line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
1
|
|
|
|
|
|
|
# it wouldn't hurt to add a new line for such a configuration. You may distribute under the terms of either the GNU General Public License |
2
|
|
|
|
|
|
|
# or the Artistic License (the same terms as Perl itself) |
3
|
|
|
|
|
|
|
# |
4
|
|
|
|
|
|
|
# (C) Paul Evans, 2009-2018 -- leonerd@leonerd.org.uk |
5
|
|
|
|
|
|
|
|
6
|
|
|
|
|
|
|
package List::UtilsBy; |
7
|
|
|
|
|
|
|
|
8
|
12
|
|
|
12
|
|
610791
|
use strict; |
|
12
|
|
|
|
|
98
|
|
|
12
|
|
|
|
|
304
|
|
9
|
12
|
|
|
12
|
|
53
|
use warnings; |
|
12
|
|
|
|
|
17
|
|
|
12
|
|
|
|
|
376
|
|
10
|
|
|
|
|
|
|
|
11
|
|
|
|
|
|
|
our $VERSION = '0.12'; |
12
|
|
|
|
|
|
|
|
13
|
12
|
|
|
12
|
|
53
|
use Exporter 'import'; |
|
12
|
|
|
|
|
17
|
|
|
12
|
|
|
|
|
17949
|
|
14
|
|
|
|
|
|
|
|
15
|
|
|
|
|
|
|
our @EXPORT_OK = qw( |
16
|
|
|
|
|
|
|
sort_by |
17
|
|
|
|
|
|
|
nsort_by |
18
|
|
|
|
|
|
|
rev_sort_by |
19
|
|
|
|
|
|
|
rev_nsort_by |
20
|
|
|
|
|
|
|
|
21
|
|
|
|
|
|
|
max_by nmax_by |
22
|
|
|
|
|
|
|
min_by nmin_by |
23
|
|
|
|
|
|
|
minmax_by nminmax_by |
24
|
|
|
|
|
|
|
|
25
|
|
|
|
|
|
|
uniq_by |
26
|
|
|
|
|
|
|
|
27
|
|
|
|
|
|
|
partition_by |
28
|
|
|
|
|
|
|
count_by |
29
|
|
|
|
|
|
|
|
30
|
|
|
|
|
|
|
zip_by |
31
|
|
|
|
|
|
|
unzip_by |
32
|
|
|
|
|
|
|
|
33
|
|
|
|
|
|
|
extract_by |
34
|
|
|
|
|
|
|
extract_first_by |
35
|
|
|
|
|
|
|
|
36
|
|
|
|
|
|
|
weighted_shuffle_by |
37
|
|
|
|
|
|
|
|
38
|
|
|
|
|
|
|
bundle_by |
39
|
|
|
|
|
|
|
); |
40
|
|
|
|
|
|
|
|
41
|
|
|
|
|
|
|
=head1 NAME |
42
|
|
|
|
|
|
|
|
43
|
|
|
|
|
|
|
C - higher-order list utility functions |
44
|
|
|
|
|
|
|
|
45
|
|
|
|
|
|
|
=head1 SYNOPSIS |
46
|
|
|
|
|
|
|
|
47
|
|
|
|
|
|
|
use List::UtilsBy qw( nsort_by min_by ); |
48
|
|
|
|
|
|
|
|
49
|
|
|
|
|
|
|
use File::stat qw( stat ); |
50
|
|
|
|
|
|
|
my @files_by_age = nsort_by { stat($_)->mtime } @files; |
51
|
|
|
|
|
|
|
|
52
|
|
|
|
|
|
|
my $shortest_name = min_by { length } @names; |
53
|
|
|
|
|
|
|
|
54
|
|
|
|
|
|
|
=head1 DESCRIPTION |
55
|
|
|
|
|
|
|
|
56
|
|
|
|
|
|
|
This module provides a number of list utility functions, all of which take an |
57
|
|
|
|
|
|
|
initial code block to control their behaviour. They are variations on similar |
58
|
|
|
|
|
|
|
core perl or C functions of similar names, but which use the block |
59
|
|
|
|
|
|
|
to control their behaviour. For example, the core Perl function C takes |
60
|
|
|
|
|
|
|
a list of values and returns them, sorted into order by their string value. |
61
|
|
|
|
|
|
|
The L function sorts them according to the string value returned by |
62
|
|
|
|
|
|
|
the extra function, when given each value. |
63
|
|
|
|
|
|
|
|
64
|
|
|
|
|
|
|
my @names_sorted = sort @names; |
65
|
|
|
|
|
|
|
|
66
|
|
|
|
|
|
|
my @people_sorted = sort_by { $_->name } @people; |
67
|
|
|
|
|
|
|
|
68
|
|
|
|
|
|
|
=cut |
69
|
|
|
|
|
|
|
|
70
|
|
|
|
|
|
|
=head1 FUNCTIONS |
71
|
|
|
|
|
|
|
|
72
|
|
|
|
|
|
|
All functions added since version 0.04 unless otherwise stated, as the |
73
|
|
|
|
|
|
|
original names for earlier versions were renamed. |
74
|
|
|
|
|
|
|
|
75
|
|
|
|
|
|
|
=cut |
76
|
|
|
|
|
|
|
|
77
|
|
|
|
|
|
|
=head2 sort_by |
78
|
|
|
|
|
|
|
|
79
|
|
|
|
|
|
|
@vals = sort_by { KEYFUNC } @vals |
80
|
|
|
|
|
|
|
|
81
|
|
|
|
|
|
|
Returns the list of values sorted according to the string values returned by |
82
|
|
|
|
|
|
|
the C block or function. A typical use of this may be to sort objects |
83
|
|
|
|
|
|
|
according to the string value of some accessor, such as |
84
|
|
|
|
|
|
|
|
85
|
|
|
|
|
|
|
sort_by { $_->name } @people |
86
|
|
|
|
|
|
|
|
87
|
|
|
|
|
|
|
The key function is called in scalar context, being passed each value in turn |
88
|
|
|
|
|
|
|
as both C<$_> and the only argument in the parameters, C<@_>. The values are |
89
|
|
|
|
|
|
|
then sorted according to string comparisons on the values returned. |
90
|
|
|
|
|
|
|
|
91
|
|
|
|
|
|
|
This is equivalent to |
