1List::MoreUtils(3) User Contributed Perl Documentation List::MoreUtils(3)
2
6 List::MoreUtils - Provide the stuff missing in List::Util
7
9 # import specific functions
10 use List::MoreUtils qw(any uniq);
12 if ( any { /foo/ } uniq @has_duplicates ) {
14 # do stuff
15 }
16 # import everything
18 use List::MoreUtils ':all';
20 # import by API
22 # has "original" any/all/none/notall behavior
24 use List::MoreUtils ':like_0.22';
25 # 0.22 + bsearch
26 use List::MoreUtils ':like_0.24';
27 # has "simplified" any/all/none/notall behavior + (n)sort_by
28 use List::MoreUtils ':like_0.33';
29
31 List::MoreUtils provides some trivial but commonly needed functionality
32 on lists which is not going to go into List::Util.
33 All of the below functions are implementable in only a couple of lines
35 of Perl code. Using the functions from this module however should give
36 slightly better performance as everything is implemented in C. The
37 pure-Perl implementation of these functions only serves as a fallback
38 in case the C portions of this module couldn't be compiled on this
39 machine.
40
42 Default behavior
43 Nothing by default. To import all of this module's symbols use the
44 ":all" tag. Otherwise functions can be imported by name as usual:
45 use List::MoreUtils ':all';
47 use List::MoreUtils qw{ any firstidx };
49 Because historical changes to the API might make upgrading
51 List::MoreUtils difficult for some projects, the legacy API is
52 available via special import tags.
53 Like version 0.22 (last release with original API)
55 This API was available from 2006 to 2009, returning undef for empty
56 lists on "all"/"any"/"none"/"notall":
57 use List::MoreUtils ':like_0.22';
59 This import tag will import all functions available as of version 0.22.
61 However, it will import "any_u" as "any", "all_u" as "all", "none_u" as
62 "none", and "notall_u" as "notall".
63 Like version 0.24 (first incompatible change)
65 This API was available from 2010 to 2011. It changed the return value
66 of "none" and added the "bsearch" function.
67 use List::MoreUtils ':like_0.24';
69 This import tag will import all functions available as of version 0.24.
71 However it will import "any_u" as "any", "all_u" as "all", and
72 "notall_u" as "notall". It will import "none" as described in the
73 documentation below (true for empty list).
74 Like version 0.33 (second incompatible change)
76 This API was available from 2011 to 2014. It is widely used in several
77 CPAN modules and thus it's closest to the current API. It changed the
78 return values of "any", "all", and "notall". It added the "sort_by"
79 and "nsort_by" functions and the "distinct" alias for "uniq". It
80 omitted "bsearch".
81 use List::MoreUtils ':like_0.33';
83 This import tag will import all functions available as of version 0.33.
85 Note: it will not import "bsearch" for consistency with the 0.33 API.
86
88 Junctions
89 Treatment of an empty list
90 There are two schools of thought for how to evaluate a junction on an
92 empty list:
93 • Reduction to an identity (boolean)
95 • Result is undefined (three-valued)
97 In the first case, the result of the junction applied to the empty list
99 is determined by a mathematical reduction to an identity depending on
100 whether the underlying comparison is "or" or "and". Conceptually:
101 "any are true" "all are true"
103 -------------- --------------
104 2 elements: A || B || 0 A && B && 1
105 1 element: A || 0 A && 1
106 0 elements: 0 1
107 In the second case, three-value logic is desired, in which a junction
109 applied to an empty list returns "undef" rather than true or false
110 Junctions with a "_u" suffix implement three-valued logic. Those
112 without are boolean.
113 all BLOCK LIST
115 all_u BLOCK LIST
117 Returns a true value if all items in LIST meet the criterion given
119 through BLOCK. Sets $_ for each item in LIST in turn:
120 print "All values are non-negative"
122 if all { $_ >= 0 } ($x, $y, $z);
123 For an empty LIST, "all" returns true (i.e. no values failed the
125 condition) and "all_u" returns "undef".
126 Thus, all_u(@list) is equivalent to "@list ? all(@list) : undef".
128 Note: because Perl treats "undef" as false, you must check the return
130 value of "all_u" with "defined" or you will get the opposite result of
131 what you expect.
132 any BLOCK LIST
134 any_u BLOCK LIST
136 Returns a true value if any item in LIST meets the criterion given
138 through BLOCK. Sets $_ for each item in LIST in turn:
139 print "At least one non-negative value"
141 if any { $_ >= 0 } ($x, $y, $z);
142 For an empty LIST, "any" returns false and "any_u" returns "undef".
144 Thus, any_u(@list) is equivalent to "@list ? any(@list) : undef".
146 none BLOCK LIST
148 none_u BLOCK LIST
150 Logically the negation of "any". Returns a true value if no item in
152 LIST meets the criterion given through BLOCK. Sets $_ for each item in
153 LIST in turn:
154 print "No non-negative values"
156 if none { $_ >= 0 } ($x, $y, $z);
157 For an empty LIST, "none" returns true (i.e. no values failed the
159 condition) and "none_u" returns "undef".
