Sort using a custom comparator: Difference between revisions
Line 600: | Line 600: | ||
function(const Left, Right: string): Integer |
function(const Left, Right: string): Integer |
||
begin |
begin |
||
//Returns ('Here', 'are', 'be', 'sample', 'some', 'sorted', 'strings', 'to') |
|||
⚫ | |||
//Result := CompareStr(Left, Right); |
|||
//Returns ('are', 'be', 'Here', 'sample', 'some', 'sorted', 'strings', 'to') |
|||
⚫ | |||
end)); |
end)); |
||
end.</lang> |
end.</lang> |
Revision as of 07:26, 30 June 2017
You are encouraged to solve this task according to the task description, using any language you may know.
- Task
Sort an array (or list) of strings in order of descending length, and in ascending lexicographic order for strings of equal length.
Use a sorting facility provided by the language/library, combined with your own callback comparison function.
Note: Lexicographic order is case-insensitive.
Ada
<lang ada> with Ada.Text_Io; use Ada.Text_Io; with Ada.Strings.Unbounded; use Ada.Strings.Unbounded; with Gnat.Heap_Sort_G;
procedure Custom_Compare is
type StringArrayType is array (Natural range <>) of Unbounded_String; Strings : StringArrayType := (Null_Unbounded_String, To_Unbounded_String("this"), To_Unbounded_String("is"), To_Unbounded_String("a"), To_Unbounded_String("set"), To_Unbounded_String("of"), To_Unbounded_String("strings"), To_Unbounded_String("to"), To_Unbounded_String("sort"), To_Unbounded_String("This"), To_Unbounded_String("Is"), To_Unbounded_String("A"), To_Unbounded_String("Set"), To_Unbounded_String("Of"), To_Unbounded_String("Strings"), To_Unbounded_String("To"), To_Unbounded_String("Sort")); procedure Move (From, To : in Natural) is begin Strings(To) := Strings(From); end Move; function UpCase (Char : in Character) return Character is Temp : Character; begin if Char >= 'a' and Char <= 'z' then Temp := Character'Val(Character'Pos(Char) - Character'Pos('a') + Character'Pos('A')); else Temp := Char; end if; return Temp; end UpCase; function Lt (Op1, Op2 : Natural) return Boolean is Temp, Len : Natural; begin Len := Length(Strings(Op1)); Temp := Length(Strings(Op2)); if Len < Temp then return False; elsif Len > Temp then return True; end if;
declare S1, S2 : String(1..Len); begin S1 := To_String(Strings(Op1)); S2 := To_String(Strings(Op2)); Put("Same size: "); Put(S1); Put(" "); Put(S2); Put(" "); for I in S1'Range loop Put(UpCase(S1(I))); Put(UpCase(S2(I))); if UpCase(S1(I)) = UpCase(S2(I)) then null; elsif UpCase(S1(I)) < UpCase(S2(I)) then Put(" LT"); New_Line; return True; else return False; end if; end loop; Put(" GTE"); New_Line; return False; end; end Lt; procedure Put (Arr : in StringArrayType) is begin for I in 1..Arr'Length-1 loop Put(To_String(Arr(I))); New_Line; end loop; end Put; package Heap is new Gnat.Heap_Sort_G(Move, Lt); use Heap;
begin
Put_Line("Unsorted list:"); Put(Strings); New_Line; Sort(16); New_Line; Put_Line("Sorted list:"); Put(Strings);
end Custom_Compare;</lang>
- Output:
Unsorted list: this is a set of strings to sort This Is A Set Of Strings To Sort Sorted list: strings Strings sort Sort this This Set set is Is Of of to To a A
AutoHotkey
