Harvard-only trial pre-release of kwant
=======================================

kwant is made available as a special preview **exclusively** to the audience of
the 30 April 2013 presentation at Harvard University by Anton Akhmerov.  Please
do not spread the link to this page.  Instead, you are welcome show the link to
the `kwant website <http://kwant-project.org/>`_.

You may use kwant for your personal evaluation.  Any other use, **explicitly
including further distribution**, is not allowed.  Please wait for the release
of the final version that we expect to occur very soon.

So far, kwant is running on the following platforms: GNU/Linux, Mac OS, and
Windows.  The usage experience on GNU/Linux is best (this is the platform
recommended by the authors of kwant), followed by Mac OS.  Support for Windows
is highly experimental (help wanted).


Installation from source
------------------------

This method of installation is described in the documentation.  It is extremely
easy on Debian-based GNU/Linux distributions (e.g. Ubuntu).  On Mac OS and
Windows it is more involved, but documented as well.  Please download the files
and follow the `installation instructions </docs/install.html>`_.

* kwant: `tar-gz-archive <kwant-0.3pre4.tar.gz>`__, `zip-archive <kwant-0.3pre4.zip>`__
* tinyarray (needed by kwant): `tar-gz-archive <tinyarray-0.0.tar.gz>`__, `zip-archive <tinyarray-0.0.zip>`__


Binary installation instructions for Windows
--------------------------------------------

**Warning:** These packages are highly experimental and likely not to work
fully.  In a test with Windows XP, most tutorial examples ran.  In another test
with Windows 7, most did not.  We would appreciate the help of people
experienced with the Windows platform.

**Important note**: The provided binary packages are compiled for 32-bit
Microsoft Windows (win32).  They can be installed on a 64-bit Windows as well,
but for them to work all the other required packages (like Python itself) have
to be present in their win32 versions.

1. Install the official Python 2.7 distribution from the official `Python
   download page <http://python.org/download/>`_. Bear in mind the note above:
   Choose "Python 2.7.x Windows Installer", not the 64 bit version.

2. Install the Python packages required by kwant (NumPy, SciPy, matplotlib).  An
   easy way to do so is to install the `scipy-stack package
   <http://www.lfd.uci.edu/~gohlke/pythonlibs/#scipy-stack>`_ (win32-py2.7)
   kindly provided by Christoph Gohlke.  Please also note the instructions
   at the top of that page.

3. Download and execute the `tinyarray installer <tinyarray-0.0.win32-py2.7.exe>`__.

4. Download and execute the `kwant installer <kwant-0.3pre2.win32-py2.7.exe>`__.

5. kwant and tinyarray require four libraries (DLLs) that can be downloaded
   `here <kwant-dlls.zip>`__.  For these libraries to be found by kwant, they
   have either to be in the current directory, or in a directory that belongs to
   ``PATH``, or in ``C:\WINDOWS\SYSTEM32``.  You have the following alternative
   possibilities:

   (a) Unpack the 4 files into some directory and launch all your kwant Python
       scripts from the same directory.

   (b) Like (a), but add that directory to the ``PATH`` environment variable
       (The Web knows how to do that.).  Now you should be able to launch kwant
       scripts from everywhere.

   (c) Copy the four DLL files directly to ``C:\WINDOWS\SYSTEM32``.

Please note that the kwant binary packages do not include any kwant examples or
documentation.  Use the `online documentation </docs/>`_.