92
|
|
|
|
|
|
|
|
93
|
|
|
|
|
|
|
sort { $a->name cmp $b->name } @people |
94
|
|
|
|
|
|
|
|
95
|
|
|
|
|
|
|
except that it guarantees the C accessor will be executed only once per |
96
|
|
|
|
|
|
|
value. |
97
|
|
|
|
|
|
|
|
98
|
|
|
|
|
|
|
One interesting use-case is to sort strings which may have numbers embedded in |
99
|
|
|
|
|
|
|
them "naturally", rather than lexically. |
100
|
|
|
|
|
|
|
|
101
|
|
|
|
|
|
|
sort_by { s/(\d+)/sprintf "%09d", $1/eg; $_ } @strings |
102
|
|
|
|
|
|
|
|
103
|
|
|
|
|
|
|
This sorts strings by generating sort keys which zero-pad the embedded numbers |
104
|
|
|
|
|
|
|
to some level (9 digits in this case), helping to ensure the lexical sort puts |
105
|
|
|
|
|
|
|
them in the correct order. |
106
|
|
|
|
|
|
|
|
107
|
|
|
|
|
|
|
=cut |
108
|
|
|
|
|
|
|
|
109
|
|
|
|
|
|
|
sub sort_by(&@) |
110
|
|
|
|
|
|
|
{ |
111
|
7
|
|
|
7
|
1
|
77
|
my $keygen = shift; |
112
|
|
|
|
|
|
|
|
113
|
7
|
|
|
|
|
11
|
my @keys = map { local $_ = $_; scalar $keygen->( $_ ) } @_; |
|
10
|
|
|
|
|
25
|
|
|
10
|
|
|
|
|
13
|
|
114
|
7
|
|
|
|
|
43
|
return @_[ sort { $keys[$a] cmp $keys[$b] } 0 .. $#_ ]; |
|
4
|
|
|
|
|
21
|
|
115
|
|
|
|
|
|
|
} |
116
|
|
|
|
|
|
|
|
117
|
|
|
|
|
|
|
=head2 nsort_by |
118
|
|
|
|
|
|
|
|
119
|
|
|
|
|
|
|
@vals = nsort_by { KEYFUNC } @vals |
120
|
|
|
|
|
|
|
|
121
|
|
|
|
|
|
|
Similar to L but compares its key values numerically. |
122
|
|
|
|
|
|
|
|
123
|
|
|
|
|
|
|
=cut |
124
|
|
|
|
|
|
|
|
125
|
|
|
|
|
|
|
sub nsort_by(&@) |
126
|
|
|
|
|
|
|
{ |
127
|
8
|
|
|
8
|
1
|
80
|
my $keygen = shift; |
128
|
|
|
|
|
|
|
|
129
|
8
|
|
|
|
|
16
|
my @keys = map { local $_ = $_; scalar $keygen->( $_ ) } @_; |
|
14
|
|
|
|
|
59
|
|
|
14
|
|
|
|
|
25
|
|
130
|
8
|
|
|
|
|
46
|
return @_[ sort { $keys[$a] <=> $keys[$b] } 0 .. $#_ ]; |
|
9
|
|
|
|
|
34
|
|
131
|
|
|
|
|
|
|
} |
132
|
|
|
|
|
|
|
|
133
|
|
|
|
|
|
|
=head2 rev_sort_by |
134
|
|
|
|
|
|
|
|
135
|
|
|
|
|
|
|
=head2 rev_nsort_by |
136
|
|
|
|
|
|
|
|
137
|
|
|
|
|
|
|
@vals = rev_sort_by { KEYFUNC } @vals |
138
|
|
|
|
|
|
|
|
139
|
|
|
|
|
|
|
@vals = rev_nsort_by { KEYFUNC } @vals |
140
|
|
|
|
|
|
|
|
141
|
|
|
|
|
|
|
I |
142
|
|
|
|
|
|
|
|
143
|
|
|
|
|
|
|
Similar to L and L but returns the list in the reverse |
144
|
|
|
|
|
|
|
order. Equivalent to |
145
|
|
|
|
|
|
|
|
146
|
|
|
|
|
|
|
@vals = reverse sort_by { KEYFUNC } @vals |
147
|
|
|
|
|
|
|
|
148
|
|
|
|
|
|
|
except that these functions are slightly more efficient because they avoid |
149
|
|
|
|
|
|
|
the final C operation. |
150
|
|
|
|
|
|
|
|
151
|
|
|
|
|
|
|
=cut |
152
|
|
|
|
|
|
|
|
153
|
|
|
|
|
|
|
sub rev_sort_by(&@) |
154
|
|
|
|
|
|
|
{ |
155
|
1
|
|
|
1
|
1
|
2
|
my $keygen = shift; |
156
|
|
|
|
|
|
|
|
157
|
1
|
|
|
|
|
3
|
my @keys = map { local $_ = $_; scalar $keygen->( $_ ) } @_; |
|
2
|
|
|
|
|
4
|
|
|
2
|
|
|
|
|
4
|
|
158
|
1
|
|
|
|
|
5
|
return @_[ sort { $keys[$b] cmp $keys[$a] } 0 .. $#_ ]; |
|
1
|
|
|
|
|
6
|
|
159
|
|
|
|
|
|
|
} |
160
|
|
|
|
|
|
|
|
161
|
|
|
|
|
|
|
sub rev_nsort_by(&@) |
162
|
|
|
|
|
|
|
{ |
163
|
1
|
|
|
1
|
1
|
2
|
my $keygen = shift; |
164
|
|
|
|
|
|
|
|
165
|
1
|
|
|
|
|
2
|
my @keys = map { local $_ = $_; scalar $keygen->( $_ ) } @_; |
|
3
|
|
|
|
|
7
|
|
|
3
|
|
|
|
|
5
|
|
166
|
1
|
|
|
|
|
5
|
return @_[ sort { $keys[$b] <=> $keys[$a] } 0 .. $#_ ]; |
|
3
|
|
|
|
|
8
|
|
167
|
|
|
|
|
|
|
} |
168
|
|
|
|
|
|
|
|
169
|
|
|
|
|
|
|
=head2 max_by |
170
|
|
|
|
|
|
|
|
171
|
|
|
|
|
|
|
$optimal = max_by { KEYFUNC } @vals |
172
|
|
|
|
|
|
|
|
173
|
|
|
|
|
|
|
@optimal = max_by { KEYFUNC } @vals |
174
|
|
|
|
|
|
|
|
175
|
|
|
|
|
|
|
Returns the (first) value from C<@vals> that gives the numerically largest |
176
|
|
|
|
|
|
|
result from the key function. |
177
|
|
|
|
|
|
|
|
178
|
|
|
|
|
|
|
my $tallest = max_by { $_->height } @people |
179
|
|
|
|
|
|
|
|
180
|
|
|
|
|
|
|
use File::stat qw( stat ); |
181
|
|
|
|
|
|
|
my $newest = max_by { stat($_)->mtime } @files; |
