-
Notifications
You must be signed in to change notification settings - Fork 55
/
macros.jl
1786 lines (1413 loc) · 43.5 KB
/
macros.jl
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
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
##############################################################################
##
## @col
##
##############################################################################
"""
@col(kw)
`@col` transforms an expression of the form `:z = :x + :y` into it's equivalent in
DataFrames's `source => fun => destination` syntax.
### Details
Parsing follows the same convention as other DataFramesMeta.jl macros, such as `@with`. All
terms in the expression that are `Symbol`s are treated as columns in the data frame, except
`Symbol`s wrapped in `^`. To use a variable representing a column name, wrap the variable
in `\$`.
`@col` constructs an anonymous function `fun` based on the given expression. It then creates
a `source => fun => destination` pair that is suitable for the `select`, `transform`, and
`combine` functions in DataFrames.jl.
### Examples
```julia
julia> @col :z = :x + :y
[:x, :y] => (##595 => :z)
```
In the above example, `##595` is an anonymous function equivalent to the following
```julia
(_x, _y) -> _x + _y
```
```jldoctest
julia> using DataFramesMeta;
julia> df = DataFrame(x = [1, 2], y = [3, 4]);
julia> import DataFramesMeta: @col;
julia> DataFrames.transform(df, @col :z = :x .* :y)
2×3 DataFrame
Row │ x y z
│ Int64 Int64 Int64
─────┼─────────────────────
1 │ 1 3 3
2 │ 2 4 8
```
"""
macro col(kw)
esc(fun_to_vec(kw))
end
##############################################################################
##
## @byrow
##
##############################################################################
"""
@byrow
Broadcast operations within DataFramesMeta.jl macros.
`@byrow` is not a "real" Julia macro but rather serves as a "flag"
to indicate that the anonymous function created by DataFramesMeta
to represent an operation should be applied "by-row".
If an expression starts with `@byrow`, either of the form `@byrow :y = f(:x)`
in transformations or `@byrow f(:x)` in `@orderby`, `@subset`, and `@with`,
then the anonymous function created by DataFramesMeta is wrapped in the
`DataFrames.ByRow` function wrapper, which broadcasts the function so that it run on each row.
### Examples
```julia
julia> df = DataFrame(a = [1, 2, 3, 4], b = [5, 6, 7, 8]);
julia> @transform(df, @byrow :c = :a * :b)
4×3 DataFrame
Row │ a b c
│ Int64 Int64 Int64
─────┼─────────────────────
1 │ 1 5 5
2 │ 2 6 12
3 │ 3 7 21
4 │ 4 8 32
julia> @subset(df, @byrow :a == 1 ? true : false)
1×2 DataFrame
Row │ a b
│ Int64 Int64
─────┼──────────────
1 │ 1 5
```
To avoid writing `@byrow` multiple times when performing multiple
operations, it is allowed to use`@byrow` at the beginning of a block of
operations. All transformations in the block will operate by row.
```julia
julia> @subset df @byrow begin
:a > 1
:b < 5
end
1×2 DataFrame
Row │ a b
│ Int64 Int64
─────┼──────────────
1 │ 2 4
```
### Comparison with `@eachrow`
To re-cap, the `@eachrow` macro roughly transforms
```julia
@eachrow df begin
:a * :b
end
```
to
```julia
begin
function tempfun(a, b)
for i in eachindex(a)
a[i] * b[i]
end
end
tempfun(df.a, df.b)
df
end
```
The function `*` is applied by-row. But the result of those operations
is not stored anywhere, as with `for`-loops in Base Julia.
Rather, `@eachrow` and `@eachrow!` return data frames.
Now consider `@byrow`. `@byrow` transforms
```julia
@with df @byrow begin
:a * :b
end
```
to
```julia
tempfun(a, b) = a * b
tempfun.(df.a, df.b)
```
In contrast to `@eachrow`, `@with` combined with `@byrow` returns a vector of the
broadcasted multiplication and not a data frame.
Additionally, transformations applied using `@eachrow!` modify the input
data frame. On the contrary, `@byrow` does not update columns.
