Getting Started =============== Caveat Emptor ------------- I developed this code with the bleeding edge versions of everything listed below, so don't be surprised if you try this out on an older version of NumPy (for example) and it doesn't work. When in doubt set up a `virtualenv `_ and install the git versions of all the dependencies. This code has only been used and tested on `Arch Linux `_ and `Red Hat Enterprise Linux 5 `_. If you set up a virtualenv and follow the install instructions there should be no problems. Dependencies ------------ 1. `Python `_ >= 2.6 2. `NumPy `_ 3. `SciPy `_ 4. `Cython `_ 5. `pandas `_ 6. `six `_ Optional Dependencies --------------------- 1. `nose `_ if you want to run the tests 2. `sphinx `_ if you want to build the documentation 3. `numpydoc `_ if you want to build the documentation with NumPy formatting support (recommended) Installation ------------ You should set up a `virtualenv `_ when using this code, so that you don't break anything. Personally, I prefer the excellent `virtualenvwrapper `_ tool to set up Python environments. First things first: .. code-block:: bash # change to span's directory cd wherever/you/downloaded/span # install dependencies pip install -r requirements.txt .. code-block:: bash # install span python setup.py install or if you don't want to install it and you want to use ``span`` from its directory then you can do .. code-block:: bash python setup.py build_ext --inplace python setup.py build For the last one you must set the ``PYTHONPATH`` environment variable or mess around with ``sys.path`` (not recommended).