182
|
|
|
|
|
|
|
|
183
|
|
|
|
|
|
|
In scalar context, the first maximal value is returned. In list context, a |
184
|
|
|
|
|
|
|
list of all the maximal values is returned. This may be used to obtain |
185
|
|
|
|
|
|
|
positions other than the first, if order is significant. |
186
|
|
|
|
|
|
|
|
187
|
|
|
|
|
|
|
If called on an empty list, an empty list is returned. |
188
|
|
|
|
|
|
|
|
189
|
|
|
|
|
|
|
For symmetry with the L function, this is also provided under the |
190
|
|
|
|
|
|
|
name C since it behaves numerically. |
191
|
|
|
|
|
|
|
|
192
|
|
|
|
|
|
|
=cut |
193
|
|
|
|
|
|
|
|
194
|
|
|
|
|
|
|
sub max_by(&@) |
195
|
|
|
|
|
|
|
{ |
196
|
9
|
|
|
9
|
1
|
96
|
my $code = shift; |
197
|
|
|
|
|
|
|
|
198
|
9
|
100
|
|
|
|
23
|
return unless @_; |
199
|
|
|
|
|
|
|
|
200
|
8
|
|
|
|
|
10
|
local $_; |
201
|
|
|
|
|
|
|
|
202
|
8
|
|
|
|
|
16
|
my @maximal = $_ = shift @_; |
203
|
8
|
|
|
|
|
13
|
my $max = $code->( $_ ); |
204
|
|
|
|
|
|
|
|
205
|
8
|
|
|
|
|
26
|
foreach ( @_ ) { |
206
|
11
|
|
|
|
|
15
|
my $this = $code->( $_ ); |
207
|
11
|
100
|
100
|
|
|
41
|
if( $this > $max ) { |
|
|
100
|
|
|
|
|
|
208
|
6
|
|
|
|
|
10
|
@maximal = $_; |
209
|
6
|
|
|
|
|
9
|
$max = $this; |
210
|
|
|
|
|
|
|
} |
211
|
|
|
|
|
|
|
elsif( wantarray and $this == $max ) { |
212
|
1
|
|
|
|
|
2
|
push @maximal, $_; |
213
|
|
|
|
|
|
|
} |
214
|
|
|
|
|
|
|
} |
215
|
|
|
|
|
|
|
|
216
|
8
|
100
|
|
|
|
32
|
return wantarray ? @maximal : $maximal[0]; |
217
|
|
|
|
|
|
|
} |
218
|
|
|
|
|
|
|
|
219
|
|
|
|
|
|
|
*nmax_by = \&max_by; |
220
|
|
|
|
|
|
|
|
221
|
|
|
|
|
|
|
=head2 min_by |
222
|
|
|
|
|
|
|
|
223
|
|
|
|
|
|
|
$optimal = min_by { KEYFUNC } @vals |
224
|
|
|
|
|
|
|
|
225
|
|
|
|
|
|
|
@optimal = min_by { KEYFUNC } @vals |
226
|
|
|
|
|
|
|
|
227
|
|
|
|
|
|
|
Similar to L but returns values which give the numerically smallest |
228
|
|
|
|
|
|
|
result from the key function. Also provided as C |
229
|
|
|
|
|
|
|
|
230
|
|
|
|
|
|
|
=cut |
231
|
|
|
|
|
|
|
|
232
|
|
|
|
|
|
|
sub min_by(&@) |
233
|
|
|
|
|
|
|
{ |
234
|
9
|
|
|
9
|
1
|
449
|
my $code = shift; |
235
|
|
|
|
|
|
|
|
236
|
9
|
100
|
|
|
|
29
|
return unless @_; |
237
|
|
|
|
|
|
|
|
238
|
8
|
|
|
|
|
11
|
local $_; |
239
|
|
|
|
|
|
|
|
240
|
8
|
|
|
|
|
15
|
my @minimal = $_ = shift @_; |
241
|
8
|
|
|
|
|
15
|
my $min = $code->( $_ ); |
242
|
|
|
|
|
|
|
|
243
|
8
|
|
|
|
|
24
|
foreach ( @_ ) { |
244
|
12
|
|
|
|
|
16
|
my $this = $code->( $_ ); |
245
|
12
|
50
|
100
|
|
|
47
|
if( $this < $min ) { |
|
|
100
|
|
|
|
|
|
246
|
0
|
|
|
|
|
0
|
@minimal = $_; |
247
|
0
|
|
|
|
|
0
|
$min = $this; |
248
|
|
|
|
|
|
|
} |
249
|
|
|
|
|
|
|
elsif( wantarray and $this == $min ) { |
250
|
1
|
|
|
|
|
3
|
push @minimal, $_; |
251
|
|
|
|
|
|
|
} |
252
|
|
|
|
|
|
|
} |
253
|
|
|
|
|
|
|
|
254
|
8
|
100
|
|
|
|
32
|
return wantarray ? @minimal : $minimal[0]; |
255
|
|
|
|
|
|
|
} |
256
|
|
|
|
|
|
|
|
257
|
|
|
|
|
|
|
*nmin_by = \&min_by; |
258
|
|
|
|
|
|
|
|
259
|
|
|
|
|
|
|
=head2 minmax_by |
260
|
|
|
|
|
|
|
|
261
|
|
|
|
|
|
|
( $minimal, $maximal ) = minmax_by { KEYFUNC } @vals |
262
|
|
|
|
|
|
|
|
263
|
|
|
|
|
|
|
I |
264
|
|
|
|
|
|
|
|
265
|
|
|
|
|
|
|
Similar to calling both L and L with the same key function |
266
|
|
|
|
|
|
|
on the same list. This version is more efficient than calling the two other |
267
|
|
|
|
|
|
|
functions individually, as it has less work to perform overall. In the case of |
268
|
|
|
|
|
|
|
ties, only the first optimal element found in each case is returned. Also |
269
|
|
|
|
|
|
|
provided as C. |
270
|
|
|
|
|
|
|
|
271
|
|
|
|
|
|
|
=cut |
272
|
|
|
|
|
|
|
|
273
|
|
|
|
|
|
|
sub minmax_by(&@) |
274
|
|
|
|
|
|
|
{ |
275
|
7
|
|
|
7
|
1
|
461
|
my $code = shift; |
276
|
|
|
|
|
|
|
|
277
|
7
|
100
|
|
|
|
19
|
return unless @_; |
278
|
|
|
|
|
|
|
|
279
|
6
|
|
|
|
|
9
|
my $minimal = $_ = shift @_; |
280
|
6
|
|
|
|
|
11
|
my $min = $code->( $_ ); |
281
|
|
|
|
|
|
|
|
282
|
6
|
100
|
|
|
|
24
|
