red_amber 0.3.0 → 0.4.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/.rubocop.yml +39 -20
- data/.yardopts +2 -0
- data/CHANGELOG.md +113 -0
- data/Gemfile +1 -1
- data/LICENSE +1 -1
- data/README.md +25 -26
- data/benchmark/basic.yml +2 -2
- data/benchmark/combine.yml +2 -2
- data/benchmark/dataframe.yml +2 -2
- data/benchmark/group.yml +2 -2
- data/benchmark/reshape.yml +2 -2
- data/benchmark/vector.yml +3 -0
- data/doc/DataFrame.md +32 -12
- data/doc/DataFrame_Comparison.md +65 -0
- data/doc/SubFrames.md +11 -0
- data/doc/Vector.md +207 -1
- data/doc/yard-templates/default/fulldoc/html/css/common.css +6 -0
- data/lib/red_amber/data_frame.rb +429 -75
- data/lib/red_amber/data_frame_combinable.rb +516 -66
- data/lib/red_amber/data_frame_displayable.rb +244 -14
- data/lib/red_amber/data_frame_indexable.rb +121 -18
- data/lib/red_amber/data_frame_loadsave.rb +78 -10
- data/lib/red_amber/data_frame_reshaping.rb +184 -14
- data/lib/red_amber/data_frame_selectable.rb +622 -66
- data/lib/red_amber/data_frame_variable_operation.rb +446 -34
- data/lib/red_amber/group.rb +187 -22
- data/lib/red_amber/helper.rb +70 -10
- data/lib/red_amber/refinements.rb +12 -5
- data/lib/red_amber/subframes.rb +1066 -0
- data/lib/red_amber/vector.rb +385 -11
- data/lib/red_amber/vector_aggregation.rb +312 -0
- data/lib/red_amber/vector_binary_element_wise.rb +387 -0
- data/lib/red_amber/vector_selectable.rb +217 -12
- data/lib/red_amber/vector_unary_element_wise.rb +436 -0
- data/lib/red_amber/vector_updatable.rb +278 -34
- data/lib/red_amber/version.rb +2 -1
- data/lib/red_amber.rb +13 -1
- data/red_amber.gemspec +2 -2
- metadata +13 -8
- data/doc/image/dataframe/reshaping_DataFrames.png +0 -0
- data/lib/red_amber/vector_functions.rb +0 -242
@@ -1,17 +1,36 @@
|
|
1
1
|
# frozen_string_literal: true
|
2
2
|
|
3
3
|
module RedAmber
|
4
|
-
#
|
4
|
+
# Mix-in for the class DataFrame
|
5
5
|
module DataFrameCombinable
|
6
6
|
# Refinements for Arrow::Table
|
7
7
|
using RefineArrowTable
|
8
8
|
|
9
|
-
# Concatenate other
|
9
|
+
# Concatenate other dataframes or tables onto the bottom of self.
|
10
10
|
#
|
11
|
+
# @note the `#types` must be same as `other#types`.
|
11
12
|
# @param other [DataFrame, Arrow::Table, Array<DataFrame, Arrow::Table>]
|
12
|
-
#
|
13
|
+
# DataFrames or Tables to concatenate.
|
13
14
|
# @return [DataFrame]
|
14
|
-
#
|
15
|
+
# concatenated dataframe.
|
16
|
+
# @example
|
17
|
+
# df = DataFrame.new(x: [1, 2], y: ['A', 'B'])
|
18
|
+
# other = DataFrame.new(x: [3, 4], y: ['C', 'D'])
|
19
|
+
# [df.types, other.types]
|
20
|
+
#
|
21
|
+
# # =>
|
22
|
+
# [[:uint8, :string], [:uint8, :string]]
|
23
|
+
#
|
24
|
+
# df.concatenate(other)
|
25
|
+
#
|
26
|
+
# # =>
|
27
|
+
# x y
|
28
|
+
# <uint8> <string>
|
29
|
+
# 0 1 A
|
30
|
+
# 1 2 B
|
31
|
+
# 2 3 C
|
32
|
+
# 3 4 D
|
33
|
+
#
|
15
34
|
def concatenate(*other)
|
16
35
|
case other
|
17
36
|
in [] | [nil] | [[]]
