daoBase is the core library of DAO (Durham Adaptive Optics), a real-time control framework for adaptive optics instruments. It provides the shared-memory (dao.shm) and messaging primitives that every other DAO repository builds on: fixed-layout IMAGE/IMAGE_METADATA shared-memory buffers for passing frames, slopes, and commands between real-time processes with minimal latency, a ZMQ-based command/event/logging layer, and a component/state-machine framework (daoComponent) for structuring long-running RTC processes.
The core is written in C, with a C++ layer built on top for the component/threading/state-machine framework, and bindings for Python, Rust, Julia, and MATLAB — so a pipeline can mix languages freely while every process still talks over the same shared-memory buffers. See daoTools for the application layer (centroiding, reconstruction, loop control, GUIs) built on top of this library.
The complete documentation for installation, usage, and API reference can be found here:
If you use this software in your research or work, please cite it using the following DOI:
@software{barr_2025_17264152,
author = {Barr, David and
Cetre, Sylvain and
Connolly, John and
Thomas Davies},
title = {Durham-Adaptive-Optics/daoBase: Initial Release},
month = oct,
year = 2025,
publisher = {Zenodo},
version = {v0.0.1},
doi = {10.5281/zenodo.17264152},
url = {https://doi.org/10.5281/zenodo.17264152},
swhid = {swh:1:dir:e28d715b004461ad5ac69d7ac13d5a4a0f8fc196
;origin=https://doi.org/10.5281/zenodo.17264151;vi
sit=swh:1:snp:c39ecfb0c6dc79d3ad9523b5cdd88dc1a14f
43e2;anchor=swh:1:rel:b822be04b6dda609af461bb157fc
b1ec74853023;path=Durham-Adaptive-Optics-daoBase-
ccd63ee
},
}For a first install, run the installer from the repo root:
./install.shIt builds everything, end to end: detects your OS and installs the system
packages (apt/dnf/yum/pacman/zypper/brew), reuses Miniconda if you
already have it or installs it otherwise, proposes a dedicated dao conda env
(optional — decline it and it'll ask which existing env to use instead), gets
waf, asks where DAOROOT/DAODATA should live (default /opt/dao/...,
created with sudo and handed back to you if that needs root the first
time), installs the Python dependencies, and finally builds and installs
daoBase itself (waf configure && waf && waf install) — finishing with a
real import daoShm + shared-memory round-trip to confirm it actually works.
./install.sh --helplists every option (skip a step, pick a different prefix, non-interactive
--yes, --dry-run, ...). It's idempotent — safe to re-run any time; it
reuses whatever it finds (existing conda, existing env, existing waf, ...)
rather than redoing work.
If daoBase is already installed and only the source changed (no new system dependency), you don't need the full installer again — from the repo root:
waf install(or waf configure --prefix=$DAOROOT && waf && waf install if you changed
something configure cares about, e.g. enabling a newly-available optional
dependency like BLAS or CUDA). Re-running ./install.sh in full is always
fine too — being idempotent, it will just confirm everything is already in
place; that's slower, not harmful, so use whichever you prefer.
test/install/ holds a Docker-based harness that
runs install.sh on a disposable, fresh Ubuntu/Rocky container as a
non-root sudo user. It's there for anyone changing install.sh itself, to
check it still works end to end before it reaches a real machine — not
something you need as a regular user.
The installer above automates everything below; this is kept as a reference for doing it by hand, or for platform quirks (e.g. arm64 protobuf) it doesn't cover.
(for build instructions on Windows, see windows-build.md)
Package for Redhat based distribution... to be adapted for other Linux distrib
yum install openssl-devel protobuf-devel gsl-devel numactl-devel glibc-devel ncurses-devel redis gtest-devel zeromq zeromq-devel
for ubuntu distribution:
sudo apt install -y libtool pkg-config build-essential autoconf automake python3 python-is-python3 libssl-dev libncurses5-dev libncursesw5-dev redis libgtest-dev libgsl-dev libzmq3-dev protobuf-compiler numactl libnuma-dev
You need homebrew for the packages.