160 Thus, none_u(@list) is equivalent to "@list ? none(@list) : undef".
162 Note: because Perl treats "undef" as false, you must check the return
164 value of "none_u" with "defined" or you will get the opposite result of
165 what you expect.
166 notall BLOCK LIST
168 notall_u BLOCK LIST
170 Logically the negation of "all". Returns a true value if not all items
172 in LIST meet the criterion given through BLOCK. Sets $_ for each item
173 in LIST in turn:
174 print "Not all values are non-negative"
176 if notall { $_ >= 0 } ($x, $y, $z);
177 For an empty LIST, "notall" returns false and "notall_u" returns
179 "undef".
180 Thus, notall_u(@list) is equivalent to "@list ? notall(@list) : undef".
182 one BLOCK LIST
184 one_u BLOCK LIST
186 Returns a true value if precisely one item in LIST meets the criterion
188 given through BLOCK. Sets $_ for each item in LIST in turn:
189 print "Precisely one value defined"
191 if one { defined($_) } @list;
192 Returns false otherwise.
194 For an empty LIST, "one" returns false and "one_u" returns "undef".
196 The expression "one BLOCK LIST" is almost equivalent to "1 == true
198 BLOCK LIST", except for short-cutting. Evaluation of BLOCK will
199 immediately stop at the second true value.
200 Transformation
202 apply BLOCK LIST
203 Applies BLOCK to each item in LIST and returns a list of the values
205 after BLOCK has been applied. In scalar context, the last element is
206 returned. This function is similar to "map" but will not modify the
207 elements of the input list:
208 my @list = (1 .. 4);
210 my @mult = apply { $_ *= 2 } @list;
211 print "\@list = @list\n";
212 print "\@mult = @mult\n";
213 __END__
214 @list = 1 2 3 4
215 @mult = 2 4 6 8
216 Think of it as syntactic sugar for
218 for (my @mult = @list) { $_ *= 2 }
220 insert_after BLOCK VALUE LIST
222 Inserts VALUE after the first item in LIST for which the criterion in
224 BLOCK is true. Sets $_ for each item in LIST in turn.
225 my @list = qw/This is a list/;
227 insert_after { $_ eq "a" } "longer" => @list;
228 print "@list";
229 __END__
230 This is a longer list
231 insert_after_string STRING VALUE LIST
233 Inserts VALUE after the first item in LIST which is equal to STRING.
235 my @list = qw/This is a list/;
237 insert_after_string "a", "longer" => @list;
238 print "@list";
239 __END__
240 This is a longer list
241 pairwise BLOCK ARRAY1 ARRAY2
243 Evaluates BLOCK for each pair of elements in ARRAY1 and ARRAY2 and
245 returns a new list consisting of BLOCK's return values. The two
246 elements are set to $a and $b. Note that those two are aliases to the
247 original value so changing them will modify the input arrays.
248 @a = (1 .. 5);
250 @b = (11 .. 15);
251 @x = pairwise { $a + $b } @a, @b; # returns 12, 14, 16, 18, 20
252 # mesh with pairwise
254 @a = qw/a b c/;
255 @b = qw/1 2 3/;
256 @x = pairwise { ($a, $b) } @a, @b; # returns a, 1, b, 2, c, 3
257 mesh ARRAY1 ARRAY2 [ ARRAY3 ... ]
259 zip ARRAY1 ARRAY2 [ ARRAY3 ... ]
261 Returns a list consisting of the first elements of each array, then the
263 second, then the third, etc, until all arrays are exhausted.
264 Examples:
266 @x = qw/a b c d/;
268 @y = qw/1 2 3 4/;
269 @z = mesh @x, @y; # returns a, 1, b, 2, c, 3, d, 4
270 @a = ('x');
272 @b = ('1', '2');
273 @c = qw/zip zap zot/;
274 @d = mesh @a, @b, @c; # x, 1, zip, undef, 2, zap, undef, undef, zot
275 "zip" is an alias for "mesh".
277 zip6
279 zip_unflatten
281 Returns a list of arrays consisting of the first elements of each
283 array, then the second, then the third, etc, until all arrays are
284 exhausted.
285 @x = qw/a b c d/;
287 @y = qw/1 2 3 4/;
288 @z = zip6 @x, @y; # returns [a, 1], [b, 2], [c, 3], [d, 4]
289 @a = ('x');
291 @b = ('1', '2');
292 @c = qw/zip zap zot/;
293 @d = zip6 @a, @b, @c; # [x, 1, zip], [undef, 2, zap], [undef, undef, zot]
294 "zip_unflatten" is an alias for "zip6".
