Windows GDI API.
HWControl.h: provides support for working with Windows Controls
HWMessage.h: provides the Message class, a simple abstraction for the MSG structure
HWMenu.h: provides support for working with Windows Menus.
HWForm.h: provides support for working with top level Windows Controls using the Form class.
HWApplication.h: provides an Application class and a MessageLoop class.
HWDXApplication.h: declares some Direct3D 12 utility classes:
- class DXMessageLoop: The render/message loop.
- class DXContext: Initializes and manages interaction with the D3D12Device, the DXGI::SwapChain3, the D3D12GraphicsCommandList, and the few other objects required for interaction with the Direct3D 12 API.
- class DXApplication: Derived from Windows::Application, tailored for developing Direct3D 12 using Windows::Form.
HWStdCtrls.h: provides a number of classes for working with standard Windows controls. The provided functionality is, unfortunately, quite limited.
HWFileDialog.h: provides support for the Open and Save As functionality from the Common Dialog Box Library.
HWEnvironment.h: provides easy access to system information.
HWSkia.h: provides support for creating Skia based applications.
HWImGui.h: provides support for creating Dear ImGui based applications.

HWControl.h, HWMessage.h, HWMenu.h, HWForm.h, HWApplication.h, HWStdCtrls.h, HWFileDialog.h, HWSkia.h are meant to be used together, while HWDXGI.h, HWGraphicsD3D12.h, HWGraphics.h, HWImaging.h and HWHandles.h can be used on their own.

These days I often use the slightly modified version of Dear ImGui and ImPlot located under 3rdParty\Harlinn.ImGui to create user interfaces for my Harlinn::Windows::Form based applications.

#include "HWDXApplication.h"
#include "HWImGui.h"
#include "HWmenu.h"
#include "HWstdctrls.h"
#include <iostream>

using namespace Harlinn;

int main()
{
    try
    {
        Windows::ApplicationOptions applicationOptions;
        applicationOptions.Load( );
        Windows::ImGui::Application application( applicationOptions );
        application.Start( );

        Windows::ImGui::Form form;
        form.SetText( L"ImGui & ImPlot Demo" );

        bool showImGuiDemoWindow = true;
        bool showImPlotDemoWindow = true;
        bool showAnotherWindow = false;

        form.OnRender.connect( [&showImGuiDemoWindow,&showImPlotDemoWindow, &showAnotherWindow]( Windows::ImGui::Form* sender )
        {
            if ( showImGuiDemoWindow )
            {
                ImGui::ShowDemoWindow( &showImGuiDemoWindow );
            }
            if ( showImPlotDemoWindow )
            {
                ImPlot::ShowDemoWindow( &showImPlotDemoWindow );
            }

        } );

        auto result = application.Run( form );

        application.Stop( );

        return result;
    }
    catch ( std::exception& exc )
    {
        std::cout << exc.what( ) << std::endl;
    }
    catch ( ... )
    {
        std::cout << "Unknown exception" << std::endl;
    }
    return 0;
}

Harlinn.OCI

Harlinn.OCI provides high performance access to the Oracle RDBMS using the Oracle OCI API

Harlinn.ODBC

Harlinn.ODBC provides high performance access to database servers and engines through the ODBC API.

Harlinn.Julia

This library provides easy access to Julia.

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Overview.md

Overview.md

Overview

Introduction

Why did I not use cmake and vcpkg?

Why put Harlinn in front of every dll name?

What about the unit tests?

What about the examples?

How stable is this?

Harlinn.Common.Core

Dependencies

Other 3rd party code included in Harlinn.Common.Core

About `char`, `wchar_t`, `signed char` and `unsigned char`

Main header files:

Harlinn.Windows

Main header files:

Harlinn.OCI

Harlinn.ODBC

Harlinn.Julia

Files

Overview.md

Latest commit

History

Overview.md

File metadata and controls

Overview

Introduction

Why did I not use cmake and vcpkg?

Why put Harlinn in front of every dll name?

What about the unit tests?

What about the examples?

How stable is this?

Harlinn.Common.Core

Dependencies

Other 3rd party code included in Harlinn.Common.Core

About char, wchar_t, signed char and unsigned char

Main header files:

Harlinn.Windows

Main header files:

Harlinn.OCI

Harlinn.ODBC

Harlinn.Julia

About `char`, `wchar_t`, `signed char` and `unsigned char`