return ( $minimal, $minimal ) unless @_; |
283
|
|
|
|
|
|
|
|
284
|
4
|
|
|
|
|
7
|
my $maximal = $_ = shift @_; |
285
|
4
|
|
|
|
|
6
|
my $max = $code->( $_ ); |
286
|
|
|
|
|
|
|
|
287
|
4
|
100
|
|
|
|
13
|
if( $max < $min ) { |
288
|
1
|
|
|
|
|
5
|
( $maximal, $minimal ) = ( $minimal, $maximal ); |
289
|
1
|
|
|
|
|
2
|
( $max, $min ) = ( $min, $max ); |
290
|
|
|
|
|
|
|
} |
291
|
|
|
|
|
|
|
|
292
|
|
|
|
|
|
|
# Minmax algorithm is faster than naïve min + max individually because it |
293
|
|
|
|
|
|
|
# takes pairs of values |
294
|
4
|
|
|
|
|
7
|
while( @_ ) { |
295
|
7
|
|
|
|
|
10
|
my $try_minimal = $_ = shift @_; |
296
|
7
|
|
|
|
|
11
|
my $try_min = $code->( $_ ); |
297
|
|
|
|
|
|
|
|
298
|
7
|
|
|
|
|
14
|
my $try_maximal = $try_minimal; |
299
|
7
|
|
|
|
|
8
|
my $try_max = $try_min; |
300
|
7
|
100
|
|
|
|
10
|
if( @_ ) { |
301
|
3
|
|
|
|
|
4
|
$try_maximal = $_ = shift @_; |
302
|
3
|
|
|
|
|
6
|
$try_max = $code->( $_ ); |
303
|
|
|
|
|
|
|
|
304
|
3
|
100
|
|
|
|
9
|
if( $try_max < $try_min ) { |
305
|
1
|
|
|
|
|
3
|
( $try_minimal, $try_maximal ) = ( $try_maximal, $try_minimal ); |
306
|
1
|
|
|
|
|
1
|
( $try_min, $try_max ) = ( $try_max, $try_min ); |
307
|
|
|
|
|
|
|
} |
308
|
|
|
|
|
|
|
} |
309
|
|
|
|
|
|
|
|
310
|
7
|
100
|
|
|
|
11
|
if( $try_min < $min ) { |
311
|
2
|
|
|
|
|
2
|
$minimal = $try_minimal; |
312
|
2
|
|
|
|
|
2
|
$min = $try_min; |
313
|
|
|
|
|
|
|
} |
314
|
7
|
100
|
|
|
|
13
|
if( $try_max > $max ) { |
315
|
4
|
|
|
|
|
4
|
$maximal = $try_maximal; |
316
|
4
|
|
|
|
|
7
|
$max = $try_max; |
317
|
|
|
|
|
|
|
} |
318
|
|
|
|
|
|
|
} |
319
|
|
|
|
|
|
|
|
320
|
4
|
|
|
|
|
21
|
return ( $minimal, $maximal ); |
321
|
|
|
|
|
|
|
} |
322
|
|
|
|
|
|
|
|
323
|
|
|
|
|
|
|
*nminmax_by = \&minmax_by; |
324
|
|
|
|
|
|
|
|
325
|
|
|
|
|
|
|
=head2 uniq_by |
326
|
|
|
|
|
|
|
|
327
|
|
|
|
|
|
|
@vals = uniq_by { KEYFUNC } @vals |
328
|
|
|
|
|
|
|
|
329
|
|
|
|
|
|
|
Returns a list of the subset of values for which the key function block |
330
|
|
|
|
|
|
|
returns unique values. The first value yielding a particular key is chosen, |
331
|
|
|
|
|
|
|
subsequent values are rejected. |
332
|
|
|
|
|
|
|
|
333
|
|
|
|
|
|
|
my @some_fruit = uniq_by { $_->colour } @fruit; |
334
|
|
|
|
|
|
|
|
335
|
|
|
|
|
|
|
To select instead the last value per key, reverse the input list. If the order |
336
|
|
|
|
|
|
|
of the results is significant, don't forget to reverse the result as well: |
337
|
|
|
|
|
|
|
|
338
|
|
|
|
|
|
|
my @some_fruit = reverse uniq_by { $_->colour } reverse @fruit; |
339
|
|
|
|
|
|
|
|
340
|
|
|
|
|
|
|
Because the values returned by the key function are used as hash keys, they |
341
|
|
|
|
|
|
|
ought to either be strings, or at least well-behaved as strings (such as |
342
|
|
|
|
|
|
|
numbers, or object references which overload stringification in a suitable |
343
|
|
|
|
|
|
|
manner). |
344
|
|
|
|
|
|
|
|
345
|
|
|
|
|
|
|
=cut |
346
|
|
|
|
|
|
|
|
347
|
|
|
|
|
|
|
sub uniq_by(&@) |
348
|
|
|
|
|
|
|
{ |
349
|
7
|
|
|
7
|
1
|
77
|
my $code = shift; |
350
|
|
|
|
|
|
|
|
351
|
7
|
|
|
|
|
9
|
my %present; |
352
|
|
|
|
|
|
|
return grep { |
353
|
7
|
|
|
|
|
15
|
my $key = $code->( local $_ = $_ ); |
|
13
|
|
|
|
|
23
|
|
354
|
13
|
|
|
|
|
64
|
!$present{$key}++ |
355
|
|
|
|
|
|
|
} @_; |
356
|
|
|
|
|
|
|
} |
357
|
|
|
|
|
|
|
|
358
|
|
|
|
|
|
|
=head2 partition_by |
359
|
|
|
|
|
|
|
|
360
|
|
|
|
|
|
|
%parts = partition_by { KEYFUNC } @vals |
361
|
|
|
|
|
|
|
|
362
|
|
|
|
|
|
|
Returns a key/value list of ARRAY refs containing all the original values |
363
|
|
|
|
|
|
|
distributed according to the result of the key function block. Each value will |
364
|
|
|
|
|
|
|
be an ARRAY ref containing all the values which returned the string from the |
365
|
|
|
|
|
|
|
key function, in their original order. |
366
|
|
|
|
|
|
|
|
367
|
|
|
|
|
|
|
my %balls_by_colour = partition_by { $_->colour } @balls; |
368
|
|
|
|
|
|
|
|
369
|
|
|
|
|
|
|