<lang AutoHotkey>numbers = 5,3,7,9,1,13,999,-4 strings = Here,are,some,sample,strings,to,be,sorted Sort, numbers, F IntegerSort D, Sort, strings, F StringLengthSort D, msgbox % numbers msgbox % strings
IntegerSort(a1, a2) { return a2 - a1 }
StringLengthSort(a1, a2){ return strlen(a1) - strlen(a2) }</lang>
AWK
<lang AWK>
- syntax: GAWK -f SORT_USING_A_CUSTOM_COMPARATOR.AWK
- sorting:
- PROCINFO["sorted_in"] is used by GAWK
- SORTTYPE is used by Thompson Automation's TAWK
BEGIN {
words = "This Is A Set Of Strings To Sort duplicated" n = split(words " " tolower(words),tmp_arr," ") print("unsorted:") for (i=1; i<=n; i++) { word = tmp_arr[i] arr[length(word)][word]++ print(word) } print("\nsorted:") PROCINFO["sorted_in"] = "@ind_num_desc" ; SORTTYPE = 9 for (i in arr) { PROCINFO["sorted_in"] = "@ind_str_asc" ; SORTTYPE = 2 for (j in arr[i]) { for (k=1; k<=arr[i][j]; k++) { print(j) } } } exit(0)
} </lang>
- Output:
unsorted: This Is A Set Of Strings To Sort duplicated this is a set of strings to sort duplicated sorted: duplicated duplicated Strings strings Sort This sort this Set set Is Of To is of to A a
Babel
To sort ASCII strings, use the strsort or lexsort utilities to sort alphabetically and lexicographically, respectively.
<lang babel>babel> ("Here" "are" "some" "sample" "strings" "to" "be" "sorted") strsort ! lsstr ! ( "Here" "are" "be" "sample" "some" "sorted" "strings" "to" ) babel> ("Here" "are" "some" "sample" "strings" "to" "be" "sorted") lexsort ! lsstr ! ( "be" "to" "are" "Here" "some" "sample" "sorted" "strings" )</lang>
If you want to sort UTF-8 encoded Unicode strings, first convert to array-string form using the str2ar operator, then sort using the strcmp operator. To sort lexicographically, use the arcmp operator. The following examples illustrate each case:
<lang babel>babel> ("Here" "are" "some" "sample" "strings" "to" "be" "sorted") {str2ar} over ! {strcmp 0 lt?} lssort ! {ar2str} over ! lsstr ! ( "Here" "are" "be" "some" "sample" "sorted" "strings" "to" ) babel> ("Here" "are" "some" "sample" "strings" "to" "be" "sorted") {str2ar} over ! {arcmp 0 lt?} lssort ! {ar2str} over ! lsstr ! ( "be" "to" "are" "Here" "some" "sample" "sorted" "strings" )</lang>
You can sort a list of any kind of structure you like using the lssort utility. Use the lt? numerical comparison operator for sorting numerical lists:
<lang babel>babel> ( 5 6 8 4 5 3 9 9 4 9 ) {lt?} lssort ! lsnum ! ( 3 4 4 5 5 6 8 9 9 9 )</lang>
You can even shuffle a list with lssort using the randlf operator (your results will probably differ):
<lang babel>babel> (1 2 3 4 5 6 7 8 9) {1 randlf 2 rem} lssort ! lsnum ! ( 7 5 9 6 2 4 3 1 8 )</lang>
To sort complex objects, you need to access the relevant field in each object, and then provide the result of comparing them. For example, to sort a list of pairs by first number:
<lang babel> babel> 20 lsrange ! {1 randlf 2 rem} lssort ! 2 group ! --> this creates a shuffled list of pairs babel> dup {lsnum !} ... --> display the shuffled list, pair-by-pair ( 11 10 ) ( 15 13 ) ( 12 16 ) ( 17 3 ) ( 14 5 ) ( 4 19 ) ( 18 9 ) ( 1 7 ) ( 8 6 ) ( 0 2 ) babel> {<- car -> car lt? } lssort ! --> sort the list by first element of each pair babel> dup {lsnum !} ... --> display the sorted list, pair-by-pair ( 0 2 ) ( 1 7 ) ( 4 19 ) ( 8 6 ) ( 11 10 ) ( 12 16 ) ( 14 5 ) ( 15 13 ) ( 17 3 ) ( 18 9 )</lang>
