Installing zasim and its dependencies

Prerequisites

The only dependency beyond Python 2.6 or Python 2.7 [1] that you cannot use zasim without is NumPy, which it uses extensively for its array class.

Beyond that, to run stepfuncs compiled to C code, you need SciPy for its weave module. This also adds the requirement of a C++ compiler, such as the one from GCC.

For the graphical user interface, you need PySide, nokias Qt bindings for Python. PyQt is, in theory, api-compatible, but the code is currently only tested with PySide and is not guaranteed to work with PyQt.

There is a tutorial based on the IPython notebook and there is additional support for the IPython qtconsole.

If you have the GNU indent tool installed, generated C code will be a lot more pretty to look at and thus quite a bit easier to debug, if you run into trouble with self-made code.

[1]Almost the same subset of zasim you get if you have neither SciPy nor PySide is also usable with pypy 1.8 or newer.

Installing zasim

From a git checkout

After checking out the git repository, you can just run - as root or in a virtualenv -

$ python setup.py develop

which will create a link to zasim in your global site-packages repository.

Whenever names of packages or modules change in the git repository, you should re-run the above command, else you may get error messages about failing imports.

Running without an installation

You can also run zasim without installing it, just from the extracted source tree. In order for this to work, just cd into the directory you extracted zasim to. This folder should contain the setup.py and a zasim folder.

Table Of Contents

Previous topic

Tutorial section

Next topic

Invoking zasim from the commandline

This Page