Because the values returned by the key function are used as hash keys, they |
370
|
|
|
|
|
|
|
ought to either be strings, or at least well-behaved as strings (such as |
371
|
|
|
|
|
|
|
numbers, or object references which overload stringification in a suitable |
372
|
|
|
|
|
|
|
manner). |
373
|
|
|
|
|
|
|
|
374
|
|
|
|
|
|
|
=cut |
375
|
|
|
|
|
|
|
|
376
|
|
|
|
|
|
|
sub partition_by(&@) |
377
|
|
|
|
|
|
|
{ |
378
|
7
|
|
|
7
|
1
|
77
|
my $code = shift; |
379
|
|
|
|
|
|
|
|
380
|
7
|
|
|
|
|
8
|
my %parts; |
381
|
7
|
|
|
|
|
14
|
push @{ $parts{ $code->( local $_ = $_ ) } }, $_ for @_; |
|
13
|
|
|
|
|
40
|
|
382
|
|
|
|
|
|
|
|
383
|
7
|
|
|
|
|
53
|
return %parts; |
384
|
|
|
|
|
|
|
} |
385
|
|
|
|
|
|
|
|
386
|
|
|
|
|
|
|
=head2 count_by |
387
|
|
|
|
|
|
|
|
388
|
|
|
|
|
|
|
%counts = count_by { KEYFUNC } @vals |
389
|
|
|
|
|
|
|
|
390
|
|
|
|
|
|
|
I |
391
|
|
|
|
|
|
|
|
392
|
|
|
|
|
|
|
Returns a key/value list of integers, giving the number of times the key |
393
|
|
|
|
|
|
|
function block returned the key, for each value in the list. |
394
|
|
|
|
|
|
|
|
395
|
|
|
|
|
|
|
my %count_of_balls = count_by { $_->colour } @balls; |
396
|
|
|
|
|
|
|
|
397
|
|
|
|
|
|
|
Because the values returned by the key function are used as hash keys, they |
398
|
|
|
|
|
|
|
ought to either be strings, or at least well-behaved as strings (such as |
399
|
|
|
|
|
|
|
numbers, or object references which overload stringification in a suitable |
400
|
|
|
|
|
|
|
manner). |
401
|
|
|
|
|
|
|
|
402
|
|
|
|
|
|
|
=cut |
403
|
|
|
|
|
|
|
|
404
|
|
|
|
|
|
|
sub count_by(&@) |
405
|
|
|
|
|
|
|
{ |
406
|
5
|
|
|
5
|
1
|
98
|
my $code = shift; |
407
|
|
|
|
|
|
|
|
408
|
5
|
|
|
|
|
7
|
my %counts; |
409
|
5
|
|
|
|
|
14
|
$counts{ $code->( local $_ = $_ ) }++ for @_; |
410
|
|
|
|
|
|
|
|
411
|
5
|
|
|
|
|
48
|
return %counts; |
412
|
|
|
|
|
|
|
} |
413
|
|
|
|
|
|
|
|
414
|
|
|
|
|
|
|
=head2 zip_by |
415
|
|
|
|
|
|
|
|
416
|
|
|
|
|
|
|
@vals = zip_by { ITEMFUNC } \@arr0, \@arr1, \@arr2,... |
417
|
|
|
|
|
|
|
|
418
|
|
|
|
|
|
|
Returns a list of each of the values returned by the function block, when |
419
|
|
|
|
|
|
|
invoked with values from across each each of the given ARRAY references. Each |
420
|
|
|
|
|
|
|
value in the returned list will be the result of the function having been |
421
|
|
|
|
|
|
|
invoked with arguments at that position, from across each of the arrays given. |
422
|
|
|
|
|
|
|
|
423
|
|
|
|
|
|
|
my @transposition = zip_by { [ @_ ] } @matrix; |
424
|
|
|
|
|
|
|
|
425
|
|
|
|
|
|
|
my @names = zip_by { "$_[1], $_[0]" } \@firstnames, \@surnames; |
426
|
|
|
|
|
|
|
|
427
|
|
|
|
|
|
|
print zip_by { "$_[0] => $_[1]\n" } [ keys %hash ], [ values %hash ]; |
428
|
|
|
|
|
|
|
|
429
|
|
|
|
|
|
|
If some of the arrays are shorter than others, the function will behave as if |
430
|
|
|
|
|
|
|
they had C in the trailing positions. The following two lines are |
431
|
|
|
|
|
|
|
equivalent: |
432
|
|
|
|
|
|
|
|
433
|
|
|
|
|
|
|
zip_by { f(@_) } [ 1, 2, 3 ], [ "a", "b" ] |
434
|
|
|
|
|
|
|
f( 1, "a" ), f( 2, "b" ), f( 3, undef ) |
435
|
|
|
|
|
|
|
|
436
|
|
|
|
|
|
|
The item function is called by C |
437
|
|
|
|
|
|
|
list is included in the result. This can be useful for example, for generating |
438
|
|
|
|
|
|
|
a hash from two separate lists of keys and values |
439
|
|
|
|
|
|
|
|
440
|
|
|
|
|
|
|
my %nums = zip_by { @_ } [qw( one two three )], [ 1, 2, 3 ]; |
441
|
|
|
|
|
|
|
# %nums = ( one => 1, two => 2, three => 3 ) |
442
|
|
|
|
|
|
|
|
443
|
|
|
|
|
|
|
(A function having this behaviour is sometimes called C, e.g. in |
444
|
|
|
|
|
|
|
Haskell, but that name would not fit the naming scheme used by this module). |
445
|
|
|
|
|
|
|
|
446
|
|
|
|
|
|
|
=cut |
447
|
|
|
|
|
|
|
|
448
|
|
|
|
|
|
|
sub zip_by(&@) |
449
|
|
|
|
|
|
|
{ |
450
|
7
|
|
|
7
|
1
|
3481
|
my $code = shift; |
451
|
|
|
|
|
|
|
|
452
|
7
|
100
|
|
|
|
22
|
@_ or return; |
453
|
|
|
|
|
|
|
|
454
|
6
|
|
|
|
|
8
|
