def rule(
*,
targets: Union[Sequence[Bundle[Ex]], _OmittedArgument] = (),
target: Optional[Bundle[Ex]] = None,
**kwargs: SearchStrategy,
) -> Union[_RuleWrapper[Ex], Callable[[Callable[..., None]], Callable[...,
None]]]:
"""Decorator for RuleBasedStateMachine. Any Bundle present in ``target`` or
``targets`` will define where the end result of this function should go. If
both are empty then the end result will be discarded.
``target`` must be a Bundle, or if the result should go to multiple
bundles you can pass a tuple of them as the ``targets`` argument.
It is invalid to use both arguments for a single rule. If the result
should go to exactly one of several bundles, define a separate rule for
each case.
kwargs then define the arguments that will be passed to the function
invocation. If their value is a Bundle, or if it is ``consumes(b)``
where ``b`` is a Bundle, then values that have previously been produced
for that bundle will be provided. If ``consumes`` is used, the value
will also be removed from the bundle.
Any other kwargs should be strategies and values from them will be
provided.
"""
converted_targets = _convert_targets(targets, target)
for k, v in kwargs.items():
check_strategy(v, name=k)
def accept(f):
if getattr(f, INVARIANT_MARKER, None):
raise InvalidDefinition(
"A function cannot be used for both a rule and an invariant.",
Settings.default,
)
existing_rule = getattr(f, RULE_MARKER, None)
existing_initialize_rule = getattr(f, INITIALIZE_RULE_MARKER, None)
if existing_rule is not None or existing_initialize_rule is not None:
raise InvalidDefinition(
"A function cannot be used for two distinct rules. ",
Settings.default)
preconditions = getattr(f, PRECONDITIONS_MARKER, ())
rule = Rule(
targets=converted_targets,
arguments=kwargs,
function=f,
preconditions=preconditions,
)
@proxies(f)
def rule_wrapper(*args, **kwargs):
return f(*args, **kwargs)
setattr(rule_wrapper, RULE_MARKER, rule)
return rule_wrapper
return accept
def fill_for(elements, unique, fill, name=""):
if fill is None:
if unique or not elements.has_reusable_values:
fill = st.nothing()
else:
fill = elements
else:
check_strategy(fill, f"{name}.fill" if name else "fill")
return fill
def elements_and_dtype(elements, dtype, source=None):
if source is None:
prefix = ""
else:
prefix = f"{source}."
if elements is not None:
check_strategy(elements, f"{prefix}elements")
else:
with check("dtype is not None"):
if dtype is None:
raise InvalidArgument(
f"At least one of {prefix}elements or {prefix}dtype must be provided."
)
with check("is_categorical_dtype"):
if is_categorical_dtype(dtype):
raise InvalidArgument(
f"{prefix}dtype is categorical, which is currently unsupported"
)
if isinstance(
dtype,
type) and np.dtype(dtype).kind == "O" and dtype is not object:
note_deprecation(
f"Passed dtype={dtype!r} is not a valid Pandas dtype. We'll treat it as "
"dtype=object for now, but this will be an error in a future version.",
since="2021-12-31",
has_codemod=False,
)
dtype = try_convert(np.dtype, dtype, "dtype")
if elements is None:
elements = npst.from_dtype(dtype)
elif dtype is not None:
def convert_element(value):
name = f"draw({prefix}elements)"
try:
return np.array([value], dtype=dtype)[0]
except TypeError:
raise InvalidArgument(
"Cannot convert %s=%r of type %s to dtype %s" %
(name, value, type(value).__name__, dtype.str)) from None
except ValueError:
raise InvalidArgument(
f"Cannot convert {name}={value!r} to type {dtype.str}"
) from None
elements = elements.map(convert_element)
assert elements is not None
return elements, dtype
def tuples(*args: SearchStrategy) -> SearchStrategy[tuple]:
"""Return a strategy which generates a tuple of the same length as args by
generating the value at index i from args[i].
e.g. tuples(integers(), integers()) would generate a tuple of length
two with both values an integer.
Examples from this strategy shrink by shrinking their component parts.