|
@@ -39,14 +58,27 @@ module RedAmber
|
|
39
58
|
alias_method :concat, :concatenate
|
40
59
|
alias_method :bind_rows, :concatenate
|
41
60
|
|
42
|
-
# Merge other
|
43
|
-
#
|
44
|
-
#
|
45
|
-
#
|
61
|
+
# Merge other DataFrames or Tables.
|
62
|
+
#
|
63
|
+
# @note the `#size` must be same as `other#size`.
|
64
|
+
# @note self and other must not share the same key.
|
46
65
|
# @param other [DataFrame, Arrow::Table, Array<DataFrame, Arrow::Table>]
|
47
|
-
#
|
66
|
+
# DataFrames or Tables to merge.
|
67
|
+
# @raise [DataFrameArgumentError]
|
68
|
+
# if size is not same or self and other shares the same key.
|
48
69
|
# @return [DataFrame]
|
49
|
-
#
|
70
|
+
# merged dataframe.
|
71
|
+
# @example
|
72
|
+
# df = DataFrame.new(x: [1, 2], y: [3, 4])
|
73
|
+
# other = DataFrame.new(a: ['A', 'B'], b: ['C', 'D'])
|
74
|
+
# df.merge(other)
|
75
|
+
#
|
76
|
+
# # =>
|
77
|
+
# x y a b
|
78
|
+
# <uint8> <uint8> <string> <string>
|
79
|
+
# 0 1 3 A C
|
80
|
+
# 1 2 4 B D
|
81
|
+
#
|
50
82
|
def merge(*other)
|
51
83
|
case other
|
52
84
|
in [] | [nil] | [[]]
