Test your AI Agents with the all-new Agent to Agent Testing Platform.Learn More
How to use dtype method in pandera
Best Python code snippet using pandera_python
common.py
Source:common.py 
...38 is_scalar,39 is_sequence,40)41_POSSIBLY_CAST_DTYPES = {42 np.dtype(t).name43 for t in [44 "O",45 "int8",46 "uint8",47 "int16",48 "uint16",49 "int32",50 "uint32",51 "int64",52 "uint64",53 ]54}55DT64NS_DTYPE = conversion.DT64NS_DTYPE56TD64NS_DTYPE = conversion.TD64NS_DTYPE57INT64_DTYPE = np.dtype(np.int64)58# oh the troubles to reduce import time59_is_scipy_sparse = None60ensure_float64 = algos.ensure_float6461ensure_float32 = algos.ensure_float3262def ensure_float(arr):63 """64 Ensure that an array object has a float dtype if possible.65 Parameters66 ----------67 arr : array-like68 The array whose data type we want to enforce as float.69 Returns70 -------71 float_arr : The original array cast to the float dtype if72 possible. Otherwise, the original array is returned.73 """74 if issubclass(arr.dtype.type, (np.integer, np.bool_)):75 arr = arr.astype(float)76 return arr77ensure_uint64 = algos.ensure_uint6478ensure_int64 = algos.ensure_int6479ensure_int32 = algos.ensure_int3280ensure_int16 = algos.ensure_int1681ensure_int8 = algos.ensure_int882ensure_platform_int = algos.ensure_platform_int83ensure_object = algos.ensure_object84def ensure_str(value: Union[bytes, Any]) -> str:85 """86 Ensure that bytes and non-strings get converted into ``str`` objects.87 """88 if isinstance(value, bytes):89 value = value.decode("utf-8")90 elif not isinstance(value, str):91 value = str(value)92 return value93def ensure_int_or_float(arr: ArrayLike, copy: bool = False) -> np.array:94 """95 Ensure that an dtype array of some integer dtype96 has an int64 dtype if possible.97 If it's not possible, potentially because of overflow,98 convert the array to float64 instead.99 Parameters100 ----------101 arr : array-like102 The array whose data type we want to enforce.103 copy: bool104 Whether to copy the original array or reuse105 it in place, if possible.106 Returns107 -------108 out_arr : The input array cast as int64 if109 possible without overflow.110 Otherwise the input array cast to float64.111 Notes112 -----113 If the array is explicitly of type uint64 the type114 will remain unchanged.115 """116 # TODO: GH27506 potential bug with ExtensionArrays117 try:118 return arr.astype("int64", copy=copy, casting="safe") # type: ignore119 except TypeError:120 pass121 try:122 return arr.astype("uint64", copy=copy, casting="safe") # type: ignore123 except TypeError:124 if is_extension_array_dtype(arr.dtype):125 return arr.to_numpy(dtype="float64", na_value=np.nan)126 return arr.astype("float64", copy=copy)127def ensure_python_int(value: Union[int, np.integer]) -> int:128 """129 Ensure that a value is a python int.130 Parameters131 ----------132 value: int or numpy.integer133 Returns134 -------135 int136 Raises137 ------138 TypeError: if the value isn't an int or can't be converted to one.139 """140 if not is_scalar(value):141 raise TypeError(142 f"Value needs to be a scalar value, was type {type(value).__name__}"143 )144 try:145 new_value = int(value)146 assert new_value == value147 except (TypeError, ValueError, AssertionError) as err:148 raise TypeError(f"Wrong type {type(value)} for value {value}") from err149 return new_value150def classes(*klasses) -> Callable:151 """ evaluate if the tipo is a subclass of the klasses """152 return lambda tipo: issubclass(tipo, klasses)153def classes_and_not_datetimelike(*klasses) -> Callable:154 """155 evaluate if the tipo is a subclass of the klasses156 and not a datetimelike157 """158 return lambda tipo: (159 issubclass(tipo, klasses)160 and not issubclass(tipo, (np.datetime64, np.timedelta64))161 )162def is_object_dtype(arr_or_dtype) -> bool:163 """164 Check whether an array-like or dtype is of the object dtype.165 Parameters166 ----------167 arr_or_dtype : array-like168 The array-like or dtype to check.169 Returns170 -------171 boolean172 Whether or not the array-like or dtype is of the object dtype.173 Examples174 --------175 >>> is_object_dtype(object)176 True177 >>> is_object_dtype(int)178 False179 >>> is_object_dtype(np.array([], dtype=object))180 True181 >>> is_object_dtype(np.array([], dtype=int))182 False183 >>> is_object_dtype([1, 2, 3])184 False185 """186 return _is_dtype_type(arr_or_dtype, classes(np.object_))187def is_sparse(arr) -> bool:188 """189 Check whether an array-like is a 1-D pandas sparse array.190 Check that the one-dimensional array-like is a pandas sparse array.191 Returns True if it is a pandas sparse array, not another type of192 sparse array.193 Parameters194 ----------195 arr : array-like196 Array-like to check.197 Returns198 -------199 bool200 Whether or not the array-like is a pandas sparse array.201 Examples202 --------203 Returns `True` if the parameter is a 1-D pandas sparse array.204 >>> is_sparse(pd.arrays.SparseArray([0, 0, 1, 0]))205 True206 >>> is_sparse(pd.Series(pd.arrays.SparseArray([0, 0, 1, 0])))207 True208 Returns `False` if the parameter is not sparse.209 >>> is_sparse(np.array([0, 0, 1, 0]))210 False211 >>> is_sparse(pd.Series([0, 1, 0, 0]))212 False213 Returns `False` if the parameter is not a pandas sparse array.214 >>> from scipy.sparse import bsr_matrix215 >>> is_sparse(bsr_matrix([0, 1, 0, 0]))216 False217 Returns `False` if the parameter has more than one dimension.218 """219 from pandas.core.arrays.sparse import SparseDtype220 dtype = getattr(arr, "dtype", arr)221 return isinstance(dtype, SparseDtype)222def is_scipy_sparse(arr) -> bool:223 """224 Check whether an array-like is a scipy.sparse.spmatrix instance.225 Parameters226 ----------227 arr : array-like228 The array-like to check.229 Returns230 -------231 boolean232 Whether or not the array-like is a scipy.sparse.spmatrix instance.233 Notes234 -----235 If scipy is not installed, this function will always return False.236 Examples237 --------238 >>> from scipy.sparse import bsr_matrix239 >>> is_scipy_sparse(bsr_matrix([1, 2, 3]))240 True241 >>> is_scipy_sparse(pd.arrays.SparseArray([1, 2, 3]))242 False243 """244 global _is_scipy_sparse245 if _is_scipy_sparse is None:246 try:247 from scipy.sparse import issparse as _is_scipy_sparse248 except ImportError:249 _is_scipy_sparse = lambda _: False250 assert _is_scipy_sparse is not None251 return _is_scipy_sparse(arr)252def is_categorical(arr) -> bool:253 """254 Check whether an array-like is a Categorical instance.255 Parameters256 ----------257 arr : array-like258 The array-like to check.259 Returns260 -------261 boolean262 Whether or not the array-like is of a Categorical instance.263 Examples264 --------265 >>> is_categorical([1, 2, 3])266 False267 Categoricals, Series Categoricals, and CategoricalIndex will return True.268 >>> cat = pd.Categorical([1, 2, 3])269 >>> is_categorical(cat)270 True271 >>> is_categorical(pd.Series(cat))272 True273 >>> is_categorical(pd.CategoricalIndex([1, 2, 3]))274 True275 """276 warnings.warn(277 "is_categorical is deprecated and will be removed in a future version. "278 "Use is_categorical_dtype instead",279 FutureWarning,280 stacklevel=2,281 )282 return isinstance(arr, ABCCategorical) or is_categorical_dtype(arr)283def is_datetime64_dtype(arr_or_dtype) -> bool:284 """285 Check whether an array-like or dtype is of the datetime64 dtype.286 Parameters287 ----------288 arr_or_dtype : array-like289 The array-like or dtype to check.290 Returns291 -------292 boolean293 Whether or not the array-like or dtype is of the datetime64 dtype.294 Examples295 --------296 >>> is_datetime64_dtype(object)297 False298 >>> is_datetime64_dtype(np.datetime64)299 True300 >>> is_datetime64_dtype(np.array([], dtype=int))301 False302 >>> is_datetime64_dtype(np.array([], dtype=np.datetime64))303 True304 >>> is_datetime64_dtype([1, 2, 3])305 False306 """307 if isinstance(arr_or_dtype, np.dtype):308 # GH#33400 fastpath for dtype object309 return arr_or_dtype.kind == "M"310 return _is_dtype_type(arr_or_dtype, classes(np.datetime64))311def is_datetime64tz_dtype(arr_or_dtype) -> bool:312 """313 Check whether an array-like or dtype is of a DatetimeTZDtype dtype.314 Parameters315 ----------316 arr_or_dtype : array-like317 The array-like or dtype to check.318 Returns319 -------320 boolean321 Whether or not the array-like or dtype is of a DatetimeTZDtype dtype.322 Examples323 --------324 >>> is_datetime64tz_dtype(object)325 False326 >>> is_datetime64tz_dtype([1, 2, 3])327 False328 >>> is_datetime64tz_dtype(pd.DatetimeIndex([1, 2, 3])) # tz-naive329 False330 >>> is_datetime64tz_dtype(pd.DatetimeIndex([1, 2, 3], tz="US/Eastern"))331 True332 >>> dtype = DatetimeTZDtype("ns", tz="US/Eastern")333 >>> s = pd.Series([], dtype=dtype)334 >>> is_datetime64tz_dtype(dtype)335 True336 >>> is_datetime64tz_dtype(s)337 True338 """339 if isinstance(arr_or_dtype, ExtensionDtype):340 # GH#33400 fastpath for dtype object341 return arr_or_dtype.kind == "M"342 if arr_or_dtype is None:343 return False344 return DatetimeTZDtype.is_dtype(arr_or_dtype)345def is_timedelta64_dtype(arr_or_dtype) -> bool:346 """347 Check whether an array-like or dtype is of the timedelta64 dtype.348 Parameters349 ----------350 arr_or_dtype : array-like351 The array-like or dtype to check.352 Returns353 -------354 boolean355 Whether or not the array-like or dtype is of the timedelta64 dtype.356 Examples357 --------358 >>> is_timedelta64_dtype(object)359 False360 >>> is_timedelta64_dtype(np.timedelta64)361 True362 >>> is_timedelta64_dtype([1, 2, 3])363 False364 >>> is_timedelta64_dtype(pd.Series([], dtype="timedelta64[ns]"))365 True366 >>> is_timedelta64_dtype('0 days')367 False368 """369 if isinstance(arr_or_dtype, np.dtype):370 # GH#33400 fastpath for dtype object371 return arr_or_dtype.kind == "m"372 return _is_dtype_type(arr_or_dtype, classes(np.timedelta64))373def is_period_dtype(arr_or_dtype) -> bool:374 """375 Check whether an array-like or dtype is of the Period dtype.376 Parameters377 ----------378 arr_or_dtype : array-like379 The array-like or dtype to check.380 Returns381 -------382 boolean383 Whether or not the array-like or dtype is of the Period dtype.384 Examples385 --------386 >>> is_period_dtype(object)387 False388 >>> is_period_dtype(PeriodDtype(freq="D"))389 True390 >>> is_period_dtype([1, 2, 3])391 False392 >>> is_period_dtype(pd.Period("2017-01-01"))393 False394 >>> is_period_dtype(pd.PeriodIndex([], freq="A"))395 True396 """397 if isinstance(arr_or_dtype, ExtensionDtype):398 # GH#33400 fastpath for dtype object399 return arr_or_dtype.type is Period400 if arr_or_dtype is None:401 return False402 return PeriodDtype.is_dtype(arr_or_dtype)403def is_interval_dtype(arr_or_dtype) -> bool:404 """405 Check whether an array-like or dtype is of the Interval dtype.406 Parameters407 ----------408 arr_or_dtype : array-like409 The array-like or dtype to check.410 Returns411 -------412 boolean413 Whether or not the array-like or dtype is of the Interval dtype.414 Examples415 --------416 >>> is_interval_dtype(object)417 False418 >>> is_interval_dtype(IntervalDtype())419 True420 >>> is_interval_dtype([1, 2, 3])421 False422 >>>423 >>> interval = pd.Interval(1, 2, closed="right")424 >>> is_interval_dtype(interval)425 False426 >>> is_interval_dtype(pd.IntervalIndex([interval]))427 True428 """429 if isinstance(arr_or_dtype, ExtensionDtype):430 # GH#33400 fastpath for dtype object431 return arr_or_dtype.type is Interval432 if arr_or_dtype is None:433 return False434 return IntervalDtype.is_dtype(arr_or_dtype)435def is_categorical_dtype(arr_or_dtype) -> bool:436 """437 Check whether an array-like or dtype is of the Categorical dtype.438 Parameters439 ----------440 arr_or_dtype : array-like441 The array-like or dtype to check.442 Returns443 -------444 boolean445 Whether or not the array-like or dtype is of the Categorical dtype.446 Examples447 --------448 >>> is_categorical_dtype(object)449 False450 >>> is_categorical_dtype(CategoricalDtype())451 True452 >>> is_categorical_dtype([1, 2, 3])453 False454 >>> is_categorical_dtype(pd.Categorical([1, 2, 3]))455 True456 >>> is_categorical_dtype(pd.CategoricalIndex([1, 2, 3]))457 True458 """459 if