```julia
julia> df = DataFrame(a = [1, 2], b = [3, 4]);
julia> @with df @byrow begin
:a = 500
end
2-element Vector{Int64}:
500
500
julia> df
2×2 DataFrame
Row │ a b
│ Int64 Int64
─────┼──────────────
1 │ 1 3
2 │ 2 4
```
### Comparison with `@.` and Base broadcasting
Base Julia provides the broadasting macro `@.` and in many cases `@.`
and `@byrow` will give equivalent results. But there are important
deviations in behavior. Consider the setup
```julia
df = DataFrame(a = [1, 2], b = [3, 4])
```
* Control flow. `@byrow` allows for operations of the form `if ... else`
and `a ? b : c` to be applied by row. These expressions cannot be
broadcasted in Base Julia. `@byrow` also allows for expressions of
the form `a && b` and `a || b` to be applied by row, something that
is not possible in Julia versions below 1.7.
```
julia> @with df @byrow begin
if :a == 1
5
else
10
end
end
2-element Vector{Int64}:
5
10
julia> @with df @. begin
if :a == 1
5
else
10
end
end # will error
```
* Broadcasting objects that are not columns. `@byrow` constructs an
anonymous function which accepts only the columns of the input data frame
and broadcasts that function. Consequently, it does not broadcast
referenced objects which are not columns.
```julia
julia> df = DataFrame(a = [1, 2], b = [3, 4]);
julia> @with df @byrow :x + [5, 6]
```
will error, because the `:x` in the above expression refers
to a scalar `Int`, and you cannot do `1 + [5, 6]`.
On the other hand
```julia
@with df @. :x + [5, 6]
```
will succeed, as `df.x` is a 2-element vector as is `[5, 6]`.
Because `ByRow` inside `transform` blocks does not internally
use broadcasting in all circumstances, in the rare instance
that a column in a data frame is a custom vector type that
implements custom broadcasting, this custom behavior will
not be called with `@byrow`.
* Broadcasting expensive calls. In Base Julia, broadcasting
evaluates calls first and then broadcasts the result. Because
`@byrow` constructs an anonymous function and evaluates
that function for every row in the data frame, expensive functions
will be evaluated many times.
```julia
julia> function expensive()
sleep(.5)
return 1
end;
julia> @time @with df @byrow :a + expensive();
1.037073 seconds (51.67 k allocations: 3.035 MiB, 3.19% compilation time)
julia> @time @with df :a .+ expensive();
0.539900 seconds (110.67 k allocations: 6.525 MiB, 7.05% compilation time)
```
This problem comes up when using the `@.` macro as well,
but can easily be fixed with `\$`.
```julia
julia> @time @with df @. :a + expensive();
1.036888 seconds (97.55 k allocations: 5.617 MiB, 3.20% compilation time)
julia> @time @with df @. :a + \$expensive();
0.537961 seconds (110.68 k allocations: 6.525 MiB, 6.73% compilation time)
```
No such solution currently exists with `@byrow`.
"""
macro byrow(args...)
throw(ArgumentError("@byrow is deprecated outside of DataFramesMeta macros."))
end
"""
passmissing(args...)
Propograte missing values inside DataFramesMeta.jl macros.
`@passmissing` is not a "real" Julia macro but rather serves as a "flag"
to indicate that the anonymous function created by DataFramesMeta.jl
to represent an operation should be wrapped in `passmissing` from Missings.jl.
`@passmissing` can only be combined with `@byrow` or the row-wise versions of macros
such as `@rtransform` and `@rselect`, etc. If any of the arguments passed
to the row-wise anonymous function created by DataFramesMeta.jl with `@byrow`, the
result will automatically be `missing`.
In the below example, `@transform` would throw an error without the `@passmissing`
flag.
`@passmissing` is especially useful for functions which operate on strings, such as
`parse`.
### Examples
```
julia> no_missing(x::Int, y::Int) = x + y;
julia> df = DataFrame(a = [1, 2, missing], b = [4, 5, 6])
3×2 DataFrame
Row │ a b
│ Int64? Int64
─────┼────────────────
1 │ 1 4
2 │ 2 5
3 │ missing 6
julia> @transform df @passmissing @byrow c = no_missing(:a, :b)
3×3 DataFrame
Row │ a b c
│ Int64? Int64 Int64?
