GitHub - andy-slac/h5py: HDF5 for Python -- The h5py package is a Pythonic interface to the HDF5 binary data format.

andy-slac / h5py Public

forked from h5py/h5py

Notifications You must be signed in to change notification settings
Fork 0
Star 0

HDF5 for Python -- The h5py package is a Pythonic interface to the HDF5 binary data format.

www.h5py.org

0 stars 530 forks Branches Tags Activity

Star

Notifications

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 834 Commits
docs		docs
examples		examples
h5py		h5py
licenses		licenses
lzf		lzf
other		other
win_include		win_include
.gitignore		.gitignore
.hgignore		.hgignore
.travis.yml		.travis.yml
ANN.txt		ANN.txt
MANIFEST.in		MANIFEST.in
README.txt		README.txt
configure.py		configure.py
setup.py		setup.py

Repository files navigation

README file for h5py version 2.1.2
==================================

Websites
--------

  * Main website: http://www.h5py.org
  * Source code: http://github.com/h5py/h5py
  * Mailing list: h5py at googlegroups

Prerequisites
-------------

You need, at a minimum:

  * HDF5 1.8.3 or later on Linux (On Windows, h5py ships with HDF5)
  * Python 2.6, 2.7 or 3.2
  * Any modern version of NumPy

Optionally:

  * Cython 0.13 or later, to build from a git checkout
  * If using Python 2.6, unittest2 is needed to run the tests

Installing from tarball
-----------------------

Run the following commands:

  * python setup.py build [--hdf5=/path/to/hdf5]
  * python setup.py test   # optional
  * [sudo] python setup.py install

Installing via easy_install
---------------------------

Run the following commands:
 
  * export HDF5_DIR=/path/to/hdf5   # optional
  * [sudo] easy_install h5py

Building from a Git checkout (UNIX)
-----------------------------------------

We have switched development to GitHub.  Here's how to build
h5py from source:

  * Clone the project:
    $ git clone https://github.com/h5py/h5py.git

  * Generate the Cython files which talk to HDF5:
    $ cd h5py/h5py
    $ python api_gen.py

  * Build the project (this step also auto-compiles the .c files)
    $ cd ..
    $ python setup.py build [--hdf5=/path/to/hdf5]

  * Run the unit tests (optional)
    $ python setup.py test

Reporting bugs
--------------

  * Bug reports are always welcome at the GitHub tracker.  Please don't be
    offended if it takes a while to respond to your report... we value user
    input and take all bugs seriously.

  * If you're not sure you have a bug, or want to ask any question at all
    about h5py or HDF5, post to the mailing list (h5py at Google Groups).
    This list is read by the main developers of h5py, as well as the user
    community.  As a bonus, posting to the list means that other people with
    similar problems can find your question and the responses.

  * You're always free to email the author directly at [andrew dot collette
    at gmail dot com].