isinstance(arr_or_dtype, ExtensionDtype):460 # GH#33400 fastpath for dtype object461 return arr_or_dtype.name == "category"462 if arr_or_dtype is None:463 return False464 return CategoricalDtype.is_dtype(arr_or_dtype)465def is_string_dtype(arr_or_dtype) -> bool:466 """467 Check whether the provided array or dtype is of the string dtype.468 Parameters469 ----------470 arr_or_dtype : array-like471 The array or dtype to check.472 Returns473 -------474 boolean475 Whether or not the array or dtype is of the string dtype.476 Examples477 --------478 >>> is_string_dtype(str)479 True480 >>> is_string_dtype(object)481 True482 >>> is_string_dtype(int)483 False484 >>>485 >>> is_string_dtype(np.array(['a', 'b']))486 True487 >>> is_string_dtype(pd.Series([1, 2]))488 False489 """490 # TODO: gh-15585: consider making the checks stricter.491 def condition(dtype) -> bool:492 return dtype.kind in ("O", "S", "U") and not is_excluded_dtype(dtype)493 def is_excluded_dtype(dtype) -> bool:494 """495 These have kind = "O" but aren't string dtypes so need to be explicitly excluded496 """497 is_excluded_checks = (is_period_dtype, is_interval_dtype, is_categorical_dtype)498 return any(is_excluded(dtype) for is_excluded in is_excluded_checks)499 return _is_dtype(arr_or_dtype, condition)500def is_dtype_equal(source, target) -> bool:501 """502 Check if two dtypes are equal.503 Parameters504 ----------505 source : The first dtype to compare506 target : The second dtype to compare507 Returns508 -------509 boolean510 Whether or not the two dtypes are equal.511 Examples512 --------513 >>> is_dtype_equal(int, float)514 False515 >>> is_dtype_equal("int", int)516 True517 >>> is_dtype_equal(object, "category")518 False519 >>> is_dtype_equal(CategoricalDtype(), "category")520 True521 >>> is_dtype_equal(DatetimeTZDtype(tz="UTC"), "datetime64")522 False523 """524 try:525 source = _get_dtype(source)526 target = _get_dtype(target)527 return source == target528 except (TypeError, AttributeError):529 # invalid comparison530 # object == category will hit this531 return False532def is_any_int_dtype(arr_or_dtype) -> bool:533 """534 Check whether the provided array or dtype is of an integer dtype.535 In this function, timedelta64 instances are also considered "any-integer"536 type objects and will return True.537 This function is internal and should not be exposed in the public API.538 .. versionchanged:: 0.24.0539 The nullable Integer dtypes (e.g. pandas.Int64Dtype) are also considered540 as integer by this function.541 Parameters542 ----------543 arr_or_dtype : array-like544 The array or dtype to check.545 Returns546 -------547 boolean548 Whether or not the array or dtype is of an integer dtype.549 Examples550 --------551 >>> is_any_int_dtype(str)552 False553 >>> is_any_int_dtype(int)554 True555 >>> is_any_int_dtype(float)556 False557 >>> is_any_int_dtype(np.uint64)558 True559 >>> is_any_int_dtype(np.datetime64)560 False561 >>> is_any_int_dtype(np.timedelta64)562 True563 >>> is_any_int_dtype(np.array(['a', 'b']))564 False565 >>> is_any_int_dtype(pd.Series([1, 2]))566 True567 >>> is_any_int_dtype(np.array([], dtype=np.timedelta64))568 True569 >>> is_any_int_dtype(pd.Index([1, 2.])) # float570 False571 """572 return _is_dtype_type(arr_or_dtype, classes(np.integer, np.timedelta64))573def is_integer_dtype(arr_or_dtype) -> bool:574 """575 Check whether the provided array or dtype is of an integer dtype.576 Unlike in `in_any_int_dtype`, timedelta64 instances will return False.577 .. versionchanged:: 0.24.0578 The nullable Integer dtypes (e.g. pandas.Int64Dtype) are also considered579 as integer by this function.580 Parameters581 ----------582 arr_or_dtype : array-like583 The array or dtype to check.584 Returns585 -------586 boolean587 Whether or not the array or dtype is of an integer dtype and588 not an instance of timedelta64.589 Examples590 --------591 >>> is_integer_dtype(str)592 False593 >>> is_integer_dtype(int)594 True595 >>> is_integer_dtype(float)596 False597 >>> is_integer_dtype(np.uint64)598 True599 >>> is_integer_dtype('int8')600 True601 >>> is_integer_dtype('Int8')602 True603 >>> is_integer_dtype(pd.Int8Dtype)604 True605 >>> is_integer_dtype(np.datetime64)606 False607 >>> is_integer_dtype(np.timedelta64)608 False609 >>> is_integer_dtype(np.array(['a', 'b']))610 False611 >>> is_integer_dtype(pd.Series([1, 2]))612 True613 >>> is_integer_dtype(np.array([], dtype=np.timedelta64))614 False615 >>> is_integer_dtype(pd.Index([1, 2.])) # float616 False617 """618 return _is_dtype_type(arr_or_dtype, classes_and_not_datetimelike(np.integer))619def is_signed_integer_dtype(arr_or_dtype) -> bool:620 """621 Check whether the provided array or dtype is of a signed integer dtype.622 Unlike in `in_any_int_dtype`, timedelta64 instances will return False.623 .. versionchanged:: 0.24.0624 The nullable Integer dtypes (e.g. pandas.Int64Dtype) are also considered625 as integer by this function.626 Parameters627 ----------628 arr_or_dtype : array-like629 The array or dtype to check.630 Returns631 -------632 boolean633 Whether or not the array or dtype is of a signed integer dtype634 and not an instance of timedelta64.635 Examples636 --------637 >>> is_signed_integer_dtype(str)638 False639 >>> is_signed_integer_dtype(int)640 True641 >>> is_signed_integer_dtype(float)642 False643 >>> is_signed_integer_dtype(np.uint64) # unsigned644 False645 >>> is_signed_integer_dtype('int8')646 True647 >>> is_signed_integer_dtype('Int8')648 True649 >>> is_signed_integer_dtype(pd.Int8Dtype)650 True651 >>> is_signed_integer_dtype(np.datetime64)652 False653 >>> is_signed_integer_dtype(np.timedelta64)654 False655 >>> is_signed_integer_dtype(np.array(['a', 'b']))656 False657 >>> is_signed_integer_dtype(pd.Series([1, 2]))658 True659 >>> is_signed_integer_dtype(np.array([], dtype=np.timedelta64))660 False661 >>> is_signed_integer_dtype(pd.Index([1, 2.])) # float662 False663 >>> is_signed_integer_dtype(np.array([1, 2], dtype=np.uint32)) # unsigned664 False665 """666 return _is_dtype_type(arr_or_dtype, classes_and_not_datetimelike(np.signedinteger))667def is_unsigned_integer_dtype(arr_or_dtype) -> bool:668 """669 Check whether the provided array or dtype is of an unsigned integer dtype.670 .. versionchanged:: 0.24.0671 The nullable Integer dtypes (e.g. pandas.UInt64Dtype) are also672 considered as integer by this function.673 Parameters674 ----------675 arr_or_dtype : array-like676 The array or dtype to check.677 Returns678 -------679 boolean680 Whether or not the array or dtype is of an unsigned integer dtype.681 Examples682 --------683 >>> is_unsigned_integer_dtype(str)684 False685 >>> is_unsigned_integer_dtype(int) # signed686 False687 >>> is_unsigned_integer_dtype(float)688 False689 >>> is_unsigned_integer_dtype(np.uint64)690 True691 >>> is_unsigned_integer_dtype('uint8')692 True693 >>> is_unsigned_integer_dtype('UInt8')694 True695 >>> is_unsigned_integer_dtype(pd.UInt8Dtype)696 True697 >>> is_unsigned_integer_dtype(np.array(['a', 'b']))698 False699 >>> is_unsigned_integer_dtype(pd.Series([1, 2])) # signed700 False701 >>> is_unsigned_integer_dtype(pd.Index([1, 2.])) # float702 False703 >>> is_unsigned_integer_dtype(np.array([1, 2], dtype=np.uint32))704 True705 """706 return _is_dtype_type(707 arr_or_dtype, classes_and_not_datetimelike(np.unsignedinteger)708 )709def is_int64_dtype(arr_or_dtype) -> bool:710 """711 Check whether the provided array or dtype is of the int64 dtype.712 Parameters713 ----------714 arr_or_dtype : array-like715 The array or dtype to check.716 Returns717 -------718 boolean719 Whether or not the array or dtype is of the int64 dtype.720 Notes721 -----722 Depending on system architecture, the return value of `is_int64_dtype(723 int)` will be True if the OS uses 64-bit integers and False if the OS724 uses 32-bit integers.725 Examples726 --------727 >>> is_int64_dtype(str)728 False729 >>> is_int64_dtype(np.int32)730 False731 >>> is_int64_dtype(np.int64)732 True733 >>> is_int64_dtype('int8')734 False735 >>> is_int64_dtype('Int8')736 False737 >>> is_int64_dtype(pd.Int64Dtype)738 True739 >>> is_int64_dtype(float)740 False741 >>> is_int64_dtype(np.uint64) # unsigned742 False743 >>> is_int64_dtype(np.array(['a', 'b']))744 False745 >>> is_int64_dtype(np.array([1, 2], dtype=np.int64))746 True747 >>> is_int64_dtype(pd.Index([1, 2.])) # float748 False749 >>> is_int64_dtype(np.array([1, 2], dtype=np.uint32)) # unsigned750 False751 """752 return _is_dtype_type(arr_or_dtype, classes(np.int64))753def is_datetime64_any_dtype(arr_or_dtype) -> bool:754 """755 Check whether the provided array or dtype is of the datetime64 dtype.756 Parameters757 ----------758 arr_or_dtype : array-like759 The array or dtype to check.760 Returns761 -------762 bool763 Whether or not the array or dtype is of the datetime64 dtype.764 Examples765 --------766 >>> is_datetime64_any_dtype(str)767 False768 >>> is_datetime64_any_dtype(int)769 False770 >>> is_datetime64_any_dtype(np.datetime64) # can be tz-naive771 True772 >>> is_datetime64_any_dtype(DatetimeTZDtype("ns", "US/Eastern"))773 True774 >>> is_datetime64_any_dtype(np.array(['a', 'b']))775 False776 >>> is_datetime64_any_dtype(np.array([1, 2]))777 False778 >>> is_datetime64_any_dtype(np.array([], dtype="datetime64[ns]"))779 True780 >>> is_datetime64_any_dtype(pd.DatetimeIndex([1, 2, 3], dtype="datetime64[ns]"))781 True782 """783 if isinstance(arr_or_dtype, (np.dtype, ExtensionDtype)):784 # GH#33400 fastpath for dtype object785 return arr_or_dtype.kind == "M"786 if arr_or_dtype is None:787 return False788 return is_datetime64_dtype(arr_or_dtype) or is_datetime64tz_dtype(arr_or_dtype)789def is_datetime64_ns_dtype(arr_or_dtype) -> bool:790 """791 Check whether the provided array or dtype is of the datetime64[ns] dtype.792 Parameters793 ----------794 arr_or_dtype : array-like795 The array or dtype to check.796 Returns797 -------798 bool799 Whether or not the array or dtype is of the datetime64[ns] dtype.800 Examples801 --------802 >>> is_datetime64_ns_dtype(str)803 False804 >>> is_datetime64_ns_dtype(int)805 False806 >>> is_datetime64_ns_dtype(np.datetime64) # no unit807 False808 >>> is_datetime64_ns_dtype(DatetimeTZDtype("ns", "US/Eastern"))809 True810 >>> is_datetime64_ns_dtype(np.array(['a', 'b']))811 False812 >>> is_datetime64_ns_dtype(np.array([1, 2]))813 False814 >>> is_datetime64_ns_dtype(np.array([], dtype="datetime64")) # no unit815 False816 >>> is_datetime64_ns_dtype(np.array([], dtype="datetime64[ps]")) # wrong unit817 False818 >>> is_datetime64_ns_dtype(pd.DatetimeIndex([1, 2, 3], dtype="datetime64[ns]"))819 True820 """821 if arr_or_dtype is None:822 return False823 try:824 tipo = _get_dtype(arr_or_dtype)825 except TypeError:826 if is_datetime64tz_dtype(arr_or_dtype):827 tipo = _get_dtype(arr_or_dtype.dtype)828 else:829 return False830 return tipo == DT64NS_DTYPE or getattr(tipo, "base", None) == DT64NS_DTYPE831def is_timedelta64_ns_dtype(arr_or_dtype) -> bool:832 """833 Check whether the provided array or dtype is of the timedelta64[ns] dtype.834 This is a very specific dtype, so generic ones like `np.timedelta64`835 will return False if passed into this function.836 Parameters837 ----------838 arr_or_dtype : array-like839 The array or dtype to check.840 Returns841 -------842 boolean843 Whether or not the array or dtype is of the timedelta64[ns] dtype.844 Examples845 --------846 >>> is_timedelta64_ns_dtype(np.dtype('m8[ns]'))847 True848 >>> is_timedelta64_ns_dtype(np.dtype('m8[ps]')) # Wrong frequency849 False850 >>> is_timedelta64_ns_dtype(np.array([1, 2], dtype='m8[ns]'))851 True852 >>> is_timedelta64_ns_dtype(np.array([1, 2], dtype=np.timedelta64))853 False854 """855 return _is_dtype(arr_or_dtype, lambda dtype: dtype == TD64NS_DTYPE)856def is_datetime_or_timedelta_dtype(arr_or_dtype) -> bool:857 """858 Check whether the provided array or dtype is of859 a timedelta64 or datetime64 dtype.860 Parameters861 ----------862 arr_or_dtype : array-like863 The array or dtype to check.864 Returns865 -------866 boolean867 Whether or not the array or dtype is of a timedelta64,868 or datetime64 dtype.869 Examples870 --------871 >>> is_datetime_or_timedelta_dtype(str)872 False873 >>> is_datetime_or_timedelta_dtype(int)874 False875 >>> is_datetime_or_timedelta_dtype(np.datetime64)876 True877 >>> is_datetime_or_timedelta_dtype(np.timedelta64)878 True879 >>> is_datetime_or_timedelta_dtype(np.array(['a', 'b']))880 False881 >>> is_datetime_or_timedelta_dtype(pd.Series([1, 2]))882 False883 >>> is_datetime_or_timedelta_dtype(np.array([], dtype=np.timedelta64))884 True885 >>> is_datetime_or_timedelta_dtype(np.array([], dtype=np.datetime64))886 True887 """888 return _is_dtype_type(arr_or_dtype, classes(np.datetime64, np.timedelta64))889# This exists to silence numpy deprecation warnings, see GH#29553890def is_numeric_v_string_like(a, b):891 """892 Check if we are comparing a string-like object to a numeric ndarray.893 NumPy doesn't like to compare such objects, especially numeric arrays894 and scalar string-likes.895 Parameters896 ----------897 a : array-like, scalar898 The first object to check.899 b : array-like, scalar900 The second object to check.901 Returns902 -------903 boolean904 Whether we return a comparing a string-like object to a numeric array.905 Examples906 --------907 >>> is_numeric_v_string_like(1, 1)908 False909 >>> is_numeric_v_string_like("foo", "foo")910 False911 >>> is_numeric_v_string_like(1, "foo") # non-array numeric912 False913 >>> is_numeric_v_string_like(np.array([1]), "foo")914 True915 >>> is_numeric_v_string_like("foo", np.array([1])) # symmetric check916 True917 >>> is_numeric_v_string_like(np.array([1, 2]), np.array(["foo"]))918 True919 >>> is_numeric_v_string_like(np.array(["foo"]), np.array([1, 2]))920 True921 >>> is_numeric_v_string_like(np.array([1]), np.array([2]))922 False923 >>> is_numeric_v_string_like(np.array(["foo"]), np.array(["foo"]))924 False925 """926 is_a_array = isinstance(a, np.ndarray)927 is_b_array = isinstance(b, np.ndarray)928 is_a_numeric_array = is_a_array and is_numeric_dtype(a)929 is_b_numeric_array = is_b_array and is_numeric_dtype(b)930 is_a_string_array = is_a_array and is_string_like_dtype(a)931 is_b_string_array = is_b_array and is_string_like_dtype(b)932 is_a_scalar_string_like = not is_a_array and isinstance(a, str)933 is_b_scalar_string_like = not is_b_array and isinstance(b, str)934 return (935 (is_a_numeric_array and is_b_scalar_string_like)936 or (is_b_numeric_array and is_a_scalar_string_like)937 or (is_a_numeric_array and is_b_string_array)938 or (is_b_numeric_array and is_a_string_array)939 )940# This exists to silence numpy deprecation warnings, see GH#29553941def is_datetimelike_v_numeric(a, b):942 """943 Check if we are comparing a datetime-like object to a numeric object.944 By "numeric," we mean an object that is either of an int or float dtype.945 Parameters946 ----------947 a : array-like, scalar948 The first object to check.949 b : array-like, scalar950 The second object to check.951 Returns952 -------953 boolean954 Whether we return a comparing a datetime-like to a numeric object.955 Examples956 --------957 >>> from datetime import datetime958 >>> dt = np.datetime64(datetime(2017, 1, 1))959 >>>960 >>> is_datetimelike_v_numeric(1, 1)961 False962 >>> is_datetimelike_v_numeric(dt, dt)963 False964 >>> is_datetimelike_v_numeric(1, dt)965 True966 >>> is_datetimelike_v_numeric(dt, 1) # symmetric check967 True968 >>> is_datetimelike_v_numeric(np.array([dt]), 1)969 True970 >>> is_datetimelike_v_numeric(np.array([1]), dt)971 True972 >>> is_datetimelike_v_numeric(np.array([dt]), np.array([1]))973 True974 >>> is_datetimelike_v_numeric(np.array([1]), np.array([2]))975 False976 >>> is_datetimelike_v_numeric(np.array([dt]), np.array([dt]))977 False978 """979 if not hasattr(a, "dtype"):980 a = np.asarray(a)981 if not hasattr(b, "dtype"):982 b = np.asarray(b)983 def is_numeric(x):984 """985 Check if an object has a numeric dtype (i.e. integer or float).986 """987 return is_integer_dtype(x) or is_float_dtype(x)988 return (needs_i8_conversion(a) and is_numeric(b)) or (989 needs_i8_conversion(b) and is_numeric(a)990 )991def needs_i8_conversion(arr_or_dtype) -> bool:992 """993 Check whether the array or dtype should be converted to int64.994 An array-like or dtype "needs" such a conversion if the array-like995 or dtype is of a datetime-like dtype996 Parameters997 ----------998 arr_or_dtype : array-like999 The array or dtype to check.1000 Returns1001 -------1002 boolean1003 Whether or not the array or dtype should be converted to int64.1004 Examples1005 --------1006 >>> needs_i8_conversion(str)1007 False1008 >>> needs_i8_conversion(np.int64)1009 False1010 >>> needs_i8_conversion(np.datetime64)1011 True1012 >>> needs_i8_conversion(np.array(['a', 'b']))1013 False1014 >>> needs_i8_conversion(pd.Series([1, 2]))1015 False1016 >>> needs_i8_conversion(pd.Series([], dtype="timedelta64[ns]"))1017 True1018 >>> needs_i8_conversion(pd.DatetimeIndex([1, 2, 3], tz="US/Eastern"))1019 True1020 """1021 if arr_or_dtype is None:1022 return False1023 return (1024 is_datetime_or_timedelta_dtype(arr_or_dtype)1025 or is_datetime64tz_dtype(arr_or_dtype)1026 or is_period_dtype(arr_or_dtype)1027 )1028def is_numeric_dtype(arr_or_dtype) -> bool:1029 """1030 Check whether the provided array or dtype is of a numeric dtype.1031 Parameters1032 ----------1033 arr_or_dtype : array-like1034 The array or dtype to check.1035 Returns1036 -------1037 boolean1038 Whether or not the array or dtype is of a numeric dtype.1039 Examples1040 --------1041 >>> is_numeric_dtype(str)1042 False1043 >>> is_numeric_dtype(int)1044 True1045 >>> is_numeric_dtype(float)1046 True1047 >>> is_numeric_dtype(np.uint64)1048 True1049 >>> is_numeric_dtype(np.datetime64)1050 False1051 >>> is_numeric_dtype(np.timedelta64)1052 False1053 >>> is_numeric_dtype(np.array(['a', 'b']))1054 False1055 >>> is_numeric_dtype(pd.Series([1, 2]))1056 True1057 >>> is_numeric_dtype(pd.Index([1, 2.]))1058 True1059 >>> is_numeric_dtype(np.array([], dtype=np.timedelta64))1060 False1061 """1062 return _is_dtype_type(1063 arr_or_dtype, classes_and_not_datetimelike(np.number, np.bool_)1064 )1065def is_string_like_dtype(arr_or_dtype) -> bool:1066 """1067 Check whether the provided array or dtype is of a string-like dtype.1068 Unlike `is_string_dtype`, the object dtype is excluded because it1069 is a mixed dtype.1070 Parameters1071 ----------1072 arr_or_dtype : array-like1073 The array or dtype to check.1074 Returns1075 -------1076 boolean1077 Whether or not the array or dtype is of the string dtype.1078 Examples1079 --------1080 >>> is_string_like_dtype(str)1081 True1082 >>> is_string_like_dtype(object)1083 False1084 >>> is_string_like_dtype(np.array(['a', 'b']))1085 True1086 >>> is_string_like_dtype(pd.Series([1, 2]))1087 False1088 """1089 return _is_dtype(arr_or_dtype, lambda dtype: dtype.kind in ("S", "U"))1090def is_float_dtype(arr_or_dtype) -> bool:1091 """1092 Check whether the provided array or dtype is of a float dtype.1093 This function is internal and should not be exposed in the public API.1094 Parameters1095 ----------1096 arr_or_dtype : array-like1097 The array or dtype to check.1098 Returns1099 -------1100 boolean1101 Whether or not the array or dtype is of a float dtype.1102 Examples1103 --------1104 >>> is_float_dtype(str)1105 False1106 >>> is_float_dtype(int)1107 False1108 >>> is_float_dtype(float)1109 True1110 >>> is_float_dtype(np.array(['a', 'b']))1111 False1112 >>> is_float_dtype(pd.Series([1, 2]))1113 False1114 >>> is_float_dtype(pd.Index([1, 2.]))1115 True1116 """1117 return _is_dtype_type(arr_or_dtype, classes(np.floating))1118def is_bool_dtype(arr_or_dtype) -> bool:1119 """1120 Check whether the provided array or dtype is of a boolean dtype.1121 Parameters1122 ----------1123 arr_or_dtype : array-like1124 The array or dtype to check.1125 Returns1126 -------1127 boolean1128 Whether or not the array or dtype is of a boolean dtype.1129 Notes1130 -----1131 An ExtensionArray is considered boolean when the ``_is_boolean``1132 attribute is set to True.1133 Examples1134 --------1135 >>> is_bool_dtype(str)1136 False1137 >>> is_bool_dtype(int)1138 False1139 >>> is_bool_dtype(bool)1140 True1141 >>> is_bool_dtype(np.bool_)1142 True1143 >>> is_bool_dtype(np.array(['a', 'b']))1144 False1145 >>> is_bool_dtype(pd.Series([1, 2]))1146 False1147 >>> is_bool_dtype(np.array([True, False]))1148 True1149 >>> is_bool_dtype(pd.Categorical([True, False]))1150 True1151 >>> is_bool_dtype(pd.arrays.SparseArray([True, False]))1152 True1153 """1154 if arr_or_dtype is None:1155 return False1156 try:1157 dtype = _get_dtype(arr_or_dtype)1158 except TypeError:1159 return False1160 if isinstance(arr_or_dtype, CategoricalDtype):1161 arr_or_dtype = arr_or_dtype.categories1162 # now we use the special definition for Index1163 if isinstance(arr_or_dtype, ABCIndexClass):1164 # TODO(jreback)1165 # we don't have a boolean Index class1166 # so its object, we need to infer to1167 # guess this1168 return arr_or_dtype.is_object and arr_or_dtype.inferred_type == "boolean"1169 elif is_extension_array_dtype(arr_or_dtype):1170 dtype = getattr(arr_or_dtype, "dtype", arr_or_dtype)1171 return dtype._is_boolean1172 return issubclass(dtype.type, np.bool_)1173def is_extension_type(arr) -> bool:1174 """1175 Check whether an array-like is of a pandas extension class instance.1176 .. deprecated:: 1.0.01177 Use ``is_extension_array_dtype`` instead.1178 Extension classes include categoricals, pandas sparse objects (i.e.1179 classes represented within the pandas library and not ones external1180 to it like scipy sparse matrices), and datetime-like arrays.1181 Parameters1182 ----------1183 arr : array-like1184 The array-like to check.1185 Returns1186 -------1187 boolean1188 Whether or not the array-like is of a pandas extension class instance.1189 Examples1190 --------1191 >>> is_extension_type([1, 2, 3])1192 False1193 >>> is_extension_type(np.array([1, 2, 3]))1194 False1195 >>>1196 >>> cat = pd.Categorical([1, 2, 3])1197 >>>1198 >>> is_extension_type(cat)1199 True1200 >>> is_extension_type(pd.Series(cat))1201 True1202 >>> is_extension_type(pd.arrays.SparseArray([1, 2, 3]))1203 True1204 >>> from scipy.sparse import bsr_matrix1205 >>> is_extension_type(bsr_matrix([1, 2, 3]))1206 False1207 >>> is_extension_type(pd.DatetimeIndex([1, 2, 3]))1208 False1209 >>> is_extension_type(pd.DatetimeIndex([1, 2, 3], tz="US/Eastern"))1210 True1211 >>>1212 >>> dtype = DatetimeTZDtype("ns", tz="US/Eastern")1213 >>> s = pd.Series([], dtype=dtype)1214 >>> is_extension_type(s)1215 True1216 """1217 warnings.warn(1218 "'is_extension_type' is deprecated and will be removed in a future "1219 "version. Use 'is_extension_array_dtype' instead.",1220 FutureWarning,1221 stacklevel=2,1222 )1223 if is_categorical_dtype(arr):1224 return True1225 elif is_sparse(arr):1226 return True1227 elif is_datetime64tz_dtype(arr):1228 return True1229 return False1230def is_extension_array_dtype(arr_or_dtype) -> bool:1231 """1232 Check if an object is a pandas extension array type.1233 See the :ref:`Use Guide <extending.extension-types>` for more.1234 Parameters1235 ----------1236 arr_or_dtype : object1237 For array-like input, the ``.dtype`` attribute will1238 be extracted.1239 Returns1240 -------1241 bool1242 Whether the `arr_or_dtype` is an extension array type.1243 Notes1244 -----1245 This checks whether an object implements the pandas extension1246 array interface. In pandas, this includes:1247 * Categorical1248 * Sparse1249 * Interval1250 * Period1251 * DatetimeArray1252 * TimedeltaArray1253 Third-party libraries may implement arrays or types satisfying1254 this interface as well.1255 Examples1256 --------1257 >>> from pandas.api.types import is_extension_array_dtype1258 >>> arr = pd.Categorical(['a', 'b'])1259 >>> is_extension_array_dtype(arr)1260 True1261 >>> is_extension_array_dtype(arr.dtype)1262 True1263 >>> arr = np.array(['a', 'b'])1264 >>> is_extension_array_dtype(arr.dtype)1265 False1266 """1267 dtype = getattr(arr_or_dtype, "dtype", arr_or_dtype)1268 return isinstance(dtype, ExtensionDtype) or registry.find(dtype) is not None1269def is_complex_dtype(arr_or_dtype) -> bool:1270 """1271 Check whether the provided array or dtype is of a complex dtype.1272 Parameters1273 ----------1274 arr_or_dtype : array-like1275 The array or dtype to check.1276 Returns1277 -------1278 boolean1279 Whether or not the