Hypothesis strategy for generating Variable objects (#8404)

* copied files defining strategies over to this branch * placed testing functions in their own directory * moved hypothesis strategies into new testing directory * begin type hinting strategies * renamed strategies for consistency with hypothesis conventions * added strategies to public API (with experimental warning) * strategies for chunking patterns * rewrote variables strategy to have same signature as Variable constructor * test variables strategy * fixed most tests * added helpers so far to API docs * add hypothesis to docs CI env * add todo about attrs * draft of new user guide page on testing * types for dataarrays strategy * draft for chained chunking example * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * only accept strategy objects * fixed failure with passing in two custom strategies that must be compatible * syntax error in example * allow sizes dict as argument to variables * copied subsequences_of strategy * coordinate_variables generates non-dimensional coords * dataarrays strategy given nothing working! * improved docstrings * datasets strategy works (given nothing) * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * pass dims or data to dataarrays() strategy * importorskip hypothesis in tests * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * added warning about inefficient example generation * remove TODO about deterministic examples in docs * un-restrict names strategy * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * removed convert kwarg * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * avoid using subsequences_of * refactored into separate function for unique subset of dims * removed subsequences_of * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix draw(st.booleans()) * remove all references to chunking until chunks strategy merged upstream in dask * added example of complicated strategy for dims dict * remove superfluous utils file * removed elements strategy * removed np_arrays strategy from public API * min_ndims -> min_dims * forbid non-matching dims and data completely * simple test for data_variables strategy * passing arguments to datasets strategy * whatsnew * add attrs strategy * autogenerate attrs for all objects * attempt to make attrs strategy quicker * extend deadline * attempt to speed up attrs strategy * promote all strategies to be functions * valid_dtypes -> numeric_dtypes * changed hypothesis error type * make all strategies keyword-arg only * min_length -> min_side * correct error type * remove coords kwarg * test different types of coordinates are sometimes generated * zip dict Co-authored-by: Zac Hatfield-Dodds <[email protected]> * add dim_names kwarg to dimension_sizes strategy * return a dict from _alignable_variables * add coord_names arg to coordinate_variables strategy * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * change typing of dims arg * support dims as list to datasets strat when data not given * put coord and data var generation in optional branch to try to improve shrinking * improve simple test example * add documentation on creating duck arrays * okexcept for sparse examples * fix sparse dataarrays example * todo about building a duck array dataset * fix imports and cross-links * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * add hypothesis library to intersphinx mapping * fix many links * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fixed all local mypy errors * move numpy strategies import * reduce sizes * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * fix some api links in docs * remove every strategy beyond variables * variable strategy now accepts callable generating array strategies * use only readable unicode characters in names * examples * only use unicode characters that docs can deal with * docs: dataarrays -> variables * update tests for variables strategy * test values in attrs dict * duck array type examples * altered whatsnew * maybe fix mypy * fix some mypy errors * more typing changes * fix import * skip doctests in docstrings * fix link to duckarrays page * don't actually try to run cupy in docs env * missed a skip * okwarning * just remove the cupy example * ensure shape is always passed to array_strategy_fn * test using make_strategies_namespace * test catching array_strategy_fn that returns different dtype * test catching array_strategy_fn that returns different shape * generalise test of attrs strategy * remove misguided comments * save working version of test_mean * expose unique_subset_of * generalize unique_subset_of to handle iterables * type hint unique_subset_of using overloads * use iterables in test_mean example * test_mean example in docs now uses iterable of dimension_names * fix some warnings in docs build * example of passing list to unique_subset_of * fix import in docs page * try to satisfy sphinx * Minor corrections to docs * Add supported_dtypes to list of public strategies in docs * Generate number of dimensions in test_given_arbitrary_dims_list Co-authored-by: Zac Hatfield-Dodds <[email protected]> * Update minimum version of hypothesis Co-authored-by: Zac Hatfield-Dodds <[email protected]> * fix incorrect indentation in autosummary * link to docs page on testing * use warning imperative for array API non-compliant dtypes * fix bugs in sparse examples * add tag for array API standard info * move no-dependencies-on-other-values-inputs to given decorator * generate everything that can be generated * fix internal link to page on strategies * split up TypeError messages for each arg * use hypothesis.errors.InvalidArgument * generalize tests for generating specific number of dimensions * fix some typing errors * test that reduction example in docs actually works * fix typing errors * simply generation of sparse arrays in example * fix impot in docs example * correct type hints in sparse example * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Use .copy in convert_to_sparse Co-authored-by: Justus Magin <[email protected]> * Use st.builds in sparse example Co-authored-by: Justus Magin <[email protected]> * correct intersphinx link in whatsnew * rename module containing assertion functions * clarify sentence * add general ImportError if hypothesis not installed * add See Also link to strategies docs page from docstring of every strategy * typo in ImportError message * remove extra blank lines in examples * remove smallish_arrays --------- Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Zac Hatfield-Dodds <[email protected]> Co-authored-by: Justus Magin <[email protected]>
pydata · Dec 5, 2023 · ab6a255 · ab6a255
1 parent 1f94829
commit ab6a255
Show file tree