Burlesque
<lang burlesque> blsq ) {"acb" "Abc" "Acb" "acc" "ADD"}>< {"ADD" "Abc" "Acb" "acb" "acc"} blsq ) {"acb" "Abc" "Acb" "acc" "ADD"}(zz)CMsb {"Abc" "acb" "Acb" "acc" "ADD"} </lang>
C
<lang c>#include <stdlib.h> /* for qsort */
- include <string.h> /* for strlen */
- include <strings.h> /* for strcasecmp */
int mycmp(const void *s1, const void *s2) {
const char *l = *(const char **)s1, *r = *(const char **)s2; size_t ll = strlen(l), lr = strlen(r);
if (ll > lr) return -1; if (ll < lr) return 1; return strcasecmp(l, r);
}
int main() {
const char *strings[] = { "Here", "are", "some", "sample", "strings", "to", "be", "sorted" };
qsort(strings, sizeof(strings)/sizeof(*strings), sizeof(*strings), mycmp); return 0;
}</lang>
C#
C# allows you to specify a custom compare to the built in sort method on a list
<lang csharp>using System; using System.Collections.Generic;
namespace RosettaCode {
class SortCustomComparator { // Driver program public void CustomSort() { String[] items = { "Here", "are", "some", "sample", "strings", "to", "be", "sorted" }; List<String> list = new List<string>(items);
DisplayList("Unsorted", list); list.Sort(CustomCompare); DisplayList("Descending Length", list);
list.Sort(); DisplayList("Ascending order", list); }
// Custom compare public int CustomCompare(String x, String y) { int result = -x.Length.CompareTo(y.Length); if (result == 0) { result = x.ToLower().CompareTo(y.ToLower()); }
return result; }
// Output routine public void DisplayList(String header, List<String> theList) { Console.WriteLine(header); Console.WriteLine("".PadLeft(header.Length, '*')); foreach (String str in theList) { Console.WriteLine(str); } Console.WriteLine(); } }
}</lang>
- Output:
Unsorted ******** Here are some sample strings to be sorted Descending Length ***************** strings sample sorted Here some are be to Ascending order *************** are be Here sample some sorted strings to
Alternative using Linq (.NET 3.5)
<lang csharp>using System; using System.Collections.Generic; using System.Linq;
namespace RosettaCode { class SortCustomComparator { // Driver program public void CustomSort() { List<string> list = new List<string> { "Here", "are", "some", "sample", "strings", "to", "be", "sorted" };
DisplayList("Unsorted", list);
var descOrdered = from l in list orderby l.Length descending select l; DisplayList("Descending Length", descOrdered);
var ascOrdered = from l in list orderby l select l; DisplayList("Ascending order", ascOrdered); }
// Output routine public void DisplayList(String header, IEnumerable<string> theList) { Console.WriteLine(header); Console.WriteLine("".PadLeft(header.Length, '*')); foreach (String str in theList) { Console.WriteLine(str); } Console.WriteLine(); } } } </lang>
C++
<lang cpp>#include <algorithm>
- include <string>
- include <cctype>
// compare character case-insensitive struct icompare_char {
bool operator()(char c1, char c2) { return std::toupper(c1) < std::toupper(c2); }
};
// return true if s1 comes before s2 struct compare {
bool operator()(std::string const& s1, std::string const& s2) { if (s1.length() > s2.length()) return true; if (s1.length() < s2.length()) return false; return std::lexicographical_compare(s1.begin(), s1.end(), s2.begin(), s2.end(), icompare_char()); }