296 listcmp ARRAY0 ARRAY1 [ ARRAY2 ... ]
298 Returns an associative list of elements and every id of the list it was
300 found in. Allows easy implementation of @a & @b, @a | @b, @a ^ @b and
301 so on. Undefined entries in any given array are skipped.
302 my @a = qw(one two three four five six seven eight nine ten eleven twelve thirteen);
304 my @b = qw(two three five seven eleven thirteen seventeen);
305 my @c = qw(one one two three five eight thirteen twentyone);
306 my %cmp = listcmp @a, @b, @c; # returns (one => [0, 2], two => [0, 1, 2], three => [0, 1, 2], four => [0], ...)
307 my @seq = (1, 2, 3);
309 my @prim = (undef, 2, 3, 5);
310 my @fib = (1, 1, 2);
311 my %cmp = listcmp @seq, @prim, @fib;
312 # returns ( 1 => [0, 2], 2 => [0, 1, 2], 3 => [0, 1], 5 => [1] )
313 arrayify LIST[,LIST[,LIST...]]
315 Returns a list consisting of each element of given arrays. Recursive
317 arrays are flattened, too.
318 @a = (1, [[2], 3], 4, [5], 6, [7], 8, 9);
320 @l = arrayify @a; # returns 1, 2, 3, 4, 5, 6, 7, 8, 9
321 uniq LIST
323 distinct LIST
325 Returns a new list by stripping duplicate values in LIST by comparing
327 the values as hash keys, except that undef is considered separate from
328 ''. The order of elements in the returned list is the same as in LIST.
329 In scalar context, returns the number of unique elements in LIST.
330 my @x = uniq 1, 1, 2, 2, 3, 5, 3, 4; # returns 1 2 3 5 4
332 my $x = uniq 1, 1, 2, 2, 3, 5, 3, 4; # returns 5
333 # returns "Mike", "Michael", "Richard", "Rick"
334 my @n = distinct "Mike", "Michael", "Richard", "Rick", "Michael", "Rick"
335 # returns "A8", "", undef, "A5", "S1"
336 my @s = distinct "A8", "", undef, "A5", "S1", "A5", "A8"
337 # returns "Giulia", "Giulietta", undef, "", 156, "GTA", "GTV", 159, "Brera", "4C"
338 my @w = uniq "Giulia", "Giulietta", undef, "", 156, "GTA", "GTV", 159, "Brera", "4C", "Giulietta", "Giulia"
339 "distinct" is an alias for "uniq".
341 RT#49800 can be used to give feedback about this behavior.
343 singleton LIST
345 Returns a new list by stripping values in LIST occurring more than once
347 by comparing the values as hash keys, except that undef is considered
348 separate from ''. The order of elements in the returned list is the
349 same as in LIST. In scalar context, returns the number of elements
350 occurring only once in LIST.
351 my @x = singleton 1,1,2,2,3,4,5 # returns 3 4 5
353 duplicates LIST
355 Returns a new list by stripping values in LIST occurring less than
357 twice by comparing the values as hash keys, except that undef is
358 considered separate from ''. The order of elements in the returned
359 list is the same as in LIST. In scalar context, returns the number of
360 elements occurring more than once in LIST.
361 my @y = duplicates 1,1,2,4,7,2,3,4,6,9; #returns 1,2,4
363 frequency LIST
365 Returns an associative list of distinct values and the corresponding
367 frequency.
368 my @f = frequency values %radio_nrw; # returns (
370 # 'Deutschlandfunk (DLF)' => 9, 'WDR 3' => 10,
371 # 'WDR 4' => 11, 'WDR 5' => 14, 'WDR Eins Live' => 14,
372 # 'Deutschlandradio Kultur' => 8,...)
373 occurrences LIST
375 Returns a new list of frequencies and the corresponding values from
377 LIST.
378 my @o = occurrences ((1) x 3, (2) x 4, (3) x 2, (4) x 7, (5) x 2, (6) x 4);
380 # @o = (undef, undef, [3, 5], [1], [2, 6], undef, undef, [4]);
381 mode LIST
383 Returns the modal value of LIST. In scalar context, just the modal
385 value is returned, in list context all probes occurring modal times are
386 returned, too.
387 my @m = mode ((1) x 3, (2) x 4, (3) x 2, (4) x 7, (5) x 2, (6) x 4, (7) x 3, (8) x 7);
389 # @m = (7, 4, 8) - bimodal LIST
390 slide BLOCK LIST
392 The function "slide" operates on pairs of list elements like:
394 my @s = slide { "$a and $b" } (0..3);
396 # @s = ("0 and 1", "1 and 2", "2 and 3")
397 The idea behind this function is a kind of magnifying glass that is
399 moved along a list and calls "BLOCK" every time the next list item is
400 reached.