"""
for arg in args:
check_strategy(arg)
return TupleStrategy(args)
def initialize(
*,
targets: Union[Sequence[Bundle[Ex]], _OmittedArgument] = (),
target: Optional[Bundle[Ex]] = None,
**kwargs: SearchStrategy,
) -> Union[_RuleWrapper[Ex], Callable[[Callable[..., None]], Callable[...,
None]]]:
"""Decorator for RuleBasedStateMachine.
An initialize decorator behaves like a rule, but all ``@initialize()`` decorated
methods will be called before any ``@rule()`` decorated methods, in an arbitrary
order. Each ``@initialize()`` method will be called exactly once per run, unless
one raises an exception - after which only the ``.teardown()`` method will be run.
``@initialize()`` methods may not have preconditions.
"""
converted_targets = _convert_targets(targets, target)
for k, v in kwargs.items():
check_strategy(v, name=k)
def accept(f):
if getattr(f, INVARIANT_MARKER, None):
raise InvalidDefinition(
"A function cannot be used for both a rule and an invariant.",
Settings.default,
)
existing_rule = getattr(f, RULE_MARKER, None)
existing_initialize_rule = getattr(f, INITIALIZE_RULE_MARKER, None)
if existing_rule is not None or existing_initialize_rule is not None:
raise InvalidDefinition(
"A function cannot be used for two distinct rules. ",
Settings.default)
preconditions = getattr(f, PRECONDITIONS_MARKER, ())
if preconditions:
raise InvalidDefinition(
"An initialization rule cannot have a precondition. ",
Settings.default)
rule = Rule(
targets=converted_targets,
arguments=kwargs,
function=f,
preconditions=preconditions,
)
@proxies(f)
def rule_wrapper(*args, **kwargs):
return f(*args, **kwargs)
setattr(rule_wrapper, INITIALIZE_RULE_MARKER, rule)
return rule_wrapper
return accept
def wrapped_strategy(self):
if self.__wrapped_strategy is None:
if not inspect.isfunction(self.__definition):
raise InvalidArgument(
f"Expected definition to be a function but got {self.__definition!r} "
f"of type {type(self.__definition).__name__} instead.")
result = self.__definition()
if result is self:
raise InvalidArgument(
"Cannot define a deferred strategy to be itself")
check_strategy(result, "definition()")
self.__wrapped_strategy = result
self.__definition = None
return self.__wrapped_strategy
def elements_and_dtype(elements, dtype, source=None):
if source is None:
prefix = ""
else:
prefix = f"{source}."
if elements is not None:
check_strategy(elements, f"{prefix}elements")
else:
with check("dtype is not None"):
if dtype is None:
raise InvalidArgument(
f"At least one of {prefix}elements or {prefix}dtype must be provided."
)
with check("is_categorical_dtype"):
if is_categorical_dtype(dtype):
raise InvalidArgument(
f"{prefix}dtype is categorical, which is currently unsupported"
)
dtype = try_convert(np.dtype, dtype, "dtype")
if elements is None:
elements = npst.from_dtype(dtype)
elif dtype is not None:
def convert_element(value):
name = f"draw({prefix}elements)"
try:
return np.array([value], dtype=dtype)[0]
except TypeError:
raise InvalidArgument(
"Cannot convert %s=%r of type %s to dtype %s" %
(name, value, type(value).__name__, dtype.str)) from None
except ValueError:
raise InvalidArgument(
f"Cannot convert {name}={value!r} to type {dtype.str}"
) from None
elements = elements.map(convert_element)
assert elements is not None
return elements, dtype
def initialize(*, targets=(), target=None, **kwargs):
"""Decorator for RuleBasedStateMachine.
An initialize decorator behaves like a rule, but the decorated
method is called at most once in a run. All initialize decorated
methods will be called before any rule decorated methods, in an
arbitrary order.
