/
num_ops.py
498 lines (402 loc) · 20.7 KB
/
num_ops.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
#
# Licensed to the Apache Software Foundation (ASF) under one or more
# contributor license agreements. See the NOTICE file distributed with
# this work for additional information regarding copyright ownership.
# The ASF licenses this file to You under the Apache License, Version 2.0
# (the "License"); you may not use this file except in compliance with
# the License. You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
import numbers
from typing import Any, Union
import numpy as np
import pandas as pd
from pandas.api.types import is_bool_dtype, is_integer_dtype, CategoricalDtype
from pyspark.pandas._typing import Dtype, IndexOpsLike, SeriesOrIndex
from pyspark.pandas.base import column_op, IndexOpsMixin, numpy_column_op
from pyspark.pandas.data_type_ops.base import (
DataTypeOps,
is_valid_operand_for_numeric_arithmetic,
transform_boolean_operand_to_numeric,
_as_bool_type,
_as_categorical_type,
_as_other_type,
_as_string_type,
_sanitize_list_like,
)
from pyspark.pandas.spark import functions as SF
from pyspark.pandas.typedef.typehints import extension_dtypes, pandas_on_spark_type
from pyspark.sql import functions as F
from pyspark.sql.column import Column
from pyspark.sql.types import (
BooleanType,
DataType,
StringType,
)
def _non_fractional_astype(
index_ops: IndexOpsLike, dtype: Dtype, spark_type: DataType
) -> IndexOpsLike:
if isinstance(dtype, CategoricalDtype):
return _as_categorical_type(index_ops, dtype, spark_type)
elif isinstance(spark_type, BooleanType):
return _as_bool_type(index_ops, dtype)
elif isinstance(spark_type, StringType):
return _as_string_type(index_ops, dtype, null_str=str(np.nan))
else:
return _as_other_type(index_ops, dtype, spark_type)
class NumericOps(DataTypeOps):
"""The class for binary operations of numeric pandas-on-Spark objects."""
@property
def pretty_name(self) -> str:
return "numerics"
def add(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
if not is_valid_operand_for_numeric_arithmetic(right):
raise TypeError("Addition can not be applied to given types.")
right = transform_boolean_operand_to_numeric(right, spark_type=left.spark.data_type)
return column_op(Column.__add__)(left, right)
def sub(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
if not is_valid_operand_for_numeric_arithmetic(right):
raise TypeError("Subtraction can not be applied to given types.")
right = transform_boolean_operand_to_numeric(right, spark_type=left.spark.data_type)
return column_op(Column.__sub__)(left, right)
def mod(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
if not is_valid_operand_for_numeric_arithmetic(right):
raise TypeError("Modulo can not be applied to given types.")
def mod(left: Column, right: Any) -> Column:
return ((left % right) + right) % right
right = transform_boolean_operand_to_numeric(right, spark_type=left.spark.data_type)
return column_op(mod)(left, right)
def pow(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
if not is_valid_operand_for_numeric_arithmetic(right):
raise TypeError("Exponentiation can not be applied to given types.")
def pow_func(left: Column, right: Any) -> Column:
return (
F.when(left == 1, left)
.when(SF.lit(right) == 0, 1)
.otherwise(Column.__pow__(left, right))
)
right = transform_boolean_operand_to_numeric(right, spark_type=left.spark.data_type)
return column_op(pow_func)(left, right)
def radd(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
if not isinstance(right, numbers.Number):
raise TypeError("Addition can not be applied to given types.")
right = transform_boolean_operand_to_numeric(right)
return column_op(Column.__radd__)(left, right)
def rsub(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
if not isinstance(right, numbers.Number):
raise TypeError("Subtraction can not be applied to given types.")
right = transform_boolean_operand_to_numeric(right)
return column_op(Column.__rsub__)(left, right)
def rmul(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
if not isinstance(right, numbers.Number):
raise TypeError("Multiplication can not be applied to given types.")
right = transform_boolean_operand_to_numeric(right)
return column_op(Column.__rmul__)(left, right)
def rpow(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
if not isinstance(right, numbers.Number):
raise TypeError("Exponentiation can not be applied to given types.")
def rpow_func(left: Column, right: Any) -> Column:
return F.when(SF.lit(right == 1), right).otherwise(Column.__rpow__(left, right))
right = transform_boolean_operand_to_numeric(right)
return column_op(rpow_func)(left, right)
def rmod(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
if not isinstance(right, numbers.Number):
raise TypeError("Modulo can not be applied to given types.")