401 Partitioning
403 after BLOCK LIST
404 Returns a list of the values of LIST after (and not including) the
406 point where BLOCK returns a true value. Sets $_ for each element in
407 LIST in turn.
408 @x = after { $_ % 5 == 0 } (1..9); # returns 6, 7, 8, 9
410 after_incl BLOCK LIST
412 Same as "after" but also includes the element for which BLOCK is true.
414 before BLOCK LIST
416 Returns a list of values of LIST up to (and not including) the point
418 where BLOCK returns a true value. Sets $_ for each element in LIST in
419 turn.
420 before_incl BLOCK LIST
422 Same as "before" but also includes the element for which BLOCK is true.
424 part BLOCK LIST
426 Partitions LIST based on the return value of BLOCK which denotes into
428 which partition the current value is put.
429 Returns a list of the partitions thusly created. Each partition created
431 is a reference to an array.
432 my $i = 0;
434 my @part = part { $i++ % 2 } 1 .. 8; # returns [1, 3, 5, 7], [2, 4, 6, 8]
435 You can have a sparse list of partitions as well where non-set
437 partitions will be undef:
438 my @part = part { 2 } 1 .. 10; # returns undef, undef, [ 1 .. 10 ]
440 Be careful with negative values, though:
442 my @part = part { -1 } 1 .. 10;
444 __END__
445 Modification of non-creatable array value attempted, subscript -1 ...
446 Negative values are only ok when they refer to a partition previously
448 created:
449 my @idx = ( 0, 1, -1 );
451 my $i = 0;
452 my @part = part { $idx[$i++ % 3] } 1 .. 8; # [1, 4, 7], [2, 3, 5, 6, 8]
453 samples COUNT LIST
455 Returns a new list containing COUNT random samples from LIST. Is
457 similar to "shuffle" in List::Util, but stops after COUNT.
458 @r = samples 10, 1..10; # same as shuffle
460 @r2 = samples 5, 1..10; # gives 5 values from 1..10;
461 Iteration
463 each_array ARRAY1 ARRAY2 ...
464 Creates an array iterator to return the elements of the list of arrays
466 ARRAY1, ARRAY2 throughout ARRAYn in turn. That is, the first time it
467 is called, it returns the first element of each array. The next time,
468 it returns the second elements. And so on, until all elements are
469 exhausted.
470 This is useful for looping over more than one array at once:
472 my $ea = each_array(@a, @b, @c);
474 while ( my ($a, $b, $c) = $ea->() ) { .... }
475 The iterator returns the empty list when it reached the end of all
477 arrays.
478 If the iterator is passed an argument of '"index"', then it returns the
480 index of the last fetched set of values, as a scalar.
481 each_arrayref LIST
483 Like each_array, but the arguments are references to arrays, not the
485 plain arrays.
486 natatime EXPR, LIST
488 Creates an array iterator, for looping over an array in chunks of $n
490 items at a time. (n at a time, get it?). An example is probably a
491 better explanation than I could give in words.
492 Example:
494 my @x = ('a' .. 'g');
496 my $it = natatime 3, @x;
497 while (my @vals = $it->())
498 {
499 print "@vals\n";
500 }
501 This prints
503 a b c
505 d e f
506 g
507 slideatatime STEP, WINDOW, LIST
509 Creates an array iterator, for looping over an array in chunks of
511 "$windows-size" items at a time.
512 The idea behind this function is a kind of magnifying glass (finer
514 controllable compared to "slide") that is moved along a list.
515 Example:
517 my @x = ('a' .. 'g');
519 my $it = slideatatime 2, 3, @x;
520 while (my @vals = $it->())
521 {
522 print "@vals\n";
523 }
524 This prints
526 a b c
528 c d e
529 e f g
530 g
531 Searching
533 firstval BLOCK LIST
534 first_value BLOCK LIST
536 Returns the first element in LIST for which BLOCK evaluates to true.
538 Each element of LIST is set to $_ in turn. Returns "undef" if no such
539 element has been found.
540 "first_value" is an alias for "firstval".
542 onlyval BLOCK LIST
544 only_value BLOCK LIST
546 Returns the only element in LIST for which BLOCK evaluates to true.
548 Sets $_ for each item in LIST in turn. Returns "undef" if no such
549 element has been found.
550 "only_value" is an alias for "onlyval".
552 lastval BLOCK LIST
554 last_value BLOCK LIST
556 Returns the last value in LIST for which BLOCK evaluates to true. Each
558 element of LIST is set to $_ in turn. Returns "undef" if no such
559 element has been found.
560 "last_value" is an alias for "lastval".
562 firstres BLOCK LIST
564 first_result BLOCK LIST
566 Returns the result of BLOCK for the first element in LIST for which
568 BLOCK evaluates to true. Each element of LIST is set to $_ in turn.
569 Returns "undef" if no such element has been found.
570 "first_result" is an alias for "firstres".