array or dtype is of a complex dtype.1280 Examples1281 --------1282 >>> is_complex_dtype(str)1283 False1284 >>> is_complex_dtype(int)1285 False1286 >>> is_complex_dtype(np.complex_)1287 True1288 >>> is_complex_dtype(np.array(['a', 'b']))1289 False1290 >>> is_complex_dtype(pd.Series([1, 2]))1291 False1292 >>> is_complex_dtype(np.array([1 + 1j, 5]))1293 True1294 """1295 return _is_dtype_type(arr_or_dtype, classes(np.complexfloating))1296def _is_dtype(arr_or_dtype, condition) -> bool:1297 """1298 Return a boolean if the condition is satisfied for the arr_or_dtype.1299 Parameters1300 ----------1301 arr_or_dtype : array-like, str, np.dtype, or ExtensionArrayType1302 The array-like or dtype object whose dtype we want to extract.1303 condition : callable[Union[np.dtype, ExtensionDtype]]1304 Returns1305 -------1306 bool1307 """1308 if arr_or_dtype is None:1309 return False1310 try:1311 dtype = _get_dtype(arr_or_dtype)1312 except (TypeError, ValueError, UnicodeEncodeError):1313 return False1314 return condition(dtype)1315def _get_dtype(arr_or_dtype) -> DtypeObj:1316 """1317 Get the dtype instance associated with an array1318 or dtype object.1319 Parameters1320 ----------1321 arr_or_dtype : array-like1322 The array-like or dtype object whose dtype we want to extract.1323 Returns1324 -------1325 obj_dtype : The extract dtype instance from the1326 passed in array or dtype object.1327 Raises1328 ------1329 TypeError : The passed in object is None.1330 """1331 if arr_or_dtype is None:1332 raise TypeError("Cannot deduce dtype from null object")1333 # fastpath1334 elif isinstance(arr_or_dtype, np.dtype):1335 return arr_or_dtype1336 elif isinstance(arr_or_dtype, type):1337 return np.dtype(arr_or_dtype)1338 # if we have an array-like1339 elif hasattr(arr_or_dtype, "dtype"):1340 arr_or_dtype = arr_or_dtype.dtype1341 return pandas_dtype(arr_or_dtype)1342def _is_dtype_type(arr_or_dtype, condition) -> bool:1343 """1344 Return a boolean if the condition is satisfied for the arr_or_dtype.1345 Parameters1346 ----------1347 arr_or_dtype : array-like1348 The array-like or dtype object whose dtype we want to extract.1349 condition : callable[Union[np.dtype, ExtensionDtypeType]]1350 Returns1351 -------1352 bool : if the condition is satisfied for the arr_or_dtype1353 """1354 if arr_or_dtype is None:1355 return condition(type(None))1356 # fastpath1357 if isinstance(arr_or_dtype, np.dtype):1358 return condition(arr_or_dtype.type)1359 elif isinstance(arr_or_dtype, type):1360 if issubclass(arr_or_dtype, ExtensionDtype):1361 arr_or_dtype = arr_or_dtype.type1362 return condition(np.dtype(arr_or_dtype).type)1363 # if we have an array-like1364 if hasattr(arr_or_dtype, "dtype"):1365 arr_or_dtype = arr_or_dtype.dtype1366 # we are not possibly a dtype1367 elif is_list_like(arr_or_dtype):1368 return condition(type(None))1369 try:1370 tipo = pandas_dtype(arr_or_dtype).type1371 except (TypeError, ValueError, UnicodeEncodeError):1372 if is_scalar(arr_or_dtype):1373 return condition(type(None))1374 return False1375 return condition(tipo)1376def infer_dtype_from_object(dtype):1377 """1378 Get a numpy dtype.type-style object for a dtype object.1379 This methods also includes handling of the datetime64[ns] and1380 datetime64[ns, TZ] objects.1381 If no dtype can be found, we return ``object``.1382 Parameters1383 ----------1384 dtype : dtype, type1385 The dtype object whose numpy dtype.type-style1386 object we want to extract.1387 Returns1388 -------1389 dtype_object : The extracted numpy dtype.type-style object.1390 """1391 if isinstance(dtype, type) and issubclass(dtype, np.generic):1392 # Type object from a dtype1393 return dtype1394 elif isinstance(dtype, (np.dtype, ExtensionDtype)):1395 # dtype object1396 try:1397 _validate_date_like_dtype(dtype)1398 except TypeError:1399 # Should still pass if we don't have a date-like1400 pass1401 return dtype.type1402 try:1403 dtype = pandas_dtype(dtype)1404 except TypeError:1405 pass1406 if is_extension_array_dtype(dtype):1407 return dtype.type1408 elif isinstance(dtype, str):1409 # TODO(jreback)1410 # should deprecate these1411 if dtype in ["datetimetz", "datetime64tz"]:1412 return DatetimeTZDtype.type1413 elif dtype in ["period"]:1414 raise NotImplementedError1415 if dtype == "datetime" or dtype == "timedelta":1416 dtype += "64"1417 try:1418 return infer_dtype_from_object(getattr(np, dtype))1419 except (AttributeError, TypeError):1420 # Handles cases like _get_dtype(int) i.e.,1421 # Python objects that are valid dtypes1422 # (unlike user-defined types, in general)1423 #1424 # TypeError handles the float16 type code of 'e'1425 # further handle internal types1426 pass1427 return infer_dtype_from_object(np.dtype(dtype))1428def _validate_date_like_dtype(dtype) -> None:1429 """1430 Check whether the dtype is a date-like dtype. Raises an error if invalid.1431 Parameters1432 ----------1433 dtype : dtype, type1434 The dtype to check.1435 Raises1436 ------1437 TypeError : The dtype could not be casted to a date-like dtype.1438 ValueError : The dtype is an illegal date-like dtype (e.g. the1439 the frequency provided is too specific)1440 """1441 try:1442 typ = np.datetime_data(dtype)[0]1443 except ValueError as e:1444 raise TypeError(e) from e1445 if typ != "generic" and typ != "ns":1446 raise ValueError(1447 f"{repr(dtype.name)} is too specific of a frequency, "1448 f"try passing {repr(dtype.type.__name__)}"1449 )1450def pandas_dtype(dtype) -> DtypeObj:1451 """1452 Convert input into a pandas only dtype object or a numpy dtype object.1453 Parameters1454 ----------1455 dtype : object to be converted1456 Returns1457 -------1458 np.dtype or a pandas dtype1459 Raises1460 ------1461 TypeError if not a dtype1462 """1463 # short-circuit1464 if isinstance(dtype, np.ndarray):1465 return dtype.dtype1466 elif isinstance(dtype, (np.dtype, ExtensionDtype)):1467 return dtype1468 # registered extension types1469 result = registry.find(dtype)1470 if result is not None:1471 return result1472 # try a numpy dtype1473 # raise a consistent TypeError if failed1474 try:1475 npdtype = np.dtype(dtype)1476 except SyntaxError as err:1477 # np.dtype uses `eval` which can raise SyntaxError1478 raise TypeError(f"data type '{dtype}' not understood") from err1479 # Any invalid dtype (such as pd.Timestamp) should raise an error.1480 # np.dtype(invalid_type).kind = 0 for such objects. However, this will1481 # also catch some valid dtypes such as object, np.object_ and 'object'1482 # which we safeguard against by catching them earlier and returning1483 # np.dtype(valid_dtype) before this condition is evaluated.1484 if is_hashable(dtype) and dtype in [object, np.object_, "object", "O"]:1485 # check hashability to avoid errors/DeprecationWarning when we get1486 # here and `dtype` is an array1487 return npdtype1488 elif npdtype.kind == "O":1489 raise TypeError(f"dtype '{dtype}' not understood")...
test_dtypes.py
Source:test_dtypes.py
...45 msg = "|".join(46 ["data type not understood", "Cannot interpret '.*' as a data type"]47 )48 with pytest.raises(TypeError, match=msg):49 np.dtype(dtype)50 assert not dtype == np.str_51 assert not np.str_ == dtype52 def test_pickle(self, dtype):53 # make sure our cache is NOT pickled54 # clear the cache55 type(dtype).reset_cache()56 assert not len(dtype._cache)57 # force back to the cache58 result = tm.round_trip_pickle(dtype)59 if not isinstance(dtype, PeriodDtype):60 # Because PeriodDtype has a cython class as a base class,61 # it has different pickle semantics, and its cache is re-populated62 # on un-pickling.63 assert not len(dtype._cache)64 assert result == dtype65class TestCategoricalDtype(Base):66 @pytest.fixture67 def dtype(self):68 """69 Class level fixture of dtype for TestCategoricalDtype70 """71 return CategoricalDtype()72 def test_hash_vs_equality(self, dtype):73 dtype2 = CategoricalDtype()74 assert dtype == dtype275 assert dtype2 == dtype76 assert hash(dtype) == hash(dtype2)77 def test_equality(self, dtype):78 assert is_dtype_equal(dtype, "category")79 assert is_dtype_equal(dtype, CategoricalDtype())80 assert not is_dtype_equal(dtype, "foo")81 def test_construction_from_string(self, dtype):82 result = CategoricalDtype.construct_from_string("category")83 assert is_dtype_equal(dtype, result)84 msg = "Cannot construct a 'CategoricalDtype' from 'foo'"85 with pytest.raises(TypeError, match=msg):86 CategoricalDtype.construct_from_string("foo")87 def test_constructor_invalid(self):88 msg = "Parameter 'categories' must be list-like"89 with pytest.raises(TypeError, match=msg):90 CategoricalDtype("category")91 dtype1 = CategoricalDtype(["a", "b"], ordered=True)92 dtype2 = CategoricalDtype(["x", "y"], ordered=False)93 c = Categorical([0, 1], dtype=dtype1, fastpath=True)94 @pytest.mark.parametrize(95 "values, categories, ordered, dtype, expected",96 [97 [None, None, None, None, CategoricalDtype()],98 [None, ["a", "b"], True, None, dtype1],99 [c, None, None, dtype2, dtype2],100 [c, ["x", "y"], False, None, dtype2],101 ],102 )103 def test_from_values_or_dtype(self, values, categories, ordered, dtype, expected):104 result = CategoricalDtype._from_values_or_dtype(105 values, categories, ordered, dtype106 )107 assert result == expected108 @pytest.mark.parametrize(109 "values, categories, ordered, dtype",110 [111 [None, ["a", "b"], True, dtype2],112 [None, ["a", "b"], None, dtype2],113 [None, None, True, dtype2],114 ],115 )116 def test_from_values_or_dtype_raises(self, values, categories, ordered, dtype):117 msg = "Cannot specify `categories` or `ordered` together with `dtype`."118 with pytest.raises(ValueError, match=msg):119 CategoricalDtype._from_values_or_dtype(values, categories, ordered, dtype)120 def test_from_values_or_dtype_invalid_dtype(self):121 msg = "Cannot not construct CategoricalDtype from <class 'object'>"122 with pytest.raises(ValueError, match=msg):123 CategoricalDtype._from_values_or_dtype(None, None, None, object)124 def test_is_dtype(self, dtype):125 assert CategoricalDtype.is_dtype(dtype)126 assert CategoricalDtype.is_dtype("category")127 assert CategoricalDtype.is_dtype(CategoricalDtype())128 assert not CategoricalDtype.is_dtype("foo")129 assert not CategoricalDtype.is_dtype(np.float64)130 def test_basic(self, dtype):131 assert is_categorical_dtype(dtype)132 factor = Categorical(["a", "b", "b", "a", "a", "c", "c", "c"])133 s = Series(factor, name="A")134 # dtypes135 assert is_categorical_dtype(s.dtype)136 assert is_categorical_dtype(s)137 assert not is_categorical_dtype(np.dtype("float64"))138 with tm.assert_produces_warning(FutureWarning):139 # GH#33385 deprecated140 assert is_categorical(s.dtype)141 assert is_categorical(s)142 assert not is_categorical(np.dtype("float64"))143 assert not is_categorical(1.0)144 def test_tuple_categories(self):145 categories = [(1, "a"), (2, "b"), (3, "c")]146 result = CategoricalDtype(categories)147 assert all(result.categories == categories)148 @pytest.mark.parametrize(149 "categories, expected",150 [151 ([True, False], True),152 ([True, False, None], True),153 ([True, False, "a", "b'"], False),154 ([0, 1], False),155 ],156 )157 def test_is_boolean(self, categories, expected):158 cat = Categorical(categories)159 assert cat.dtype._is_boolean is expected160 assert is_bool_dtype(cat) is expected161 assert is_bool_dtype(cat.dtype) is expected162 def test_dtype_specific_categorical_dtype(self):163 expected = "datetime64[ns]"164 result = str(Categorical(DatetimeIndex([])).categories.dtype)165 assert result == expected166 def test_not_string(self):167 # though CategoricalDtype has object kind, it cannot be string168 assert not is_string_dtype(CategoricalDtype())169class TestDatetimeTZDtype(Base):170 @pytest.fixture171 def dtype(self):172 """173 Class level fixture of dtype for TestDatetimeTZDtype174 """175 return DatetimeTZDtype("ns", "US/Eastern")176 def test_alias_to_unit_raises(self):177 # 23990178 with pytest.raises(ValueError, match="Passing a dtype alias"):179 DatetimeTZDtype("datetime64[ns, US/Central]")180 def test_alias_to_unit_bad_alias_raises(self):181 # 23990182 with pytest.raises(TypeError, match=""):183 DatetimeTZDtype("this is a bad string")184 with pytest.raises(TypeError, match=""):185 