All platforms: - rename scripts/ to tools/: Bazelisk expects to find its wrapper script (used by the Mac changes below) in tools/. Rather than have a separate scripts/ and tools/, it's simpler to just move everything into tools/. - wheel outputs and binary bundles now go into .bazel/out/dist. While not technically Bazel build products, doing it this way ensures they get cleaned up when 'bazel clean' is run, and it keeps them out of the source folder. - update to the latest Bazel Windows changes: - bazel.bat has been removed, and tools\setup-env.bat has been added. Other scripts like .\run.bat will automatically call it to set up the environment. - because Bazel is now on the path, you can 'bazel test ...' from any folder, instead of having to do \anki\bazel. - the bat files can handle being called from any working directory, so things like running "\anki\tools\python" from c:\ will work. - build installer as part of bundling process Mac changes: - `arch -arch x86_64 bazel ...` will now automatically use a different build root, so that it is cheap to switch back and forth between archs on a new Mac. - tools/run-qt* will now automatically use Rosetta - disable jemalloc in Mac x86 build for now, as it won't build under Rosetta (perhaps due to its build scripts using $host_cpu instead of $target_cpu) - create app bundle as part of bundling process Linux changes: - remove arm64 orjson workaround in Linux bundle, as without a readily-available, relatively distro-agonstic PyQt/Qt build we can use, the arm64 Linux bundle is of very limited usefulness. - update Docker files for release build - include fcitx5 in both the qt5 and qt6 bundles - create tarballs as part of the bundling process
4.2 KiB
Linux
Requirements
These instructions are written for Debian/Ubuntu; adjust for your distribution. Some extra notes have been provided by a forum member: https://forums.ankiweb.net/t/guide-how-to-build-and-run-anki-from-source-with-xubuntu-20-04/12865
You can see a full list of buildtime and runtime requirements by looking at the Dockerfiles used to build the official releases.
Glibc is required - if you are on a distro like Alpine that uses musl, you'll need to contribute fixes to the upstream Rust rules, then follow the steps in Other Platforms.
Users on ARM64, see the notes at the bottom of this file before proceeding.
Ensure some basic tools are installed:
$ sudo apt install bash grep findutils curl gcc g++ git
The 'find' utility is 'findutils' on Debian.
Python:
For building and running from source, Python is required, but the version is
flexible - any version from 2.7 onwards should work. The build system expects to
find the command python
, so if your system only has a python3
, you'll need
to link it to python
, or do something like sudo apt install python-is-python3
.
The system Python is only used for running scripts, and the build system will download a copy of Python 3.9 into a local folder as part of the build.
You can have it use a locally installed Python instead, by putting something like the following into a file called user.bazelrc at the top of this repo before proceeding:
build --action_env=PYO3_PYTHON=/usr/local/bin/python3.9
Install Bazelisk:
Download it under the name 'bazel':
$ curl -L https://github.com/bazelbuild/bazelisk/releases/download/v1.10.1/bazelisk-linux-amd64 -o ./bazel
And put it on your path:
$ chmod +x bazel && sudo mv bazel /usr/local/bin/
Running Anki during development
From the top level of Anki's source folder:
./run
This will build Anki and run it in place.
The first build will take a while, as it downloads and builds a bunch of dependencies. When the build is complete, Anki will automatically start.
The Javascript build code is currently a bit flaky, so the initial build may fail with an error. If you get an error when running/building, try repeating the command once or twice - it should pick up where it left off.
To play and record audio, install mpv and lame.
If you or your distro has made ccache the standard compiler, you will need to set CC and CXX to point directly to gcc/g++ or clang/clang++ prior to building Anki.
Missing Libraries
If you get errors during startup, try starting with
QT_DEBUG_PLUGINS=1 ./run
It will likely complain about missing libraries, which you can install with your package manager. Some of the libraries that might be required on Debian for example:
sudo apt install libxcb-icccm4 libxcb-image0 libxcb-keysyms1 \
libxcb-randr0 libxcb-render-util0
Optimized builds
The ./run
command will create a non-optimized build by default. This is faster
to compile, but will mean Anki will run considerably slower.
To run Anki in optimized mode, use:
./tools/runopt
ARM64 support
Other platforms download PyQt binary wheels from PyPI. There are no PyQt wheels available for ARM Linux, so you will need to rely on your system-provided libraries instead. As Anki requires Python 3.9, this means you will need a fairly up-to-date distro such as Debian 11.
After installing the system libraries (eg 'sudo apt install python3-pyqt5.qtwebengine'), find the place they are installed (eg '/usr/lib/python3/dist-packages'). Then before running any commands like './run', tell Anki where they can be found:
export PYTHON_SITE_PACKAGES=/usr/lib/python3/dist-packages/
Note: the trailing slash at the end is required.
There are a few things to be aware of:
- You should use ./run and not tools/run-qt5*, even if your system libraries are Qt5.
- If your system libraries are Qt5, when creating an aqt wheel, the wheel will not work on Qt6 environments.
- Some of the tests only work with PyQt6, and will show failures when run under PyQt5.
More
For info on running tests, building wheels and so on, please see Development.