More strategies for NumPy

CONTRIBUTING.rst

@@ -160,3 +160,4 @@ their individual contributions.
 * `Tariq Khokhar <https://www.github.com/tkb>`_ (`[email protected] <mailto:[email protected]>`_)
 * `Will Hall <https://www.github.com/wrhall>`_ (`[email protected] <mailto:[email protected]>`_)
 * `Will Thompson <https://www.github.com/wjt>`_ (`[email protected] <mailto:[email protected]>`_)
+* `Zac Hatfield-Dodds <https://www.github.com/Zac-HD>`_ (`[email protected] <mailto:[email protected]>`_)

docs/conf.py

@@ -64,6 +64,8 @@
     'python': ('http://docs.python.org/', None),
 }
 
+autodoc_mock_imports = ['numpy']
+
 
 # -- Options for HTML output ----------------------------------------------
 

docs/extras.rst

@@ -188,56 +188,9 @@ It's large enough that it is :doc:`documented elsewhere <django>`.
 hypothesis[numpy]
 ------------------
 
-hypothesis.extra.numpy adds support for testing 
-`NumPy <http://www.numpy.org/>`_-based implementations with Hypothesis by 
-providing an ``arrays`` function.
+hypothesis.extra.numpy adds support for testing your Numpy code with Hypothesis.
 
-It lives in the ``hypothesis.extra.numpy`` package.
+This includes generating arrays, array shapes, and both scalar or compound dtypes.
 
-.. method:: arrays(dtype, shape, elements=None)
+Like the Django extra, :doc:`Numpy has it's own page <numpy>`.
 
-    Arrays of specified `dtype` and `shape` are generated for example 
-    like this:
-
-    .. code-block:: pycon
-
-      >>> import numpy as np
-      >>> arrays(np.int8, (2, 3)).example()
-      array([[-8,  6,  3],
-             [-6,  4,  6]], dtype=int8)
-
-
-    However, to obtain more fine grained control over the elements, use
-    the `elements` keyword (see also :doc:`What you can generate and how <data>`):
-
-    .. code-block:: pycon
-
-      >>> import numpy as np
-      >>> from hypothesis.strategies import floats
-      >>> arrays(np.float, 3, elements=floats(min_value=0, max_value=1)).example()
-      array([ 0.88974794,  0.77387938,  0.1977879 ])
-
-    By combining different strategies, the shape of the array can be modified as well:
-
-    .. code-block:: pycon
-
-      >>> import numpy as np
-      >>> from hypothesis.strategies import integers, floats
-      >>>
-      >>> def rnd_len_arrays(dtype, min_len=0, max_len=3, elements=None):
-      ...     lengths = integers(min_value=min_len, max_value=max_len)
-      ...     return lengths.flatmap(lambda n: arrays(dtype, n, elements=elements))
-      ... 
-      >>> 
-      >>> rla = rnd_len_arrays(np.int8)
-      >>> rla.example()
-      array([], dtype=int8)
-      >>> rla.example()
-      array([-2], dtype=int8)
-      >>> rla.example()
-      array([ 7, -6, -2], dtype=int8)
-
-**Note**: To generate large arrays/matrices, or even medium-sized matrices that
-have 3 or more dimensions, it's necessary to increase the ``buffer_size``
-setting to be greater than its default of ``8192`` (see the 
-:doc:`Settings documentation <settings>` for details on how to do this).

docs/index.rst

@@ -8,7 +8,7 @@ finding edge cases in your code you wouldn't have thought to look for. It is
 stable, powerful and easy to add to any existing test suite.
 
 It works by letting you write tests that assert that something should be true
-for every case, not just the ones you happen to think of. 
+for every case, not just the ones you happen to think of.
 
 Think of a normal unit test as being something like the following:
 
@@ -57,11 +57,12 @@ check out some of the
   :hidden:
 
   quickstart
-  django
   details
   settings
   data
   extras
+  django
+  numpy
   healthchecks
   database
   stateful

docs/numpy.rst

@@ -0,0 +1,18 @@
+.. _hypothesis-numpy:
+
+=================================
+Scientific Hypothesis (for NumPy)
+=================================
+
+Hypothesis offers a number of strategies for `NumPy <http://www.numpy.org/>`_ testing,
+available in the :mod:`hypothesis[numpy]` :doc:`extra </extras>`.
+It lives in the ``hypothesis.extra.numpy`` package.
+
+The centerpiece is the ``arrays`` strategy, which generates arrays with
+any dtype, shape, and contents you can specify or give a strategy for.
+To make this as useful as possible, strategies are provided to generate array
+shapes and generate all kinds of fixed-size or compound dtypes.
+
+
+.. automodule:: hypothesis.extra.numpy
+   :members: arrays, array_shapes, scalar_dtypes, boolean_dtypes, unsigned_integer_dtypes, integer_dtypes, floating_dtypes, complex_number_dtypes, datetime64_dtypes, timedelta64_dtypes, byte_string_dtypes, unicode_string_dtypes, array_dtypes, nested_dtypes

src/hypothesis/errors.py

@@ -115,7 +115,7 @@ class Flaky(HypothesisException):
            depends sensitively on where it's been called from.
         3. The function is timing sensitive and can fail or pass depending on
            how long it takes. Try breaking it up into smaller functions which
-           dont' do that and testing those instead.
+           don't do that and testing those instead.
 
     """