Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Updated install.rst to make it easier to find official binary downloads. #30

Closed
wants to merge 1 commit into from
Closed
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
43 changes: 34 additions & 9 deletions www/install.rst
Expand Up @@ -20,6 +20,18 @@ which includes all the key packages:
including the Scipy stack. Windows only.
* `Pyzo <http://www.pyzo.org/>`_: A free distribution based on Python 3 (see
:ref:`python3`) with the IEP editor. Supports Linux and Windows.
* Christoph Gohlke provides `pre-built Windows installers <http://www.lfd.uci.edu/~gohlke/pythonlibs/>`_ for many Python packages, including all of the core Scipy stack.

"Official" binaries
-------------------------------

The maintainers of many of the packages in the Scipy stack provide binary installers for common Windows and OS-X systems.

* `numpy <http://sourceforge.net/projects/numpy/files/NumPy/>`_
* `scipy <http://sourceforge.net/projects/scipy/files/scipy/>`_
* `matplotlib <http://matplotlib.org/downloads.html>`_
* `ipython <https://github.com/ipython/ipython/releases>`_
* and others -- check teh project's web site
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

typo "teh"

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Maybe the numpy and scipy links shouldnt go directly to sf, but to the local download page in numpy and scipy section of the page?


Linux packages
--------------
Expand Down Expand Up @@ -79,20 +91,33 @@ To install the scipy stack for Python 2.7 with `Macports

sudo port install py27-numpy py27-scipy py27-matplotlib py27-ipython +notebook py27-pandas py27-sympy py27-nose

Homebrew
~~~~~~~~

To install the scipy stack for Python with `Homebrew
<http://brew.sh/>`_ : First intall homebrew, then::
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Does this actually work? I thought you needed to add a tap. @samueljohn?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ah, pip not brew. I think that's a recipe for disaster. Especially -10 on pip install matplotlib.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

typo "intall" also

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think the main question about the homebrew recipe is: does it work out of the box, and does it work in a sensible way. If not, it's better to not add it.


brew install python
pip install numpy
brew install gfortran
pip install scipy
brew install freetype
pip install matplotlib
pip install ipython

And you should be able to `pip install` other packages as well.

Custom
------

You can assemble the Scipy stack from individual packages. For details of what
you need, see :ref:`the specification <stackspec>`. Packages are typically on
`the Python Package Index <http://pypi.python.org/pypi/>`_, and projects' sites
may also offer official binary packages (e.g. `numpy
<http://sourceforge.net/projects/numpy/files/NumPy/>`_, `scipy library
<http://sourceforge.net/projects/scipy/files/scipy/>`_).
`the Python Package Index <http://pypi.python.org/pypi/>`_, and many can be
installed with a simple::

pip install packagename
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I really don't think you want to recommend using pip.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ever? pip is great for pure-python packages, and remarkably useful for others once your system is set up to compile.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Not ever, but not for any of numpy/scipy/matplotlib/ipython/pandas/statsmodels/sklearn/skimage. So close to never:)


Christoph Gohlke provides `pre-built Windows installers
<http://www.lfd.uci.edu/~gohlke/pythonlibs/>`_ for many Python packages,
including all of the core Scipy stack.
In addition, many projects' sites may also offer official binary packages

You can also build any of the Scipy packages from source, for instance if you
want to get involved with development. This is easy for packages written
Expand All @@ -106,8 +131,8 @@ Note on Python 3

The Python language is moving from the 2.x series to Python 3. As of late 2012,
all of the core Scipy Stack packages support Python 3, but some more specialist
packages still only work on Python 2. The methods above will mostly install
Python 2, with the exception of `Pyzo <http://www.pyzo.org/>`_.
packages still only work on Python 2. The full distribution listed above will
mostly install Python 2, with the exception of `Pyzo <http://www.pyzo.org/>`_.

If you choose to use the new version of the language, it should be easy to find
Python 3 versions of packages in your package manager.