"""
converted_targets = _convert_targets(targets, target)
for k, v in kwargs.items():
check_strategy(v, name=k)
def accept(f):
existing_rule = getattr(f, RULE_MARKER, None)
existing_initialize_rule = getattr(f, INITIALIZE_RULE_MARKER, None)
if existing_rule is not None or existing_initialize_rule is not None:
raise InvalidDefinition(
"A function cannot be used for two distinct rules. ",
Settings.default)
precondition = getattr(f, PRECONDITION_MARKER, None)
if precondition:
raise InvalidDefinition(
"An initialization rule cannot have a precondition. ",
Settings.default)
rule = Rule(
targets=converted_targets,
arguments=kwargs,
function=f,
precondition=precondition,
)
@proxies(f)
def rule_wrapper(*args, **kwargs):
return f(*args, **kwargs)
setattr(rule_wrapper, INITIALIZE_RULE_MARKER, rule)
return rule_wrapper
return accept
def _arrays(
xp: Any,
dtype: Union[DataType, str, st.SearchStrategy[DataType],
st.SearchStrategy[str]],
shape: Union[int, Shape, st.SearchStrategy[Shape]],
*,
elements: Optional[Union[Mapping[str, Any], st.SearchStrategy]] = None,
fill: Optional[st.SearchStrategy[Any]] = None,
unique: bool = False,
) -> st.SearchStrategy:
"""Returns a strategy for :xp-ref:`arrays <array_object.html>`.
* ``dtype`` may be a :xp-ref:`valid dtype <data_types.html>` object or name,
or a strategy that generates such values.
* ``shape`` may be an integer >= 0, a tuple of such integers, or a strategy
that generates such values.
* ``elements`` is a strategy for values to put in the array. If ``None``
then a suitable value will be inferred based on the dtype, which may give
any legal value (including e.g. NaN for floats). If a mapping, it will be
passed as ``**kwargs`` to :func:`from_dtype()` when inferring based on the dtype.
* ``fill`` is a strategy that may be used to generate a single background
value for the array. If ``None``, a suitable default will be inferred
based on the other arguments. If set to
:func:`~hypothesis.strategies.nothing` then filling behaviour will be
disabled entirely and every element will be generated independently.
* ``unique`` specifies if the elements of the array should all be distinct
from one another; if fill is also set, the only valid values for fill to
return are NaN values.
Arrays of specified ``dtype`` and ``shape`` are generated for example
like this:
.. code-block:: pycon
>>> from numpy import array_api as xp
>>> xps.arrays(xp, xp.int8, (2, 3)).example()
Array([[-8, 6, 3],
[-6, 4, 6]], dtype=int8)
Specifying element boundaries by a :obj:`python:dict` of the kwargs to pass
to :func:`from_dtype` will ensure ``dtype`` bounds will be respected.
.. code-block:: pycon
>>> xps.arrays(xp, xp.int8, 3, elements={"min_value": 10}).example()
Array([125, 13, 79], dtype=int8)
Refer to :doc:`What you can generate and how <data>` for passing
your own elements strategy.
.. code-block:: pycon
>>> xps.arrays(xp, xp.float32, 3, elements=floats(0, 1, width=32)).example()
Array([ 0.88974794, 0.77387938, 0.1977879 ], dtype=float32)
Array values are generated in two parts:
1. A single value is drawn from the fill strategy and is used to create a
filled array.
2. Some subset of the coordinates of the array are populated with a value
drawn from the elements strategy (or its inferred form).
You can set ``fill`` to :func:`~hypothesis.strategies.nothing` if you want
to disable this behaviour and draw a value for every element.
By default ``arrays`` will attempt to infer the correct fill behaviour: if
``unique`` is also ``True``, no filling will occur. Otherwise, if it looks
safe to reuse the values of elements across multiple coordinates (this will
be the case for any inferred strategy, and for most of the builtins, but is
not the case for mutable values or strategies built with flatmap, map,
composite, etc.) then it will use the elements strategy as the fill, else it
will default to having no fill.
Having a fill helps Hypothesis craft high quality examples, but its
main importance is when the array generated is large: Hypothesis is
primarily designed around testing small examples. If you have arrays with
hundreds or more elements, having a fill value is essential if you want
your tests to run in reasonable time.