DatetimeTZDtype("datetime64[ns, US/NotATZ]")186 def test_hash_vs_equality(self, dtype):187 # make sure that we satisfy is semantics188 dtype2 = DatetimeTZDtype("ns", "US/Eastern")189 dtype3 = DatetimeTZDtype(dtype2)190 assert dtype == dtype2191 assert dtype2 == dtype192 assert dtype3 == dtype193 assert hash(dtype) == hash(dtype2)194 assert hash(dtype) == hash(dtype3)195 dtype4 = DatetimeTZDtype("ns", "US/Central")196 assert dtype2 != dtype4197 assert hash(dtype2) != hash(dtype4)198 def test_construction(self):199 msg = "DatetimeTZDtype only supports ns units"200 with pytest.raises(ValueError, match=msg):201 DatetimeTZDtype("ms", "US/Eastern")202 def test_subclass(self):203 a = DatetimeTZDtype.construct_from_string("datetime64[ns, US/Eastern]")204 b = DatetimeTZDtype.construct_from_string("datetime64[ns, CET]")205 assert issubclass(type(a), type(a))206 assert issubclass(type(a), type(b))207 def test_compat(self, dtype):208 assert is_datetime64tz_dtype(dtype)209 assert is_datetime64tz_dtype("datetime64[ns, US/Eastern]")210 assert is_datetime64_any_dtype(dtype)211 assert is_datetime64_any_dtype("datetime64[ns, US/Eastern]")212 assert is_datetime64_ns_dtype(dtype)213 assert is_datetime64_ns_dtype("datetime64[ns, US/Eastern]")214 assert not is_datetime64_dtype(dtype)215 assert not is_datetime64_dtype("datetime64[ns, US/Eastern]")216 def test_construction_from_string(self, dtype):217 result = DatetimeTZDtype.construct_from_string("datetime64[ns, US/Eastern]")218 assert is_dtype_equal(dtype, result)219 @pytest.mark.parametrize(220 "string",221 [222 "foo",223 "datetime64[ns, notatz]",224 # non-nano unit225 "datetime64[ps, UTC]",226 # dateutil str that returns None from gettz227 "datetime64[ns, dateutil/invalid]",228 ],229 )230 def test_construct_from_string_invalid_raises(self, string):231 msg = f"Cannot construct a 'DatetimeTZDtype' from '{string}'"232 with pytest.raises(TypeError, match=re.escape(msg)):233 DatetimeTZDtype.construct_from_string(string)234 def test_construct_from_string_wrong_type_raises(self):235 msg = "'construct_from_string' expects a string, got <class 'list'>"236 with pytest.raises(TypeError, match=msg):237 DatetimeTZDtype.construct_from_string(["datetime64[ns, notatz]"])238 def test_is_dtype(self, dtype):239 assert not DatetimeTZDtype.is_dtype(None)240 assert DatetimeTZDtype.is_dtype(dtype)241 assert DatetimeTZDtype.is_dtype("datetime64[ns, US/Eastern]")242 assert DatetimeTZDtype.is_dtype("M8[ns, US/Eastern]")243 assert not DatetimeTZDtype.is_dtype("foo")244 assert DatetimeTZDtype.is_dtype(DatetimeTZDtype("ns", "US/Pacific"))245 assert not DatetimeTZDtype.is_dtype(np.float64)246 def test_equality(self, dtype):247 assert is_dtype_equal(dtype, "datetime64[ns, US/Eastern]")248 assert is_dtype_equal(dtype, "M8[ns, US/Eastern]")249 assert is_dtype_equal(dtype, DatetimeTZDtype("ns", "US/Eastern"))250 assert not is_dtype_equal(dtype, "foo")251 assert not is_dtype_equal(dtype, DatetimeTZDtype("ns", "CET"))252 assert not is_dtype_equal(253 DatetimeTZDtype("ns", "US/Eastern"), DatetimeTZDtype("ns", "US/Pacific")254 )255 # numpy compat256 assert is_dtype_equal(np.dtype("M8[ns]"), "datetime64[ns]")257 assert dtype == "M8[ns, US/Eastern]"258 def test_basic(self, dtype):259 assert is_datetime64tz_dtype(dtype)260 dr = date_range("20130101", periods=3, tz="US/Eastern")261 s = Series(dr, name="A")262 # dtypes263 assert is_datetime64tz_dtype(s.dtype)264 assert is_datetime64tz_dtype(s)265 assert not is_datetime64tz_dtype(np.dtype("float64"))266 assert not is_datetime64tz_dtype(1.0)267 def test_dst(self):268 dr1 = date_range("2013-01-01", periods=3, tz="US/Eastern")269 s1 = Series(dr1, name="A")270 assert is_datetime64tz_dtype(s1)271 dr2 = date_range("2013-08-01", periods=3, tz="US/Eastern")272 s2 = Series(dr2, name="A")273 assert is_datetime64tz_dtype(s2)274 assert s1.dtype == s2.dtype275 @pytest.mark.parametrize("tz", ["UTC", "US/Eastern"])276 @pytest.mark.parametrize("constructor", ["M8", "datetime64"])277 def test_parser(self, tz, constructor):278 # pr #11245279 dtz_str = f"{constructor}[ns, {tz}]"280 result = DatetimeTZDtype.construct_from_string(dtz_str)281 expected = DatetimeTZDtype("ns", tz)282 assert result == expected283 def test_empty(self):284 with pytest.raises(TypeError, match="A 'tz' is required."):285 DatetimeTZDtype()286 def test_tz_standardize(self):287 # GH 24713288 tz = pytz.timezone("US/Eastern")289 dr = date_range("2013-01-01", periods=3, tz="US/Eastern")290 dtype = DatetimeTZDtype("ns", dr.tz)291 assert dtype.tz == tz292 dtype = DatetimeTZDtype("ns", dr[0].tz)293 assert dtype.tz == tz294class TestPeriodDtype(Base):295 @pytest.fixture296 def dtype(self):297 """298 Class level fixture of dtype for TestPeriodDtype299 """300 return PeriodDtype("D")301 def test_hash_vs_equality(self, dtype):302 # make sure that we satisfy is semantics303 dtype2 = PeriodDtype("D")304 dtype3 = PeriodDtype(dtype2)305 assert dtype == dtype2306 assert dtype2 == dtype307 assert dtype3 == dtype308 assert dtype is dtype2309 assert dtype2 is dtype310 assert dtype3 is dtype311 assert hash(dtype) == hash(dtype2)312 assert hash(dtype) == hash(dtype3)313 def test_construction(self):314 with pytest.raises(ValueError, match="Invalid frequency: xx"):315 PeriodDtype("xx")316 for s in ["period[D]", "Period[D]", "D"]:317 dt = PeriodDtype(s)318 assert dt.freq == pd.tseries.offsets.Day()319 assert is_period_dtype(dt)320 for s in ["period[3D]", "Period[3D]", "3D"]:321 dt = PeriodDtype(s)322 assert dt.freq == pd.tseries.offsets.Day(3)323 assert is_period_dtype(dt)324 for s in [325 "period[26H]",326 "Period[26H]",327 "26H",328 "period[1D2H]",329 "Period[1D2H]",330 "1D2H",331 ]:332 dt = PeriodDtype(s)333 assert dt.freq == pd.tseries.offsets.Hour(26)334 assert is_period_dtype(dt)335 def test_subclass(self):336 a = PeriodDtype("period[D]")337 b = PeriodDtype("period[3D]")338 assert issubclass(type(a), type(a))339 assert issubclass(type(a), type(b))340 def test_identity(self):341 assert PeriodDtype("period[D]") == PeriodDtype("period[D]")342 assert PeriodDtype("period[D]") is PeriodDtype("period[D]")343 assert PeriodDtype("period[3D]") == PeriodDtype("period[3D]")344 assert PeriodDtype("period[3D]") is PeriodDtype("period[3D]")345 assert PeriodDtype("period[1S1U]") == PeriodDtype("period[1000001U]")346 assert PeriodDtype("period[1S1U]") is PeriodDtype("period[1000001U]")347 def test_compat(self, dtype):348 assert not is_datetime64_ns_dtype(dtype)349 assert not is_datetime64_ns_dtype("period[D]")350 assert not is_datetime64_dtype(dtype)351 assert not is_datetime64_dtype("period[D]")352 def test_construction_from_string(self, dtype):353 result = PeriodDtype("period[D]")354 assert is_dtype_equal(dtype, result)355 result = PeriodDtype.construct_from_string("period[D]")356 assert is_dtype_equal(dtype, result)357 with pytest.raises(TypeError, match="list"):358 PeriodDtype.construct_from_string([1, 2, 3])359 @pytest.mark.parametrize(360 "string",361 [362 "foo",363 "period[foo]",364 "foo[D]",365 "datetime64[ns]",366 "datetime64[ns, US/Eastern]",367 ],368 )369 def test_construct_dtype_from_string_invalid_raises(self, string):370 msg = f"Cannot construct a 'PeriodDtype' from '{string}'"371 with pytest.raises(TypeError, match=re.escape(msg)):372 PeriodDtype.construct_from_string(string)373 def test_is_dtype(self, dtype):374 assert PeriodDtype.is_dtype(dtype)375 assert PeriodDtype.is_dtype("period[D]")376 assert PeriodDtype.is_dtype("period[3D]")377 assert PeriodDtype.is_dtype(PeriodDtype("3D"))378 assert PeriodDtype.is_dtype("period[U]")379 assert PeriodDtype.is_dtype("period[S]")380 assert PeriodDtype.is_dtype(PeriodDtype("U"))381 assert PeriodDtype.is_dtype(PeriodDtype("S"))382 assert not PeriodDtype.is_dtype("D")383 assert not PeriodDtype.is_dtype("3D")384 assert not PeriodDtype.is_dtype("U")385 assert not PeriodDtype.is_dtype("S")386 assert not PeriodDtype.is_dtype("foo")387 assert not PeriodDtype.is_dtype(np.object_)388 assert not PeriodDtype.is_dtype(np.int64)389 assert not PeriodDtype.is_dtype(np.float64)390 def test_equality(self, dtype):391 assert is_dtype_equal(dtype, "period[D]")392 assert is_dtype_equal(dtype, PeriodDtype("D"))393 assert is_dtype_equal(dtype, PeriodDtype("D"))394 assert is_dtype_equal(PeriodDtype("D"), PeriodDtype("D"))395 assert not is_dtype_equal(dtype, "D")396 assert not is_dtype_equal(PeriodDtype("D"), PeriodDtype("2D"))397 def test_basic(self, dtype):398 assert is_period_dtype(dtype)399 pidx = pd.period_range("2013-01-01 09:00", periods=5, freq="H")400 assert is_period_dtype(pidx.dtype)401 assert is_period_dtype(pidx)402 s = Series(pidx, name="A")403 assert is_period_dtype(s.dtype)404 assert is_period_dtype(s)405 assert not is_period_dtype(np.dtype("float64"))406 assert not is_period_dtype(1.0)407 def test_empty(self):408 dt = PeriodDtype()409 msg = "object has no attribute 'freqstr'"410 with pytest.raises(AttributeError, match=msg):411 str(dt)412 def test_not_string(self):413 # though PeriodDtype has object kind, it cannot be string414 assert not is_string_dtype(PeriodDtype("D"))415class TestIntervalDtype(Base):416 @pytest.fixture417 def dtype(self):418 """419 Class level fixture of dtype for TestIntervalDtype420 """421 return IntervalDtype("int64")422 def test_hash_vs_equality(self, dtype):423 # make sure that we satisfy is semantics424 dtype2 = IntervalDtype("int64")425 dtype3 = IntervalDtype(dtype2)426 assert dtype == dtype2427 assert dtype2 == dtype428 assert dtype3 == dtype429 assert dtype is dtype2430 assert dtype2 is dtype3431 assert dtype3 is dtype432 assert hash(dtype) == hash(dtype2)433 assert hash(dtype) == hash(dtype3)434 dtype1 = IntervalDtype("interval")435 dtype2 = IntervalDtype(dtype1)436 dtype3 = IntervalDtype("interval")437 assert dtype2 == dtype1438 assert dtype2 == dtype2439 assert dtype2 == dtype3440 assert dtype2 is dtype1441 assert dtype2 is dtype2442 assert dtype2 is dtype3443 assert hash(dtype2) == hash(dtype1)444 assert hash(dtype2) == hash(dtype2)445 assert hash(dtype2) == hash(dtype3)446 @pytest.mark.parametrize(447 "subtype", ["interval[int64]", "Interval[int64]", "int64", np.dtype("int64")]448 )449 def test_construction(self, subtype):450 i = IntervalDtype(subtype)451 assert i.subtype == np.dtype("int64")452 assert is_interval_dtype(i)453 @pytest.mark.parametrize("subtype", [None, "interval", "Interval"])454 def test_construction_generic(self, subtype):455 # generic456 i = IntervalDtype(subtype)457 assert i.subtype is None458 assert is_interval_dtype(i)459 @pytest.mark.parametrize(460 "subtype",461 [462 CategoricalDtype(list("abc"), False),463 CategoricalDtype(list("wxyz"), True),464 object,465 str,466 "<U10",467 "interval[category]",468 "interval[object]",469 ],470 )471 def test_construction_not_supported(self, subtype):472 # GH 19016473 msg = (474 "category, object, and string subtypes are not supported "475 "for IntervalDtype"476 )477 with pytest.raises(TypeError, match=msg):478 IntervalDtype(subtype)479 @pytest.mark.parametrize("subtype", ["xx", "IntervalA", "Interval[foo]"])480 def test_construction_errors(self, subtype):481 msg = "could not construct IntervalDtype"482 with pytest.raises(TypeError, match=msg):483 IntervalDtype(subtype)484 def test_construction_from_string(self, dtype):485 result = IntervalDtype("interval[int64]")486 assert is_dtype_equal(dtype, result)487 result = IntervalDtype.construct_from_string("interval[int64]")488 assert is_dtype_equal(dtype, result)489 @pytest.mark.parametrize("string", [0, 3.14, ("a", "b"), None])490 def test_construction_from_string_errors(self, string):491 # these are invalid entirely492 msg = f"'construct_from_string' expects a string, got {type(string)}"493 with pytest.raises(TypeError, match=re.escape(msg)):494 IntervalDtype.construct_from_string(string)495 @pytest.mark.parametrize("string", ["foo", "foo[int64]", "IntervalA"])496 def test_construction_from_string_error_subtype(self, string):497 # this is an invalid subtype498 msg = (499 "Incorrectly formatted string passed to constructor. "500 r"Valid formats include Interval or Interval\[dtype\] "501 "where dtype is numeric, datetime, or timedelta"502 )503 with pytest.raises(TypeError, match=msg):504 IntervalDtype.construct_from_string(string)505 