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.


  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)


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:

# change to span's directory
cd wherever/you/downloaded/span

# install dependencies
pip install -r requirements.txt
# install span
python install

or if you don’t want to install it and you want to use span from its directory then you can do

python build_ext --inplace
python build

For the last one you must set the PYTHONPATH environment variable or mess around with sys.path (not recommended).

Project Versions

Table Of Contents

Previous topic


Next topic

Reading in TDT Files

This Page