def from_date_range(cls,
start: DateInitializer,
stop: DateInitializer,
step: int = 1,
*,
name: tp.Hashable = None
):
'''
Get an IndexDate instance over a range of dates, where start and stop is inclusive.
'''
labels = np.arange(
to_datetime64(start, _DT64_DAY),
to_datetime64(stop, _DT64_DAY) + _TD64_DAY,
np.timedelta64(step, 'D'))
labels.flags.writeable = False
return cls(labels, name=name)
def from_year_month_range(cls,
start: YearMonthInitializer,
stop: YearMonthInitializer,
step: int = 1,
*,
name: tp.Hashable = None):
'''
Get an IndexDate instance over a range of months, where start and end are inclusive.
'''
labels = np.arange(
to_datetime64(start, _DT64_MONTH),
to_datetime64(stop, _DT64_MONTH) + _TD64_MONTH,
step=np.timedelta64(step, 'D'),
dtype=_DT64_DAY)
labels.flags.writeable = False
return cls(labels, name=name)
def _ufunc_binary_operator(self, *, operator: tp.Callable[..., tp.Any],
other: object) -> np.ndarray:
if self._recache:
self._update_array_cache()
if operator.__name__ == 'matmul' or operator.__name__ == 'rmatmul':
raise NotImplementedError('matrix multiplication not supported')
if isinstance(other, Index):
other = other.values # operate on labels to labels
elif isinstance(other, str):
# do not pass dtype, as want to coerce to this parsed type, not the type of sled
other = to_datetime64(other)
if isinstance(other, np.datetime64):
# convert labels to other's datetime64 type to enable matching on month, year, etc.
array = operator(self._labels.astype(other.dtype), other)
elif isinstance(other, datetime.timedelta):
array = operator(self._labels, to_timedelta64(other))
else:
# np.timedelta64 should work fine here
array = operator(self._labels, other)
array.flags.writeable = False
return array
def from_date_range(cls,
start: DateInitializer,
stop: DateInitializer,
step: int = 1,
*,
name: tp.Hashable = None):
'''
Get an IndexYearMonth instance over a range of dates, where start and stop are inclusive.
'''
labels = np.arange(
to_datetime64(start, _DT64_DAY),
to_datetime64(stop, _DT64_DAY).astype(_DT64_YEAR) + _TD64_YEAR,
np.timedelta64(step, 'Y'),
dtype=_DT64_YEAR)
labels.flags.writeable = False
return cls(labels, name=name)
def from_year_range(cls: tp.Type[I],
start: YearInitializer,
stop: YearInitializer,
step: int = 1,
*,
name: tp.Optional[tp.Hashable] = None) -> I:
'''
Get an IndexDate instance over a range of years, where start and end are inclusive.
'''
labels = np.arange(
to_datetime64(start, _DT64_YEAR),
to_datetime64(stop, _DT64_YEAR) + _TD64_YEAR,
step=np.timedelta64(step, 'Y'),
)
labels.flags.writeable = False
return cls(labels, name=name)
def from_year_month_range(cls: tp.Type[I],
start: YearMonthInitializer,
stop: YearMonthInitializer,
step: int = 1,
*,
name: tp.Optional[tp.Hashable] = None) -> I:
'''
Get an IndexYearMonth instance over a range of months, where start and end are inclusive.
'''
labels = np.arange(to_datetime64(start, DT64_MONTH),
to_datetime64(stop, DT64_MONTH) + TD64_MONTH,
np.timedelta64(step, 'M'),
dtype=DT64_MONTH)
labels.flags.writeable = False
return cls(labels, name=name)
def from_date_range(cls: tp.Type[I],
start: DateInitializer,
stop: DateInitializer,
step: int = 1,
*,
name: tp.Optional[tp.Hashable] = None) -> I:
'''
Get an IndexYearMonth instance over a range of dates, where start and stop is inclusive.
'''
labels = np.arange(to_datetime64(start, _DT64_DAY),
to_datetime64(stop, _DT64_DAY).astype(_DT64_MONTH) +
_TD64_MONTH,
np.timedelta64(step, 'M'),
dtype=_DT64_MONTH)
labels.flags.writeable = False
return cls(labels, name=name)
def append(self, value: tp.Hashable) -> None:
'''Specialize for fixed-typed indices: convert `value` argument; do not need to resolve_dtype with each addition; do not need to check for _loc_is_iloc.
