drbunbury / build-wheel

Introduction

This repository contains the build-wheel tool, which produces Android .whl files for Chaquopy.

build-wheel can build .whl files for all Android ABIs (armeabi-v7a, arm64-v8a, x86 and x86_64). However, the tool itself only runs on Linux x86-64. If you don't already have a Linux machine available, a cheap virtual server from somewhere like DigitalOcean will do just fine.

Adding a package

Create a recipe directory in packages. Its name must be in PyPI normalized form (PEP 503). Alternatively, you can create this directory somewhere else, and pass its path when calling build-wheel.py.

Inside the recipe directory, add the following files:

• A meta.yaml file. This supports a subset of Conda syntax, defined in meta-schema.yaml.
• For non-Python packages, a build.sh script. See build-wheel.py for environment variables which are passed to it.
• If necessary, a patches subdirectory containing patch files.

The following examples are included:

• multidict: a minimal example, downloaded from PyPI.
• cython-example: a minimal example, built from a local directory.
• cryptography: a package with a build-time requirement.
• python-example: a pybind11-based package, downloaded from a Git repository.
• cmake-example: similar to python-example, but uses cmake. A patch is used to help cmake find the Android toolchain file.
• chaquopy-libzmq: a non-Python library, downloaded from a URL.
• pyzmq: a Python package which depends on a non-Python library. A patch is used to help setup.py find the library.
• scikit-learn: lists several build-time requirements in meta.yaml:
  • The "build" requirement (Cython) will be installed automatically.
  • The "host" requirements (NumPy etc.) should be downloaded from the public repository and copied into server/pypi/dist before running the build. A patch is used to allow NumPy and SciPy to be imported during the build.

Building with Docker

Docker is the easiest way to set up all of build-wheel's dependencies.

If necessary, install Docker using the instructions on its website.

Build the Docker images:

docker build -t chaquopy-base -f base.dockerfile .
docker build -t chaquopy-target target
docker build -t build-wheel server/pypi

If the server/pypi directory changes, you'll need to rerun the last of these commands to update the build-wheel image.

Then run build-wheel from the server/pypi directory as follows:

docker run -v $(pwd)/dist:/root/pypi/dist build-wheel --toolchain target/toolchains/<abi> <package>

Where:

• <abi> is an Android ABI.
• <package> is the name of the recipe directory created above.

The resulting .whl files will be generated in server/pypi/dist.

Building without Docker

This is more convenient for development, but you'll have to set up all of build-wheel's dependencies manually. Use the Dockerfiles as a guide.

Then run build-wheel from the server/pypi directory as follows:

./build-wheel.py --toolchain ../../target/toolchains/<abi> <package>

<abi> and <package> are described above. The toolchains directory can be copied out of the chaquopy-target image.

The resulting .whl files will be generated in server/pypi/dist.

Using the generated .whl files in your app

.whl files can be built into your app using the pip block in your build.gradle file. First, add an options line to pass --extra-index-url with the location of the dist directory mentioned above. Either an HTTP URL or a local path can be used. Then add an install line giving the name of your package.