"""
check_xp_attributes(xp, [
"finfo", "asarray", "zeros", "full", "all", "isnan", "isfinite",
"reshape"
])
if isinstance(dtype, st.SearchStrategy):
return dtype.flatmap(lambda d: _arrays(
xp, d, shape, elements=elements, fill=fill, unique=unique))
elif isinstance(dtype, str):
dtype = dtype_from_name(xp, dtype)
if isinstance(shape, st.SearchStrategy):
return shape.flatmap(lambda s: _arrays(
xp, dtype, s, elements=elements, fill=fill, unique=unique))
elif isinstance(shape, int):
shape = (shape, )
elif not isinstance(shape, tuple):
raise InvalidArgument(
f"shape={shape} is not a valid shape or strategy")
check_argument(
all(isinstance(x, int) and x >= 0 for x in shape),
f"shape={shape!r}, but all dimensions must be non-negative integers.",
)
if elements is None:
elements = _from_dtype(xp, dtype)
elif isinstance(elements, Mapping):
elements = _from_dtype(xp, dtype, **elements)
check_strategy(elements, "elements")
if fill is None:
assert isinstance(elements, st.SearchStrategy) # for mypy
if unique or not elements.has_reusable_values:
fill = st.nothing()
else:
fill = elements
check_strategy(fill, "fill")
return ArrayStrategy(xp, elements, dtype, shape, fill, unique)
def do_validate(self):
check_strategy(self.base, "base")
extended = self.extend(self.limited_base)
check_strategy(extended, f"extend({self.limited_base!r})")
self.limited_base.validate()
extended.validate()
def data_frames(
columns: Optional[Sequence[column]] = None,
*,
rows: Optional[st.SearchStrategy[Union[dict, Sequence[Any]]]] = None,
index: Optional[st.SearchStrategy[Ex]] = None,
) -> st.SearchStrategy[pandas.DataFrame]:
"""Provides a strategy for producing a :class:`pandas.DataFrame`.
Arguments:
* columns: An iterable of :class:`column` objects describing the shape
of the generated DataFrame.
* rows: A strategy for generating a row object. Should generate
either dicts mapping column names to values or a sequence mapping
column position to the value in that position (note that unlike the
:class:`pandas.DataFrame` constructor, single values are not allowed
here. Passing e.g. an integer is an error, even if there is only one
column).
At least one of rows and columns must be provided. If both are
provided then the generated rows will be validated against the
columns and an error will be raised if they don't match.
Caveats on using rows:
* In general you should prefer using columns to rows, and only use
rows if the columns interface is insufficiently flexible to
describe what you need - you will get better performance and
example quality that way.
* If you provide rows and not columns, then the shape and dtype of
the resulting DataFrame may vary. e.g. if you have a mix of int
and float in the values for one column in your row entries, the
column will sometimes have an integral dtype and sometimes a float.
* index: If not None, a strategy for generating indexes for the
resulting DataFrame. This can generate either :class:`pandas.Index`
objects or any sequence of values (which will be passed to the
Index constructor).
You will probably find it most convenient to use the
:func:`~hypothesis.extra.pandas.indexes` or
:func:`~hypothesis.extra.pandas.range_indexes` function to produce
values for this argument.
Usage:
The expected usage pattern is that you use :class:`column` and
:func:`columns` to specify a fixed shape of the DataFrame you want as
follows. For example the following gives a two column data frame:
.. code-block:: pycon
>>> from hypothesis.extra.pandas import column, data_frames
>>> data_frames([
... column('A', dtype=int), column('B', dtype=float)]).example()
A B
0 2021915903 1.793898e+232
1 1146643993 inf
2 -2096165693 1.000000e+07
If you want the values in different columns to interact in some way you
can use the rows argument. For example the following gives a two column
DataFrame where the value in the first column is always at most the value
in the second:
.. code-block:: pycon
>>> from hypothesis.extra.pandas import column, data_frames
>>> import hypothesis.strategies as st
>>> data_frames(
... rows=st.tuples(st.floats(allow_nan=False),
... st.floats(allow_nan=False)).map(sorted)
... ).example()
0 1
0 -3.402823e+38 9.007199e+15
1 -1.562796e-298 5.000000e-01
You can also combine the two:
.. code-block:: pycon
>>> from hypothesis.extra.pandas import columns, data_frames
>>> import hypothesis.strategies as st
>>> data_frames(
... columns=columns(["lo", "hi"], dtype=float),
... rows=st.tuples(st.floats(allow_nan=False),
... st.floats(allow_nan=False)).map(sorted)
... ).example()
lo hi
0 9.314723e-49 4.353037e+45
1 -9.999900e-01 1.000000e+07
2 -2.152861e+134 -1.069317e-73
(Note that the column dtype must still be specified and will not be
inferred from the rows. This restriction may be lifted in future).