def rmod(left: Column, right: Any) -> Column:
return ((right % left) + left) % left
right = transform_boolean_operand_to_numeric(right)
return column_op(rmod)(left, right)
def neg(self, operand: IndexOpsLike) -> IndexOpsLike:
return operand._with_new_scol(-operand.spark.column, field=operand._internal.data_fields[0])
def abs(self, operand: IndexOpsLike) -> IndexOpsLike:
return operand._with_new_scol(
F.abs(operand.spark.column), field=operand._internal.data_fields[0]
)
def lt(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
return column_op(Column.__lt__)(left, right)
def le(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
return column_op(Column.__le__)(left, right)
def ge(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
return column_op(Column.__ge__)(left, right)
def gt(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
return column_op(Column.__gt__)(left, right)
class IntegralOps(NumericOps):
"""
The class for binary operations of pandas-on-Spark objects with spark types:
LongType, IntegerType, ByteType and ShortType.
"""
@property
def pretty_name(self) -> str:
return "integrals"
def mul(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
if isinstance(right, IndexOpsMixin) and isinstance(right.spark.data_type, StringType):
return column_op(SF.repeat)(right, left)
if not is_valid_operand_for_numeric_arithmetic(right):
raise TypeError("Multiplication can not be applied to given types.")
right = transform_boolean_operand_to_numeric(right, spark_type=left.spark.data_type)
return column_op(Column.__mul__)(left, right)
def truediv(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
if not is_valid_operand_for_numeric_arithmetic(right):
raise TypeError("True division can not be applied to given types.")
def truediv(left: Column, right: Any) -> Column:
return F.when(
SF.lit(right != 0) | SF.lit(right).isNull(), left.__div__(right)
).otherwise(SF.lit(np.inf).__div__(left))
right = transform_boolean_operand_to_numeric(right, spark_type=left.spark.data_type)
return numpy_column_op(truediv)(left, right)
def floordiv(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
if not is_valid_operand_for_numeric_arithmetic(right):
raise TypeError("Floor division can not be applied to given types.")
def floordiv(left: Column, right: Any) -> Column:
return F.when(SF.lit(right is np.nan), np.nan).otherwise(
F.when(
SF.lit(right != 0) | SF.lit(right).isNull(), F.floor(left.__div__(right))
).otherwise(SF.lit(np.inf).__div__(left))
)
right = transform_boolean_operand_to_numeric(right, spark_type=left.spark.data_type)
return numpy_column_op(floordiv)(left, right)
def rtruediv(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
if not isinstance(right, numbers.Number):
raise TypeError("True division can not be applied to given types.")
def rtruediv(left: Column, right: Any) -> Column:
return F.when(left == 0, SF.lit(np.inf).__div__(right)).otherwise(
SF.lit(right).__truediv__(left)
)
right = transform_boolean_operand_to_numeric(right, spark_type=left.spark.data_type)
return numpy_column_op(rtruediv)(left, right)
def rfloordiv(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
if not isinstance(right, numbers.Number):
raise TypeError("Floor division can not be applied to given types.")
def rfloordiv(left: Column, right: Any) -> Column:
return F.when(SF.lit(left == 0), SF.lit(np.inf).__div__(right)).otherwise(
F.floor(SF.lit(right).__div__(left))
)
right = transform_boolean_operand_to_numeric(right, spark_type=left.spark.data_type)
return numpy_column_op(rfloordiv)(left, right)
def invert(self, operand: IndexOpsLike) -> IndexOpsLike:
return operand._with_new_scol(
F.bitwise_not(operand.spark.column), field=operand._internal.data_fields[0]
)
def astype(self, index_ops: IndexOpsLike, dtype: Union[str, type, Dtype]) -> IndexOpsLike:
dtype, spark_type = pandas_on_spark_type(dtype)
return _non_fractional_astype(index_ops, dtype, spark_type)
class FractionalOps(NumericOps):
"""
The class for binary operations of pandas-on-Spark objects with spark types:
FloatType, DoubleType.