|
@@ -85,32 +117,105 @@ module RedAmber
|
|
85
117
|
|
86
118
|
# Mutating joins (#inner_join, #full_join, #left_join, #right_join)
|
87
119
|
|
88
|
-
# Join another DataFrame or Table, leaving only the matching records.
|
89
|
-
# - Same as `#join` with `type: :inner`
|
90
|
-
# - A kind of mutating join.
|
91
|
-
#
|
92
120
|
# @!macro join_before
|
93
121
|
# @param other [DataFrame, Arrow::Table]
|
94
122
|
# A DataFrame or a Table to be joined with self.
|
95
123
|
#
|
124
|
+
# @!macro join_dorce_order
|
125
|
+
# @param force_order [Boolean]
|
126
|
+
# wheather force order of the output always same.
|
127
|
+
# - This option is used in `:full_outer` and `:right_outer`.
|
128
|
+
# - If this option is true (by default) it will append index to the source
|
129
|
+
# and sort after joining. It will cause some degradation in performance.
|
130
|
+
#
|
96
131
|
# @!macro join_after
|
97
132
|
# @param suffix [#succ]
|
98
133
|
# a suffix to rename keys when key names conflict as a result of join.
|
99
134
|
# `suffix` must be responsible to `#succ`.
|
100
135
|
# @return [DataFrame]
|
101
|
-
#
|
136
|
+
# joined dataframe.
|
102
137
|
#
|
103
138
|
# @!macro join_key_in_array
|
104
139
|
# @param join_keys [String, Symbol, Array<String, Symbol>]
|
105
|
-
#
|
140
|
+
# a key or keys to match.
|
106
141
|
#
|
107
142
|
# @!macro join_key_in_hash
|
108
143
|
# @param join_key_pairs [Hash]
|
109
|
-
#
|
144
|
+
# pairs of a key name or key names to match in left and right.
|
110
145
|
# @option join_key_pairs [String, Symbol, Array<String, Symbol>] :left
|
111
|
-
#
|
146
|
+
# join keys in `self`.
|
112
147
|
# @option join_key_pairs [String, Symbol, Array<String, Symbol>] :right
|
113
|
-
#
|
148
|
+
# join keys in `other`.
|
149
|
+
#
|
150
|
+
# @!macro join_common_example_1
|
151
|
+
# @example
|
152
|
+
# df = DataFrame.new(KEY: %w[A B C], X1: [1, 2, 3])
|
153
|
+
#
|
154
|
+
# # =>
|
155
|
+
# KEY X1
|
156
|
+
# <string> <uint8>
|
157
|
+
# 0 A 1
|
158
|
+
# 1 B 2
|
159
|
+
# 2 C 3
|
160
|
+
#
|
161
|
+
# other = DataFrame.new(KEY: %w[A B D], X2: [true, false, nil])
|
162
|
+
#
|
163
|
+
# # =>
|
164
|
+
# KEY X2
|
165
|
+
# <string> <boolean>
|
166
|
+
# 0 A true
|
167
|
+
# 1 B false
|
168
|
+
# 2 D (nil)
|
169
|
+
#
|
170
|
+
# @!macro join_common_example_2
|
171
|
+
# @example
|
172
|
+
# df2 = DataFrame.new(KEY1: %w[A B C], X1: [1, 2, 3])
|
173
|
+
#
|
174
|
+
# # =>
|
175
|
+
# KEY1 X1
|
176
|
+
# <string> <uint8>
|
177
|
+
# 0 A 1
|
178
|
+
# 1 B 2
|
179
|
+
# 2 C 3
|
180
|
+
#
|
181
|
+
# other2 = DataFrame.new(KEY2: %w[A B D], X2: [true, false, nil])
|
182
|
+
#
|
183
|
+
# # =>
|
184
|
+
# KEY2 X2
|
185
|
+
# <string> <boolean>
|
186
|
+
# 0 A true
|
187
|
+
# 1 B false
|
188
|
+
# 2 D (nil)
|
189
|
+
#
|
190
|
+
# @!macro join_common_example_3
|
191
|
+
# @example
|
192
|
+
# df3 = DataFrame.new(
|
193
|
+
# KEY1: %w[A B C],
|
194
|
+
# KEY2: [1, 2, 3]
|
195
|
+
# )
|
196
|
+
#
|
197
|
+
# # =>
|
198
|
+
# KEY1 KEY2
|
199
|
+
# <string> <uint8>
|
200
|
+
# 0 A 1
|
201
|
+
# 1 B 2
|
202
|
+
# 2 C 3
|
203
|
+
#
|
204
|
+
# other3 = DataFrame.new(
|
205
|
+
# KEY1: %w[A B D],
|
206
|
+
# KEY2: [1, 4, 5]
|
207
|
+
# )
|
208
|
+
#
|
209
|
+
# # =>
|
210
|
+
# KEY1 KEY2
|
211
|
+
# <string> <uint8>
|
212
|
+
# 0 A 1
|
213
|
+
# 1 B 4
|
214
|
+
# 2 D 5
|
215
|
+
|
216
|
+
# Join another DataFrame or Table, leaving only the matching records.
|
217
|
+
# - Same as `#join` with `type: :inner`
|
218
|
+
# - A kind of mutating join.
|
114
219
|
#
|
115
220
|
# @overload inner_join(other, suffix: '.1')
|
116
221
|
# If `join_key` is not specified, common keys in self and other are used
|
@@ -118,18 +223,45 @@ module RedAmber
|
|
118
223
|
#
|
119
224
|
# @macro join_before
|
120
225
|
# @macro join_after
|
226
|
+
# @macro join_common_example_1
|
227
|
+
# @example without key (use implicit common key)
|
228
|
+
# df.inner_join(other)
|
229
|
+
#
|
230
|
+
# # =>
|
231
|
+
# KEY X1 X2
|
232
|
+
# <string> <uint8> <boolean>
|
233
|
+
# 0 A 1 true
|
234
|
+
# 1 B 2 false
|
121
235
|
#
|
122
236
|
# @overload inner_join(other, join_keys, suffix: '.1')
|
123
237
|
#
|
124
238
|
# @macro join_before
|
125
239
|
# @macro join_key_in_array
|
126
240
|
# @macro join_after
|
241
|
+
# @macro join_common_example_1
|
242
|
+
# @example with a key
|
243
|
+
# df.inner_join(other, :KEY)
|
244
|
+
#
|
245
|
+
# # =>
|
246
|
+
# KEY X1 X2
|
247
|
+
# <string> <uint8> <boolean>
|
248
|
+
# 0 A 1 true
|
249
|
+
# 1 B 2 false
|
127
250
|
#
|
128
251
|
# @overload inner_join(other, join_key_pairs, suffix: '.1')
|
129
252
|
#
|
130
253
|
# @macro join_before
|
131
254
|
# @macro join_key_in_hash
|
132
255
|
# @macro join_after
|
256
|
+
# @macro join_common_example_2
|
257
|
+
# @example with key pairs
|
258
|
+
# df2.inner_join(other2, { left: :KEY1, right: :KEY2 })
|
259
|
+
#
|
260
|
+
# # =>
|
261
|
+
# KEY1 X1 X2
|
262
|
+
# <string> <uint8> <boolean>
|
263
|
+
# 0 A 1 true
|
264
|
+
# 1 B 2 false
|
133
265
|
#
|
134
266
|
def inner_join(other, join_keys = nil, suffix: '.1')
|
135
267
|
join(other, join_keys, type: :inner, suffix: suffix)