Combining rows and columns has the following behaviour:
* The column names and dtypes will be used.
* If the column is required to be unique, this will be enforced.
* Any values missing from the generated rows will be provided using the
column's fill.
* Any values in the row not present in the column specification (if
dicts are passed, if there are keys with no corresponding column name,
if sequences are passed if there are too many items) will result in
InvalidArgument being raised.
"""
if index is None:
index = range_indexes()
else:
check_strategy(index, "index")
index_strategy = index
if columns is None:
if rows is None:
raise InvalidArgument(
"At least one of rows and columns must be provided")
else:
@st.composite
def rows_only(draw):
index = draw(index_strategy)
@check_function
def row():
result = draw(rows)
check_type(abc.Iterable, result, "draw(row)")
return result
if len(index) > 0:
return pandas.DataFrame([row() for _ in index],
index=index)
else:
# If we haven't drawn any rows we need to draw one row and
# then discard it so that we get a consistent shape for the
# DataFrame.
base = pandas.DataFrame([row()])
return base.drop(0)
return rows_only()
assert columns is not None
cols = try_convert(tuple, columns, "columns")
rewritten_columns = []
column_names: Set[str] = set()
for i, c in enumerate(cols):
check_type(column, c, f"columns[{i}]")
c = copy(c)
if c.name is None:
label = f"columns[{i}]"
c.name = i
else:
label = c.name
try:
hash(c.name)
except TypeError:
raise InvalidArgument(
"Column names must be hashable, but columns[%d].name was "
"%r of type %s, which cannot be hashed." %
(i, c.name, type(c.name).__name__)) from None
if c.name in column_names:
raise InvalidArgument(
f"duplicate definition of column name {c.name!r}")
column_names.add(c.name)
c.elements, c.dtype = elements_and_dtype(c.elements, c.dtype, label)
if c.dtype is None and rows is not None:
raise InvalidArgument(
"Must specify a dtype for all columns when combining rows with columns."
)
c.fill = npst.fill_for(fill=c.fill,
elements=c.elements,
unique=c.unique,
name=label)
rewritten_columns.append(c)
if rows is None:
@st.composite
def just_draw_columns(draw):
index = draw(index_strategy)
local_index_strategy = st.just(index)
data = OrderedDict((c.name, None) for c in rewritten_columns)