"""
@property
def pretty_name(self) -> str:
return "fractions"
def mul(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
if not is_valid_operand_for_numeric_arithmetic(right):
raise TypeError("Multiplication can not be applied to given types.")
right = transform_boolean_operand_to_numeric(right, spark_type=left.spark.data_type)
return column_op(Column.__mul__)(left, right)
def truediv(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
if not is_valid_operand_for_numeric_arithmetic(right):
raise TypeError("True division can not be applied to given types.")
def truediv(left: Column, right: Any) -> Column:
return F.when(
SF.lit(right != 0) | SF.lit(right).isNull(), left.__div__(right)
).otherwise(
F.when(SF.lit(left == np.inf) | SF.lit(left == -np.inf), left).otherwise(
SF.lit(np.inf).__div__(left)
)
)
right = transform_boolean_operand_to_numeric(right, spark_type=left.spark.data_type)
return numpy_column_op(truediv)(left, right)
def floordiv(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
if not is_valid_operand_for_numeric_arithmetic(right):
raise TypeError("Floor division can not be applied to given types.")
def floordiv(left: Column, right: Any) -> Column:
return F.when(SF.lit(right is np.nan), np.nan).otherwise(
F.when(
SF.lit(right != 0) | SF.lit(right).isNull(), F.floor(left.__div__(right))
).otherwise(
F.when(SF.lit(left == np.inf) | SF.lit(left == -np.inf), left).otherwise(
SF.lit(np.inf).__div__(left)
)
)
)
right = transform_boolean_operand_to_numeric(right, spark_type=left.spark.data_type)
return numpy_column_op(floordiv)(left, right)
def rtruediv(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
if not isinstance(right, numbers.Number):
raise TypeError("True division can not be applied to given types.")
def rtruediv(left: Column, right: Any) -> Column:
return F.when(left == 0, SF.lit(np.inf).__div__(right)).otherwise(
SF.lit(right).__truediv__(left)
)
right = transform_boolean_operand_to_numeric(right, spark_type=left.spark.data_type)
return numpy_column_op(rtruediv)(left, right)
def rfloordiv(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
_sanitize_list_like(right)
if not isinstance(right, numbers.Number):
raise TypeError("Floor division can not be applied to given types.")
def rfloordiv(left: Column, right: Any) -> Column:
return F.when(SF.lit(left == 0), SF.lit(np.inf).__div__(right)).otherwise(
F.when(SF.lit(left) == np.nan, np.nan).otherwise(
F.floor(SF.lit(right).__div__(left))
)
)
right = transform_boolean_operand_to_numeric(right, spark_type=left.spark.data_type)
return numpy_column_op(rfloordiv)(left, right)
def isnull(self, index_ops: IndexOpsLike) -> IndexOpsLike:
return index_ops._with_new_scol(
index_ops.spark.column.isNull() | F.isnan(index_ops.spark.column),
field=index_ops._internal.data_fields[0].copy(
dtype=np.dtype("bool"), spark_type=BooleanType(), nullable=False
),
)
def nan_to_null(self, index_ops: IndexOpsLike) -> IndexOpsLike:
# Special handle floating point types because Spark's count treats nan as a valid value,
# whereas pandas count doesn't include nan.
return index_ops._with_new_scol(
F.nanvl(index_ops.spark.column, SF.lit(None)),
field=index_ops._internal.data_fields[0].copy(nullable=True),
)
def astype(self, index_ops: IndexOpsLike, dtype: Union[str, type, Dtype]) -> IndexOpsLike:
dtype, spark_type = pandas_on_spark_type(dtype)
if is_integer_dtype(dtype) and not isinstance(dtype, extension_dtypes):
if index_ops.hasnans:
raise ValueError(
"Cannot convert %s with missing values to integer" % self.pretty_name
)
if isinstance(dtype, CategoricalDtype):
return _as_categorical_type(index_ops, dtype, spark_type)
elif isinstance(spark_type, BooleanType):
if isinstance(dtype, extension_dtypes):
scol = index_ops.spark.column.cast(spark_type)
else:
scol = F.when(
index_ops.spark.column.isNull() | F.isnan(index_ops.spark.column),
SF.lit(True),
).otherwise(index_ops.spark.column.cast(spark_type))
return index_ops._with_new_scol(
scol.alias(index_ops._internal.data_spark_column_names[0]),
field=index_ops._internal.data_fields[0].copy(dtype=dtype, spark_type=spark_type),
)
elif isinstance(spark_type, StringType):
return _as_string_type(index_ops, dtype, null_str=str(np.nan))
else:
return _as_other_type(index_ops, dtype, spark_type)
class DecimalOps(FractionalOps):
"""
The class for decimal operations of pandas-on-Spark objects with spark type:
DecimalType.