};
int main() {
std::string strings[8] = {"Here", "are", "some", "sample", "strings", "to", "be", "sorted"}; std::sort(strings, strings+8, compare()); return 0;
}</lang>
Ceylon
<lang ceylon>shared void run() {
value strings = [ "Cat", "apple", "Adam", "zero", "Xmas", "quit", "Level", "add", "Actor", "base", "butter" ];
value sorted = strings.sort((String x, String y) => if(x.size == y.size) then increasing(x.lowercased, y.lowercased) else decreasing(x.size, y.size));
sorted.each(print); }</lang>
Clean
<lang clean>import StdEnv
less s1 s2
| size s1 > size s2 = True | size s1 < size s2 = False | otherwise = lower s1 < lower s2
where
lower :: String -> String lower s = {toLower c \\ c <-: s}
Start = sortBy less ["This", "is", "a", "set", "of", "strings", "to", "sort"]</lang>
Clojure
Clojure's sort function has a 2-argument version where the first argument is a java.util.Comparator, and the second is the collection to be sorted. Thus the heart of this version is a comparator function that satisfies the problem spec. What makes this work is that all Clojure functions (thus rosetta-code defined here) implement the java.util.Comparator interface. <lang clojure>(defn rosetta-compare [s1 s2]
(let [len1 (count s1), len2 (count s2)] (if (= len1 len2) (compare (.toLowerCase s1) (.toLowerCase s2)) (- len2 len1))))
(println
(sort rosetta-compare ["Here" "are" "some" "sample" "strings" "to" "be" "sorted"]))</lang>
- Output:
(strings sample sorted Here some are be to)
An alternative, using sort-by: <lang clojure>(sort-by (juxt (comp - count) #(.toLowerCase %))
["Here" "are" "some" "sample" "strings" "to" "be" "sorted"])</lang>
Common Lisp
In Common Lisp, the sort function takes a "less than" predicate that is used as the comparator. This parameter can be any two-argument function. Note: Common Lisp's sort function is destructive; for lists you should not use the original list afterwards, you should only use the return value. This also means you don't call it directly on constant data.
For example, to sort strings case-insensitively in ascending order:
<lang lisp>CL-USER> (defvar *strings*
(list "Cat" "apple" "Adam" "zero" "Xmas" "quit" "Level" "add" "Actor" "base" "butter"))
- STRINGS*
CL-USER> (sort *strings* #'string-lessp) ("Actor" "Adam" "add" "apple" "base" "butter" "Cat" "Level" "quit" "Xmas" "zero")</lang>
You can also provide an optional key function which maps each element to a key. The keys are then compared using the comparator. For example, to sort strings by length in descending order:
<lang lisp>CL-USER> (defvar *strings*
(list "Cat" "apple" "Adam" "zero" "Xmas" "quit" "Level" "add" "Actor" "base" "butter"))
- STRINGS*
CL-USER> (sort *strings* #'> :key #'length) ("butter" "apple" "Level" "Actor" "Adam" "zero" "Xmas" "quit" "base"
"Cat" "add")</lang>
D
<lang d>import std.stdio, std.string, std.algorithm, std.typecons;
void main() {
"here are Some sample strings to be sorted" .split .schwartzSort!q{ tuple(-a.length, a.toUpper) } .writeln;
}</lang>
- Output:
["strings", "sample", "sorted", "here", "Some", "are", "be", "to"]
Alternative Version
The more natural and efficient way to solve this problem is to use std.algorith.multiSort
.