# Depending on how the columns are going to be generated we group
# them differently to get better shrinking. For columns with fill
# enabled, the elements can be shrunk independently of the size,
# so we can just shrink by shrinking the index then shrinking the
# length and are generally much more free to move data around.
# For columns with no filling the problem is harder, and drawing
# them like that would result in rows being very far apart from
# each other in the underlying data stream, which gets in the way
# of shrinking. So what we do is reorder and draw those columns
# row wise, so that the values of each row are next to each other.
# This makes life easier for the shrinker when deleting blocks of
# data.
columns_without_fill = [
c for c in rewritten_columns if c.fill.is_empty
]
if columns_without_fill:
for c in columns_without_fill:
data[c.name] = pandas.Series(np.zeros(shape=len(index),
dtype=c.dtype),
index=index)
seen = {
c.name: set()
for c in columns_without_fill if c.unique
}
for i in range(len(index)):
for c in columns_without_fill:
if c.unique:
for _ in range(5):
value = draw(c.elements)
if value not in seen[c.name]:
seen[c.name].add(value)
break
else:
reject()
else:
value = draw(c.elements)
try:
data[c.name][i] = value
except ValueError as err: # pragma: no cover
# This just works in Pandas 1.4 and later, but gives
# a confusing error on previous versions.
if c.dtype is None and not isinstance(
value,
(float, int, str, bool, datetime, timedelta)):
raise ValueError(
f"Failed to add value={value!r} to column "
f"{c.name} with dtype=None. Maybe passing "
"dtype=object would help?") from err
# Unclear how this could happen, but users find a way...
raise
for c in rewritten_columns:
if not c.fill.is_empty:
data[c.name] = draw(
series(
index=local_index_strategy,
dtype=c.dtype,
elements=c.elements,
fill=c.fill,
unique=c.unique,
))
return pandas.DataFrame(data, index=index)
return just_draw_columns()
else:
@st.composite
def assign_rows(draw):
index = draw(index_strategy)
result = pandas.DataFrame(
OrderedDict((
c.name,
pandas.Series(np.zeros(dtype=c.dtype, shape=len(index)),
dtype=c.dtype),
) for c in rewritten_columns),
index=index,
)
fills = {}
any_unique = any(c.unique for c in rewritten_columns)
if any_unique:
all_seen = [
set() if c.unique else None for c in rewritten_columns
]
while all_seen[-1] is None:
all_seen.pop()
for row_index in range(len(index)):
for _ in range(5):
original_row = draw(rows)
row = original_row
if isinstance(row, dict):
as_list = [None] * len(rewritten_columns)
for i, c in enumerate(rewritten_columns):
try:
as_list[i] = row[c.name]
except KeyError:
try:
as_list[i] = fills[i]
except KeyError:
if c.fill.is_empty:
raise InvalidArgument(
f"Empty fill strategy in {c!r} cannot "
f"complete row {original_row!r}"
) from None
fills[i] = draw(c.fill)
as_list[i] = fills[i]
for k in row:
if k not in column_names:
raise InvalidArgument(
"Row %r contains column %r not in columns %r)"
% (row, k,
[c.name for c in rewritten_columns]))
row = as_list
if any_unique:
has_duplicate = False
for seen, value in zip(all_seen, row):
if seen is None:
continue
if value in seen:
has_duplicate = True
break
seen.add(value)
if has_duplicate:
continue
row = list(try_convert(tuple, row, "draw(rows)"))
if len(row) > len(rewritten_columns):
raise InvalidArgument(
f"Row {original_row!r} contains too many entries. Has "
f"{len(row)} but expected at most {len(rewritten_columns)}"
)
while len(row) < len(rewritten_columns):
c = rewritten_columns[len(row)]
if c.fill.is_empty:
raise InvalidArgument(
f"Empty fill strategy in {c!r} cannot "
f"complete row {original_row!r}")
row.append(draw(c.fill))
result.iloc[row_index] = row
break
else:
reject()
return result
return assign_rows()
def series(
*,
elements: Optional[st.SearchStrategy[Ex]] = None,
dtype: Any = None,
index: Optional[st.SearchStrategy[Union[Sequence, pandas.Index]]] = None,
fill: Optional[st.SearchStrategy[Ex]] = None,
unique: bool = False,
) -> st.SearchStrategy[pandas.Series]:
"""Provides a strategy for producing a :class:`pandas.Series`.
Arguments:
* elements: a strategy that will be used to generate the individual
values in the series. If None, we will attempt to infer a suitable
default from the dtype.
* dtype: the dtype of the resulting series and may be any value
that can be passed to :class:`numpy.dtype`. If None, will use
pandas's standard behaviour to infer it from the type of the elements
values. Note that if the type of values that comes out of your
elements strategy varies, then so will the resulting dtype of the
series.
* index: If not None, a strategy for generating indexes for the
resulting Series. This can generate either :class:`pandas.Index`
objects or any sequence of values (which will be passed to the
Index constructor).
You will probably find it most convenient to use the
:func:`~hypothesis.extra.pandas.indexes` or
:func:`~hypothesis.extra.pandas.range_indexes` function to produce
values for this argument.