|
@@ -139,27 +271,64 @@ module RedAmber
|
|
139
271
|
# - Same as `#join` with `type: :full_outer`
|
140
272
|
# - A kind of mutating join.
|
141
273
|
#
|
142
|
-
# @overload full_join(other, suffix: '.1')
|
274
|
+
# @overload full_join(other, suffix: '.1', force_order: true)
|
143
275
|
# If `join_key` is not specified, common keys in self and other are used
|
144
276
|
# (natural keys). Returns joined dataframe.
|
145
277
|
#
|
146
278
|
# @macro join_before
|
279
|
+
# @macro join_dorce_order
|
147
280
|
# @macro join_after
|
281
|
+
# @macro join_common_example_1
|
282
|
+
# @example without key (use implicit common key)
|
283
|
+
# df.full_join(other)
|
284
|
+
#
|
285
|
+
# # =>
|
286
|
+
# KEY X1 X2
|
287
|
+
# <string> <uint8> <boolean>
|
288
|
+
# 0 A 1 true
|
289
|
+
# 1 B 2 false
|
290
|
+
# 2 C 3 (nil)
|
291
|
+
# 3 D (nil) (nil)
|
148
292
|
#
|
149
|
-
# @overload full_join(other, join_keys, suffix: '.1')
|
293
|
+
# @overload full_join(other, join_keys, suffix: '.1', force_order: true)
|
150
294
|
#
|
151
295
|
# @macro join_before
|
152
296
|
# @macro join_key_in_array
|
297
|
+
# @macro join_dorce_order
|
153
298
|
# @macro join_after
|
299
|
+
# @macro join_common_example_1
|
300
|
+
# @example with a key
|
301
|
+
# df.full_join(other, :KEY)
|
154
302
|
#
|
155
|
-
#
|
303
|
+
# # =>
|
304
|
+
# KEY X1 X2
|
305
|
+
# <string> <uint8> <boolean>
|
306
|
+
# 0 A 1 true
|
307
|
+
# 1 B 2 false
|
308
|
+
# 2 C 3 (nil)
|
309
|
+
# 3 D (nil) (nil)
|
310
|
+
#
|
311
|
+
# @overload full_join(other, join_key_pairs, suffix: '.1', force_order: true)
|
156
312
|
#
|
157
313
|
# @macro join_before
|
158
314
|
# @macro join_key_in_hash
|
315
|
+
# @macro join_dorce_order
|
159
316
|
# @macro join_after
|
160
|
-
#
|
161
|
-
|
162
|
-
|
317
|
+
# @macro join_common_example_2
|
318
|
+
# @example with key pairs
|
319
|
+
# df2.full_join(other2, { left: :KEY1, right: :KEY2 })
|
320
|
+
#
|
321
|
+
# # =>
|
322
|
+
# KEY1 X1 X2
|
323
|
+
# <string> <uint8> <boolean>
|
324
|
+
# 0 A 1 true
|
325
|
+
# 1 B 2 false
|
326
|
+
# 2 C 3 (nil)
|
327
|
+
# 3 D (nil) (nil)
|
328
|
+
#
|
329
|
+
def full_join(other, join_keys = nil, suffix: '.1', force_order: true)
|
330
|
+
join(other, join_keys,
|
331
|
+
type: :full_outer, suffix: suffix, force_order: force_order)
|
163
332
|
end
|
164
333
|
|
165
334
|
alias_method :outer_join, :full_join
|
@@ -174,18 +343,48 @@ module RedAmber
|
|
174
343
|
#
|
175
344
|
# @macro join_before
|
176
345
|
# @macro join_after
|
346
|
+
# @macro join_common_example_1
|
347
|
+
# @example without key (use implicit common key)
|
348
|
+
# df.left_join(other)
|
349
|
+
#
|
350
|
+
# # =>
|
351
|
+
# KEY X1 X2
|
352
|
+
# <string> <uint8> <boolean>
|
353
|
+
# 0 A 1 true
|
354
|
+
# 1 B 2 false
|
355
|
+
# 2 C 3 (nil)
|
177
356
|
#
|
178
357
|
# @overload left_join(other, join_keys, suffix: '.1')
|
179
358
|
#
|
180
359
|
# @macro join_before
|
181
360
|
# @macro join_key_in_array
|
182
361
|
# @macro join_after
|
362
|
+
# @macro join_common_example_1
|
363
|
+
# @example with a key
|
364
|
+
# df.left_join(other, :KEY)
|
365
|
+
#
|
366
|
+
# # =>
|
367
|
+
# KEY X1 X2
|
368
|
+
# <string> <uint8> <boolean>
|
369
|
+
# 0 A 1 true
|
370
|
+
# 1 B 2 false
|
371
|
+
# 2 C 3 (nil)
|
183
372
|
#
|
184
373
|
# @overload left_join(other, join_key_pairs, suffix: '.1')
|
185
374
|
#
|
186
375
|
# @macro join_before
|
187
376
|
# @macro join_key_in_hash
|
188
377
|
# @macro join_after
|
378
|
+
# @macro join_common_example_2
|
379
|
+
# @example with key pairs
|
380
|
+
# df2.left_join(other2, { left: :KEY1, right: :KEY2 })
|
381
|
+
#
|
382
|
+
# # =>
|
383
|
+
# KEY1 X1 X2
|
384
|
+
# <string> <uint8> <boolean>
|
385
|
+
# 0 A 1 true
|
386
|
+
# 1 B 2 false
|
387
|
+
# 2 C 3 (nil)
|
189
388
|
#
|
190
389
|
def left_join(other, join_keys = nil, suffix: '.1')
|
191
390
|
join(other, join_keys, type: :left_outer, suffix: suffix)