def test_subclass(self):506 a = IntervalDtype("interval[int64]")507 b = IntervalDtype("interval[int64]")508 assert issubclass(type(a), type(a))509 assert issubclass(type(a), type(b))510 def test_is_dtype(self, dtype):511 assert IntervalDtype.is_dtype(dtype)512 assert IntervalDtype.is_dtype("interval")513 assert IntervalDtype.is_dtype(IntervalDtype("float64"))514 assert IntervalDtype.is_dtype(IntervalDtype("int64"))515 assert IntervalDtype.is_dtype(IntervalDtype(np.int64))516 assert not IntervalDtype.is_dtype("D")517 assert not IntervalDtype.is_dtype("3D")518 assert not IntervalDtype.is_dtype("U")519 assert not IntervalDtype.is_dtype("S")520 assert not IntervalDtype.is_dtype("foo")521 assert not IntervalDtype.is_dtype("IntervalA")522 assert not IntervalDtype.is_dtype(np.object_)523 assert not IntervalDtype.is_dtype(np.int64)524 assert not IntervalDtype.is_dtype(np.float64)525 def test_equality(self, dtype):526 assert is_dtype_equal(dtype, "interval[int64]")527 assert is_dtype_equal(dtype, IntervalDtype("int64"))528 assert is_dtype_equal(IntervalDtype("int64"), IntervalDtype("int64"))529 assert not is_dtype_equal(dtype, "int64")530 assert not is_dtype_equal(IntervalDtype("int64"), IntervalDtype("float64"))531 # invalid subtype comparisons do not raise when directly compared532 dtype1 = IntervalDtype("float64")533 dtype2 = IntervalDtype("datetime64[ns, US/Eastern]")534 assert dtype1 != dtype2535 assert dtype2 != dtype1536 @pytest.mark.parametrize(537 "subtype",538 [539 None,540 "interval",541 "Interval",542 "int64",543 "uint64",544 "float64",545 "complex128",546 "datetime64",547 "timedelta64",548 PeriodDtype("Q"),549 ],550 )551 def test_equality_generic(self, subtype):552 # GH 18980553 dtype = IntervalDtype(subtype)554 assert is_dtype_equal(dtype, "interval")555 assert is_dtype_equal(dtype, IntervalDtype())556 @pytest.mark.parametrize(557 "subtype",558 [559 "int64",560 "uint64",561 "float64",562 "complex128",563 "datetime64",564 "timedelta64",565 PeriodDtype("Q"),566 ],567 )568 def test_name_repr(self, subtype):569 # GH 18980570 dtype = IntervalDtype(subtype)571 expected = f"interval[{subtype}]"572 assert str(dtype) == expected573 assert dtype.name == "interval"574 @pytest.mark.parametrize("subtype", [None, "interval", "Interval"])575 def test_name_repr_generic(self, subtype):576 # GH 18980577 dtype = IntervalDtype(subtype)578 assert str(dtype) == "interval"579 assert dtype.name == "interval"580 def test_basic(self, dtype):581 assert is_interval_dtype(dtype)582 ii = IntervalIndex.from_breaks(range(3))583 assert is_interval_dtype(ii.dtype)584 assert is_interval_dtype(ii)585 s = Series(ii, name="A")586 assert is_interval_dtype(s.dtype)587 assert is_interval_dtype(s)588 def test_basic_dtype(self):589 assert is_interval_dtype("interval[int64]")590 assert is_interval_dtype(IntervalIndex.from_tuples([(0, 1)]))591 assert is_interval_dtype(IntervalIndex.from_breaks(np.arange(4)))592 assert is_interval_dtype(593 IntervalIndex.from_breaks(date_range("20130101", periods=3))594 )595 assert not is_interval_dtype("U")596 assert not is_interval_dtype("S")597 assert not is_interval_dtype("foo")598 assert not is_interval_dtype(np.object_)599 assert not is_interval_dtype(np.int64)600 assert not is_interval_dtype(np.float64)601 def test_caching(self):602 IntervalDtype.reset_cache()603 dtype = IntervalDtype("int64")604 assert len(IntervalDtype._cache) == 1605 IntervalDtype("interval")606 assert len(IntervalDtype._cache) == 2607 IntervalDtype.reset_cache()608 tm.round_trip_pickle(dtype)609 assert len(IntervalDtype._cache) == 0610 def test_not_string(self):611 # GH30568: though IntervalDtype has object kind, it cannot be string612 assert not is_string_dtype(IntervalDtype())613class TestCategoricalDtypeParametrized:614 @pytest.mark.parametrize(615 "categories",616 [617 list("abcd"),618 np.arange(1000),619 ["a", "b", 10, 2, 1.3, True],620 [True, False],621 pd.date_range("2017", periods=4),622 ],623 )624 def test_basic(self, categories, ordered):625 c1 = CategoricalDtype(categories, ordered=ordered)626 tm.assert_index_equal(c1.categories, pd.Index(categories))627 assert c1.ordered is ordered628 def test_order_matters(self):629 categories = ["a", "b"]630 c1 = CategoricalDtype(categories, ordered=True)631 c2 = CategoricalDtype(categories, ordered=False)632 c3 = CategoricalDtype(categories, ordered=None)633 assert c1 is not c2634 assert c1 is not c3635 @pytest.mark.parametrize("ordered", [False, None])636 def test_unordered_same(self, ordered):637 c1 = CategoricalDtype(["a", "b"], ordered=ordered)638 c2 = CategoricalDtype(["b", "a"], ordered=ordered)639 assert hash(c1) == hash(c2)640 def test_categories(self):641 result = CategoricalDtype(["a", "b", "c"])642 tm.assert_index_equal(result.categories, pd.Index(["a", "b", "c"]))643 assert result.ordered is False644 def test_equal_but_different(self, ordered):645 c1 = CategoricalDtype([1, 2, 3])646 c2 = CategoricalDtype([1.0, 2.0, 3.0])647 assert c1 is not c2648 assert c1 != c2649 @pytest.mark.parametrize("v1, v2", [([1, 2, 3], [1, 2, 3]), ([1, 2, 3], [3, 2, 1])])650 def test_order_hashes_different(self, v1, v2):651 c1 = CategoricalDtype(v1, ordered=False)652 c2 = CategoricalDtype(v2, ordered=True)653 c3 = CategoricalDtype(v1, ordered=None)654 assert c1 is not c2655 assert c1 is not c3656 def test_nan_invalid(self):657 msg = "Categorical categories cannot be null"658 with pytest.raises(ValueError, match=msg):659 CategoricalDtype([1, 2, np.nan])660 def test_non_unique_invalid(self):661 msg = "Categorical categories must be unique"662 with pytest.raises(ValueError, match=msg):663 CategoricalDtype([1, 2, 1])664 def test_same_categories_different_order(self):665 c1 = CategoricalDtype(["a", "b"], ordered=True)666 c2 = CategoricalDtype(["b", "a"], ordered=True)667 assert c1 is not c2668 @pytest.mark.parametrize("ordered1", [True, False, None])669 @pytest.mark.parametrize("ordered2", [True, False, None])670 def test_categorical_equality(self, ordered1, ordered2):671 # same categories, same order672 # any combination of None/False are equal673 # True/True is the only combination with True that are equal674 c1 = CategoricalDtype(list("abc"), ordered1)675 c2 = CategoricalDtype(list("abc"), ordered2)676 result = c1 == c2677 expected = bool(ordered1) is bool(ordered2)678 assert result is expected679 # same categories, different order680 # any combination of None/False are equal (order doesn't matter)681 # any combination with True are not equal (different order of cats)682 c1 = CategoricalDtype(list("abc"), ordered1)683 c2 = CategoricalDtype(list("cab"), ordered2)684 result = c1 == c2685 expected = (bool(ordered1) is False) and (bool(ordered2) is False)686 assert result is expected687 # different categories688 c2 = CategoricalDtype([1, 2, 3], ordered2)689 assert c1 != c2690 # none categories691 c1 = CategoricalDtype(list("abc"), ordered1)692 c2 = CategoricalDtype(None, ordered2)693 c3 = CategoricalDtype(None, ordered1)694 assert c1 == c2695 assert c2 == c1696 assert c2 == c3697 @pytest.mark.parametrize("categories", [list("abc"), None])698 @pytest.mark.parametrize("other", ["category", "not a category"])699 def test_categorical_equality_strings(self, categories, ordered, other):700 c1 = CategoricalDtype(categories, ordered)701 result = c1 == other702 expected = other == "category"703 assert result is expected704 def test_invalid_raises(self):705 with pytest.raises(TypeError, match="ordered"):706 CategoricalDtype(["a", "b"], ordered="foo")707 with pytest.raises(TypeError, match="'categories' must be list-like"):708 CategoricalDtype("category")709 def test_mixed(self):710 a = CategoricalDtype(["a", "b", 1, 2])711 b = CategoricalDtype(["a", "b", "1", "2"])712 assert hash(a) != hash(b)713 def test_from_categorical_dtype_identity(self):714 c1 = Categorical([1, 2], categories=[1, 2, 3], ordered=True)715 # Identity test for no changes716 c2 = CategoricalDtype._from_categorical_dtype(c1)717 assert c2 is c1718 def test_from_categorical_dtype_categories(self):719 c1 = Categorical([1, 2], categories=[1, 2, 3], ordered=True)720 # override categories721 result = CategoricalDtype._from_categorical_dtype(c1, categories=[2, 3])722 assert result == CategoricalDtype([2, 3], ordered=True)723 def test_from_categorical_dtype_ordered(self):724 c1 = Categorical([1, 2], categories=[1, 2, 3], ordered=True)725 # override ordered726 result = CategoricalDtype._from_categorical_dtype(c1, ordered=False)727 assert result == CategoricalDtype([1, 2, 3], ordered=False)728 def test_from_categorical_dtype_both(self):729 c1 = Categorical([1, 2], categories=[1, 2, 3], ordered=True)730 # override ordered731 result = CategoricalDtype._from_categorical_dtype(732 c1, categories=[1, 2], ordered=False733 )734 assert result == CategoricalDtype([1, 2], ordered=False)735 def test_str_vs_repr(self, ordered):736 c1 = CategoricalDtype(["a", "b"], ordered=ordered)737 assert str(c1) == "category"738 # Py2 will have unicode prefixes739 pat = r"CategoricalDtype\(categories=\[.*\], ordered={ordered}\)"740 assert re.match(pat.format(ordered=ordered), repr(c1))741 def test_categorical_categories(self):742 # GH17884743 c1 = CategoricalDtype(Categorical(["a", "b"]))744 tm.assert_index_equal(c1.categories, pd.Index(["a", "b"]))745 c1 = CategoricalDtype(CategoricalIndex(["a", "b"]))746 tm.assert_index_equal(c1.categories, pd.Index(["a", "b"]))747 @pytest.mark.parametrize(748 "new_categories", [list("abc"), list("cba"), list("wxyz"), None]749 )750 @pytest.mark.parametrize("new_ordered", [True, False, None])751 def test_update_dtype(self, ordered, new_categories, new_ordered):752 original_categories = list("abc")753 dtype = CategoricalDtype(original_categories, ordered)754 new_dtype = CategoricalDtype(new_categories, new_ordered)755 result = dtype.update_dtype(new_dtype)756 expected_categories = pd.Index(new_categories or original_categories)757 expected_ordered = new_ordered if new_ordered is not None else dtype.ordered758 tm.assert_index_equal(result.categories, expected_categories)759 assert result.ordered is expected_ordered760 def test_update_dtype_string(self, ordered):761 dtype = CategoricalDtype(list("abc"), ordered)762 expected_categories = dtype.categories763 expected_ordered = dtype.ordered764 result = dtype.update_dtype("category")765 tm.assert_index_equal(result.categories, expected_categories)766 assert result.ordered is expected_ordered767 @pytest.mark.parametrize("bad_dtype", ["foo", object, np.int64, PeriodDtype("Q")])768 def test_update_dtype_errors(self, bad_dtype):769 dtype = CategoricalDtype(list("abc"), False)770 msg = "a CategoricalDtype must be passed to perform an update, "771 with pytest.raises(ValueError, match=msg):772 dtype.update_dtype(bad_dtype)773@pytest.mark.parametrize(774 "dtype", [CategoricalDtype, IntervalDtype, DatetimeTZDtype, PeriodDtype]775)776def test_registry(dtype):777 assert dtype in registry.dtypes778@pytest.mark.parametrize(779 "dtype, expected",780 [781 ("int64", None),782 ("interval", IntervalDtype()),783 ("interval[int64]", IntervalDtype()),784 ("interval[datetime64[ns]]", IntervalDtype("datetime64[ns]")),785 ("period[D]", PeriodDtype("D")),786 ("category", CategoricalDtype()),787 ("datetime64[ns, US/Eastern]", DatetimeTZDtype("ns", "US/Eastern")),788 ],789)790def test_registry_find(dtype, expected):791 assert registry.find(dtype) == expected792@pytest.mark.parametrize(793 "dtype, expected",794 [795 (str, False),796 (int, False),797 (bool, True),798 (np.bool_, True),799 (np.array(["a", "b"]), False),800 (pd.Series([1, 2]), False),801 (np.array([True, False]), True),802 (pd.Series([True, False]), True),803 (SparseArray([True, False]), True),804 (SparseDtype(bool), True),805 ],806)807def test_is_bool_dtype(dtype, expected):808 result = is_bool_dtype(dtype)809 assert result is expected810def test_is_bool_dtype_sparse():811 result = is_bool_dtype(pd.Series(SparseArray([True, False])))812 assert result is True813@pytest.mark.parametrize(814 "check",815 [816 is_categorical_dtype,817 is_datetime64tz_dtype,818 is_period_dtype,819 is_datetime64_ns_dtype,820 is_datetime64_dtype,821 is_interval_dtype,822 is_datetime64_any_dtype,823 is_string_dtype,824 is_bool_dtype,825 ],...