Usage:
.. code-block:: pycon
>>> series(dtype=int).example()
0 -2001747478
1 1153062837
"""
if index is None:
index = range_indexes()
else:
check_strategy(index, "index")
elements, dtype = elements_and_dtype(elements, dtype)
index_strategy = index
@st.composite
def result(draw):
index = draw(index_strategy)
if len(index) > 0:
if dtype is not None:
result_data = draw(
npst.arrays(
dtype=dtype,
elements=elements,
shape=len(index),
fill=fill,
unique=unique,
))
else:
result_data = list(
draw(
npst.arrays(
dtype=object,
elements=elements,
shape=len(index),
fill=fill,
unique=unique,
)))
return pandas.Series(result_data, index=index, dtype=dtype)
else:
return pandas.Series(
(),
index=index,
dtype=dtype if dtype is not None else draw(
dtype_for_elements_strategy(elements)),
)
return result()
def do_draw(self, data):
source = data.draw(self.flatmapped_strategy)
expanded_source = self.expand(source)
check_strategy(expanded_source)
return data.draw(expanded_source)
def check_strategy_(*args):
return check_strategy(*args)
def integer_array_indices(
shape: Shape,
*,
result_shape: st.SearchStrategy[Shape] = array_shapes(),
dtype: np.dtype = "int",
) -> st.SearchStrategy[Tuple[np.ndarray, ...]]:
"""Return a search strategy for tuples of integer-arrays that, when used
to index into an array of shape ``shape``, given an array whose shape
was drawn from ``result_shape``.
Examples from this strategy shrink towards the tuple of index-arrays::
len(shape) * (np.zeros(drawn_result_shape, dtype), )
* ``shape`` a tuple of integers that indicates the shape of the array,
whose indices are being generated.
* ``result_shape`` a strategy for generating tuples of integers, which
describe the shape of the resulting index arrays. The default is
:func:`~hypothesis.extra.numpy.array_shapes`. The shape drawn from
this strategy determines the shape of the array that will be produced
when the corresponding example from ``integer_array_indices`` is used
as an index.
* ``dtype`` the integer data type of the generated index-arrays. Negative
integer indices can be generated if a signed integer type is specified.
Recall that an array can be indexed using a tuple of integer-arrays to
access its members in an arbitrary order, producing an array with an
arbitrary shape. For example:
.. code-block:: pycon
>>> from numpy import array
>>> x = array([-0, -1, -2, -3, -4])
>>> ind = (array([[4, 0], [0, 1]]),) # a tuple containing a 2D integer-array
>>> x[ind] # the resulting array is commensurate with the indexing array(s)
array([[-4, 0],
[0, -1]])
Note that this strategy does not accommodate all variations of so-called
'advanced indexing', as prescribed by NumPy's nomenclature. Combinations
of basic and advanced indexes are too complex to usefully define in a
standard strategy; we leave application-specific strategies to the user.
Advanced-boolean indexing can be defined as ``arrays(shape=..., dtype=bool)``,
and is similarly left to the user.
"""
check_type(tuple, shape, "shape")
check_argument(
shape and all(isinstance(x, int) and x > 0 for x in shape),
f"shape={shape!r} must be a non-empty tuple of integers > 0",
)
check_strategy(result_shape, "result_shape")
check_argument(np.issubdtype(dtype, np.integer),
f"dtype={dtype!r} must be an integer dtype")
signed = np.issubdtype(dtype, np.signedinteger)
def array_for(index_shape, size):
return arrays(
dtype=dtype,
shape=index_shape,
elements=st.integers(-size if signed else 0, size - 1),
)
return result_shape.flatmap(lambda index_shape: st.tuples(*(
array_for(index_shape, size) for size in shape)))
def arrays(
dtype: Any,
shape: Union[int, st.SearchStrategy[int], Shape, st.SearchStrategy[Shape]],
*,
elements: Optional[Union[st.SearchStrategy, Mapping[str, Any]]] = None,
fill: Optional[st.SearchStrategy[Any]] = None,
unique: bool = False,
) -> st.SearchStrategy[np.ndarray]:
r"""Returns a strategy for generating :class:`numpy:numpy.ndarray`\ s.