─────┼─────────────────────────
1 │ 1 4 5
2 │ 2 5 7
3 │ missing 6 missing
julia> df = DataFrame(x_str = ["1", "2", missing])
3×1 DataFrame
Row │ x_str
│ String?
─────┼─────────
1 │ 1
2 │ 2
3 │ missing
julia> @rtransform df @passmissing x = parse(Int, :x_str)
3×2 DataFrame
Row │ x_str x
│ String? Int64?
─────┼──────────────────
1 │ 1 1
2 │ 2 2
3 │ missing missing
```
"""
macro passmissing(args...)
throw(ArgumentError("@passmissing only works inside DataFramesMeta macros."))
end
##############################################################################
##
## @with
##
##############################################################################
function exec(df, p::Pair)
cols = first(p)
fun = last(p)
fun(map(c -> DataFramesMeta.getsinglecolumn(df, c), cols)...)
end
exec(df, s::Union{Symbol, AbstractString}) = df[!, s]
getsinglecolumn(df, s::DataFrames.ColumnIndex) = df[!, s]
getsinglecolumn(df, s) = throw(ArgumentError("Only indexing with Symbols, strings and integers " *
"is currently allowed with \$"))
function with_helper(d, body)
# Make body an expression to force the
# complicated method of fun_to_vec
# in the case of QuoteNode
t = fun_to_vec(Expr(:block, body); no_dest=true)
:(DataFramesMeta.exec($d, $t))
end
"""
@with(d, expr)
`@with` allows DataFrame columns keys to be referenced as symbols.
### Arguments
* `d` : an AbstractDataFrame type
* `expr` : the expression to evaluate in `d`
### Details
`@with` works by parsing the expression body for all columns indicated
by symbols (e.g. `:colA`). Then, a function is created that wraps the
body and passes the columns as function arguments. This function is
then called. Operations are efficient because:
- A pseudo-anonymous function is defined, so types are stable.
- Columns are passed as references, eliminating DataFrame indexing.
The following
```julia
@with(d, :a .+ :b .+ 1)
```
becomes
```julia
tempfun(a, b) = a .+ b .+ 1
tempfun(d[!, :a], d[!, :b])
```
If an expression is wrapped in `^(expr)`, `expr` gets passed through untouched.
If an expression is wrapped in `\$(expr)`, the column is referenced by the
variable `expr` rather than a symbol.
If the expression provide to `@with` begins with `@byrow`, the function
created by the `@with` block is broadcasted along the columns of the
data frame.
### Examples
```jldoctest
julia> using DataFramesMeta
julia> y = 3;
julia> df = DataFrame(x = 1:3, y = [2, 1, 2]);
julia> x = [2, 1, 0];
julia> @with(df, :y .+ 1)
3-element Vector{Int64}:
3
2
3
julia> @with(df, :x + x)
3-element Vector{Int64}:
3
3
3
julia> @with df begin
res = 0.0
for i in 1:length(:x)
res += :x[i] * :y[i]
end
res
end
10.0
julia> @with(df, df[:x .> 1, ^(:y)]) # The ^ means leave the :y alone
2-element Vector{Int64}:
1
2
julia> colref = :x;
julia> @with(df, :y + \$colref) # Equivalent to df[!, :y] + df[!, colref]
3-element Vector{Int64}:
3
3
5
julia> @with df @byrow :x * :y
3-element Vector{Int64}:
2
2
6
```
!!! note
`@with` creates a function, so the scope within `@with` is a local scope.
Variables in the parent can be read. Writing to variables in the parent scope
differs depending on the type of scope of the parent. If the parent scope is a
global scope, then a variable cannot be assigned without using the `global` keyword.
If the parent scope is a local scope (inside a function or let block for example),
the `global` keyword is not needed to assign to that parent scope.