572 onlyres BLOCK LIST
574 only_result BLOCK LIST
576 Returns the result of BLOCK for the first element in LIST for which
578 BLOCK evaluates to true. Sets $_ for each item in LIST in turn. Returns
579 "undef" if no such element has been found.
580 "only_result" is an alias for "onlyres".
582 lastres BLOCK LIST
584 last_result BLOCK LIST
586 Returns the result of BLOCK for the last element in LIST for which
588 BLOCK evaluates to true. Each element of LIST is set to $_ in turn.
589 Returns "undef" if no such element has been found.
590 "last_result" is an alias for "lastres".
592 indexes BLOCK LIST
594 Evaluates BLOCK for each element in LIST (assigned to $_) and returns a
596 list of the indices of those elements for which BLOCK returned a true
597 value. This is just like "grep" only that it returns indices instead of
598 values:
599 @x = indexes { $_ % 2 == 0 } (1..10); # returns 1, 3, 5, 7, 9
601 firstidx BLOCK LIST
603 first_index BLOCK LIST
605 Returns the index of the first element in LIST for which the criterion
607 in BLOCK is true. Sets $_ for each item in LIST in turn:
608 my @list = (1, 4, 3, 2, 4, 6);
610 printf "item with index %i in list is 4", firstidx { $_ == 4 } @list;
611 __END__
612 item with index 1 in list is 4
613 Returns -1 if no such item could be found.
615 "first_index" is an alias for "firstidx".
617 onlyidx BLOCK LIST
619 only_index BLOCK LIST
621 Returns the index of the only element in LIST for which the criterion
623 in BLOCK is true. Sets $_ for each item in LIST in turn:
624 my @list = (1, 3, 4, 3, 2, 4);
626 printf "uniqe index of item 2 in list is %i", onlyidx { $_ == 2 } @list;
627 __END__
628 unique index of item 2 in list is 4
629 Returns -1 if either no such item or more than one of these has been
631 found.
632 "only_index" is an alias for "onlyidx".
634 lastidx BLOCK LIST
636 last_index BLOCK LIST
638 Returns the index of the last element in LIST for which the criterion
640 in BLOCK is true. Sets $_ for each item in LIST in turn:
641 my @list = (1, 4, 3, 2, 4, 6);
643 printf "item with index %i in list is 4", lastidx { $_ == 4 } @list;
644 __END__
645 item with index 4 in list is 4
646 Returns -1 if no such item could be found.
648 "last_index" is an alias for "lastidx".
650 Sorting
652 sort_by BLOCK LIST
653 Returns the list of values sorted according to the string values
655 returned by the KEYFUNC block or function. A typical use of this may be
656 to sort objects according to the string value of some accessor, such as
657 sort_by { $_->name } @people
659 The key function is called in scalar context, being passed each value
661 in turn as both $_ and the only argument in the parameters, @_. The
662 values are then sorted according to string comparisons on the values
663 returned. This is equivalent to
664 sort { $a->name cmp $b->name } @people
666 except that it guarantees the name accessor will be executed only once
668 per value. One interesting use-case is to sort strings which may have
669 numbers embedded in them "naturally", rather than lexically.
670 sort_by { s/(\d+)/sprintf "%09d", $1/eg; $_ } @strings
672 This sorts strings by generating sort keys which zero-pad the embedded
674 numbers to some level (9 digits in this case), helping to ensure the
675 lexical sort puts them in the correct order.
676 nsort_by BLOCK LIST
678 Similar to sort_by but compares its key values numerically.
680 qsort BLOCK ARRAY
682 This sorts the given array in place using the given compare code.
684 Except for tiny compare code like "$a <=> $b", qsort is much faster
685 than Perl's "sort" depending on the version.
686 Compared 5.8 and 5.26:
688 my @rl;
690 for(my $i = 0; $i < 1E6; ++$i) { push @rl, rand(1E5) }
691 my $idx;
692 sub ext_cmp { $_[0] <=> $_[1] }
694 cmpthese( -60, {
696 'qsort' => sub {
697 my @qrl = @rl;
698 qsort { ext_cmp($a, $b) } @qrl;
699 $idx = bsearchidx { ext_cmp($_, $rl[0]) } @qrl
700 },
701 'reverse qsort' => sub {
702 my @qrl = @rl;
703 qsort { ext_cmp($b, $a) } @qrl;
704 $idx = bsearchidx { ext_cmp($rl[0], $_) } @qrl
705 },
706 'sort' => sub {
707 my @srl = @rl;
708 @srl = sort { ext_cmp($a, $b) } @srl;
709 $idx = bsearchidx { ext_cmp($_, $rl[0]) } @srl
710 },
711 'reverse sort' => sub {
712 my @srl = @rl;
713 @srl = sort { ext_cmp($b, $a) } @srl;
714 $idx = bsearchidx { ext_cmp($rl[0], $_) } @srl
715 },
716 });