But currently it's less convenient because it can't be used with the UFCSyntax (same output):
<lang d>void main() {
import std.stdio, std.string, std.algorithm;
auto parts = "here are Some sample strings to be sorted".split; parts.multiSort!(q{a.length > b.length}, q{a.toUpper < b.toUpper}); parts.writeln;
}</lang>
Delphi
<lang Delphi>program SortWithCustomComparator;
{$APPTYPE CONSOLE}
uses SysUtils, Types, Generics.Collections, Generics.Defaults;
var
lArray: TStringDynArray;
begin
lArray := TStringDynArray.Create('Here', 'are', 'some', 'sample', 'strings', 'to', 'be', 'sorted');
TArray.Sort<string>(lArray , TDelegatedComparer<string>.Construct( function(const Left, Right: string): Integer begin //Returns ('Here', 'are', 'be', 'sample', 'some', 'sorted', 'strings', 'to') //Result := CompareStr(Left, Right);
//Returns ('are', 'be', 'Here', 'sample', 'some', 'sorted', 'strings', 'to') Result := CompareText(Left, Right); end));
end.</lang>
E
<lang e>/** returns a if it is nonzero, otherwise b() */ def nonzeroOr(a, b) { return if (a.isZero()) { b() } else { a } }
["Here", "are", "some", "sample", "strings", "to", "be", "sorted"] \
.sort(fn a, b { nonzeroOr(b.size().op__cmp(a.size()), fn { a.compareToIgnoreCase(b) }) })</lang>
EGL
<lang EGL>program SortExample
function main() test1 string[] = ["Here", "are", "some", "sample", "strings", "to", "be", "sorted"];
test1.sort(sortFunction);
SysLib.writeStdout("Test 1:"); for(i int from 1 to test1.getSize())
SysLib.writeStdout(test1[i]);
end
test2 string[] = ["Cat", "apple", "Adam", "zero", "Xmas", "quit", "Level", "add", "Actor", "base", "butter"]; test2.sort(sortFunction);
SysLib.writeStdout("Test 2:"); for(i int from 1 to test2.getSize()) SysLib.writeStdout(test2[i]); end
end function sortFunction(a any in, b any in) returns (int)
result int = (b as string).length() - (a as string).length(); if (result == 0)
case
when ((a as string).toLowerCase() > (b as string).toLowerCase()) result = 1; when ((a as string).toLowerCase() < (b as string).toLowerCase()) result = -1; otherwise result = 0; end end
return result; end
end</lang>
- Output:
Test 1: strings sample sorted Here some are be to Test 2: butter Actor apple Level Adam base quit Xmas zero add Cat
Elixir
<lang elixir>strs = ~w[this is a set of strings to sort This Is A Set Of Strings To Sort]
comparator = fn s1,s2 -> if String.length(s1)==String.length(s2),
do: String.downcase(s1) <= String.downcase(s2), else: String.length(s1) >= String.length(s2) end
IO.inspect Enum.sort(strs, comparator)
- or
IO.inspect Enum.sort_by(strs, fn str -> {-String.length(str), String.downcase(str)} end)</lang>
- Output:
["strings", "Strings", "sort", "Sort", "this", "This", "set", "Set", "is", "Is", "of", "Of", "to", "To", "a", "A"]
Erlang
<lang Erlang> -module( sort_using_custom_comparator ).
-export( [task/0] ).
task() -> lists:sort( fun longest_first_case_insensitive/2, ["this", "is", "a", "set", "of", "strings", "to", "sort", "This", "Is", "A", "Set", "Of", "Strings", "To", "Sort"] ).