|
@@ -195,27 +394,66 @@ module RedAmber
|
|
195
394
|
# - Same as `#join` with `type: :right_outer`
|
196
395
|
# - A kind of mutating join.
|
197
396
|
#
|
198
|
-
# @overload right_join(other, suffix: '.1')
|
397
|
+
# @overload right_join(other, suffix: '.1', force_order: true)
|
199
398
|
# If `join_key` is not specified, common keys in self and other are used
|
200
399
|
# (natural keys). Returns joined dataframe.
|
201
400
|
#
|
202
401
|
# @macro join_before
|
402
|
+
# @macro join_dorce_order
|
203
403
|
# @macro join_after
|
404
|
+
# @macro join_common_example_1
|
405
|
+
# @example without key (use implicit common key)
|
406
|
+
# df.right_join(other)
|
407
|
+
#
|
408
|
+
# # =>
|
409
|
+
# KEY X1 X2
|
410
|
+
# <string> <uint8> <boolean>
|
411
|
+
# 0 A 1 true
|
412
|
+
# 1 B 2 false
|
413
|
+
# 2 D (nil) (nil)
|
204
414
|
#
|
205
|
-
# @overload right_join(other, join_keys, suffix: '.1')
|
415
|
+
# @overload right_join(other, join_keys, suffix: '.1', force_order: true)
|
206
416
|
#
|
207
417
|
# @macro join_before
|
208
418
|
# @macro join_key_in_array
|
419
|
+
# @macro join_dorce_order
|
209
420
|
# @macro join_after
|
421
|
+
# @macro join_common_example_1
|
422
|
+
# @example with a key
|
423
|
+
# df.right_join(other, :KEY)
|
210
424
|
#
|
211
|
-
#
|
425
|
+
# # =>
|
426
|
+
# KEY X1 X2
|
427
|
+
# <string> <uint8> <boolean>
|
428
|
+
# 0 A 1 true
|
429
|
+
# 1 B 2 false
|
430
|
+
# 2 D (nil) (nil)
|
431
|
+
#
|
432
|
+
# @overload right_join(other, join_key_pairs, suffix: '.1', force_order: true)
|
212
433
|
#
|
213
434
|
# @macro join_before
|
214
435
|
# @macro join_key_in_hash
|
436
|
+
# @macro join_dorce_order
|
215
437
|
# @macro join_after
|
216
|
-
#
|
217
|
-
|
218
|
-
|
438
|
+
# @macro join_common_example_2
|
439
|
+
# @example with key pairs
|
440
|
+
# df2.right_join(other2, { left: :KEY1, right: :KEY2 })
|
441
|
+
#
|
442
|
+
# # =>
|
443
|
+
# KEY1 X1 X2
|
444
|
+
# <string> <uint8> <boolean>
|
445
|
+
# 0 A 1 true
|
446
|
+
# 1 B 2 false
|
447
|
+
# 2 D (nil) (nil)
|
448
|
+
#
|
449
|
+
def right_join(other, join_keys = nil, suffix: '.1', force_order: true)
|
450
|
+
join(
|
451
|
+
other,
|
452
|
+
join_keys,
|
453
|
+
type: :right_outer,
|
454
|
+
suffix: suffix,
|
455
|
+
force_order: force_order
|
456
|
+
)
|
219
457
|
end
|
220
458
|
|
221
459
|
# Filtering joins (#semi_join, #anti_join)
|
@@ -230,18 +468,45 @@ module RedAmber
|
|
230
468
|
#
|
231
469
|
# @macro join_before
|
232
470
|
# @macro join_after
|
471
|
+
# @macro join_common_example_1
|
472
|
+
# @example without key (use implicit common key)
|
473
|
+
# df.semi_join(other)
|
474
|
+
#
|
475
|
+
# # =>
|
476
|
+
# KEY X1
|
477
|
+
# <string> <uint8>
|
478
|
+
# 0 A 1
|
479
|
+
# 1 B 2
|
233
480
|
#
|
234
481
|
# @overload semi_join(other, join_keys, suffix: '.1')
|
235
482
|
#
|
236
483
|
# @macro join_before
|
237
484
|
# @macro join_key_in_array
|
238
485
|
# @macro join_after
|
486
|
+
# @macro join_common_example_1
|
487
|
+
# @example with a key
|
488
|
+
# df.semi_join(other, :KEY)
|
489
|
+
#
|
490
|
+
# # =>
|
491
|
+
# KEY X1
|
492
|
+
# <string> <uint8>
|
493
|
+
# 0 A 1
|
494
|
+
# 1 B 2
|
239
495
|
#
|
240
496
|
# @overload semi_join(other, join_key_pairs, suffix: '.1')
|
241
497
|
#
|
242
498
|
# @macro join_before
|
243
499
|
# @macro join_key_in_hash
|
244
500
|
# @macro join_after
|
501
|
+
# @macro join_common_example_2
|
502
|
+
# @example with key pairs
|
503
|
+
# df2.semi_join(other2, { left: :KEY1, right: :KEY2 })
|
504
|
+
#
|
505
|
+
# # =>
|
506
|
+
# KEY1 X1
|
507
|
+
# <string> <uint8>
|
508
|
+
# 0 A 1
|
509
|
+
# 1 B 2
|
245
510
|
#
|
246
511
|
def semi_join(other, join_keys = nil, suffix: '.1')