717 5.8 results
719 s/iter reverse sort sort reverse qsort qsort
721 reverse sort 6.21 -- -0% -8% -10%
722 sort 6.19 0% -- -7% -10%
723 reverse qsort 5.73 8% 8% -- -2%
724 qsort 5.60 11% 11% 2% --
725 5.26 results
727 s/iter reverse sort sort reverse qsort qsort
729 reverse sort 4.54 -- -0% -96% -96%
730 sort 4.52 0% -- -96% -96%
731 reverse qsort 0.203 2139% 2131% -- -19%
732 qsort 0.164 2666% 2656% 24% --
733 Use it where external data sources might have to be compared (think of
735 Unix::Statgrab "tables").
736 "qsort" is available from List::MoreUtils::XS only. It's insane to
738 maintain a wrapper around Perl's sort nor having a pure Perl
739 implementation. One could create a flip-book in same speed as PP runs a
740 qsort.
741 Searching in sorted Lists
743 bsearch BLOCK LIST
744 Performs a binary search on LIST which must be a sorted list of values.
746 BLOCK must return a negative value if the current element (stored in
747 $_) is smaller, a positive value if it is bigger and zero if it
748 matches.
749 Returns a boolean value in scalar context. In list context, it returns
751 the element if it was found, otherwise the empty list.
752 bsearchidx BLOCK LIST
754 bsearch_index BLOCK LIST
756 Performs a binary search on LIST which must be a sorted list of values.
758 BLOCK must return a negative value if the current element (stored in
759 $_) is smaller, a positive value if it is bigger and zero if it
760 matches.
761 Returns the index of found element, otherwise -1.
763 "bsearch_index" is an alias for "bsearchidx".
765 lower_bound BLOCK LIST
767 Returns the index of the first element in LIST which does not compare
769 less than val. Technically it's the first element in LIST which does
770 not return a value below zero when passed to BLOCK.
771 @ids = (1, 1, 2, 2, 3, 3, 3, 3, 3, 3, 4, 4, 4, 4, 5, 5, 6, 7, 7, 7, 8, 8, 9, 9, 9, 9, 9, 11, 13, 13, 13, 17);
773 $lb = lower_bound { $_ <=> 2 } @ids; # returns 2
774 $lb = lower_bound { $_ <=> 4 } @ids; # returns 10
775 lower_bound has a complexity of O(log n).
777 upper_bound BLOCK LIST
779 Returns the index of the first element in LIST which does not compare
781 greater than val. Technically it's the first element in LIST which does
782 not return a value below or equal to zero when passed to BLOCK.
783 @ids = (1, 1, 2, 2, 3, 3, 3, 3, 3, 3, 4, 4, 4, 4, 5, 5, 6, 7, 7, 7, 8, 8, 9, 9, 9, 9, 9, 11, 13, 13, 13, 17);
785 $lb = upper_bound { $_ <=> 2 } @ids; # returns 4
786 $lb = upper_bound { $_ <=> 4 } @ids; # returns 14
787 upper_bound has a complexity of O(log n).
789 equal_range BLOCK LIST
791 Returns a pair of indices containing the lower_bound and the
793 upper_bound.
794 Operations on sorted Lists
796 binsert BLOCK ITEM LIST
797 bsearch_insert BLOCK ITEM LIST
799 Performs a binary search on LIST which must be a sorted list of values.
801 BLOCK must return a negative value if the current element (stored in
802 $_) is smaller, a positive value if it is bigger and zero if it
803 matches.
804 ITEM is inserted at the index where the ITEM should be placed (based on
806 above search). That means, it's inserted before the next bigger
807 element.
808 @l = (2,3,5,7);
810 binsert { $_ <=> 4 } 4, @l; # @l = (2,3,4,5,7)
811 binsert { $_ <=> 6 } 42, @l; # @l = (2,3,4,42,7)
812 You take care that the inserted element matches the compare result.
814 bremove BLOCK LIST
816 bsearch_remove BLOCK LIST
818 Performs a binary search on LIST which must be a sorted list of values.
820 BLOCK must return a negative value if the current element (stored in
821 $_) is smaller, a positive value if it is bigger and zero if it
822 matches.
823 The item at the found position is removed and returned.
825 @l = (2,3,4,5,7);
827 bremove { $_ <=> 4 }, @l; # @l = (2,3,5,7);
828 Counting and calculation
830 true BLOCK LIST
831 Counts the number of elements in LIST for which the criterion in BLOCK
833 is true. Sets $_ for each item in LIST in turn:
834 printf "%i item(s) are defined", true { defined($_) } @list;
836 false BLOCK LIST
838 Counts the number of elements in LIST for which the criterion in BLOCK
840 is false. Sets $_ for each item in LIST in turn:
841 printf "%i item(s) are not defined", false { defined($_) } @list;
843 reduce_0 BLOCK LIST
845 Reduce LIST by calling BLOCK in scalar context for each element of
847 LIST. $a contains the progressional result and is initialized with 0.
848 $b contains the current processed element of LIST and $_ contains the
849 index of the element in $b.