my $len = 0; |
455
|
6
|
|
100
|
|
|
20
|
scalar @$_ > $len and $len = scalar @$_ for @_; |
456
|
|
|
|
|
|
|
|
457
|
|
|
|
|
|
|
return map { |
458
|
6
|
|
|
|
|
14
|
my $idx = $_; |
|
14
|
|
|
|
|
33
|
|
459
|
14
|
|
|
|
|
20
|
$code->( map { $_[$_][$idx] } 0 .. $#_ ) |
|
26
|
|
|
|
|
42
|
|
460
|
|
|
|
|
|
|
} 0 .. $len-1; |
461
|
|
|
|
|
|
|
} |
462
|
|
|
|
|
|
|
|
463
|
|
|
|
|
|
|
=head2 unzip_by |
464
|
|
|
|
|
|
|
|
465
|
|
|
|
|
|
|
$arr0, $arr1, $arr2, ... = unzip_by { ITEMFUNC } @vals |
466
|
|
|
|
|
|
|
|
467
|
|
|
|
|
|
|
I |
468
|
|
|
|
|
|
|
|
469
|
|
|
|
|
|
|
Returns a list of ARRAY references containing the values returned by the |
470
|
|
|
|
|
|
|
function block, when invoked for each of the values given in the input list. |
471
|
|
|
|
|
|
|
Each of the returned ARRAY references will contain the values returned at that |
472
|
|
|
|
|
|
|
corresponding position by the function block. That is, the first returned |
473
|
|
|
|
|
|
|
ARRAY reference will contain all the values returned in the first position by |
474
|
|
|
|
|
|
|
the function block, the second will contain all the values from the second |
475
|
|
|
|
|
|
|
position, and so on. |
476
|
|
|
|
|
|
|
|
477
|
|
|
|
|
|
|
my ( $firstnames, $lastnames ) = unzip_by { m/^(.*?) (.*)$/ } @names; |
478
|
|
|
|
|
|
|
|
479
|
|
|
|
|
|
|
If the function returns lists of differing lengths, the result will be padded |
480
|
|
|
|
|
|
|
with C in the missing elements. |
481
|
|
|
|
|
|
|
|
482
|
|
|
|
|
|
|
This function is an inverse of L, if given a corresponding inverse |
483
|
|
|
|
|
|
|
function. |
484
|
|
|
|
|
|
|
|
485
|
|
|
|
|
|
|
=cut |
486
|
|
|
|
|
|
|
|
487
|
|
|
|
|
|
|
sub unzip_by(&@) |
488
|
|
|
|
|
|
|
{ |
489
|
5
|
|
|
5
|
1
|
77
|
my $code = shift; |
490
|
|
|
|
|
|
|
|
491
|
5
|
|
|
|
|
7
|
my @ret; |
492
|
5
|
|
|
|
|
12
|
foreach my $idx ( 0 .. $#_ ) { |
493
|
12
|
|
|
|
|
24
|
my @slice = $code->( local $_ = $_[$idx] ); |
494
|
12
|
100
|
|
|
|
48
|
$#slice = $#ret if @slice < @ret; |
495
|
12
|
|
|
|
|
38
|
$ret[$_][$idx] = $slice[$_] for 0 .. $#slice; |
496
|
|
|
|
|
|
|
} |
497
|
|
|
|
|
|
|
|
498
|
5
|
|
|
|
|
25
|
return @ret; |
499
|
|
|
|
|
|
|
} |
500
|
|
|
|
|
|
|
|
501
|
|
|
|
|
|
|
=head2 extract_by |
502
|
|
|
|
|
|
|
|
503
|
|
|
|
|
|
|
@vals = extract_by { SELECTFUNC } @arr |
504
|
|
|
|
|
|
|
|
505
|
|
|
|
|
|
|
I |
506
|
|
|
|
|
|
|
|
507
|
|
|
|
|
|
|
Removes elements from the referenced array on which the selection function |
508
|
|
|
|
|
|
|
returns true, and returns a list containing those elements. This function is |
509
|
|
|
|
|
|
|
similar to C, except that it modifies the referenced array to remove the |
510
|
|
|
|
|
|
|
selected values from it, leaving only the unselected ones. |
511
|
|
|
|
|
|
|
|
512
|
|
|
|
|
|
|
my @red_balls = extract_by { $_->color eq "red" } @balls; |
513
|
|
|
|
|
|
|
|
514
|
|
|
|
|
|
|
# Now there are no red balls in the @balls array |
515
|
|
|
|
|
|
|
|
516
|
|
|
|
|
|
|
This function modifies a real array, unlike most of the other functions in this |
517
|
|
|
|
|
|
|
module. Because of this, it requires a real array, not just a list. |
518
|
|
|
|
|
|
|
|
519
|
|
|
|
|
|
|
This function is implemented by invoking C on the array, not by |
520
|
|
|
|
|
|
|
constructing a new list and assigning it. One result of this is that weak |
521
|
|
|
|
|
|
|
references will not be disturbed. |
522
|
|
|
|
|
|
|
|
523
|
|
|
|
|
|
|
extract_by { !defined $_ } @refs; |
524
|
|
|
|
|
|
|
|
525
|
|
|
|
|
|
|
will leave weak references weakened in the C<@refs> array, whereas |
526
|
|
|
|
|
|
|
|
527
|
|
|
|
|
|
|
@refs = grep { defined $_ } @refs; |
528
|
|
|
|
|
|
|
|
529
|
|
|
|
|
|
|
will strengthen them all again. |
530
|
|
|
|
|
|
|
|
531
|
|
|
|
|
|
|
=cut |
532
|
|
|
|
|
|
|
|
533
|
|
|
|
|
|
|
sub extract_by(&\@) |
534
|
|
|
|
|
|
|
{ |
535
|
6
|
|
|
6
|
1
|
493
|
my $code = shift; |
536
|
6
|
|
|
|
|
8
|
my ( $arrref ) = @_; |
537
|
|
|
|
|
|
|
|
538
|
6
|
|
|
|
|
9
|
my @ret; |
539