|
247
512
|
join(other, join_keys, type: :left_semi, suffix: suffix)
|
@@ -257,18 +522,42 @@ module RedAmber
|
|
257
522
|
#
|
258
523
|
# @macro join_before
|
259
524
|
# @macro join_after
|
525
|
+
# @macro join_common_example_1
|
526
|
+
# @example without key (use implicit common key)
|
527
|
+
# df.anti_join(other)
|
528
|
+
#
|
529
|
+
# # =>
|
530
|
+
# KEY X1
|
531
|
+
# <string> <uint8>
|
532
|
+
# 0 C 3
|
260
533
|
#
|
261
534
|
# @overload anti_join(other, join_keys, suffix: '.1')
|
262
535
|
#
|
263
536
|
# @macro join_before
|
264
537
|
# @macro join_key_in_array
|
265
538
|
# @macro join_after
|
539
|
+
# @macro join_common_example_1
|
540
|
+
# @example with a key
|
541
|
+
# df.anti_join(other, :KEY)
|
542
|
+
#
|
543
|
+
# # =>
|
544
|
+
# KEY X1
|
545
|
+
# <string> <uint8>
|
546
|
+
# 0 C 3
|
266
547
|
#
|
267
548
|
# @overload anti_join(other, join_key_pairs, suffix: '.1')
|
268
549
|
#
|
269
550
|
# @macro join_before
|
270
551
|
# @macro join_key_in_hash
|
271
552
|
# @macro join_after
|
553
|
+
# @macro join_common_example_2
|
554
|
+
# @example with key pairs
|
555
|
+
# df2.anti_join(other2, { left: :KEY1, right: :KEY2 })
|
556
|
+
#
|
557
|
+
# # =>
|
558
|
+
# KEY1 X1
|
559
|
+
# <string> <uint8>
|
560
|
+
# 0 C 3
|
272
561
|
#
|
273
562
|
def anti_join(other, join_keys = nil, suffix: '.1')
|
274
563
|
join(other, join_keys, type: :left_anti, suffix: suffix)
|
@@ -279,8 +568,11 @@ module RedAmber
|
|
279
568
|
# Check if set operation with self and other is possible.
|
280
569
|
#
|
281
570
|
# @macro join_before
|
282
|
-
#
|
283
|
-
#
|
571
|
+
# @return [Boolean]
|
572
|
+
# true if set operation is possible.
|
573
|
+
# @macro join_common_example_3
|
574
|
+
# @example
|
575
|
+
# df3.set_operable?(other3) # => true
|
284
576
|
#
|
285
577
|
def set_operable?(other) # rubocop:disable Naming/AccessorMethodName
|
286
578
|
keys == other.keys.map(&:to_sym)
|
@@ -291,8 +583,16 @@ module RedAmber
|
|
291
583
|
# - A kind of set operations.
|
292
584
|
#
|
293
585
|
# @macro join_before
|
586
|
+
# @return [DataFrame]
|
587
|
+
# joined dataframe.
|
588
|
+
# @macro join_common_example_3
|
589
|
+
# @example
|
590
|
+
# df3.intersect(other3)
|
294
591
|
#
|
295
|
-
#
|
592
|
+
# # =>
|
593
|
+
# KEY1 KEY2
|
594
|
+
# <string> <uint8>
|
595
|
+
# 0 A 1
|
296
596
|
#
|
297
597
|
def intersect(other)
|
298
598
|
unless keys == other.keys.map(&:to_sym)
|
@@ -307,8 +607,20 @@ module RedAmber
|
|
307
607
|
# - A kind of set operations.
|
308
608
|
#
|
309
609
|
# @macro join_before
|
310
|
-
#
|
311
|
-
#
|
610
|
+
# @return [DataFrame]
|
611
|
+
# joined dataframe.
|
612
|
+
# @macro join_common_example_3
|
613
|
+
# @example
|
614
|
+
# df3.intersect(other3)
|
615
|
+
#
|
616
|
+
# # =>
|
617
|
+
# KEY1 KEY2
|
618
|
+
# <string> <uint8>
|
619
|
+
# 0 A 1
|
620
|
+
# 1 B 2
|
621
|
+
# 2 C 3
|
622
|
+
# 3 B 4
|
623
|
+
# 4 D 5
|
312
624
|
#
|
313
625
|
def union(other)
|
314
626
|
unless keys == other.keys.map(&:to_sym)
|
@@ -323,8 +635,25 @@ module RedAmber
|
|
323
635
|
# - A kind of set operations.
|
324
636
|
#
|
325
637
|
# @macro join_before
|
638
|
+
# @return [DataFrame]
|
639
|
+
# joined dataframe.
|
640
|
+
# @macro join_common_example_3
|
641
|
+
# @example
|
642
|
+
# df3.intersect(other3)
|
643
|
+
#
|
644
|
+
# # =>
|
645
|
+
# KEY1 KEY2
|
646
|
+
# <string> <uint8>
|
647
|
+
# 0 B 2
|
648
|
+
# 1 C 3
|
326
649
|
#
|
327
|
-
#
|
650
|
+
# other.intersect(df)
|
651
|
+
#
|
652
|
+
# # =>
|
653
|
+
# KEY1 KEY2
|
654
|
+
# <string> <uint8>
|
655
|
+
# 0 B 4
|
656
|
+
# 1 D 5
|
328
657
|
#
|
329
658
|
def difference(other)
|
330
659
|
unless keys == other.keys.map(&:to_sym)