850 The idea behind reduce_0 is summation (addition of a sequence of
852 numbers).
853 reduce_1 BLOCK LIST
855 Reduce LIST by calling BLOCK in scalar context for each element of
857 LIST. $a contains the progressional result and is initialized with 1.
858 $b contains the current processed element of LIST and $_ contains the
859 index of the element in $b.
860 The idea behind reduce_1 is product of a sequence of numbers.
862 reduce_u BLOCK LIST
864 Reduce LIST by calling BLOCK in scalar context for each element of
866 LIST. $a contains the progressional result and is uninitialized. $b
867 contains the current processed element of LIST and $_ contains the
868 index of the element in $b.
869 This function has been added if one might need the extra of the index
871 value but need an individual initialization.
872 Use with caution: In most cases "reduce" in List::Util will do the job
874 better.
875 minmax LIST
877 Calculates the minimum and maximum of LIST and returns a two element
879 list with the first element being the minimum and the second the
880 maximum. Returns the empty list if LIST was empty.
881 The "minmax" algorithm differs from a naive iteration over the list
883 where each element is compared to two values being the so far
884 calculated min and max value in that it only requires 3n/2 - 2
885 comparisons. Thus it is the most efficient possible algorithm.
886 However, the Perl implementation of it has some overhead simply due to
888 the fact that there are more lines of Perl code involved. Therefore,
889 LIST needs to be fairly big in order for "minmax" to win over a naive
890 implementation. This limitation does not apply to the XS version.
891 minmaxstr LIST
893 Computes the minimum and maximum of LIST using string compare and
895 returns a two element list with the first element being the minimum and
896 the second the maximum. Returns the empty list if LIST was empty.
897 The implementation is similar to "minmax".
899
901 When "LIST_MOREUTILS_PP" is set, the module will always use the pure-
902 Perl implementation and not the XS one. This environment variable is
903 really just there for the test-suite to force testing the Perl
904 implementation, and possibly for reporting of bugs. I don't see any
905 reason to use it in a production environment.
906
908 The maintenance goal is to preserve the documented semantics of the
909 API; bug fixes that bring actual behavior in line with semantics are
910 allowed. New API functions may be added over time. If a backwards
911 incompatible change is unavoidable, we will attempt to provide support
912 for the legacy API using the same export tag mechanism currently in
913 place.
914 This module attempts to use few non-core dependencies. Non-core
916 configuration and testing modules will be bundled when reasonable; run-
917 time dependencies will be added only if they deliver substantial
918 benefit.
919
921 While contributions are appreciated, a contribution should not cause
922 more effort for the maintainer than the contribution itself saves (see
923 Open Source Contribution Etiquette
924 <http://tirania.org/blog/archive/2010/Dec-31.html>).
925 To get more familiar where help could be needed - see
927 List::MoreUtils::Contributing.
928
930 There is a problem with a bug in 5.6.x perls. It is a syntax error to
931 write things like:
932 my @x = apply { s/foo/bar/ } qw{ foo bar baz };
934 It has to be written as either
936 my @x = apply { s/foo/bar/ } 'foo', 'bar', 'baz';
938 or
940 my @x = apply { s/foo/bar/ } my @dummy = qw/foo bar baz/;
942 Perl 5.5.x and Perl 5.8.x don't suffer from this limitation.
944 If you have a functionality that you could imagine being in this
946 module, please drop me a line. This module's policy will be less strict
947 than List::Util's when it comes to additions as it isn't a core module.
948 When you report bugs, it would be nice if you could additionally give
950 me the output of your program with the environment variable
951 "LIST_MOREUTILS_PP" set to a true value. That way I know where to look
952 for the problem (in XS, pure-Perl or possibly both).
953
955 Bugs should always be submitted via the CPAN bug tracker.
956 You can find documentation for this module with the perldoc command.
958 perldoc List::MoreUtils
960 You can also look for information at:
962 • RT: CPAN's request tracker
964 <https://rt.cpan.org/Dist/Display.html?Name=List-MoreUtils>
966 • AnnoCPAN: Annotated CPAN documentation
968 <http://annocpan.org/dist/List-MoreUtils>
970 • CPAN Ratings
972 <http://cpanratings.perl.org/dist/List-MoreUtils>
974 • MetaCPAN
976 <https://metacpan.org/release/List-MoreUtils>
978 • CPAN Search
980 <http://search.cpan.org/dist/List-MoreUtils/>
982 • Git Repository
984 <https://github.com/perl5-utils/List-MoreUtils>
986 Where can I go for help?
988 If you have a bug report, a patch or a suggestion, please open a new
989 report ticket at CPAN (but please check previous reports first in case
990 your issue has already been addressed) or open an issue on GitHub.