Hide file tree

Showing 14 changed files with 1,077 additions and 10 deletions.
diff --git a/ci/requirements/doc.yml b/ci/requirements/doc.yml
@@ -9,6 +9,7 @@ dependencies:
   - cartopy
   - cfgrib
   - dask-core>=2022.1
+  - hypothesis>=6.75.8
   - h5netcdf>=0.13
   - ipykernel
   - ipywidgets  # silence nbsphinx warning

diff --git a/doc/api.rst b/doc/api.rst
@@ -1069,6 +1069,27 @@ Testing
    testing.assert_allclose
    testing.assert_chunks_equal
 
+Hypothesis Testing Strategies
+=============================
+
+.. currentmodule:: xarray
+
+See the :ref:`documentation page on testing <testing.hypothesis>` for a guide on how to use these strategies.
+
+.. warning::
+    These strategies should be considered highly experimental, and liable to change at any time.
+
+.. autosummary::
+   :toctree: generated/
+
+   testing.strategies.supported_dtypes
+   testing.strategies.names
+   testing.strategies.dimension_names
+   testing.strategies.dimension_sizes
+   testing.strategies.attrs
+   testing.strategies.variables
+   testing.strategies.unique_subset_of
+
 Exceptions
 ==========
 

diff --git a/doc/conf.py b/doc/conf.py
@@ -326,6 +326,7 @@
     "dask": ("https://docs.dask.org/en/latest", None),
     "cftime": ("https://unidata.github.io/cftime", None),
     "sparse": ("https://sparse.pydata.org/en/latest/", None),
+    "hypothesis": ("https://hypothesis.readthedocs.io/en/latest/", None),
     "cubed": ("https://tom-e-white.com/cubed/", None),
     "datatree": ("https://xarray-datatree.readthedocs.io/en/latest/", None),
     "xarray-tutorial": ("https://tutorial.xarray.dev/", None),

diff --git a/doc/internals/duck-arrays-integration.rst b/doc/internals/duck-arrays-integration.rst
@@ -31,6 +31,8 @@ property needs to obey `numpy's broadcasting rules <https://numpy.org/doc/stable
 (see also the `Python Array API standard's explanation <https://data-apis.org/array-api/latest/API_specification/broadcasting.html>`_
 of these same rules).
 
+.. _internals.duckarrays.array_api_standard:
+
 Python Array API standard support
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 

