libcaffeine

This is a library for setting up broadcasts and streaming on Caffeine.tv.

Libcaffeine is licensed under the GNU GPL version 2. See LICENSE.txt for details.

For information on contributing to this project, see CONTRIBUTING.md.

TODO: More Documentation

Using the library

Minimal Example:

// Callbacks
void started(void *myContext) {
    // Start capturing audio & video
}

void failed(void *myContext, caff_Result result) {
    // Handle error
    // Stop capturing if necessary
}

// Once on startup
caff_initialize("example", "1.0", caff_SeverityDebug, NULL);

// Starting a broadcast
caff_InstanceHandle instance = caff_createInstance();
caff_Result result = caff_signIn(instance, "Myuser", "|\\/|Yp455WYrd", NULL);
if (result != caff_ResultSuccess) return -1;
result = caff_startBroadcast(instance, &myContext, "My title!", caff_RatingNone, NULL, started, failed);

// On your video/audio capture thread(s)
caff_sendVideo(
    instance, caff_VideoFormatNv12, pixels, pixelBytes, width, height, caff_TimestampGenerate);
caff_sendAudio(instance, samples, 2);

// When it's over:
caff_endBroadcast(instance);
caff_freeInstance(&instance);

Build instructions

Windows

Prereqs:

• Visual Studio 2019 with "Desktop development with C++" selected and "C++ Clang tools for Windows" installed.

Steps:

• Download the latest version of webrtc-prebuilt-windows.7z from https://github.com/caffeinetv/webrtc/releases
• Extract the file somewhere convenient
• Set WEBRTC_ROOT_DIR environment variable to the directory you extracted the files to
• Download obsdeps "https://obsproject.com/downloads/dependencies2019.zip" . Extract file
• Set DepsPath environment variable to directory you extracted files to. This is needed to get curl
• In the libcaffeine root directory:
  • mkdir build64
  • cmake -H. -Bbuild64 -G "Visual Studio 16 2019" -A "x64" -T "ClangCL" -D BUILD_TESTS=OFF -DWEBRTC_ROOT_DIR="%WEBRTC_ROOT_DIR%" -DDepsPath="%DepsPath%\win64\include" -DCMAKE_INSTALL_PREFIX=dist -DCMAKE_C_COMPILER="%VSINSTALLATION%/VC/Tools/Llvm/bin/clang.exe" -DCMAKE_CXX_COMPILER="%VSINSTALLATION%/VC/Tools/Llvm/bin/clang++.exe"
  • start libcaffeine.sln
• Build the solution in Visual Studio

Mac

Prereqs:

• Xcode
• CMAKE

Steps:

• Download the latest version of webrtc-prebuilt-macos.7z from https://github.com/caffeinetv/webrtc/releases
• Extract the archive somewhere convenient
• Set WEBRTC_ROOT_DIR environment variable to the directory you extracted the files to
• In the libcaffeine root directory:
  • mkdir build
  • cd build
  • cmake .. -G Xcode
  • Build the project in Xcode

Note: if the cmake command above fails, it may be due to the XCode commandline tools being selected. To resolve this, reset your XCode selection and retry: sudo xcode-select -r

TODO: Linux

Packaging Instructions

Prerequisites

To run the build

Windows

• Download 7z
• Add 7z.exe to system Path env variable.
• Add msbuild binary to system PATH env var
• Add cmake binary to system PATH env var.
• Add LLVM bin to system PATH env var.

Mac

• Install using homebrew

  brew install p7zip 

  Automated steps:

Change directory to be the libcaffeine directory.

Windows:

• Run the automated script in command prompt

  build.bat [OPTION] 

  Mac

• Run the automated script in Terminal

  sh build.sh [OPTION] 

  Supported options

  Options	Usage
  -check	To check project prerequisites
  -build	Builds the project
  -package	Packages libcaffein with 7z
  -buildAndPackage	Builds and packages libcaffeine.

The packaged output will be in build directory in Mac and dist directory in Windows.

Manual steps:

Windows:

• Build libcaffeine for both 32- and 64-bit windows
• Create a folder named libcaffeine-vX.Y-windows
• Inside that create these folders:
  • bin32
  • bin64
  • include
  • lib32
  • lib64
• Copy caffeine.h into include
• Copy the .dll and .pdb files into the binXX folders
• Copy the .lib files into the libXX folders
• From CMakeFiles/Export/libXX/cmake folder copy libcaffeineTarget.cmake, libcaffeineTarget-debug.cmake , libcaffeineTarget-debug.cmake into libXX/cmake
• Copy libcaffeineConfig.cmake into libXX/cmake
• Copy libcaffeineConfigVersion.cmake into libXX/cmake
• 7z a libcaffeine-vX.Y-windows libcaffeine-vX.Y-windows.7z

Mac:

• Build for macOS
• Create a folder named libcaffeine-vX.Y-macos
• Inside that create these folders:
  • binXX
  • include
  • libXX
• Copy caffeine.h into include
• Copy libcaffeine.dylib into binXX
• Copy libcaffeined.a (static lib) and libcaffeined.dylib into libXX
• Copy libcaffeine.a (static lib) and libcaffeine.dylib into libXX
• From CMakeFiles/Export/libXX/cmake folder copy libcaffeineTarget.cmake, libcaffeineTarget-debug.cmake , libcaffeineTarget-relwithdebinfo.cmake into libXX/cmake
• Copy libcaffeineConfig.cmake into libXX/cmake
• Copy libcaffeineConfigVersion.cmake into libXX/cmake
• 7z a libcaffeine-vX.Y-macos.7z libcaffeine-vX.Y-macos

Metadata

• Team: Frontend / Desktop
• Product Owner: Viral Desai
• Engineer: Prateeksha Das
• Engineer: David Chen