@@ -13,9 +13,9 @@ the details (and potential complications) on this page, the easiest
1313thing for you to do is use one of the pre-packaged python
1414distributions that already provide matplotlib built-in.  The Continuum.io
1515Python distribution `(Anaconda)
16- <https://store.continuum.io/cshop/anaconda/>`_ and the Enthought  
17- distribution `(Canopy) <https://www.enthought.com/products/canopy/>`_   
18- are both excellent choices that "just work" out of the box for any  
16+ <https://store.continuum.io/cshop/anaconda/>`_ and the Enthought
17+ distribution `(Canopy) <https://www.enthought.com/products/canopy/>`_
18+ are both excellent choices that "just work" out of the box for any
1919platform.  Both of these packages include matplotlib and
2020*lots* of other useful tools.
2121Another excellent alternative for Windows users is `Python (x, y)
@@ -87,31 +87,77 @@ python console, there are some issues relating to user interface
8787toolkits and interactive settings that are discussed in
8888:ref:`mpl-shell`.
8989
90+ 
9091Installing on Windows
9192---------------------
9293
93- If you don't already have python installed, you may want to consider
94- using the Enthought edition of python, which has scipy, numpy, and
95- wxpython, plus many other useful packages, preinstalled - `Enthought
96- Python <http://www.enthought.com/python>`_.  With the Enthought
97- edition of python + matplotlib installer, the following backends
98- should work out of the box: agg, wx, wxagg, tkagg, ps, pdf and svg.
99- 
100- For standard python installations, you will also need to install numpy
101- in addition to the matplotlib installer.  On some systems you will
102- also need to download msvcp71.dll library, which you can download from
103- http://www.dll-files.com/dllindex/dll-files.shtml?msvcp71 or other
104- sites.  You will need to unzip the archive and drag the dll into
105- :file:`c:\windows\system32`.
106- 
107- All of the GUI backends run on Windows, but TkAgg is probably the
108- best for interactive use from the standard python shell or ipython.
109- The Windows installer (:file:`*.exe`) on the download page contains all the
110- code you need to get up and running.  However, there are many
111- examples that are not included in the Windows installer.  If you
112- want to try the many demos that come in the matplotlib source
94+ If you don't already have Python installed, we recommend using
95+ one of the `scipy-stack compatible Python distributions
96+ <http://www.scipy.org/install.html>`_ such as Python(x,y),
97+ Enthought Canopy, or Continuum Anaconda, which have matplotlib and
98+ many of its dependencies, plus other useful packages, preinstalled.
99+ 
100+ For `standard Python <http://www.python.org/download/>`_ installations
101+ you will also need to install compatible versions of
102+ `setuptools <https://pypi.python.org/pypi/setuptools/>`_,
103+ `numpy <https://pypi.python.org/pypi/numpy/>`_,
104+ `python-dateutil <https://pypi.python.org/pypi/python-dateutil/>`_,
105+ `pytz <https://pypi.python.org/pypi/pytz>`_,
106+ `pyparsing <https://pypi.python.org/pypi/pyparsing/>`_ and
107+ `six <https://pypi.python.org/pypi/six>`_
108+ in addition to
109+ `matplotlib <https://pypi.python.org/pypi/matplotlib>`_.
110+ 
111+ In case Python is not installed for all users (not the default), the
112+ Microsoft Visual C++ 2008 (
113+ `64 bit <http://www.microsoft.com/download/en/details.aspx?id=15336>`_
114+ or
115+ `32 bit <http://www.microsoft.com/download/en/details.aspx?id=29>`_
116+ for Python 2.6 to 3.2) or Microsoft Visual C++ 2010 (
117+ `64 bit <http://www.microsoft.com/en-us/download/details.aspx?id=14632>`_
118+ or
119+ `32 bit <http://www.microsoft.com/en-us/download/details.aspx?id=5555>`_
120+ for Python 3.3 and 3.4) redistributable packages need to be installed.
121+ 
122+ Matplotlib depends on `Pillow <https://pypi.python.org/pypi/Pillow>`_
123+ for reading and saving JPEG, BMP, and TIFF image files.
124+ Matplotlib requires `MiKTeX <http://miktex.org/>`_ and
125+ `GhostScript <http://www.ghostscript.com/download/>`_ for rendering text
126+ with LaTeX.
127+ `FFmpeg <https://www.ffmpeg.org/>`_, `avconv <https://libav.org/>`_,
128+ `mencoder <http://www.mplayerhq.hu>`_, or
129+ `ImageMagick <http://www.imagemagick.org/>`_ are required for the
130+ animation module.
131+ 
132+ The following backends should work out of the box: agg, tkagg, ps,
133+ pdf and svg.
134+ For other backends you may need to install
135+ `pycairo <https://pypi.python.org/pypi/pycairo>`_,
136+ `PyQt4 <https://pypi.python.org/pypi/PyQt4>`_,
137+ `PyQt5 <https://pypi.python.org/pypi/PyQt5>`_,
138+ `PySide <https://pypi.python.org/pypi/PySide>`_,
139+ `wxPython <https://pypi.python.org/pypi/wxPython>`_,
140+ `PyGTK <https://pypi.python.org/pypi/PyGTK>`_,
141+ `Tornado <https://pypi.python.org/pypi/tornado>`_,
142+ or GhostScript.
143+ 
144+ TkAgg is probably the best backend for interactive use from the
145+ standard Python shell or IPython. It is enabled as the default backend
146+ for the official binaries.  GTK3 is not supported on Windows.
147+ 
148+ The Windows installers (:file:`*.exe`) and wheels (:file:`*.whl`) on
149+ the download page do not contain test data or example code.
150+ If you want to try the many demos that come in the matplotlib source
113151distribution, download the zip file and look in the :file:`examples`
114152subdirectory.
153+ To run the test suite, copy the lib\matplotlib\tests and
154+ lib\mpl_toolkits\tests directories from the source distribution to
155+ sys.prefix\Lib\site-packages\matplotlib and
156+ sys.prefix\Lib\site-packages\mpl_toolkits respectively, and install
157+ `nose <https://pypi.python.org/pypi/nose>`_,
158+ `mock <https://pypi.python.org/pypi/mock>`_,
159+ Pillow, MiKTeX, GhostScript, ffmpeg, avconv, mencoder, ImageMagick, and
160+ `Inkscape <http://inkscape.org/>`_.
115161
116162.. _install_from_source:
117163
0 commit comments