longest_first_case_insensitive( String1, String2 ) when erlang:length(String1) =:= erlang:length(String2) -> string:to_lower(String1) < string:to_lower(String2); longest_first_case_insensitive( String1, String2 ) when erlang:length(String1) =< erlang:length(String2) -> false; longest_first_case_insensitive( _String1, _String2 ) -> true. </lang>
- Output:
9> sort_using_custom_comparator:task(). ["Strings","strings","Sort","sort","This","this","Set", "set","Is","is","Of","of","To","to","A","a"]
Euphoria
<lang euphoria>include sort.e include wildcard.e include misc.e
function my_compare(sequence a, sequence b)
if length(a)!=length(b) then return -compare(length(a),length(b)) else return compare(lower(a),lower(b)) end if
end function
sequence strings strings = reverse({ "Here", "are", "some", "sample", "strings", "to", "be", "sorted" })
puts(1,"Unsorted:\n") pretty_print(1,strings,{2})
puts(1,"\n\nSorted:\n") pretty_print(1,custom_sort(routine_id("my_compare"),strings),{2})</lang>
- Output:
Unsorted: { "sorted", "be", "to", "strings", "sample", "some", "are", "Here" } Sorted: { "strings", "sample", "sorted", "Here", "some", "are", "be", "to" }
F#
<lang fsharp>let myCompare (s1:string) (s2:string) =
match compare s2.Length s1.Length with | 0 -> compare (s1.ToLower()) (s2.ToLower()) | X -> X
let strings = ["Here"; "are"; "some"; "sample"; "strings"; "to"; "be"; "sorted"]
let sortedStrings = List.sortWith myCompare strings
printfn "%A" sortedStrings</lang>
- Output:
["strings"; "sample"; "sorted"; "Here"; "some"; "are"; "be"; "to"]
Factor
<lang factor>: my-compare ( s1 s2 -- <=> )
2dup [ length ] compare invert-comparison dup +eq+ = [ drop [ >lower ] compare ] [ 2nip ] if ;
{ "this" "is" "a" "set" "of" "strings" "to" "sort" } [ my-compare ] sort</lang>
Fantom
The List's sort method can be customised using a custom comparator. This is a method which returns an Int: -1 for less than, 0 for equal, +1 for greater than.
<lang fantom> class Main {
public static Void main () { // sample strings from Lisp example strs := ["Cat", "apple", "Adam", "zero", "Xmas", "quit", "Level", "add", "Actor", "base", "butter"]
sorted := strs.dup // make a copy of original list sorted.sort |Str a, Str b -> Int| // sort using custom comparator { if (b.size == a.size) // if size is same return a.compareIgnoreCase(b) // then sort in ascending lexicographic order, ignoring case else return b.size <=> a.size // else sort in descending size order } echo ("Started with : " + strs.join(" ")) echo ("Finished with: " + sorted.join(" ")) }
} </lang>
- Output:
$ fan comparator-sort.fan Started with : Cat apple Adam zero Xmas quit Level add Actor base butter Finished with: butter Actor apple Level Adam base quit Xmas zero add Cat
Fortran
Fortran does not have builtin to sort arrays (of numbers or strings), with or without custom comparator; so we need modifying e.g. this code in order to handle strings and to accept a custom comparator.
<lang fortran>module sorts_with_custom_comparator
implicit none
contains
subroutine a_sort(a, cc) character(len=*), dimension(:), intent(inout) :: a interface integer function cc(a, b) character(len=*), intent(in) :: a, b end function cc end interface integer :: i, j, increment character(len=max(len(a), 10)) :: temp increment = size(a) / 2 do while ( increment > 0 ) do i = increment+1, size(a) j = i temp = a(i) do while ( j >= increment+1 .and. cc(a(j-increment), temp) > 0) a(j) = a(j-increment) j = j - increment end do a(j) = temp end do if ( increment == 2 ) then increment = 1 else increment = increment * 5 / 11 end if end do end subroutine a_sort
end module sorts_with_custom_comparator</lang>
Then we have to put our custom comparator in a module (to_lower is defined here):
<lang fortran>module comparators
implicit none
contains
integer function my_compare(a, b) character(len=*), intent(in) :: a, b