'''
value = to_datetime64(value, self._DTYPE)
if value in self._map:
raise KeyError(f'duplicate key append attempted: {value}')
# the new value is the count
self._map[value] = self._positions_mutable_count
self._labels_mutable.append(value)
self._positions_mutable_count += 1 #pylint: disable=E0237
self._recache = True #pylint: disable=E0237
def append(self, value: tp.Hashable) -> None:
'''Specialize for fixed-typed indices: convert `value` argument; do not need to resolve_dtype with each addition; self._map is never None
'''
value = to_datetime64(value, self._DTYPE)
if self._map is not None:
try:
self._map.add(value)
except ValueError:
raise KeyError(f'duplicate key append attempted: {value}')
self._labels_mutable.append(value)
self._positions_mutable_count += 1 #pylint: disable=E0237
self._recache = True #pylint: disable=E0237
def _ufunc_binary_operator(self, *,
operator: tp.Callable[..., tp.Any],
other: object,
fill_value: object = np.nan,
) -> np.ndarray:
if self._recache:
self._update_array_cache()
if operator.__name__ == 'matmul' or operator.__name__ == 'rmatmul':
raise NotImplementedError('matrix multiplication not supported')
if isinstance(other, Index):
other = other.values # operate on labels to labels
other_is_array = True
elif isinstance(other, str):
# do not pass dtype, as want to coerce to this parsed type, not the type of sled
other = to_datetime64(other)
other_is_array = False
elif other.__class__ is np.ndarray:
other_is_array = True
else:
other_is_array = False
if isinstance(other, np.datetime64):
# convert labels to other's datetime64 type to enable matching on month, year, etc.
result = operator(self._labels.astype(other.dtype), other)
elif isinstance(other, datetime.timedelta):
result = operator(self._labels, to_timedelta64(other))
else: # np.timedelta64 should work fine here
with WarningsSilent():
result = operator(self._labels, other)
# NOTE: similar branching as in container_util.apply_binary_operator
# NOTE: all string will have been converted to dt64, or raise ValueError; comparison to same sized iterables (list, tuple) will result in an array when they are the same size
if result is False: # will never be True
if not other_is_array and hasattr(other, '__len__') and len(other) == len(self):
# NOTE: equality comparisons of an array to same sized iterable normally return an array, but with dt64 types they just return False
result = np.full(self.shape, result, dtype=DTYPE_BOOL)
elif other_is_array and other.size == 1:
# elements in arrays of 0 or more dimensions are acceptable; this is what NP does for arithmetic operators when the types are compatible
result = np.full(self.shape, result, dtype=DTYPE_BOOL)
else:
raise ValueError('operands could not be broadcast together')
# raise on unaligned shapes as is done for arithmetic operators
result.flags.writeable = False
return result
def _ufunc_binary_operator(self, *, operator: tp.Callable, other) -> np.ndarray:
if self._recache:
self._update_array_cache()
if issubclass(other.__class__, Index):
other = other.values # operate on labels to labels
elif isinstance(other, str):
# do not pass dtype, as want to coerce to this parsed type, not the type of sled
other = to_datetime64(other)
if isinstance(other, np.datetime64):
# convert labels to other's datetime64 type to enable matching on month, year, etc.
array = operator(self._labels.astype(other.dtype), other)
elif isinstance(other, datetime.timedelta):
array = operator(self._labels, to_timedelta64(other))
else:
# np.timedelta64 should work fine here
array = operator(self._labels, other)
array.flags.writeable = False
return array
def __contains__(self, value) -> bool:
'''Return True if value in the labels. Will only return True for an exact match to the type of dates stored within.
'''
return self._map.__contains__(to_datetime64(value))
def __init__(self,
labels: IndexInitializer,
*,
loc_is_iloc: bool = False,
name: tp.Hashable = None,
dtype: DtypeSpecifier = None
) -> None:
self._recache = False
self._map = None
positions = None
# resolve the targetted labels dtype, by lookin at the class attr _DTYPE and/or the passed dtype argument
if dtype is None:
dtype_extract = self._DTYPE # set in some specialized Index classes
else: # passed dtype is not None
if self._DTYPE is not None and dtype != self._DTYPE:
raise RuntimeError('invalid dtype argument for this Index',
dtype, self._DTYPE)
# self._DTYPE is None, passed dtype is not None, use dtype
dtype_extract = dtype
# handle all Index subclasses
# check isinstance(labels, IndexBase)
if issubclass(labels.__class__, IndexBase):
if labels._recache:
labels._update_array_cache()
if name is None and labels.name is not None:
name = labels.name # immutable, so no copy necessary
if labels.depth == 1: # not an IndexHierarchy
if labels.STATIC: # can take the map
self._map = labels._map
# get a reference to the immutable arrays, even if this is an IndexGO index, we can take the cached arrays, assuming they are up to date
positions = labels._positions
loc_is_iloc = labels._loc_is_iloc
labels = labels._labels
else: # IndexHierarchy