register_functions.py
Source:register_functions.py
...3 tint32, tint64, tfloat32, tfloat64, tndarray4from .ir import register_function, register_seeded_function5def register_functions():6 locusVar = tvariable("R", "locus")7 register_function("isValidContig", (dtype("str"),), dtype("bool"), (locusVar,))8 register_function("isValidLocus", (dtype("str"), dtype("int32"),), dtype("bool"), (locusVar,))9 register_function("contigLength", (dtype("str"),), dtype("int32"), (locusVar,))10 register_function("getReferenceSequenceFromValidLocus", (dtype("str"), dtype("int32"), dtype("int32"), dtype("int32"),), dtype("str"), (locusVar,))11 register_function("getReferenceSequence", (dtype("str"), dtype("int32"), dtype("int32"), dtype("int32"),), dtype("str"), (locusVar,))12 register_function("parse_json", (dtype("str"),), dtype("tuple(?T)"), (dtype("?T"),))13 register_function("flatten", (dtype("array<array<?T>>"),), dtype("array<?T>"))14 register_function("difference", (dtype("set<?T>"), dtype("set<?T>"),), dtype("set<?T>"))15 register_function("median", (dtype("set<?T:numeric>"),), dtype("?T"))16 register_function("median", (dtype("array<?T:numeric>"),), dtype("?T"))17 register_function("uniqueMinIndex", (dtype("array<?T>"),), dtype("int32"))18 register_function("mean", (dtype("array<?T:numeric>"),), dtype("float64"))19 register_function("toFloat32", (dtype("?T:numeric"),), dtype("float32"))20 register_function("uniqueMaxIndex", (dtype("array<?T>"),), dtype("int32"))21 register_function("toSet", (dtype("array<?T>"),), dtype("set<?T>"))22 def array_floating_point_divide(arg_type, ret_type):23 register_function("div", (arg_type, tarray(arg_type),), tarray(ret_type))24 register_function("div", (tarray(arg_type), arg_type), tarray(ret_type))25 register_function("div", (tarray(arg_type), tarray(arg_type)), tarray(ret_type))26 array_floating_point_divide(tint32, tfloat32)27 array_floating_point_divide(tint64, tfloat32)28 array_floating_point_divide(tfloat32, tfloat32)29 array_floating_point_divide(tfloat64, tfloat64)30 def ndarray_floating_point_divide(arg_type, ret_type):31 register_function("div", (arg_type, tndarray(arg_type, NatVariable()),), tndarray(ret_type, NatVariable()))32 register_function("div", (tndarray(arg_type, NatVariable()), arg_type), tndarray(ret_type, NatVariable()))33 register_function("div", (tndarray(arg_type, NatVariable()),34 tndarray(arg_type, NatVariable())), tndarray(ret_type, NatVariable()))35 ndarray_floating_point_divide(tint32, tfloat32)36 ndarray_floating_point_divide(tint64, tfloat32)37 ndarray_floating_point_divide(tfloat32, tfloat32)38 ndarray_floating_point_divide(tfloat64, tfloat64)39 register_function("values", (dtype("dict<?key, ?value>"),), dtype("array<?value>"))40 register_function("sliceRight", (dtype("array<?T>"), dtype("int32"),), dtype("array<?T>"))41 register_function("sliceRight", (dtype("str"), dtype("int32"),), dtype("str"))42 register_function("get", (dtype("dict<?key, ?value>"), dtype("?key"),), dtype("?value"))43 register_function("get", (dtype("dict<?key, ?value>"), dtype("?key"), dtype("?value"),), dtype("?value"))44 register_function("max", (dtype("array<?T:numeric>"),), dtype("?T"))45 register_function("nanmax", (dtype("array<?T:numeric>"),), dtype("?T"))46 register_function("max", (dtype("?T"), dtype("?T"),), dtype("?T"))47 register_function("nanmax", (dtype("?T"), dtype("?T"),), dtype("?T"))48 register_function("max_ignore_missing", (dtype("?T"), dtype("?T"),), dtype("?T"))49 register_function("nanmax_ignore_missing", (dtype("?T"), dtype("?T"),), dtype("?T"))50 register_function("product", (dtype("array<?T:numeric>"),), dtype("?T"))51 register_function("toInt32", (dtype("?T:numeric"),), dtype("int32"))52 register_function("extend", (dtype("array<?T>"), dtype("array<?T>"),), dtype("array<?T>"))53 register_function("argmin", (dtype("array<?T>"),), dtype("int32"))54 register_function("toFloat64", (dtype("?T:numeric"),), dtype("float64"))55 register_function("sort", (dtype("array<?T>"),), dtype("array<?T>"))56 register_function("sort", (dtype("array<?T>"), dtype("bool"),), dtype("array<?T>"))57 register_function("isSubset", (dtype("set<?T>"), dtype("set<?T>"),), dtype("bool"))58 register_function("slice", (dtype("str"), dtype("int32"), dtype("int32"),), dtype("str"))59 register_function("slice", (dtype("array<?T>"), dtype("int32"), dtype("int32"),), dtype("array<?T>"))60 register_function("add", (dtype("array<?T:numeric>"), dtype("array<?T>"),), dtype("array<?T>"))61 register_function("add", (dtype("array<?T:numeric>"), dtype("?T"),), dtype("array<?T>"))62 register_function("add", (dtype("?T:numeric"), dtype("array<?T>"),), dtype("array<?T>"))63 register_function("add", (dtype("ndarray<?T:numeric, ?nat>"), dtype("ndarray<?T, ?nat>"),), dtype("ndarray<?T, ?nat>"))64 register_function("add", (dtype("ndarray<?T:numeric, ?nat>"), dtype("?T"),), dtype("ndarray<?T, ?nat>"))65 register_function("add", (dtype("?T:numeric"), dtype("ndarray<?T, ?nat>"),), dtype("ndarray<?T, ?nat>"))66 register_function("pow", (dtype("array<?T:numeric>"), dtype("array<?T>"),), dtype("array<float64>"))67 register_function("pow", (dtype("array<?T:numeric>"), dtype("?T"),), dtype("array<float64>"))68 register_function("pow", (dtype("?T:numeric"), dtype("array<?T>"),), dtype("array<float64>"))69 register_function("pow", (dtype("ndarray<?T:numeric, ?nat>"), dtype("ndarray<?T, ?nat>"),), dtype("ndarray<?T, ?nat>"))70 register_function("pow", (dtype("ndarray<?T:numeric, ?nat>"), dtype("?T"),), dtype("ndarray<?T, ?nat>"))71 register_function("pow", (dtype("?T:numeric"), dtype("ndarray<?T, ?nat>"),), dtype("ndarray<?T, ?nat>"))72 register_function("append", (dtype("array<?T>"), dtype("?T"),), dtype("array<?T>"))73 register_function("sliceLeft", (dtype("str"), dtype("int32"),), dtype("str"))74 register_function("sliceLeft", (dtype("array<?T>"), dtype("int32"),), dtype("array<?T>"))75 register_function("remove", (dtype("set<?T>"), dtype("?T"),), dtype("set<?T>"))76 register_function("index", (dtype("str"), dtype("int32"),), dtype("str"))77 register_function("indexArray", (dtype("array<?T>"), dtype("int32"), dtype("str")), dtype("?T"))78 register_function("index", (dtype("dict<?key, ?value>"), dtype("?key"),), dtype("?value"))79 register_function("dictToArray", (dtype("dict<?key, ?value>"),), dtype("array<tuple(?key, ?value)>"))80 register_function("mod", (dtype("array<?T:numeric>"), dtype("array<?T>"),), dtype("array<?T>"))81 register_function("mod", (dtype("array<?T:numeric>"), dtype("?T"),), dtype("array<?T>"))82 register_function("mod", (dtype("?T:numeric"), dtype("array<?T>"),), dtype("array<?T>"))83 register_function("mod", (dtype("ndarray<?T:numeric, ?nat>"), dtype("ndarray<?T, ?nat>"),), dtype("ndarray<?T, ?nat>"))84 register_function("mod", (dtype("ndarray<?T:numeric, ?nat>"), dtype("?T"),), dtype("ndarray<?T, ?nat>"))85 register_function("mod", (dtype("?T:numeric"), dtype("ndarray<?T, ?nat>"),), dtype("ndarray<?T, ?nat>"))86 register_function("dict", (dtype("array<tuple(?key, ?value)>"),), dtype("dict<?key, ?value>"))87 register_function("dict", (dtype("set<tuple(?key, ?value)>"),), dtype("dict<?key, ?value>"))88 register_function("keys", (dtype("dict<?key, ?value>"),), dtype("array<?key>"))89 register_function("min", (dtype("array<?T:numeric>"),), dtype("?T"))90 register_function("nanmin", (dtype("array<?T:numeric>"),), dtype("?T"))91 register_function("min", (dtype("?T"), dtype("?T"),), dtype("?T"))92 register_function("nanmin", (dtype("?T"), dtype("?T"),), dtype("?T"))93 register_function("min_ignore_missing", (dtype("?T"), dtype("?T"),), dtype("?T"))94 register_function("nanmin_ignore_missing", (dtype("?T"), dtype("?T"),), dtype("?T"))95 register_function("sum", (dtype("array<?T:numeric>"),), dtype("?T"))96 register_function("toInt64", (dtype("?T:numeric"),), dtype("int64"))97 register_function("contains", (dtype("dict<?key, ?value>"), dtype("?key"),), dtype("bool"))98 register_function("contains", (dtype("array<?T>"), dtype("?T"),), dtype("bool"))99 register_function("contains", (dtype("set<?T>"), dtype("?T"),), dtype("bool"))100 register_function("-", (dtype("array<?T:numeric>"), dtype("?T"),), dtype("array<?T>"))101 register_function("-", (dtype("array<?T:numeric>"), dtype("array<?T>"),), dtype("array<?T>"))102 register_function("-", (dtype("?T:numeric"), dtype("array<?T>"),), dtype("array<?T>"))103 register_function("-", (dtype("ndarray<?T:numeric, ?nat>"), dtype("ndarray<?T, ?nat>"),), dtype("ndarray<?T, ?nat>"))104 register_function("-", (dtype("ndarray<?T:numeric, ?nat>"), dtype("?T"),), dtype("ndarray<?T, ?nat>"))105 register_function("-", (dtype("?T:numeric"), dtype("ndarray<?T, ?nat>"),), dtype("ndarray<?T, ?nat>"))106 register_function("addone", (dtype("int32"),), dtype("int32"))107 register_function("isEmpty", (dtype("dict<?key, ?value>"),), dtype("bool"))108 register_function("isEmpty", (dtype("array<?T>"),), dtype("bool"))109 register_function("isEmpty", (dtype("set<?T>"),), dtype("bool"))110 register_function("union", (dtype("set<?T>"), dtype("set<?T>"),), dtype("set<?T>"))111 register_function("mul", (dtype("array<?T:numeric>"), dtype("array<?T>"),), dtype("array<?T>"))112 register_function("mul", (dtype("array<?T:numeric>"), dtype("?T"),), dtype("array<?T>"))113 register_function("mul", (dtype("?T:numeric"), dtype("array<?T>"),), dtype("array<?T>"))114 register_function("mul", (dtype("ndarray<?T:numeric, ?nat>"), dtype("ndarray<?T, ?nat>"),), dtype("ndarray<?T, ?nat>"))115 register_function("mul", (dtype("ndarray<?T:numeric, ?nat>"), dtype("?T"),), dtype("ndarray<?T, ?nat>"))116 register_function("mul", (dtype("?T:numeric"), dtype("ndarray<?T, ?nat>"),), dtype("ndarray<?T, ?nat>"))117 register_function("intersection", (dtype("set<?T>"), dtype("set<?T>"),), dtype("set<?T>"))118 register_function("add", (dtype("set<?T>"), dtype("?T"),), dtype("set<?T>"))119 register_function("argmax", (dtype("array<?T>"),), dtype("int32"))120 register_function("floordiv", (dtype("array<?T:numeric>"), dtype("array<?T>"),), dtype("array<?T>"))121 register_function("floordiv", (dtype("array<?T:numeric>"), dtype("?T"),), dtype("array<?T>"))122 register_function("floordiv", (dtype("?T:numeric"), dtype("array<?T>"),), dtype("array<?T>"))123 register_function("floordiv", (dtype("ndarray<?T:numeric, ?nat>"), dtype("ndarray<?T, ?nat>"),), dtype("ndarray<?T, ?nat>"))124 register_function("floordiv", (dtype("ndarray<?T:numeric, ?nat>"), dtype("?T"),), dtype("ndarray<?T, ?nat>"))125 register_function("floordiv", (dtype("?T:numeric"), dtype("ndarray<?T, ?nat>"),), dtype("ndarray<?T, ?nat>"))126 register_function("keySet", (dtype("dict<?key, ?value>"),), dtype("set<?key>"))127 register_function("qnorm", (dtype("float64"),), dtype("float64"))128 register_function("oneHotAlleles", (dtype("call"), dtype("int32"),), dtype("array<int32>"))129 register_function("dpois", (dtype("float64"), dtype("float64"), dtype("bool"),), dtype("float64"))130 register_function("dpois", (dtype("float64"), dtype("float64"),), dtype("float64"))131 register_function("ploidy", (dtype("call"),), dtype("int32"))132 register_function("lor", (dtype("bool"), dtype("bool"),), dtype("bool"))133 register_function("ppois", (dtype("float64"), dtype("float64"), dtype("bool"), dtype("bool"),), dtype("float64"))134 register_function("ppois", (dtype("float64"), dtype("float64"),), dtype("float64"))135 register_function("log10", (dtype("float64"),), dtype("float64"))136 register_function("isHet", (dtype("call"),), dtype("bool"))137 register_function("isAutosomalOrPseudoAutosomal", (dtype("?T:locus"),), dtype("bool"))138 register_function("testCodeUnification", (dtype("?x:numeric"), dtype("?x:int32"),), dtype("?x"))139 register_seeded_function("rand_pois", (dtype("float64"),), dtype("float64"))140 register_seeded_function("rand_pois", (dtype("int32"), dtype("float64"),), dtype("array<float64>"))141 register_function("toFloat32", (dtype("str"),), dtype("float32"))142 register_function("toFloat32", (dtype("bool"),), dtype("float32"))143 register_function("isAutosomal", (dtype("?T:locus"),), dtype("bool"))144 register_function("isPhased", (dtype("call"),), dtype("bool"))145 register_function("isHomVar", (dtype("call"),), dtype("bool"))146 