Join GitHub today
GitHub is home to over 28 million developers working together to host and review code, manage projects, and build software together.Sign up
The main development branch (i.e. the tar balls one can download and the corresponding git "master" branch) is supposed to be stable and usable. However, in order to test and introduce new features and improves old ones (while possibly breaking immediate backwards compatibility) before a new release we are doing ongoing development on the development branch (as described under DevelopmentWorkflow).
Everyone is more than welcome to use the development version but you should be aware that you are then first and foremost a alpha-tester: Things may break in horrible ways (but your feedback through the mailing list and the Issue Tracker is much appreciated). (That said, many of the core developers are using the development code for their daily work.)
If you find yourself using the development branch a lot then consider signing up for the developer mailing list where new features and testing of the development code are primarily discussed.
A quick overview over the latest additions to the development branch is provided by the list of commits to the develop branch.
Getting the development branch
git clone https://github.com/MDAnalysis/mdanalysis.git cd mdanalysis git fetch --all git checkout -b develop origin/develop # this should switch you to the development branch
You only have to carry out the above commands once.
Let's check that you actually have switched to the development branch: The command
should show something similar to
* develop master
The asterisk in front of develop indicates that you have checked out the development version of MDAnalysis.
You can update it to the very latest version with a simple
If you want to use the latest officially released version of the source code then switch to the master branch:
git checkout master
You can go back to the development version with
git checkout develop
The MDAnalysis source code repository contains the actual MDAnalysis library and test code. When using the development branch it is a good idea to install both so that you can always run the tests and verify that the current version of the code you're using has not been broken inadvertently.
The following uses
python setup.py install to mean "install the package in whichever way you normally do". (In particular, as a developer
python setup.py develop is often a very handy way to work on the code as it it takes changes to the code immediately into account.)
Pre-requisites for development
In order to build from develop you need
- working C/C++ compiler
- Python 2.7.x or Python ≥ 3.4
- Python packages already installed
Install the MDAnalysis library
pip to build and install; see the docs for
pip on how to customize installation.
The library lives in the mdanalysis/package directory.
pip install package/
(Note: the slash
package is important!)
(Ideally, this should also install additional dependencies but you might want to install numpy, scipy, biopython, netcdf4/hdf5 libraries through your operating systems package management system or use the anaconda distribution.)
Install the test cases
The UnitTests are stored together with test data in mdanalysis/testsuite.
pip install testsuite/
More details can be found under UnitTests.
You want to see all tests pass and no ERROR or FAIL (unless the test is marked with "knownfailure").
You can report any problems encountered through the mailing list (preferrably the ones for development) or the Issue Tracker.