|
6
|
|
|
|
|
8
|
for( my $idx = 0; ; $idx++ ) { |
540
|
43
|
100
|
|
|
|
109
|
last if $idx > $#$arrref; |
541
|
37
|
100
|
|
|
|
60
|
next unless $code->( local $_ = $arrref->[$idx] ); |
542
|
|
|
|
|
|
|
|
543
|
11
|
|
|
|
|
35
|
push @ret, splice @$arrref, $idx, 1, (); |
544
|
11
|
|
|
|
|
15
|
redo; |
545
|
|
|
|
|
|
|
} |
546
|
|
|
|
|
|
|
|
547
|
6
|
|
|
|
|
28
|
return @ret; |
548
|
|
|
|
|
|
|
} |
549
|
|
|
|
|
|
|
|
550
|
|
|
|
|
|
|
=head2 extract_first_by |
551
|
|
|
|
|
|
|
|
552
|
|
|
|
|
|
|
$val = extract_first_by { SELECTFUNC } @arr |
553
|
|
|
|
|
|
|
|
554
|
|
|
|
|
|
|
I |
555
|
|
|
|
|
|
|
|
556
|
|
|
|
|
|
|
A hybrid between L and C. Removes the first |
557
|
|
|
|
|
|
|
element from the referenced array on which the selection function returns |
558
|
|
|
|
|
|
|
true, returning it. |
559
|
|
|
|
|
|
|
|
560
|
|
|
|
|
|
|
As with L, this function requires a real array and not just a |
561
|
|
|
|
|
|
|
list, and is also implemented using C so that weak references are |
562
|
|
|
|
|
|
|
not disturbed. |
563
|
|
|
|
|
|
|
|
564
|
|
|
|
|
|
|
If this function fails to find a matching element, it will return an empty |
565
|
|
|
|
|
|
|
list in list context. This allows a caller to distinguish the case between |
566
|
|
|
|
|
|
|
no matching element, and the first matching element being C. |
567
|
|
|
|
|
|
|
|
568
|
|
|
|
|
|
|
=cut |
569
|
|
|
|
|
|
|
|
570
|
|
|
|
|
|
|
sub extract_first_by(&\@) |
571
|
|
|
|
|
|
|
{ |
572
|
2
|
|
|
2
|
1
|
395
|
my $code = shift; |
573
|
2
|
|
|
|
|
5
|
my ( $arrref ) = @_; |
574
|
|
|
|
|
|
|
|
575
|
2
|
|
|
|
|
6
|
foreach my $idx ( 0 .. $#$arrref ) { |
576
|
4
|
100
|
|
|
|
14
|
next unless $code->( local $_ = $arrref->[$idx] ); |
577
|
|
|
|
|
|
|
|
578
|
1
|
|
|
|
|
9
|
return splice @$arrref, $idx, 1, (); |
579
|
|
|
|
|
|
|
} |
580
|
|
|
|
|
|
|
|
581
|
1
|
|
|
|
|
5
|
return; |
582
|
|
|
|
|
|
|
} |
583
|
|
|
|
|
|
|
|
584
|
|
|
|
|
|
|
=head2 weighted_shuffle_by |
585
|
|
|
|
|
|
|
|
586
|
|
|
|
|
|
|
@vals = weighted_shuffle_by { WEIGHTFUNC } @vals |
587
|
|
|
|
|
|
|
|
588
|
|
|
|
|
|
|
I |
589
|
|
|
|
|
|
|
|
590
|
|
|
|
|
|
|
Returns the list of values shuffled into a random order. The randomisation is |
591
|
|
|
|
|
|
|
not uniform, but weighted by the value returned by the C. The |
592
|
|
|
|
|
|
|
probability of each item being returned first will be distributed with the |
593
|
|
|
|
|
|
|
distribution of the weights, and so on recursively for the remaining items. |
594
|
|
|
|
|
|
|
|
595
|
|
|
|
|
|
|
=cut |
596
|
|
|
|
|
|
|
|
597
|
|
|
|
|
|
|
sub weighted_shuffle_by(&@) |
598
|
|
|
|
|
|
|
{ |
599
|
25
|
|
|
25
|
1
|
748
|
my $code = shift; |
600
|
25
|
|
|
|
|
39
|
my @vals = @_; |
601
|
|
|
|
|
|
|
|
602
|
25
|
|
|
|
|
43
|
my @weights = map { $code->( local $_ = $_ ) } @vals; |
|
70
|
|
|
|
|
199
|
|
603
|
|
|
|
|
|
|
|
604
|
25
|
|
|
|
|
81
|
my @ret; |
605
|
25
|
|
|
|
|
41
|
while( @vals > 1 ) { |
606
|
46
|
|
|
|
|
50
|
my $total = 0; $total += $_ for @weights; |
|
46
|
|
|
|
|
75
|
|
607
|
46
|
|
|
|
|
58
|
my $select = int rand $total; |
608
|
46
|
|
|
|
|
367
|
my $idx = 0; |
609
|
46
|
|
|
|
|
77
|
while( $select >= $weights[$idx] ) { |
610
|
41
|
|
|
|
|
57
|
$select -= $weights[$idx++]; |
611
|
|
|
|
|
|
|
} |
612
|
|
|
|
|
|
|
|
613
|
46
|
|
|
|
|
60
|
push @ret, splice @vals, $idx, 1, (); |
614
|
46
|
|
|
|
|
77
|
splice @weights, $idx, 1, (); |
615
|
|
|
|
|
|
|
} |
616
|
|
|
|
|
|
|
|
617
|
25
|
100
|
|
|
|
36
|
push @ret, @vals if @vals; |
618
|
|
|
|
|
|
|
|
619
|
25
|
|
|
|
|
58
|
return @ret; |
620
|
|
|
|
|
|
|
} |
621
|
|
|
|
|
|
|
|
622
|
|
|
|
|
|
|
=head2 bundle_by |
623
|
|
|
|
|
|
|
|
624
|
|
|
|
|
|
|
@vals = bundle_by { BLOCKFUNC } $number, @vals |
625
|
|
|
|
|
|
|
|
626
|
|
|
|
|
|
|
I |
627
|
|
|
|
|
|
|
|
628
|
|
|
|
|
|
|
Similar to a regular C |
629
|
|
|
|
|
|
|
by C. Values from the input list are given to the block function in |
630
|
|
|
|
|
|
|
bundles of C<$number>. |
631
|
|
|
|
|
|
|
|
632
|
|
|
|
|
|
|