brew install pkg-config zeromq protobuf gsl
if you use anaconda there could be a conflict between protobuf versions so uninstall using
conda uninstall libprotobuf
We encoutered several issue with the protobuf, usually related to the librptobuf vs protox version mismatch. One simple solution was to rebuild from the source
git clone https://github.com/abseil/abseil-cpp.git
cd abseil-cpp
mkdir build && cd build
cmake .. -DCMAKE_POSITION_INDEPENDENT_CODE=ON -DCMAKE_BUILD_TYPE=Release
make -j$(nproc)
sudo make install
then protobuf
wget https://github.com/protocolbuffers/protobuf/releases/download/v25.3/protobuf-25.3.tar.gz
tar zxvf protobuf-25.4.tar.gz
cd protobuf-25.3-full
mkdir -p cmake/build
cd cmake/build
cmake ../.. \
-DCMAKE_POSITION_INDEPENDENT_CODE=ON \
-Dprotobuf_BUILD_TESTS=OFF \
-Dprotobuf_ABSL_PROVIDER=package
make -j$(nproc)
sudo make install
sudo ldconfig
and
waf needs to be installed and in the path: see https://waf.io/book/.
Our typical install
mkdir ~/bin
cd ~/bin
wget https://waf.io/waf-2.0.26
ln -s waf-2.0.26 waf
chmod u+x *
and add $HOME/bin in your path
a python install is needed. We usually used miniconda but any aother install will work. here is our typical install command. For Linux x86_64
mkdir -p ~/miniconda3
wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh -O ~/miniconda3/miniconda.sh
bash ~/miniconda3/miniconda.sh -b -u -p ~/miniconda3
rm -rf ~/miniconda3/miniconda.sh
~/miniconda3/bin/conda init bash
~/miniconda3/bin/conda init zsh
Linux arm64
mkdir -p ~/miniconda3
curl -L https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-aarch64.sh -o ~/miniconda3/miniconda.sh
bash ~/miniconda3/miniconda.sh -b -u -p ~/miniconda3
rm -rf ~/miniconda3/miniconda.sh
~/miniconda3/bin/conda init bash
~/miniconda3/bin/conda init zsh
For MACOS arm64 install
mkdir -p ~/miniconda3
curl -L https://repo.anaconda.com/miniconda/Miniconda3-latest-MacOSX-arm64.sh -o ~/miniconda3/miniconda.sh
bash ~/miniconda3/miniconda.sh -b -u -p ~/miniconda3
rm -rf ~/miniconda3/miniconda.sh
~/miniconda3/bin/conda init bash
~/miniconda3/bin/conda init zsh
The following package should be installed
pip install posix_ipc zmq protobuf==3.20.0 astropy python-statemachine statemachine redis sphinx screeninfo
For MACOS, check the version of your protoc
protoc --version
libprotoc 29.3
Corresponding protobuf is
pip install protobuf==5.29.3
export DAOROOT=$HOME/DAOROOT
export DAODATA=$HOME/DAODATA
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$DAOROOT/lib:$DAOROOT/lib64
export PATH=$PATH:$HOME/bin:$HOME/DAOROOT/bin
export PYTHONPATH=$PYTHONPATH:$DAOROOT/python
export PKG_CONFIG_PATH=$PKG_CONFIG_PATH:$DAOROOT/lib/pkgconfig
For MACOS, add the following variables in your environment
export DYLD_LIBRARY_PATH=$DYLD_LIBRARY_PATH:/opt/homebrew/lib
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/opt/homebrew/lib
export CPLUS_INCLUDE_PATH=$CPLUS_INCLUDE_PATH:/opt/homebrew/include
export C_INCLUDE_PATH=$C_INCLUDE_PATH:/opt/homebrew/include
export CPATH=$CPATH:/opt/homebrew/include
export PKG_CONFIG_PATH=/opt/homebrew/lib/pkgconfig
!! BE SURE THE DAOROOT exists. In the example
mkdir $HOME/DAOROOT
waf configure --prefix=$DAOROOT
waf
waf install
Documents are built using doxygen and sphinx. To build the documents use the following command
waf build_docs
To clean the document build.
waf clean_docs
The following build options are available
Add debug flag to support debuging build and install with '-g' to support use of gdb
waf build --debug
Mostly used for developers it is available to check address sanitisation
waf build --santizer
The build include unit tests under tests. These can be built and run by running with the following
waf --test
This will build and run the tests.