Giter VIP home page Giter VIP logo

rust-skia's Introduction

Safe Rust bindings to the Skia Graphics Library.

crates.io license Build Status

Skia Submodule Status: chrome/m90 (upstream changes, our changes).

Goals

This project provides up to date safe bindings that bridge idiomatic Rust with Skia's C++ API on desktop and mobile platforms, including GPU rendering backends for Vulkan, Metal, OpenGL, and Direct3D.

Status

Documentation

Because we can't build on docs.rs, the cargo doc output for skia-safe is manually created and uploaded to rust-skia.github.io.

We (slowly) add more documentation by converting Skia's Doxygen comments to Rust.

Crate

A prerelease crate is available from crates.io and adding

[dependencies]
skia-safe = "0"

to your Cargo.toml should get you started.

On Linux you may run into trouble when OpenSSL libraries are missing. On Debian and Ubuntu they can be installed with:

sudo apt-get install pkg-config libssl-dev

For other platforms, more information is available at the OpenSSL crate documentation.

Platform Support, Build Targets, and Prebuilt Binaries

Because building Skia takes a lot of time and needs tools that may be missing, the skia-bindings crate's build.rs tries to download prebuilt binaries from the skia-binaries repository.

Platform Binaries
Windows x86_64-pc-windows-msvc
Linux Ubuntu 16, 18
CentOS 7, 8
x86_64-unknown-linux-gnu
macOS x86_64-apple-darwin
Android aarch64-linux-android
x86_64-linux-android
iOS aarch64-apple-ios
x86_64-apple-ios

There no support for WebAssembly yet. If you'd like to help out, take a look at issue #39.

Wrappers & Codecs & Supported Features

The supported wrappers, Skia codecs, and additional Skia features are documented in the skia-safe package's readme. Prebuilt binaries are available for most feature combinations.

Building

If the target platform or feature configuration is not available as a prebuilt binary, skia-bindings' build.rs will try to build Skia and generate the Rust bindings.

To prepare for that, LLVM and Python 2 are needed:

LLVM

We recommend the version that comes preinstalled with your platform, or, if not available, the latest official LLVM release. To see which version of LLVM/Clang is installed on your system, use clang --version.

Python 2

Python version 2.7 must be available.

The build script probes for python --version and python2 --version and uses the first one that looks like a version 2 executable for building Skia.

On macOS

  • Install the Command Line Tools for Xcode with

    xcode-select --install

    or download and install the Command Line Tools for Xcode.

  • macOS Mojave only: install the SDK headers:

    sudo open /Library/Developer/CommandLineTools/Packages/macOS_SDK_headers_for_macOS_10.14.pkg

    If not installed, the Skia build may fail to build SkJpegUtility.cpp and the binding generation will fail with 'TargetConditionals.h' file not found . Also note that the Command Line Tools and SDK headers should be reinstalled after an update of XCode.

  • As an alternative to Apple's XCode LLVM, install LLVM via brew install llvm or brew install llvm and then set PATH, CPPFLAGS, and LDFLAGS like instructed.

    If the environment variables are not set, bindgen will most likely use the wrong libclang.dylib and cause confusing compilation errors (see #228).

On Windows

  • Have the latest versions of git and Rust ready.

  • Install Visual Studio 2019 Build Tools or one of the IDE releases. If you installed the IDE, make sure that the Desktop Development with C++ workload is installed.

  • Install the latest LLVM distribution.

    If the environment variable LLVM_HOME is not defined, the build script will look for LLVM installations located at C:\Program Files\LLVM\, C:\LLVM\, and %USERPROFILE%\scoop\apps\llvm\current\.

  • MSYS2:

    • Install Python2 with pacman -S python2.
  • Windows Shell (Cmd.exe):

    • Download and install Python version 2 from python.org.
  • Install and select the MSVC toolchain:

    rustup default stable-msvc

On Linux

Ubuntu 16+

  • LLVM/Clang should be available already, if not, install the latest version.

  • If OpenGL libraries are missing, install the drivers for you graphics card, or a mesa package like libgl1-mesa-dev.

CentOS 7

CentOS 8

  • Install the following packages:

    sudo yum install gcc openssl-devel libX11-devel python2 clang fontconfig-devel mesa-libGL-devel
  • Set /usr/bin/python2 as the default python command:

    sudo alternatives --set python /usr/bin/python2

For Android

Cross compilation to Android is supported for targeting 64 bit ARM and Intel x86 architectures (aarch64 and x86_64) for API Level 26 (Oreo, Android 8):

For example, to compile for aarch64:

  1. Install the rust target:
    rustup target install aarch64-linux-android
  2. Download the r21e NDK for your host architecture and unzip it.
  3. Compile your package for the aarch64-linux-android target:

On macOS:

export ANDROID_NDK=:path-to-android-ndk-r21e
export PATH=$PATH:$ANDROID_NDK/toolchains/llvm/prebuilt/darwin-x86_64/bin
export CC_aarch64_linux_android=aarch64-linux-android26-clang
export CXX_aarch64_linux_android=aarch64-linux-android26-clang++
export CARGO_TARGET_AARCH64_LINUX_ANDROID_LINKER=aarch64-linux-android26-clang

cargo build -vv --target aarch64-linux-android

Note: we don't support Apple's Clang 11 to build for Android on macOS, so you need to install LLVM and set the PATH like instructed.