"""
macro with(d, body)
esc(with_helper(d, body))
end
##############################################################################
##
## @subset and subset! - select row subsets
##
##############################################################################
function subset_helper(x, args...)
exprs, outer_flags = create_args_vector(args...)
t = (fun_to_vec(ex; no_dest=true, outer_flags=outer_flags) for ex in exprs)
quote
$subset($x, $(t...); skipmissing=true)
end
end
function where_helper(x, args...)
exprs, outer_flags = create_args_vector(args...)
t = (fun_to_vec(ex; no_dest=true, outer_flags=outer_flags) for ex in exprs)
quote
$subset($x, $(t...); skipmissing=true)
end
end
"""
@subset(d, i...)
Select row subsets in `AbstractDataFrame`s and `GroupedDataFrame`s.
### Arguments
* `d` : an AbstractDataFrame or GroupedDataFrame
* `i...` : expression for selecting rows
Multiple `i` expressions are "and-ed" together.
If given a `GroupedDataFrame`, `@subset` applies transformations by
group, and returns a fresh `DataFrame` containing the rows
for which the generated values are all `true`.
Inputs to `@subset` can come in two formats: a `begin ... end` block, in which case each
line is a separate selector, or as multiple arguments.
For example the following two statements are equivalent:
```julia
@subset df begin
:x .> 1
:y .< 2
end
```
and
```
@subset(df, :x .> 1, :y .< 2)
```
!!! note
`@subset` treats `missing` values as `false` when filtering rows.
Unlike `DataFrames.subset` and other Boolean operations with
`missing`, `@subset` will *not* error on missing values, and
will only keep `true` values.
If an expression provided to `@subset` begins with `@byrow`, operations
are applied "by row" along the data frame. To avoid writing `@byrow` multiple
times, `@orderby` also allows `@byrow`to be placed at the beginning of a block of
operations. For example, the following two statements are equivalent.
```
@subset df @byrow begin
:x > 1
:y < 2
end
```
and
```
@subset df
@byrow :x > 1
@byrow :y < 2
end
```
### Examples
```jldoctest
julia> using DataFramesMeta, Statistics
julia> df = DataFrame(x = 1:3, y = [2, 1, 2]);
julia> globalvar = [2, 1, 0];
julia> @subset(df, :x .> 1)
2×2 DataFrame
Row │ x y
│ Int64 Int64
─────┼──────────────
1 │ 2 1
2 │ 3 2
julia> @subset(df, :x .> globalvar)
2×2 DataFrame
Row │ x y
│ Int64 Int64
─────┼──────────────
1 │ 2 1
2 │ 3 2
julia> @subset df begin
:x .> globalvar
:y .== 3
end
0×2 DataFrame
julia> df = DataFrame(n = 1:20, x = [3, 3, 3, 3, 1, 1, 1, 2, 1, 1,
2, 1, 1, 2, 2, 2, 3, 1, 1, 2]);
julia> g = groupby(df, :x);
julia> @subset(g, :n .> mean(:n))
8×2 DataFrame
Row │ n x
│ Int64 Int64
─────┼──────────────
1 │ 12 1
2 │ 13 1
3 │ 15 2
4 │ 16 2
5 │ 17 3
6 │ 18 1
7 │ 19 1
8 │ 20 2
julia> @subset g begin
:n .> mean(:n)
:n .< 20
end
7×2 DataFrame
Row │ n x
│ Int64 Int64
─────┼──────────────
1 │ 12 1
2 │ 13 1
3 │ 15 2
4 │ 16 2
5 │ 17 3
6 │ 18 1
7 │ 19 1
julia> df = DataFrame(a = [1, 2, missing], b = ["x", "y", missing]);
julia> @subset(df, :a .== 1)
1×2 DataFrame
Row │ a b
│ Int64? String?
─────┼─────────────────
1 │ 1 x
```
"""
macro subset(x, args...)
esc(subset_helper(x, args...))
end
function rsubset_helper(x, args...)
exprs, outer_flags = create_args_vector(args...; wrap_byrow=true)
t = (fun_to_vec(ex; no_dest=true, outer_flags=outer_flags) for ex in exprs)
quote
$subset($x, $(t...); skipmissing=true)
end
end
"""
@rsubset(d, i...)
Row-wise version of `@subset`, i.e. all operations use `@byrow` by
default. See [`@subset`](@ref) for details.
"""
macro rsubset(x, args...)
esc(rsubset_helper(x, args...))
end
"""
@subset(x, args...)