diff --git a/doc/user-guide/index.rst b/doc/user-guide/index.rst
@@ -25,4 +25,5 @@ examples that describe many common tasks that you can accomplish with xarray.
    dask
    plotting
    options
+   testing
    duckarrays
diff --git a/doc/user-guide/testing.rst b/doc/user-guide/testing.rst
@@ -0,0 +1,303 @@
+.. _testing:
+
+Testing your code
+=================
+
+.. ipython:: python
+    :suppress:
+
+    import numpy as np
+    import pandas as pd
+    import xarray as xr
+
+    np.random.seed(123456)
+
+.. _testing.hypothesis:
+
+Hypothesis testing
+------------------
+
+.. note::
+
+  Testing with hypothesis is a fairly advanced topic. Before reading this section it is recommended that you take a look
+  at our guide to xarray's :ref:`data structures`, are familiar with conventional unit testing in
+  `pytest <https://docs.pytest.org/>`_, and have seen the
+  `hypothesis library documentation <https://hypothesis.readthedocs.io/>`_.
+
+`The hypothesis library <https://hypothesis.readthedocs.io/>`_ is a powerful tool for property-based testing.
+Instead of writing tests for one example at a time, it allows you to write tests parameterized by a source of many
+dynamically generated examples. For example you might have written a test which you wish to be parameterized by the set
+of all possible integers via :py:func:`hypothesis.strategies.integers()`.
+
+Property-based testing is extremely powerful, because (unlike more conventional example-based testing) it can find bugs
+that you did not even think to look for!
+
+Strategies
+~~~~~~~~~~
+
+Each source of examples is called a "strategy", and xarray provides a range of custom strategies which produce xarray
+data structures containing arbitrary data. You can use these to efficiently test downstream code,
+quickly ensuring that your code can handle xarray objects of all possible structures and contents.
+
+These strategies are accessible in the :py:mod:`xarray.testing.strategies` module, which provides
+
+.. currentmodule:: xarray
+
+.. autosummary::
+
+   testing.strategies.supported_dtypes
+   testing.strategies.names
+   testing.strategies.dimension_names
+   testing.strategies.dimension_sizes
+   testing.strategies.attrs
+   testing.strategies.variables
+   testing.strategies.unique_subset_of
+
+These build upon the numpy and array API strategies offered in :py:mod:`hypothesis.extra.numpy` and :py:mod:`hypothesis.extra.array_api`:
+
+.. ipython:: python
+
+    import hypothesis.extra.numpy as npst
+
+Generating Examples
+~~~~~~~~~~~~~~~~~~~
+
+To see an example of what each of these strategies might produce, you can call one followed by the ``.example()`` method,
+which is a general hypothesis method valid for all strategies.
+
+.. ipython:: python
+
+    import xarray.testing.strategies as xrst
+
+    xrst.variables().example()
+    xrst.variables().example()
+    xrst.variables().example()
+
+You can see that calling ``.example()`` multiple times will generate different examples, giving you an idea of the wide
+range of data that the xarray strategies can generate.
+
+In your tests however you should not use ``.example()`` - instead you should parameterize your tests with the
+:py:func:`hypothesis.given` decorator:
+
+.. ipython:: python
+
+    from hypothesis import given
+
+.. ipython:: python
+
+    @given(xrst.variables())
+    def test_function_that_acts_on_variables(var):
+        assert func(var) == ...
+
+
+Chaining Strategies
+~~~~~~~~~~~~~~~~~~~
+
+Xarray's strategies can accept other strategies as arguments, allowing you to customise the contents of the generated
+examples.
+
+.. ipython:: python
+
+    # generate a Variable containing an array with a complex number dtype, but all other details still arbitrary
+    from hypothesis.extra.numpy import complex_number_dtypes
+
+    xrst.variables(dtype=complex_number_dtypes()).example()
+
+This also works with custom strategies, or strategies defined in other packages.
+For example you could imagine creating a ``chunks`` strategy to specify particular chunking patterns for a dask-backed array.
+
+Fixing Arguments
+~~~~~~~~~~~~~~~~
+
+If you want to fix one aspect of the data structure, whilst allowing variation in the generated examples
+over all other aspects, then use :py:func:`hypothesis.strategies.just()`.
+
+.. ipython:: python
+
+    import hypothesis.strategies as st
+
+    # Generates only variable objects with dimensions ["x", "y"]