If given a list of values whose length does not evenly divide by C<$number>, |
633
|
|
|
|
|
|
|
the final call will be passed fewer elements than the others. |
634
|
|
|
|
|
|
|
|
635
|
|
|
|
|
|
|
=cut |
636
|
|
|
|
|
|
|
|
637
|
|
|
|
|
|
|
sub bundle_by(&@) |
638
|
|
|
|
|
|
|
{ |
639
|
6
|
|
|
6
|
1
|
79
|
my $code = shift; |
640
|
6
|
|
|
|
|
7
|
my $n = shift; |
641
|
|
|
|
|
|
|
|
642
|
6
|
|
|
|
|
6
|
my @ret; |
643
|
6
|
|
|
|
|
19
|
for( my ( $pos, $next ) = ( 0, $n ); $pos < @_; $pos = $next, $next += $n ) { |
644
|
13
|
100
|
|
|
|
41
|
$next = @_ if $next > @_; |
645
|
13
|
|
|
|
|
26
|
push @ret, $code->( @_[$pos .. $next-1] ); |
646
|
|
|
|
|
|
|
} |
647
|
6
|
|
|
|
|
37
|
return @ret; |
648
|
|
|
|
|
|
|
} |
649
|
|
|
|
|
|
|
|
650
|
|
|
|
|
|
|
=head1 EXAMPLES |
651
|
|
|
|
|
|
|
|
652
|
|
|
|
|
|
|
As many of these functions both take and return lists of values, more useful |
653
|
|
|
|
|
|
|
behaviours can be created by combining them together. |
654
|
|
|
|
|
|
|
|
655
|
|
|
|
|
|
|
=head2 Modal Value |
656
|
|
|
|
|
|
|
|
657
|
|
|
|
|
|
|
To pick the modal value in a list (i.e. the value that appears most |
658
|
|
|
|
|
|
|
frequently according to some key function) combine C and C: |
659
|
|
|
|
|
|
|
|
660
|
|
|
|
|
|
|
my %counts = count_by { KEYFUNC } ITEMS; |
661
|
|
|
|
|
|
|
my $modal = max_by { $counts{$_} } keys %counts; |
662
|
|
|
|
|
|
|
|
663
|
|
|
|
|
|
|
This can be performed in a single step by use of the C function from |
664
|
|
|
|
|
|
|
L: |
665
|
|
|
|
|
|
|
|
666
|
|
|
|
|
|
|
use List::Util qw( pairs ); |
667
|
|
|
|
|
|
|
|
668
|
|
|
|
|
|
|
my $modal = ( max_by { $_->value } pairs count_by { KEYFUNC } ITEMS )->key; |
669
|
|
|
|
|
|
|
|
670
|
|
|
|
|
|
|
=cut |
671
|
|
|
|
|
|
|
|
672
|
|
|
|
|
|
|
=head1 TODO |
673
|
|
|
|
|
|
|
|
674
|
|
|
|
|
|
|
=over 4 |
675
|
|
|
|
|
|
|
|
676
|
|
|
|
|
|
|
=item * XS implementations |
677
|
|
|
|
|
|
|
|
678
|
|
|
|
|
|
|
These functions are currently all written in pure perl. Some at least, may |
679
|
|
|
|
|
|
|
benefit from having XS implementations to speed up their logic. |
680
|
|
|
|
|
|
|
|
681
|
|
|
|
|
|
|
=item * Merge into L or L |
682
|
|
|
|
|
|
|
|
683
|
|
|
|
|
|
|
This module shouldn't really exist. The functions should instead be part of |
684
|
|
|
|
|
|
|
one of the existing modules that already contain many list utility functions. |
685
|
|
|
|
|
|
|
Having Yet Another List Utilty Module just worsens the problem. |
686
|
|
|
|
|
|
|
|
687
|
|
|
|
|
|
|
I have attempted to contact the authors of both of the above modules, to no |
688
|
|
|
|
|
|
|
avail; therefore I decided it best to write and release this code here anyway |
689
|
|
|
|
|
|
|
so that it is at least on CPAN. Once there, we can then see how best to merge |
690
|
|
|
|
|
|
|
it into an existing module. |
691
|
|
|
|
|
|
|
|
692
|
|
|
|
|
|
|
I: As I am now the maintainer of L, some |
693
|
|
|
|
|
|
|
amount of merging/copying should be possible. However, given the latter's key |
694
|
|
|
|
|
|
|
position in the core F distribution and head of the "CPAN River" I am |
695
|
|
|
|
|
|
|
keen not to do this wholesale, but a selected pick of what seems best, by a |
696
|
|
|
|
|
|
|
popular consensus. |
697
|
|
|
|
|
|
|
|
698
|
|
|
|
|
|
|
=item * C and C-like functions |
699
|
|
|
|
|
|
|
|
700
|
|
|
|
|
|
|
Consider perhaps |
701
|
|
|
|
|
|
|
|
702
|
|
|
|
|
|
|
head_before { COND } LIST # excludes terminating element |
703
|
|
|
|
|
|
|
head_upto { COND } LIST # includes terminating element |
704
|
|
|
|
|
|
|
|
705
|
|
|
|
|
|
|
tail_since { COND } LIST # includes initiating element |
706
|
|
|
|
|
|
|
tail_after { COND } LIST # excludes initiating element |
707
|
|
|
|
|
|
|
|
708
|
|
|
|
|
|
|
(See also L). |
709
|
|
|
|
|
|
|
|
710
|
|
|
|
|
|
|
=back |
711
|
|
|
|
|
|
|
|
712
|
|
|
|
|
|
|
=head1 AUTHOR |
713
|
|
|
|
|
|
|
|
714
|
|
|
|
|
|
|
Paul Evans |
715
|
|
|
|
|
|
|
|
716
|
|
|
|
|
|
|
=cut |
717
|
|
|
|
|
|
|
|
718
|
|
|
|
|
|
|
0x55AA; |