991 Report tickets should contain a detailed description of the bug or
993 enhancement request and at least an easily verifiable way of
994 reproducing the issue or fix. Patches are always welcome, too - and
995 it's cheap to send pull-requests on GitHub. Please keep in mind that
996 code changes are more likely accepted when they're bundled with an
997 approving test.
998 If you think you've found a bug then please read "How to Report Bugs
1000 Effectively" by Simon Tatham:
1001 <http://www.chiark.greenend.org.uk/~sgtatham/bugs.html>.
1002 Where can I go for help with a concrete version?
1004 Bugs and feature requests are accepted against the latest version only.
1005 To get patches for earlier versions, you need to get an agreement with
1006 a developer of your choice - who may or not report the issue and a
1007 suggested fix upstream (depends on the license you have chosen).
1008 Business support and maintenance
1010 Generally, in volunteered projects, there is no right for support.
1011 While every maintainer is happy to improve the provided software, spare
1012 time is limited.
1013 For those who have a use case which requires guaranteed support, one of
1015 the maintainers should be hired or contracted. For business support
1016 you can contact Jens via his CPAN email address rehsackATcpan.org.
1017 Please keep in mind that business support is neither available for free
1018 nor are you eligible to receive any support based on the license
1019 distributed with this package.
1020
1022 Tassilo von Parseval
1023 Credits go to a number of people: Steve Purkis for giving me namespace
1024 advice and James Keenan and Terrence Branno for their effort of keeping
1025 the CPAN tidier by making List::Utils obsolete.
1026 Brian McCauley suggested the inclusion of apply() and provided the
1028 pure-Perl implementation for it.
1029 Eric J. Roode asked me to add all functions from his module
1031 "List::MoreUtil" into this one. With minor modifications, the pure-Perl
1032 implementations of those are by him.
1033 The bunch of people who almost immediately pointed out the many
1035 problems with the glitchy 0.07 release (Slaven Rezic, Ron Savage, CPAN
1036 testers).
1037 A particularly nasty memory leak was spotted by Thomas A. Lowery.
1039 Lars Thegler made me aware of problems with older Perl versions.
1041 Anno Siegel de-orphaned each_arrayref().
1043 David Filmer made me aware of a problem in each_arrayref that could
1045 ultimately lead to a segfault.
1046 Ricardo Signes suggested the inclusion of part() and provided the Perl-
1048 implementation.
1049 Robin Huston kindly fixed a bug in perl's MULTICALL API to make the XS-
1051 implementation of part() work.
1052 Jens Rehsack
1054 Credits goes to all people contributing feedback during the v0.400
1055 development releases.
1056 Special thanks goes to David Golden who spent a lot of effort to
1058 develop a design to support current state of CPAN as well as ancient
1059 software somewhere in the dark. He also contributed a lot of patches to
1060 refactor the API frontend to welcome any user of List::MoreUtils - from
1061 ancient past to recently last used.
1062 Toby Inkster provided a lot of useful feedback for sane importer code
1064 and was a nice sounding board for API discussions.
1065 Peter Rabbitson provided a sane git repository setup containing entire
1067 package history.
1068
1070 A pile of requests from other people is still pending further
1071 processing in my mailbox. This includes:
1072 • delete_index
1074 • random_item
1076 • random_item_delete_index
1078 • list_diff_hash
1080 • list_diff_inboth
1082 • list_diff_infirst
1084 • list_diff_insecond
1086 These were all suggested by Dan Muey.
1088 • listify
1090 Always return a flat list when either a simple scalar value was
1092 passed or an array-reference. Suggested by Mark Summersault.
1093
1095 List::Util, List::AllUtils, List::UtilsBy
1096
1098 Jens Rehsack <rehsack AT cpan.org>
1099 Adam Kennedy <adamk@cpan.org>
1101 Tassilo von Parseval <tassilo.von.parseval@rwth-aachen.de>
1103
1105 Some parts copyright 2011 Aaron Crane.
1106 Copyright 2004 - 2010 by Tassilo von Parseval
1108 Copyright 2013 - 2017 by Jens Rehsack
1110 All code added with 0.417 or later is licensed under the Apache
1112 License, Version 2.0 (the "License"); you may not use this file except
1113 in compliance with the License. You may obtain a copy of the License at
1114 http://www.apache.org/licenses/LICENSE-2.0
1116 Unless required by applicable law or agreed to in writing, software
1118 distributed under the License is distributed on an "AS IS" BASIS,
1119 WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
1120 implied. See the License for the specific language governing
1121 permissions and limitations under the License.
1122 All code until 0.416 is licensed under the same terms as Perl itself,
1124 either Perl version 5.8.4 or, at your option, any later version of Perl
1125 5 you may have available.
1126perl v5.36.0 2023-01-20 List::MoreUtils(3)