On Linux:

export ANDROID_NDK=:path-to-android-ndk-r21e
export PATH=$PATH:$ANDROID_NDK/toolchains/llvm/prebuilt/linux-x86_64/bin
export CC_aarch64_linux_android=aarch64-linux-android26-clang
export CXX_aarch64_linux_android=aarch64-linux-android26-clang++
export CARGO_TARGET_AARCH64_LINUX_ANDROID_LINKER=aarch64-linux-android26-clang

cargo build -vv --target aarch64-linux-android

On Windows the Android NDK clang executable must be invoked through .cmd scripts:

export ANDROID_NDK=:path-to-android-ndk-r21e
export PATH=$PATH:$ANDROID_NDK/toolchains/llvm/prebuilt/windows-x86_64/bin
export CC_aarch64_linux_android=aarch64-linux-android26-clang.cmd
export CXX_aarch64_linux_android=aarch64-linux-android26-clang++.cmd
export CARGO_TARGET_AARCH64_LINUX_ANDROID_LINKER=aarch64-linux-android26-clang.cmd

cargo build -vv --target aarch64-linux-android

Notes:

  • The CARGO_TARGET_${TARGET}_LINKER environment variable name needs to be all uppercase.
  • In some older shells (for example macOS High Sierra), environment variable replacement can not be used when the variable was defined on the same line. Therefore the ANDROID_NDK variable must be defined before it's used in the PATH variable.
  • Rebuilding skia-bindings with a different target may cause linker errors, in that case touch skia-bindings/build.rs will force a rebuild (#10).

For iOS

Compilation to iOS is supported on macOS targeting the iOS simulator (--target x86_64-apple-ios) and 64 bit ARM devices (--target aarch64-apple-ios). The ARM64e architecture is not supported yet.

Skia

For situations in which Skia does not build or needs to be configured differently, we support some customization support in skia-bindings/build.rs. For more details take a look at the README of the skia-bindings package.

Please share your build experience so that we can try to automate the build and get to the point where cargo build is sufficient to build the bindings including Skia, and if that is not possible, clearly prompts to what's missing.

Example Applications

icon

The icon example generates the rust-skia icon in the current directory. It computes the position of all the gear teeth etc. based on parameters such as the number of teeth and wheel radius.

If you were able to build the project, run

cargo run --example icon 512

It has a single optional parameter which is the size in pixels for the PNG file. Without parameters, it’ll produce PNG frames for the animated version.

skia-org

The other examples are taken from Skia's website and ported to the Rust API.

cargo run -- [OUTPUT_DIR]

to generate some Skia drawn PNG images in the directory OUTPUT_DIR. To render with OpenGL, use

cargo run -- [OUTPUT_DIR] --driver opengl

And to show the drivers that are supported

cargo run -- --help

gl-window

An example that opens an OpenGL Window and draws the rust-skia icon with skia-safe (contributed by @nornagon).

cargo run --example gl-window --features gl

On Linux the feature x11 needs to be enabled:

cargo run --example gl-window --features gl,x11

Example Images

Fill, Radial Gradients, Stroke, Stroke with Gradient, Transparency: Rust-skia icon

Fill, Stroke, Text:

Fill, Stroke, Text

Sweep Gradient:

Sweep Gradient

Dash Path Effect:

Dash Path Effect

For more, you may take a look at the rust-skia.github.io repository.

This project needs contributions!

If you'd like to help with the bindings, take a look at the Wiki to get started and create an issue to prevent duplicate work. For smaller tasks, grep for "TODO"s in the source code. And for heroic work, check out the label help wanted. And if you like to help making the Rust API nicer to use, look out for open issues with the label api ergonomics.

More details can be found at CONTRIBUTING.md.

Notable Contributions

  • Denis Kolodin (@DenisKolodin) added build support for Android.
  • Alberto González Palomo (@AlbertoGP) designed the Rust-Skia Logo and the example program that renders it.

Maintainers

License

MIT

rust-skia's People

Contributors

albertogp avatar brooooooklyn avatar deniskolodin avatar dependabot-preview[bot] avatar dependabot[bot] avatar eira-fransham avatar j4qfrost avatar katyo avatar kethku avatar messense avatar nof1000 avatar nornagon avatar palfrey avatar photex avatar pragmatrix avatar stewartcanva avatar stringke avatar superwhiskers avatar syrel avatar upsuper avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.