# will be a generator of tuples; already updated caches
labels = array2d_to_tuples(labels._labels)
elif hasattr(labels, 'values'):
# it is a Series or similar
array = labels.values
if array.ndim == 1:
labels = array
else:
labels = array2d_to_tuples(array)
if self._DTYPE is not None:
# do not need to check arrays, as will and checked to match dtype_extract in _extract_labels
if not isinstance(labels, np.ndarray):
# for now, assume that if _DTYPE is defined, we have a date
labels = (to_datetime64(v, dtype_extract) for v in labels)
else: # coerce to target type
labels = labels.astype(dtype_extract)
self._name = name if name is None else name_filter(name)
if self._map is None:
self._map = self._get_map(labels, positions)
# this might be NP array, or a list, depending on if static or grow only; if an array, dtype will be compared with passed dtype_extract
self._labels = self._extract_labels(self._map, labels, dtype_extract)
self._positions = self._extract_positions(self._map, positions)
if self._DTYPE and self._labels.dtype != self._DTYPE:
raise RuntimeError('invalid label dtype for this Index',
self._labels.dtype, self._DTYPE)
if len(self._map) != len(self._labels):
raise KeyError(f'labels ({len(self._labels)}) have non-unique values ({len(self._map)})')
# NOTE: automatic discovery is possible, but not yet implemented
self._loc_is_iloc = loc_is_iloc
def __init__(self,
labels: IndexInitializer,
*,
loc_is_iloc: bool = False,
name: NameType = NAME_DEFAULT,
dtype: DtypeSpecifier = None) -> None:
self._recache: bool = False
self._map: tp.Optional[FrozenAutoMap] = None
positions = None
is_typed = self._DTYPE is not None
# resolve the targetted labels dtype, by lookin at the class attr _DTYPE and/or the passed dtype argument
if dtype is None:
dtype_extract = self._DTYPE # set in some specialized Index classes
else: # passed dtype is not None
if is_typed and dtype != self._DTYPE:
# NOTE: should never get to this branch, as derived Index classes that set _DTYPE remove dtype from __init__
raise ErrorInitIndex('invalid dtype argument for this Index',
dtype, self._DTYPE) #pragma: no cover
# self._DTYPE is None, passed dtype is not None, use dtype
dtype_extract = dtype
#-----------------------------------------------------------------------
# handle all Index subclasses
if isinstance(labels, IndexBase):
if labels._recache:
labels._update_array_cache()
if name is NAME_DEFAULT:
name = labels.name # immutable, so no copy necessary
if isinstance(labels, Index): # not an IndexHierarchy
if (labels.STATIC and self.STATIC and dtype is None):
if not is_typed or (is_typed
and self._DTYPE == labels.dtype):
# can take the map if static and if types in the dict are the same as those in the labels (or to become the labels after conversion)
self._map = labels._map
# get a reference to the immutable arrays, even if this is an IndexGO index, we can take the cached arrays, assuming they are up to date; for datetime64 indices, we might need to translate to a different type
positions = labels._positions
loc_is_iloc = labels._map is None
labels = labels._labels
else: # IndexHierarchy
# will be a generator of tuples; already updated caches
labels = array2d_to_tuples(labels.__iter__())
elif isinstance(labels, ContainerOperand):
# it is a Series or similar
array = labels.values
if array.ndim == 1:
labels = array
else:
labels = array2d_to_tuples(array)
# else: assume an iterable suitable for labels usage
#-----------------------------------------------------------------------
if is_typed:
# do not need to check arrays, as will and checked to match dtype_extract in _extract_labels
if not isinstance(labels, np.ndarray):
# for now, assume that if _DTYPE is defined, we have a date
labels = (to_datetime64(v, dtype_extract) for v in labels)
# coerce to target type
elif labels.dtype != dtype_extract:
labels = labels.astype(dtype_extract)
labels.flags.writeable = False #type: ignore
self._name = None if name is NAME_DEFAULT else name_filter(name)
if self._map is None: # if _map not shared from another Index
if not loc_is_iloc:
try:
self._map = FrozenAutoMap(
labels) if self.STATIC else AutoMap(labels)
except ValueError: # Automap will raise ValueError of non-unique values are encountered
pass
if self._map is None:
raise ErrorInitIndex(
f'labels ({len(tuple(labels))}) have non-unique values ({len(set(labels))})'
)
size = len(self._map)
else: # must assume labels are unique
# labels must not be a generator, but we assume that internal clients that provided loc_is_iloc will not give a generator
size = len(labels) #type: ignore
if positions is None:
positions = PositionsAllocator.get(size)
else: # map shared from another Index
size = len(self._map)
# this might be NP array, or a list, depending on if static or grow only; if an array, dtype will be compared with passed dtype_extract
self._labels = self._extract_labels(self._map, labels, dtype_extract)
self._positions = self._extract_positions(size, positions)
if self._DTYPE and self._labels.dtype != self._DTYPE:
raise ErrorInitIndex(
'invalid label dtype for this Index', #pragma: no cover
self._labels.dtype,
self._DTYPE)