register_function("corr", (dtype("array<float64>"), dtype("array<float64>"),), dtype("float64"))147 register_function("log", (dtype("float64"), dtype("float64"),), dtype("float64"))148 register_function("log", (dtype("float64"),), dtype("float64"))149 register_function("foobar2", (), dtype("int32"))150 register_function("approxEqual", (dtype("float64"), dtype("float64"), dtype("float64"), dtype("bool"), dtype("bool"),), dtype("bool"))151 register_function("includesEnd", (dtype("interval<?T>"),), dtype("bool"))152 register_function("position", (dtype("?T:locus"),), dtype("int32"))153 register_seeded_function("rand_unif", (dtype("float64"), dtype("float64"),), dtype("float64"))154 register_function("showStr", (dtype("?T"), dtype("int32")), dtype("str"))155 register_function("str", (dtype("?T"),), dtype("str"))156 register_function("valuesSimilar", (dtype("?T"), dtype("?T"), dtype('float64'), dtype('bool'),), dtype("bool"))157 register_function("replace", (dtype("str"), dtype("str"), dtype("str"),), dtype("str"))158 register_function("exp", (dtype("float64"),), dtype("float64"))159 register_function("land", (dtype("bool"), dtype("bool"),), dtype("bool"))160 register_function("compare", (dtype("int32"), dtype("int32"),), dtype("int32"))161 register_function("triangle", (dtype("int32"),), dtype("int32"))162 register_function("Interval", (dtype("?T"), dtype("?T"), dtype("bool"), dtype("bool"),), dtype("interval<?T>"))163 register_function("contig", (dtype("?T:locus"),), dtype("str"))164 register_function("Call", (dtype("bool"),), dtype("call"))165 register_function("Call", (dtype("str"),), dtype("call"))166 register_function("Call", (dtype("int32"), dtype("bool"),), dtype("call"))167 register_function("Call", (dtype("int32"), dtype("int32"), dtype("bool"),), dtype("call"))168 register_function("Call", (dtype("array<int32>"), dtype("bool"),), dtype("call"))169 register_function("qchisqtail", (dtype("float64"), dtype("float64"),), dtype("float64"))170 register_function("binomTest", (dtype("int32"), dtype("int32"), dtype("float64"), dtype("int32"),), dtype("float64"))171 register_function("qpois", (dtype("float64"), dtype("float64"),), dtype("int32"))172 register_function("qpois", (dtype("float64"), dtype("float64"), dtype("bool"), dtype("bool"),), dtype("int32"))173 register_function("is_finite", (dtype("float32"),), dtype("bool"))174 register_function("is_finite", (dtype("float64"),), dtype("bool"))175 register_function("inYPar", (dtype("?T:locus"),), dtype("bool"))176 register_function("contingency_table_test", (dtype("int32"), dtype("int32"), dtype("int32"), dtype("int32"), dtype("int32"),), dtype("struct{p_value: float64, odds_ratio: float64}"))177 register_function("toInt32", (dtype("bool"),), dtype("int32"))178 register_function("toInt32", (dtype("str"),), dtype("int32"))179 register_function("foobar1", (), dtype("int32"))180 register_function("toFloat64", (dtype("str"),), dtype("float64"))181 register_function("toFloat64", (dtype("bool"),), dtype("float64"))182 register_function("dbeta", (dtype("float64"), dtype("float64"), dtype("float64"),), dtype("float64"))183 register_function("Locus", (dtype("str"),), dtype("?T:locus"))184 register_function("Locus", (dtype("str"), dtype("int32"),), dtype("?T:locus"))185 register_function("LocusAlleles", (dtype("str"),), dtype("struct{locus: ?T, alleles: array<str>}"))186 register_function("LocusInterval", (dtype("str"), dtype("bool"),), dtype("interval<?T:locus>"))187 register_function("LocusInterval", (dtype("str"), dtype("int32"), dtype("int32"), dtype("bool"), dtype("bool"), dtype("bool"),), dtype("interval<?T:locus>"))188 register_function("globalPosToLocus", (dtype("int64"),), dtype("?T:locus"))189 register_function("locusToGlobalPos", (dtype("?T:locus"),), dtype("int64"))190 register_function("liftoverLocus", (dtype("?T:locus"), dtype('float64'),), dtype("struct{result:?U:locus,is_negative_strand:bool}"))191 register_function("liftoverLocusInterval", (dtype("interval<?T:locus>"), dtype('float64'),), dtype("struct{result:interval<?U:locus>,is_negative_strand:bool}"))192 register_function("min_rep", (dtype("?T:locus"), dtype("array<str>"),), dtype("struct{locus: ?T, alleles: array<str>}"))193 register_function("locus_windows_per_contig", (dtype("array<array<float64>>"), dtype("float64"),), dtype("tuple(array<int32>, array<int32>)"))194 register_function("toBoolean", (dtype("str"),), dtype("bool"))195 register_seeded_function("rand_bool", (dtype("float64"),), dtype("bool"))196 register_function("pchisqtail", (dtype("float64"), dtype("float64"),), dtype("float64"))197 register_seeded_function("rand_cat", (dtype("array<float64>"),), dtype("int32"))198 register_function("inYNonPar", (dtype("?T:locus"),), dtype("bool"))199 register_function("concat", (dtype("str"), dtype("str"),), dtype("str"))200 register_function("pow", (dtype("float32"), dtype("float32"),), dtype("float64"))201 register_function("pow", (dtype("int32"), dtype("int32"),), dtype("float64"))202 register_function("pow", (dtype("int64"), dtype("int64"),), dtype("float64"))203 register_function("pow", (dtype("float64"), dtype("float64"),), dtype("float64"))204 register_function("length", (dtype("str"),), dtype("int32"))205 register_function("slice", (dtype("str"), dtype("int32"), dtype("int32"),), dtype("str"))206 register_function("split", (dtype("str"), dtype("str"), dtype("int32"),), dtype("array<str>"))207 register_function("split", (dtype("str"), dtype("str"),), dtype("array<str>"))208 register_seeded_function("rand_gamma", (dtype("float64"), dtype("float64"),), dtype("float64"))209 register_function("UnphasedDiploidGtIndexCall", (dtype("int32"),), dtype("call"))210 register_function("index", (dtype("call"), dtype("int32"),), dtype("int32"))211 register_function("sign", (dtype("int64"),), dtype("int64"))212 register_function("sign", (dtype("float64"),), dtype("float64"))213 register_function("sign", (dtype("float32"),), dtype("float32"))214 register_function("sign", (dtype("int32"),), dtype("int32"))215 register_function("unphasedDiploidGtIndex", (dtype("call"),), dtype("int32"))216 register_function("gamma", (dtype("float64"),), dtype("float64"))217 register_function("mod", (dtype("float64"), dtype("float64"),), dtype("float64"))218 register_function("mod", (dtype("int64"), dtype("int64"),), dtype("int64"))219 register_function("mod", (dtype("float32"), dtype("float32"),), dtype("float32"))220 register_function("mod", (dtype("int32"), dtype("int32"),), dtype("int32"))221 register_function("fisher_exact_test", (dtype("int32"), dtype("int32"), dtype("int32"), dtype("int32"),), dtype("struct{p_value: float64, odds_ratio: float64, ci_95_lower: float64, ci_95_upper: float64}"))222 register_function("floor", (dtype("float64"),), dtype("float64"))223 register_function("floor", (dtype("float32"),), dtype("float32"))224 register_function("isNonRef", (dtype("call"),), dtype("bool"))225 register_function("includesStart", (dtype("interval<?T>"),), dtype("bool"))226 register_function("isHetNonRef", (dtype("call"),), dtype("bool"))227 register_function("hardy_weinberg_test", (dtype("int32"), dtype("int32"), dtype("int32"),), dtype("struct{het_freq_hwe: float64, p_value: float64}"))228 register_function("haplotype_freq_em", (dtype("array<int32>"),), dtype("array<float64>"))229 register_function("nNonRefAlleles", (dtype("call"),), dtype("int32"))230 register_function("abs", (dtype("float64"),), dtype("float64"))231 register_function("abs", (dtype("float32"),), dtype("float32"))232 register_function("abs", (dtype("int64"),), dtype("int64"))233 register_function("abs", (dtype("int32"),), dtype("int32"))234 register_function("endswith", (dtype("str"), dtype("str"),), dtype("bool"))235 register_function("sqrt", (dtype("float64"),), dtype("float64"))236 register_function("isnan", (dtype("float32"),), dtype("bool"))237 register_function("isnan", (dtype("float64"),), dtype("bool"))238 register_function("lower", (dtype("str"),), dtype("str"))239 register_seeded_function("rand_beta", (dtype("float64"), dtype("float64"),), dtype("float64"))240 register_seeded_function("rand_beta", (dtype("float64"), dtype("float64"), dtype("float64"), dtype("float64"),), dtype("float64"))241 register_function("toInt64", (dtype("bool"),), dtype("int64"))242 register_function("toInt64", (dtype("str"),), dtype("int64"))243 register_function("testCodeUnification2", (dtype("?x"),), dtype("?x"))244 register_function("contains", (dtype("str"), dtype("str"),), dtype("bool"))245 register_function("contains", (dtype("interval<?T>"), dtype("?T"),), dtype("bool"))246 register_function("entropy", (dtype("str"),), dtype("float64"))247 register_function("filtering_allele_frequency", (dtype("int32"), dtype("int32"), dtype("float64"),), dtype("float64"))248 register_function("gqFromPL", (dtype("array<?N:int32>"),), dtype("int32"))249 register_function("startswith", (dtype("str"), dtype("str"),), dtype("bool"))250 register_function("ceil", (dtype("float32"),), dtype("float32"))251 register_function("ceil", (dtype("float64"),), dtype("float64"))252 register_function("json", (dtype("?T"),), dtype("str"))253 register_function("strip", (dtype("str"),), dtype("str"))254 register_function("firstMatchIn", (dtype("str"), dtype("str"),), dtype("array<str>"))255 register_function("isEmpty", (dtype("interval<?T>"),), dtype("bool"))256 register_function("~", (dtype("str"), dtype("str"),), dtype("bool"))257 register_function("mkString", (dtype("set<str>"), dtype("str"),), dtype("str"))258 register_function("mkString", (dtype("array<str>"), dtype("str"),), dtype("str"))259 register_function("dosage", (dtype("array<?N:float64>"),), dtype("float64"))260 register_function("upper", (dtype("str"),), dtype("str"))261 register_function("overlaps", (dtype("interval<?T>"), dtype("interval<?T>"),), dtype("bool"))262 register_function("downcode", (dtype("call"), dtype("int32"),), dtype("call"))263 register_function("inXPar", (dtype("?T:locus"),), dtype("bool"))264 register_function("format", (dtype("str"), dtype("?T:tuple"),), dtype("str"))265 register_function("pnorm", (dtype("float64"),), dtype("float64"))266 register_function("is_infinite", (dtype("float32"),), dtype("bool"))267 register_function("is_infinite", (dtype("float64"),), dtype("bool"))268 register_function("isHetRef", (dtype("call"),), dtype("bool"))269 register_function("isMitochondrial", (dtype("?T:locus"),), dtype("bool"))270 register_function("hamming", (dtype("str"), dtype("str"),), dtype("int32"))271 register_function("end", (dtype("interval<?T>"),), dtype("?T"))272 register_function("start", (dtype("interval<?T>"),), dtype("?T"))273 register_function("inXNonPar", (dtype("?T:locus"),), dtype("bool"))274 register_function("escapeString", (dtype("str"),), dtype("str"))275 register_function("isHomRef", (dtype("call"),), dtype("bool"))276 register_seeded_function("rand_norm", (dtype("float64"), dtype("float64"),), dtype("float64"))277 register_function("chi_squared_test", (dtype("int32"), dtype("int32"), dtype("int32"), dtype("int32"),), dtype("struct{p_value: float64, odds_ratio: float64}"))278 register_function("strftime", (dtype("str"), dtype("int64"), dtype("str")), dtype("str"))...
Automation Testing Tutorials
Learn to execute automation testing from scratch with LambdaTest Learning Hub. Right from setting up the prerequisites to run your first automation test, to following best practices and diving deeper into advanced test scenarios. LambdaTest Learning Hubs compile a list of step-by-step guides to help you be proficient with different test automation frameworks i.e. Selenium, Cypress, TestNG etc.
LambdaTest Learning Hubs:
- JUnit Tutorial
- TestNG Tutorial
- Webdriver Tutorial
- WebDriverIO Tutorial
- Protractor Tutorial
- Selenium 4 Tutorial
- Jenkins Tutorial
- NUnit Tutorial
- Jest Tutorial
- Playwright Tutorial
- Cypress Tutorial
- PyTest Tutorial
YouTube
You could also refer to video tutorials over LambdaTest YouTube channel to get step by step demonstration from industry experts.
Run pandera automation tests on LambdaTest cloud grid
Perform automation testing on 3000+ real desktop and mobile devices online.
Try LambdaTest Now !!
Get 100 minutes of automation test minutes FREE!!
Was this article helpful?