* ``dtype`` may be any valid input to :class:`~numpy:numpy.dtype`
(this includes :class:`~numpy:numpy.dtype` objects), or a strategy that
generates such values.
* ``shape`` may be an integer >= 0, a tuple of such integers, or a
strategy that generates such values.
* ``elements`` is a strategy for generating values to put in the array.
If it is None a suitable value will be inferred based on the dtype,
which may give any legal value (including eg ``NaN`` for floats).
If a mapping, it will be passed as ``**kwargs`` to ``from_dtype()``
* ``fill`` is a strategy that may be used to generate a single background
value for the array. If None, a suitable default will be inferred
based on the other arguments. If set to
:func:`~hypothesis.strategies.nothing` then filling
behaviour will be disabled entirely and every element will be generated
independently.
* ``unique`` specifies if the elements of the array should all be
distinct from one another. Note that in this case multiple NaN values
may still be allowed. If fill is also set, the only valid values for
it to return are NaN values (anything for which :obj:`numpy:numpy.isnan`
returns True. So e.g. for complex numbers (nan+1j) is also a valid fill).
Note that if unique is set to True the generated values must be hashable.
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)
- See :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(0, 1)).example()
array([ 0.88974794, 0.77387938, 0.1977879 ])
Array values are generated in two parts:
1. Some subset of the coordinates of the array are populated with a value
drawn from the elements strategy (or its inferred form).
2. If any coordinates were not assigned in the previous step, a single
value is drawn from the fill strategy and is assigned to all remaining
places.
You can set fill to :func:`~hypothesis.strategies.nothing` if you want to
disable this behaviour and draw a value for every element.
If fill is set to None then it will attempt to infer the correct behaviour
automatically: If unique is True, no filling will occur by default.
Otherwise, if it looks safe to reuse the values of elements across
multiple coordinates (this will be the case for any inferred strategy, and
for most of the builtins, but is not the case for mutable values or
strategies built with flatmap, map, composite, etc) then it will use the
elements strategy as the fill, else it will default to having no fill.
Having a fill helps Hypothesis craft high quality examples, but its
main importance is when the array generated is large: Hypothesis is
primarily designed around testing small examples. If you have arrays with
hundreds or more elements, having a fill value is essential if you want
your tests to run in reasonable time.
"""
# We support passing strategies as arguments for convenience, or at least
# for legacy reasons, but don't want to pay the perf cost of a composite
# strategy (i.e. repeated argument handling and validation) when it's not
# needed. So we get the best of both worlds by recursing with flatmap,
# but only when it's actually needed.
if isinstance(dtype, st.SearchStrategy):
return dtype.flatmap(lambda d: arrays(
d, shape, elements=elements, fill=fill, unique=unique))
if isinstance(shape, st.SearchStrategy):
return shape.flatmap(lambda s: arrays(
dtype, s, elements=elements, fill=fill, unique=unique))
# From here on, we're only dealing with values and it's relatively simple.
dtype = np.dtype(dtype)
if elements is None or isinstance(elements, Mapping):
if dtype.kind in ("m", "M") and "[" not in dtype.str:
# For datetime and timedelta dtypes, we have a tricky situation -
# because they *may or may not* specify a unit as part of the dtype.
# If not, we flatmap over the various resolutions so that array
# elements have consistent units but units may vary between arrays.
return (st.sampled_from(TIME_RESOLUTIONS).map(
(dtype.str + "[{}]").format).flatmap(lambda d: arrays(
d, shape=shape, fill=fill, unique=unique)))
elements = from_dtype(dtype, **(elements or {}))
check_strategy(elements, "elements")
if isinstance(shape, int):
shape = (shape, )
shape = tuple(shape)
check_argument(
all(isinstance(s, int) for s in shape),
"Array shape must be integer in each dimension, provided shape was {}",
shape,
)
fill = fill_for(elements=elements, unique=unique, fill=fill)
return ArrayStrategy(elements, shape, dtype, fill, unique)