"""
@property
def pretty_name(self) -> str:
return "decimal"
def lt(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
raise TypeError("< can not be applied to %s." % self.pretty_name)
def le(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
raise TypeError("<= can not be applied to %s." % self.pretty_name)
def gt(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
raise TypeError("> can not be applied to %s." % self.pretty_name)
def ge(self, left: IndexOpsLike, right: Any) -> SeriesOrIndex:
raise TypeError(">= can not be applied to %s." % self.pretty_name)
def isnull(self, index_ops: IndexOpsLike) -> IndexOpsLike:
return index_ops._with_new_scol(
index_ops.spark.column.isNull(),
field=index_ops._internal.data_fields[0].copy(
dtype=np.dtype("bool"), spark_type=BooleanType(), nullable=False
),
)
def nan_to_null(self, index_ops: IndexOpsLike) -> IndexOpsLike:
return index_ops.copy()
def astype(self, index_ops: IndexOpsLike, dtype: Union[str, type, Dtype]) -> IndexOpsLike:
# TODO(SPARK-36230): check index_ops.hasnans after fixing SPARK-36230
dtype, spark_type = pandas_on_spark_type(dtype)
return _non_fractional_astype(index_ops, dtype, spark_type)
class IntegralExtensionOps(IntegralOps):
"""
The class for binary operations of pandas-on-Spark objects with one of the
- spark types:
LongType, IntegerType, ByteType and ShortType
- dtypes:
Int8Dtype, Int16Dtype, Int32Dtype, Int64Dtype
"""
def restore(self, col: pd.Series) -> pd.Series:
"""Restore column when to_pandas."""
return col.astype(self.dtype)
def astype(self, index_ops: IndexOpsLike, dtype: Union[str, type, Dtype]) -> IndexOpsLike:
dtype, spark_type = pandas_on_spark_type(dtype)
if is_integer_dtype(dtype) and not isinstance(dtype, extension_dtypes):
if index_ops.hasnans:
raise ValueError(
"Cannot convert %s with missing values to integer" % self.pretty_name
)
elif is_bool_dtype(dtype) and not isinstance(dtype, extension_dtypes):
if index_ops.hasnans:
raise ValueError("Cannot convert %s with missing values to bool" % self.pretty_name)
return _non_fractional_astype(index_ops, dtype, spark_type)
class FractionalExtensionOps(FractionalOps):
"""
The class for binary operations of pandas-on-Spark objects with one of the
- spark types:
FloatType, DoubleType and DecimalType
- dtypes:
Float32Dtype, Float64Dtype
"""
def restore(self, col: pd.Series) -> pd.Series:
"""Restore column when to_pandas."""
return col.astype(self.dtype)
def astype(self, index_ops: IndexOpsLike, dtype: Union[str, type, Dtype]) -> IndexOpsLike:
dtype, spark_type = pandas_on_spark_type(dtype)
if is_integer_dtype(dtype) and not isinstance(dtype, extension_dtypes):
if index_ops.hasnans:
raise ValueError(
"Cannot convert %s with missing values to integer" % self.pretty_name
)
elif is_bool_dtype(dtype) and not isinstance(dtype, extension_dtypes):
if index_ops.hasnans:
raise ValueError("Cannot convert %s with missing values to bool" % self.pretty_name)
if isinstance(dtype, CategoricalDtype):
return _as_categorical_type(index_ops, dtype, spark_type)
elif isinstance(spark_type, BooleanType):
if isinstance(dtype, extension_dtypes):
scol = index_ops.spark.column.cast(spark_type)
else:
scol = F.when(
index_ops.spark.column.isNull() | F.isnan(index_ops.spark.column),
SF.lit(True),
).otherwise(index_ops.spark.column.cast(spark_type))
return index_ops._with_new_scol(
scol.alias(index_ops._internal.data_spark_column_names[0]),
field=index_ops._internal.data_fields[0].copy(dtype=dtype, spark_type=spark_type),
)
elif isinstance(spark_type, StringType):
return _as_string_type(index_ops, dtype, null_str=str(np.nan))
else:
return _as_other_type(index_ops, dtype, spark_type)