NOTE: this fork is slightly modified to include precompiled mex files for Linux, and include some minimal pull requests from other forks.
This API aims to bring the awesome of ZMQ to MATLAB. This project has grown out @fagg wanting to better manage large scale numerical computing experiments across a High Performance Cluster. However, this library can be used in any number of contexts across any number of machines (even 1 is OK).
- Make sure you have ZMQ 4.0.x installed. Change settings in
config.m
to point to your installation. - Ensure you have mex configured [1] [2]
- Run
make.m
- Make sure that the
lib
directory is on your MATLAB path. - Start hacking.
- Git pull master - make clean; make; make test. Try again.
- Try the dev branch. See if that works.
- If not, open an issue and include the following information:
- Versions: MATLAB, ZMQ and Operating System
- Which version of matlab-zmq you're using (i.e. which branch, commit etc).
- What you are trying to do - preferably include a succinct code example which illustrates the problem.
- What doesn't work - please describe behavior explicitly and include any error messages you encounter.
- When building with Matlab R2022b, you may need to launch Matlab with the environment variable
LD_PRELOAD
set to the path to your system'slibstdc++
binary - e.g.
LD_PRELOAD=/lib/x86_64-linux-gnu/libstdc++.so.6 matlab
on Linux systems How can I help?
Pull requests are most welcome. As a general rule, please base all pull requests in master. If in doubt, contact @fagg.
- Ashton Fagg (@fagg)
- Anderson Bravalheri (@abravalheri)
This project contains contributions from the following people:
- Ashton Fagg (@fagg)
- Anderson Bravalheri (@abravalheri)
- Matheus Svolenski (@msvolenski)