Deprecated version of `@subset`, see `?@subset` for details.
"""
macro where(x, args...)
@warn "`@where is deprecated, use `@subset` instead."
esc(where_helper(x, args...))
end
function subset!_helper(x, args...)
exprs, outer_flags = create_args_vector(args...)
t = (fun_to_vec(ex; no_dest=true, outer_flags=outer_flags) for ex in exprs)
quote
$subset!($x, $(t...); skipmissing=true)
end
end
function rsubset!_helper(x, args...)
exprs, outer_flags = create_args_vector(args...; wrap_byrow=true)
t = (fun_to_vec(ex; no_dest=true, outer_flags=outer_flags) for ex in exprs)
quote
$subset!($x, $(t...); skipmissing=true)
end
end
"""
@subset!(d, i...)
Select row subsets in `AbstractDataFrame`s and `GroupedDataFrame`s,
mutating the underlying data-frame in-place.
### Arguments
* `d` : an AbstractDataFrame or GroupedDataFrame
* `i...` : expression for selecting rows
Multiple `i` expressions are "and-ed" together.
If given a `GroupedDataFrame`, `@subset!` applies transformations by
group, and returns a fresh `DataFrame` containing the rows
for which the generated values are all `true`.
Inputs to `@subset!` can come in two formats: a `begin ... end` block, in which case each
line is a separate selector, or as multiple arguments.
For example the following two statements are equivalent:
```julia
@subset! df begin
:x .> 1
:y .< 2
end
```
and
```
@subset!(df, :x .> 1, :y .< 2)
```
!!! note
`@subset!` treats `missing` values as `false` when filtering rows.
Unlike `DataFrames.subset!` and other Boolean operations with
`missing`, `@subset!` will *not* error on missing values, and
will only keep `true` values.
If an expression provided to `@subset!` begins with `@byrow`, operations
are applied "by row" along the data frame. To avoid writing `@byrow` multiple
times, `@orderby` also allows `@byrow`to be placed at the beginning of a block of
operations. For example, the following two statements are equivalent.
```
@subset! df @byrow begin
:x > 1
:y < 2
end
```
and
```
@subset! df
@byrow :x > 1
@byrow :y < 2
end
```
### Examples
```jldoctest
julia> using DataFramesMeta, Statistics
julia> df = DataFrame(x = 1:3, y = [2, 1, 2]);
julia> globalvar = [2, 1, 0];
julia> @subset!(copy(df), :x .> 1)
2×2 DataFrame
Row │ x y
│ Int64 Int64
─────┼──────────────
1 │ 2 1
2 │ 3 2
julia> @subset!(copy(df), :x .> globalvar)
2×2 DataFrame
Row │ x y
│ Int64 Int64
─────┼──────────────
1 │ 2 1
2 │ 3 2
julia> @subset! copy(df) begin
:x .> globalvar
:y .== 3
end
0×2 DataFrame
julia> df = DataFrame(n = 1:20, x = [3, 3, 3, 3, 1, 1, 1, 2, 1, 1,
2, 1, 1, 2, 2, 2, 3, 1, 1, 2]);
julia> g = groupby(copy(df), :x);
julia> @subset!(g, :n .> mean(:n))
8×2 DataFrame
Row │ n x
│ Int64 Int64
─────┼──────────────
1 │ 12 1
2 │ 13 1
3 │ 15 2
4 │ 16 2
5 │ 17 3
6 │ 18 1
7 │ 19 1
8 │ 20 2
julia> g = groupby(copy(df), :x);
julia> @subset! g begin
:n .> mean(:n)
:n .< 20
end
7×2 DataFrame
Row │ n x
│ Int64 Int64
─────┼──────────────
1 │ 12 1
2 │ 13 1
3 │ 15 2
4 │ 16 2
5 │ 17 3
6 │ 18 1
7 │ 19 1
julia> d = DataFrame(a = [1, 2, missing], b = ["x", "y", missing]);
julia> @subset!(d, :a .== 1)
1×2 DataFrame
Row │ a b
│ Int64? String?
─────┼─────────────────
1 │ 1 x
```
"""
macro subset!(x, args...)
esc(subset!_helper(x, args...))
end
"""
@rsubset!(d, i...)