+    xrst.variables(dims=st.just(["x", "y"])).example()
+
+(This is technically another example of chaining strategies - :py:func:`hypothesis.strategies.just()` is simply a
+special strategy that just contains a single example.)
+
+To fix the length of dimensions you can instead pass ``dims`` as a mapping of dimension names to lengths
+(i.e. following xarray objects' ``.sizes()`` property), e.g.
+
+.. ipython:: python
+
+    # Generates only variables with dimensions ["x", "y"], of lengths 2 & 3 respectively
+    xrst.variables(dims=st.just({"x": 2, "y": 3})).example()
+
+You can also use this to specify that you want examples which are missing some part of the data structure, for instance
+
+.. ipython:: python
+
+    # Generates a Variable with no attributes
+    xrst.variables(attrs=st.just({})).example()
+
+Through a combination of chaining strategies and fixing arguments, you can specify quite complicated requirements on the
+objects your chained strategy will generate.
+
+.. ipython:: python
+
+    fixed_x_variable_y_maybe_z = st.fixed_dictionaries(
+        {"x": st.just(2), "y": st.integers(3, 4)}, optional={"z": st.just(2)}
+    )
+    fixed_x_variable_y_maybe_z.example()
+
+    special_variables = xrst.variables(dims=fixed_x_variable_y_maybe_z)
+
+    special_variables.example()
+    special_variables.example()
+
+Here we have used one of hypothesis' built-in strategies :py:func:`hypothesis.strategies.fixed_dictionaries` to create a
+strategy which generates mappings of dimension names to lengths (i.e. the ``size`` of the xarray object we want).
+This particular strategy will always generate an ``x`` dimension of length 2, and a ``y`` dimension of
+length either 3 or 4, and will sometimes also generate a ``z`` dimension of length 2.
+By feeding this strategy for dictionaries into the ``dims`` argument of xarray's :py:func:`~st.variables` strategy,
+we can generate arbitrary :py:class:`~xarray.Variable` objects whose dimensions will always match these specifications.
+
+Generating Duck-type Arrays
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Xarray objects don't have to wrap numpy arrays, in fact they can wrap any array type which presents the same API as a
+numpy array (so-called "duck array wrapping", see :ref:`wrapping numpy-like arrays <internals.duckarrays>`).
+
+Imagine we want to write a strategy which generates arbitrary ``Variable`` objects, each of which wraps a
+:py:class:`sparse.COO` array instead of a ``numpy.ndarray``. How could we do that? There are two ways:
+
+1. Create a xarray object with numpy data and use the hypothesis' ``.map()`` method to convert the underlying array to a
+different type:
+
+.. ipython:: python
+
+    import sparse
+
+.. ipython:: python
+
+    def convert_to_sparse(var):
+        return var.copy(data=sparse.COO.from_numpy(var.to_numpy()))
+
+.. ipython:: python
+
+    sparse_variables = xrst.variables(dims=xrst.dimension_names(min_dims=1)).map(
+        convert_to_sparse
+    )
+
+    sparse_variables.example()
+    sparse_variables.example()
+
+2. Pass a function which returns a strategy which generates the duck-typed arrays directly to the ``array_strategy_fn`` argument of the xarray strategies:
+
+.. ipython:: python
+
+    def sparse_random_arrays(shape: tuple[int]) -> sparse._coo.core.COO:
+        """Strategy which generates random sparse.COO arrays"""
+        if shape is None:
+            shape = npst.array_shapes()
+        else:
+            shape = st.just(shape)
+        density = st.integers(min_value=0, max_value=1)
+        # note sparse.random does not accept a dtype kwarg
+        return st.builds(sparse.random, shape=shape, density=density)
+
+
+    def sparse_random_arrays_fn(
+        *, shape: tuple[int, ...], dtype: np.dtype
+    ) -> st.SearchStrategy[sparse._coo.core.COO]:
+        return sparse_random_arrays(shape=shape)
+
+
+.. ipython:: python
+
+    sparse_random_variables = xrst.variables(
+        array_strategy_fn=sparse_random_arrays_fn, dtype=st.just(np.dtype("float64"))
+    )
+    sparse_random_variables.example()
+
+Either approach is fine, but one may be more convenient than the other depending on the type of the duck array which you
+want to wrap.
+
+Compatibility with the Python Array API Standard
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Xarray aims to be compatible with any duck-array type that conforms to the `Python Array API Standard <https://data-apis.org/array-api/latest/>`_
