AuroraMiddleware/Vulkan-Hpp
1
0
Fork 0
mirror of https://github.com/KhronosGroup/Vulkan-Hpp synced 2024-11-29 07:01:08 +00:00
Code Issues Releases Wiki Activity

Merge pull request #41 from mtavenrath/readme

Browse Source 
Update README.md with tiny code samples on how to use the new C++ obj…
This commit is contained in:
asuessenbach 2016-02-24 13:54:53 +01:00
commit 766f69993c
1 changed files with 30 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

50
README.md
View File

@ -143,30 +143,29 @@ vk::createImage(device, &ci, allocator, &image));
# Enhancements beyond the API # Enhancements beyond the API
While mapping the Vulkan API to C++ without adding new functions is already a big help, one can do even more by adding new functionality. For example several C++ developers tend to use std::string and std::vector in their code, therefore we have added some more optional convenience features: To provide a more object oriented feeling we're providing classes for each handle which include all Vulkan functions where the first
parameter matches the handle. In addition to this we made a few changes to the signatures of the member functions
* Use <code>std::string</code> instead of <code>const char *</code> for strings * <code>(count, T*)</code> has been replaced by <code>std::vector&lt;T&gt;</code>
* Use <code>std::vector</code> instead of <code>(count, ptr)</code> for sized arrays * <code>const char *</code> has been replaced by <std::string>
* Throw exceptions instead of error return values (in progress) * vk::Result return values have been replaced by exceptions.
* Return handles/vectors where applicable, i.e. for the create* functions * Functions with a single output value do return this value instead
As example let's examine the device extension property enumeration in Vulkan:
Here are a few code examples:
<pre> <pre>
<code> <code>
uint32_t count; try {
VK_VERIFY(vk::enumerateDeviceExtensionProperties(physicalDevice, layerName.c_str(), &count, nullptr)); vk::Instance i = ...;
std::vector<vk::PhysicalDevice> physicalDevices = i.enumeratePhysicalDevices();
vk::FormatProperties formatProperties = physicalDevices[0].getFormatProperties(vk::Format::eR8G8B8A8Unorm);
std::vector<vk::ExtensionProperties> properties(count); vk::CommandBuffer commandBuffer = ...;
VK_VERIFY(vk::enumerateDeviceExtensionProperties(physicalDevice, layerName.c_str(), &count, properties.data())); vk::Buffer buffer = ...;
</code> commandBuffer.updateBuffer(buffer, 0, {some values}); // update buffer with std::vector
</pre> }
catch (vk::Exception e)
Luckily the official Khronos-provided vk.xml has enough information to figure out which pair of values represents a sized array or strings, so that it is possible to generate a function which allows you to write the following line of code instead: {
std::cerr << "Vulkan failure: " << e.what() << std::endl;
<pre> }
<code>
std::vector&lt;ExtensionProperties&gt; properties = vk::enumerateDeviceExtensionProperties(physicalDevice, layerName);
</code> </code>
</pre> </pre>
@ -195,3 +194,14 @@ before this line:
* Update submodules: git submodule update --init --recursive * Update submodules: git submodule update --init --recursive
* Use CMake to generate a solution or makefile for your favourite build environment * Use CMake to generate a solution or makefile for your favourite build environment
* Launch the build * Launch the build
# Providing Pull Requests
NVIDIA is happy to review and consider pull requests for merging into the main tree of vkcpp for bug fixes and features. Before providing a pull request to NVIDIA, please note the following:
* A pull request provided to this repo by a developer constitutes permission from the developer for NVIDIA to merge the provided
changes or any NVIDIA modified version of these changes to the repo. NVIDIA may remove or change the code at any time and in any
way deemed appropriate. Due to the required paperwork please refrain from providing pull requests for simple changes and file an issue
describing a bug or the desired change instead.
* Not all pull requests can be or will be accepted. NVIDIA will close pull requests that it does not intend to merge.
* The modified files and any new files must include the unmodified NVIDIA copyright header seen at the top of all shipping files.