Row-wise version of `@subset!`, i.e. all operations use `@byrow` by
default. See [`@subset!`](@ref) for details.
"""
macro rsubset!(x, args...)
esc(rsubset!_helper(x, args...))
end
##############################################################################
##
## @orderby
##
##############################################################################
function orderby_helper(x, args...)
exprs, outer_flags = create_args_vector(args...)
t = (fun_to_vec(ex; gensym_names = true, outer_flags = outer_flags) for ex in exprs)
quote
$DataFramesMeta.orderby($x, $(t...))
end
end
function orderby(x::AbstractDataFrame, @nospecialize(args...))
t = DataFrames.select(x, args...; copycols = false)
x[sortperm(t), :]
end
function orderby(x::GroupedDataFrame, @nospecialize(args...))
throw(ArgumentError("@orderby with a GroupedDataFrame is reserved"))
end
function orderby(x::SubDataFrame, @nospecialize(args...))
t = DataFrames.select(x, args...)
x[sortperm(t), :]
end
"""
@orderby(d, i...)
Sort rows by values in one of several columns or a transformation of columns.
Always returns a fresh `DataFrame`. Does not accept a `GroupedDataFrame`.
When given a `DataFrame`, `@orderby` applies the transformation
given by its arguments (but does not create new columns) and sorts
the given `DataFrame` on the result, returning a new `DataFrame`.
Inputs to `@orderby` can come in two formats: a `begin ... end` block, in which case each
line in the block is a separate ordering operation, and as mulitple
arguments. For example, the following two statements are equivalent:
```julia
@orderby df begin
:x
-:y
end
```
and
```
@orderby(df, :x, -:y)
```
### Arguments
* `d` : an AbstractDataFrame
* `i...` : expression for sorting
If an expression provided to `@orderby` begins with `@byrow`, operations
are applied "by row" along the data frame. To avoid writing `@byrow` multiple
times, `@orderby` also allows `@byrow`to be placed at the beginning of a block of
operations. For example, the following two statements are equivalent.
```
@orderby df @byrow begin
:x^2
:x^3
end
```
and
```
@orderby df
@byrow :x^2
@byrow :x^3
end
```
### Examples
```jldoctest
julia> using DataFramesMeta, Statistics
julia> d = DataFrame(x = [3, 3, 3, 2, 1, 1, 1, 2, 1, 1], n = 1:10,
c = ["a", "b", "c", "d", "e", "f", "g", "h", "i", "j"]);
julia> @orderby(d, -1 .* :n)
10×3 DataFrame
Row │ x n c
│ Int64 Int64 String
─────┼──────────────────────
1 │ 1 10 j
2 │ 1 9 i
3 │ 2 8 h
4 │ 1 7 g
5 │ 1 6 f
6 │ 1 5 e
7 │ 2 4 d
8 │ 3 3 c
9 │ 3 2 b
10 │ 3 1 a
julia> @orderby(d, sortperm(:c, rev = true))
10×3 DataFrame
Row │ x n c
│ Int64 Int64 String
─────┼──────────────────────
1 │ 1 10 j
2 │ 1 9 i
3 │ 2 8 h
4 │ 1 7 g
5 │ 1 6 f
6 │ 1 5 e
7 │ 2 4 d
8 │ 3 3 c
9 │ 3 2 b
10 │ 3 1 a
julia> @orderby d begin
:x
abs.(:n .- mean(:n))
end
10×3 DataFrame
Row │ x n c
│ Int64 Int64 String
─────┼──────────────────────
1 │ 1 5 e
2 │ 1 6 f
3 │ 1 7 g
4 │ 1 9 i
5 │ 1 10 j
6 │ 2 4 d
7 │ 2 8 h
8 │ 3 3 c
9 │ 3 2 b
10 │ 3 1 a
julia> @orderby d @byrow :x^2
10×3 DataFrame
Row │ x n c
│ Int64 Int64 String
─────┼──────────────────────
1 │ 1 5 e
2 │ 1 6 f
3 │ 1 7 g
4 │ 1 9 i
5 │ 1 10 j
6 │ 2 4 d