character(len=max(len(a),len(b))) :: a1, b1
a1 = a b1 = b call to_lower(b1) call to_lower(a1) if ( len(trim(a)) > len(trim(b)) ) then my_compare = -1 elseif ( len(trim(a)) == len(trim(b)) ) then if ( a1 > b1 ) then my_compare = 1 else my_compare = -1 end if else my_compare = 1 end if end function my_compare
end module comparators</lang>
At the end, we can test these:
<lang fortran>program CustomComparator
use comparators use sorts_with_custom_comparator implicit none
character(len=100), dimension(8) :: str integer :: i
str = (/ "this", "is", "an", "array", "of", "strings", "to", "sort" /) call a_sort(str, my_compare)
do i = 1, size(str) print *, trim(str(i)) end do
end program CustomComparator</lang>
FreeBASIC
<lang freebasic>' version 23-10-2016 ' compile with: fbc -s console
- Include Once "crt/stdlib.bi" ' for qsort
Function mycmp Cdecl (s1 As Any Pointer, s2 As Any Pointer) As Long
' -1 no swap first element before second element ' 0 no swap needed, don't care ' 1 swap first element after second element
Dim As String str1 = *Cast(String Ptr, s1) Dim As String str2 = *Cast(String Ptr, s2)
Dim As Long l1 = Len(str1), l2 = Len(str2) If (l1 > l2) Then Return -1 ' descending If (l1 < l2) Then Return 1 '
' there equal length, sort ascending If UCase(str1) = UCase(str2) Then If str1 > str2 Then Return 1 Else If UCase(str1) > UCase(str2) Then Return 1 End If Return 0
End Function
' ------=< MAIN >=------
Dim As String words(0 To ...) = {"Here", "are", "some", "sample", _
"strings", "to", "be", "sorted" }
Dim As ULong array_size = UBound(words) - LBound(words) + 1
qsort(@words(0), array_size, SizeOf(String), @mycmp)
For i As Integer = 0 To UBound(words)
Print words(i)
Next Print
' empty keyboard buffer While InKey <> "" : Wend Print : Print "hit any key to end program" Sleep End</lang>
- Output:
strings sample sorted Here some are be to
FunL
<lang funl>def preceeds( a, b ) = b.length() < a.length() or b.length() == a.length() and a.compareToIgnoreCase( b ) < 0
println( ["here", "are", "Some", "sample", "strings", "to", "be", "sorted"].sortWith(preceeds) )</lang>
- Output:
["strings", "sample", "sorted", "here", "Some", "are", "be", "to"]
Go
<lang go>package main
import (
"fmt" "sort" "strings"
)
type sortable []string
func (s sortable) Len() int { return len(s) } func (s sortable) Swap(i, j int) { s[i], s[j] = s[j], s[i] } func (s sortable) Less(i, j int) bool {
a, b := s[i], s[j] if len(a) != len(b) { return len(a) > len(b) } return strings.ToLower(a) < strings.ToLower(b)
}
func main() {
var s sortable = strings.Fields("To tell your name the livelong day To an admiring bog") fmt.Println(s, "(original)")
sort.Sort(s) fmt.Println(s, "(sorted)")
}</lang>
- Output:
[To tell your name the livelong day To an admiring bog] (original) [admiring livelong name tell your bog day the an To To] (sorted)
Groovy
The "custom comparator" is just a closure attached to the sort method invocation. <lang groovy>def strings = "Here are some sample strings to be sorted".split() strings.sort { x, y ->
y.length() <=> x.length() ?: x.compareToIgnoreCase(y)
} println strings</lang>
- Output:
[strings, sample, sorted, Here, some, are, be, to]
Haskell
<lang haskell>import Data.List (sortBy) import Data.Function (on) import Data.Char (toLower)
lengthThenAZ :: String -> String -> Ordering lengthThenAZ a b
| d == EQ = on compare (toLower <$>) a b | otherwise = d where d = on compare length a b
descLengthThenAZ :: String -> String -> Ordering descLengthThenAZ a b
| d == EQ = on compare (toLower <$>) a b | otherwise = d where d = on (flip compare) length a b
xs :: [String] xs = ["Here", "are", "some", "sample", "strings", "to", "be", "sorted"]
main :: IO () main =
mapM_ putStrLn [unlines $ sortBy lengthThenAZ xs, unlines $ sortBy descLengthThenAZ xs]</lang>
- Output:
be to are Here some sample sorted strings strings sample sorted Here some are be to