|
@@ -338,46 +667,153 @@ module RedAmber
|
|
338
667
|
|
339
668
|
# Join another DataFrame or Table to self.
|
340
669
|
#
|
341
|
-
#
|
670
|
+
# @!macro join_common_type
|
671
|
+
# @param type [:left_semi, :right_semi, :left_anti, :right_anti, :inner,
|
672
|
+
# left_outer, :right_outer, :full_outer] type of join.
|
673
|
+
#
|
674
|
+
# @!macro join_common_example_4
|
675
|
+
# @example
|
676
|
+
# df4 = DataFrame.new(
|
677
|
+
# X1: %w[A B C],
|
678
|
+
# Y: %w[D E F]
|
679
|
+
# )
|
680
|
+
#
|
681
|
+
# # =>
|
682
|
+
# X1 Y1
|
683
|
+
# <string> <string>
|
684
|
+
# 0 A D
|
685
|
+
# 1 B E
|
686
|
+
# 2 C F
|
687
|
+
#
|
688
|
+
# other4 = DataFrame.new(
|
689
|
+
# X2: %w[A B D],
|
690
|
+
# Y: %w[e E E]
|
691
|
+
# )
|
692
|
+
#
|
693
|
+
# # =>
|
694
|
+
# X1 Y1
|
695
|
+
# <string> <string>
|
696
|
+
# 0 A D
|
697
|
+
# 1 B E
|
698
|
+
# 2 C F
|
699
|
+
|
700
|
+
# @note the order of joined results will be preserved by default.
|
701
|
+
# This is enabled by appending index column to sort after joining but
|
702
|
+
# it will cause some performance degradation. If you don't matter
|
703
|
+
# the order of the result, set `force_order` option to `false`.
|
704
|
+
#
|
705
|
+
# @overload join(other, type: :inner, suffix: '.1', force_order: true)
|
342
706
|
#
|
343
707
|
# If `join_key` is not specified, common keys in self and other are used
|
344
708
|
# (natural keys). Returns joined dataframe.
|
345
709
|
#
|
346
|
-
# @!macro join_common_type
|
347
|
-
# @param type [:left_semi, :right_semi, :left_anti, :right_anti, :inner,
|
348
|
-
# left_outer, :right_outer, :full_outer] type of join.
|
349
|
-
#
|
350
710
|
# @macro join_before
|
351
711
|
# @macro join_common_type
|
712
|
+
# @macro join_dorce_order
|
352
713
|
# @macro join_after
|
714
|
+
# @macro join_common_example_1
|
715
|
+
# @example
|
716
|
+
# df.join(other)
|
353
717
|
#
|
354
|
-
#
|
718
|
+
# # =>
|
719
|
+
# KEY X1 X2
|
720
|
+
# <string> <uint8> <boolean>
|
721
|
+
# 0 A 1 true
|
722
|
+
# 1 B 2 false
|
723
|
+
#
|
724
|
+
# df.join(other, type: :full_outer)
|
725
|
+
#
|
726
|
+
# # =>
|
727
|
+
# KEY X1 X2
|
728
|
+
# <string> <uint8> <boolean>
|
729
|
+
# 0 A 1 true
|
730
|
+
# 1 B 2 false
|
731
|
+
# 2 C 3 (nil)
|
732
|
+
# 3 D (nil) (nil)
|
733
|
+
#
|
734
|
+
# @overload join(other, join_keys, type: :inner, suffix: '.1', force_order: true)
|
355
735
|
#
|
356
736
|
# @macro join_before
|
357
737
|
# @macro join_key_in_array
|
358
738
|
# @macro join_common_type
|
739
|
+
# @macro join_dorce_order
|
359
740
|
# @macro join_after
|
741
|
+
# @macro join_common_example_3
|
742
|
+
# @example join keys in an Array
|
743
|
+
# df3.join(other3, [:KEY1, :KEY2])
|
744
|
+
#
|
745
|
+
# # =>
|
746
|
+
# KEY1 KEY2
|
747
|
+
# <string> <uint8>
|
748
|
+
# 0 A 1
|
360
749
|
#
|
361
|
-
#
|
750
|
+
# @example partial join key and suffix
|
751
|
+
# df3.join(other3, :KEY1, suffix: '.a')
|
752
|
+
#
|
753
|
+
# # =>
|
754
|
+
# KEY1 KEY2 KEY2.a
|
755
|
+
# <string> <uint8> <uint8>
|
756
|
+
# 0 A 1 1
|
757
|
+
# 1 B 2 4
|
758
|
+
#
|
759
|
+
# @overload join(other, join_key_pairs, type: :inner, suffix: '.1', force_order: true)
|
362
760
|
#
|
363
761
|
# @macro join_before
|
364
762
|
# @macro join_key_in_hash
|
365
763
|
# @macro join_common_type
|
764
|
+
# @macro join_dorce_order
|
366
765
|
# @macro join_after
|
367
|
-
#
|
368
|
-
|
369
|
-
|
370
|
-
|
371
|
-
|
372
|
-
|
373
|
-
|
766
|
+
# @macro join_common_example_4
|
767
|
+
# @example without options
|
768
|
+
# df4.join(other4)
|
769
|
+