+(see our :ref:`docs on Array API Standard support <internals.duckarrays.array_api_standard>`).
+
+.. warning::
+
+    The strategies defined in :py:mod:`testing.strategies` are **not** guaranteed to use array API standard-compliant
+    dtypes by default.
+    For example arrays with the dtype ``np.dtype('float16')`` may be generated by :py:func:`testing.strategies.variables`
+    (assuming the ``dtype`` kwarg was not explicitly passed), despite ``np.dtype('float16')`` not being in the
+    array API standard.
+
+If the array type you want to generate has an array API-compliant top-level namespace
+(e.g. that which is conventionally imported as ``xp`` or similar),
+you can use this neat trick:
+
+.. ipython:: python
+    :okwarning:
+
+    from numpy import array_api as xp  # available in numpy 1.26.0
+
+    from hypothesis.extra.array_api import make_strategies_namespace
+
+    xps = make_strategies_namespace(xp)
+
+    xp_variables = xrst.variables(
+        array_strategy_fn=xps.arrays,
+        dtype=xps.scalar_dtypes(),
+    )
+    xp_variables.example()
+
+Another array API-compliant duck array library would replace the import, e.g. ``import cupy as cp`` instead.
+
+Testing over Subsets of Dimensions
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+A common task when testing xarray user code is checking that your function works for all valid input dimensions.
+We can chain strategies to achieve this, for which the helper strategy :py:func:`~testing.strategies.unique_subset_of`
+is useful.
+
+It works for lists of dimension names
+
+.. ipython:: python
+
+    dims = ["x", "y", "z"]
+    xrst.unique_subset_of(dims).example()
+    xrst.unique_subset_of(dims).example()
+
+as well as for mappings of dimension names to sizes
+
+.. ipython:: python
+
+    dim_sizes = {"x": 2, "y": 3, "z": 4}
+    xrst.unique_subset_of(dim_sizes).example()
+    xrst.unique_subset_of(dim_sizes).example()
+
+This is useful because operations like reductions can be performed over any subset of the xarray object's dimensions.
+For example we can write a pytest test that tests that a reduction gives the expected result when applying that reduction
+along any possible valid subset of the Variable's dimensions.
+
+.. code-block:: python
+
+    import numpy.testing as npt
+
+
+    @given(st.data(), xrst.variables(dims=xrst.dimension_names(min_dims=1)))
+    def test_mean(data, var):
+        """Test that the mean of an xarray Variable is always equal to the mean of the underlying array."""
+
+        # specify arbitrary reduction along at least one dimension
+        reduction_dims = data.draw(xrst.unique_subset_of(var.dims, min_size=1))
+
+        # create expected result (using nanmean because arrays with Nans will be generated)
+        reduction_axes = tuple(var.get_axis_num(dim) for dim in reduction_dims)
+        expected = np.nanmean(var.data, axis=reduction_axes)
+
+        # assert property is always satisfied
+        result = var.mean(dim=reduction_dims).data
+        npt.assert_equal(expected, result)
diff --git a/doc/whats-new.rst b/doc/whats-new.rst
@@ -23,6 +23,10 @@ v2023.11.1 (unreleased)
 New Features
 ~~~~~~~~~~~~
 
+- Added hypothesis strategies for generating :py:class:`xarray.Variable` objects containing arbitrary data, useful for parametrizing downstream tests.
+  Accessible under :py:mod:`testing.strategies`, and documented in a new page on testing in the User Guide.
+  (:issue:`6911`, :pull:`8404`)
+  By `Tom Nicholas <https://github.com/TomNicholas>`_.
 - :py:meth:`rolling` uses numbagg <https://github.com/numbagg/numbagg>`_ for
   most of its computations by default. Numbagg is up to 5x faster than bottleneck
   where parallelization is possible. Where parallelization isn't possible — for

diff --git a/xarray/core/types.py b/xarray/core/types.py
@@ -173,7 +173,8 @@ def copy(
 
 # Temporary placeholder for indicating an array api compliant type.
 # hopefully in the future we can narrow this down more:
-T_DuckArray = TypeVar("T_DuckArray", bound=Any)
+T_DuckArray = TypeVar("T_DuckArray", bound=Any, covariant=True)
+
 
 ScalarOrArray = Union["ArrayLike", np.generic, np.ndarray, "DaskArray"]
 VarCompatible = Union["Variable", "ScalarOrArray"]
-Original file line number
+Diff line change
@@ Expand Up @@
        dask
        plotting
        options
+       testing
        duckarrays