#
|
770
|
+
# # =>
|
771
|
+
# X1 Y X2
|
772
|
+
# <string> <string> <string>
|
773
|
+
# 0 B E D
|
774
|
+
# 1 B E B
|
775
|
+
#
|
776
|
+
# @example join by key pairs
|
777
|
+
# df4.join(other4, { left: [:X1, :Y], right: [:X2, :Y] })
|
778
|
+
#
|
779
|
+
# # =>
|
780
|
+
# X1 Y
|
781
|
+
# <string> <string>
|
782
|
+
# 0 B E
|
783
|
+
#
|
784
|
+
# @example join by key pairs, using renaming by suffix
|
785
|
+
# df4.join(other4, { left: :X1, right: :X2 })
|
786
|
+
#
|
787
|
+
# # =>
|
788
|
+
# X1 Y Y.1
|
789
|
+
# <string> <string> <string>
|
790
|
+
# 0 A D e
|
791
|
+
# 1 B E E
|
792
|
+
#
|
793
|
+
def join(other, join_keys = nil, type: :inner, suffix: '.1', force_order: true)
|
794
|
+
right_table =
|
795
|
+
case other
|
796
|
+
when DataFrame
|
797
|
+
other.table
|
798
|
+
when Arrow::Table
|
799
|
+
other
|
800
|
+
else
|
801
|
+
raise DataFrameArgumentError, 'other must be a DataFrame or an Arrow::Table'
|
802
|
+
end
|
803
|
+
|
804
|
+
type = type.to_sym
|
805
|
+
left_index = :__LEFT_INDEX__
|
806
|
+
right_index = :__RIGHT_INDEX__
|
807
|
+
if force_order && %i[full_outer right_outer].include?(type)
|
808
|
+
left_table = assign(left_index) { indices }.table
|
809
|
+
other = DataFrame.create(other) if other.is_a?(Arrow::Table)
|
810
|
+
right_table = other.assign(right_index) { indices }.table
|
374
811
|
else
|
375
|
-
|
812
|
+
left_table = table
|
376
813
|
end
|
377
814
|
|
378
|
-
table_keys =
|
379
|
-
other_keys =
|
380
|
-
type = type.to_sym
|
815
|
+
table_keys = left_table.keys
|
816
|
+
other_keys = right_table.keys
|
381
817
|
|
382
818
|
# natural keys (implicit common keys)
|
383
819
|
join_keys ||= table_keys.intersection(other_keys)
|
@@ -407,10 +843,13 @@ module RedAmber
|
|
407
843
|
|
408
844
|
# Should we rescue errors in Arrow::Table#join for usability ?
|
409
845
|
joined_table =
|
410
|
-
|
411
|
-
|
412
|
-
|
413
|
-
|
846
|
+
left_table.join(
|
847
|
+
right_table,
|
848
|
+
join_keys,
|
849
|
+
type: type,
|
850
|
+
left_outputs: left_outputs,
|
851
|
+
right_outputs: right_outputs
|
852
|
+
)
|
414
853
|
|
415
854
|
case type
|
416
855
|
when :inner, :left_outer, :left_semi, :left_anti, :right_semi, :right_anti
|
@@ -423,20 +862,31 @@ module RedAmber
|
|
423
862
|
renamed_table = rename_table(joined_table, n_keys, suffix)
|
424
863
|
renamed_keys = renamed_table.keys
|
425
864
|
dropper = []
|
426
|
-
DataFrame.create(renamed_table).assign do |df|
|
865
|
+
dataframe = DataFrame.create(renamed_table).assign do |df|
|
427
866
|
left_keys.map do |left_key|
|
428
867
|
i_left_key = renamed_keys.index(left_key)
|
429
868
|
right_key = renamed_keys[i_left_key + table_keys.size]
|
430
869
|
dropper << right_key
|
431
870
|
[left_key.to_sym, merge_array(df[left_key].data, df[right_key].data)]
|
432
871
|
end
|
433
|
-
end
|
872
|
+
end
|
873
|
+
dataframe = dataframe.sort(left_index, right_index) if force_order
|
874
|
+
|
875
|
+
dataframe.drop(dropper, left_index, right_index)
|
434
876
|
when :right_outer
|
435
|
-
|
436
|
-
|
437
|
-
|
438
|
-
|
439
|
-
|
877
|
+
dataframe =
|
878
|
+
if joined_table.keys.uniq!
|
879
|
+
DataFrame.create(rename_table(joined_table, left_outputs.size, suffix))
|
880
|
+
else
|
881
|
+
DataFrame.create(joined_table)
|
882
|
+
end
|
883
|
+
if force_order
|
884
|
+
dataframe =
|
885
|
+
dataframe
|
886
|
+
.sort(left_index, right_index)
|
887
|
+
.drop(left_index, right_index)
|
888
|
+
end
|
889
|
+
dataframe.pick do
|
440
890
|
[right_keys, keys.map(&:to_s) - right_keys]
|
441
891
|
end
|
442
892
|
end
|