Or, we can simplify by using the Monoid instance for Ordering, to mappend (<>) one ordering to another:
<lang haskell>import Data.List (sortBy) import Data.Char (toLower) import Data.Ord (comparing) import Data.Monoid
-- the Ordering instance of mappend is defined to yield -- lexicographical ordering. -- instance Monoid Ordering where -- mempty = EQ -- LT `mappend` _ = LT -- EQ `mappend` y = y -- GT `mappend` _ = GT
xs :: [String] xs = ["Here", "are", "some", "sample", "strings", "to", "be", "sorted"]
lowerCase :: String -> String lowerCase = (toLower <$>)
main :: IO () main =
mapM_ putStrLn $ (unlines . flip sortBy xs) <$> [ comparing length <> comparing lowerCase -- Ascending length , flip (comparing length) <> comparing lowerCase -- Descending length ]</lang>
- Output:
be to are Here some sample sorted strings strings sample sorted Here some are be to
Icon and Unicon
<lang Icon>procedure main() #: demonstrate various ways to sort a list and string
write("Sorting Demo for custom comparator") L := ["Here", "are", "some", "sample", "strings", "to", "be", "sorted"] write(" Unsorted Input : ") every write(" ",image(!L)) shellsort(L,cmptask) # most of the RC sorts will work here write(" Sorted Output : ") every write(" ",image(!L))
end
procedure cmptask(a,b) # sort by descending length and ascending lexicographic order for strings of equal length if (*a > *b) | ((*a = *b) & (map(a) << map(b))) then return b end</lang>
Note(1): This example relies on the supporting procedures 'sortop', and 'demosort' in Bubble Sort.
Note(2): This example can utilize any of the sorting algorithms that share the same base code including: Bubble, Cocktail, Comb, Gnome, and Shell.
Note(3): Using 'map' in the 'cmptask' procedure would not be efficient on large lists.
- Output:
Sorting Demo for custom comparator Unsorted Input : "Here" "are" "some" "sample" "strings" "to" "be" "sorted" Sorted Output : "strings" "sample" "sorted" "Here" "some" "are" "be" "to"
J
Case-insensitivity is obtained using lower, a verb taken from Change string case. Standard utilities tolower or toupper may be substituted.
<lang j> mycmp=: 1 :'/:u'
length_and_lex =: (-@:# ; lower)&> strings=: 'Here';'are';'some';'sample';'strings';'to';'be';'sorted' length_and_lex mycmp strings
+-------+------+------+----+----+---+--+--+ |strings|sample|sorted|Here|some|are|be|to| +-------+------+------+----+----+---+--+--+</lang>
Java
<lang java5>import java.util.Comparator; import java.util.Arrays;
public class Test {
public static void main(String[] args) { String[] strings = {"Here", "are", "some", "sample", "strings", "to", "be", "sorted"};
Arrays.sort(strings, new Comparator<String>() { public int compare(String s1, String s2) { int c = s2.length() - s1.length(); if (c == 0) c = s1.compareToIgnoreCase(s2); return c; } });
for (String s: strings) System.out.print(s + " "); }
}</lang>
Same thing as above
<lang java5>import java.util.Comparator; import java.util.Arrays;
public class ComparatorTest {
public static void main(String[] args) { String[] strings = {"Here", "are", "some", "sample", "strings", "to", "be", "sorted"};
Arrays.sort(strings, (s1, s2) -> { int c = s2.length() - s1.length(); if (c == 0) c = s1.compareToIgnoreCase(s2); return c; });
for (String s: strings) System.out.print(s + " "); }
}</lang>
JavaScript
ES5
<lang javascript>function lengthSorter(a, b) {
var result = b.length - a.length; if (result == 0) result = a.localeCompare(b); return result;
}
var test = ["Here", "are", "some", "sample", "strings", "to", "be", "sorted"]; test.sort(lengthSorter); alert( test.join(' ') ); // strings sample sorted Here some are be to</lang>
Or, abstracting a little for simpler composition of compound and derived searches (ASC and DESC, secondary sorts):
<lang javascript>(function () {
'use strict';
// GENERIC FUNCTIONS FOR COMPARISONS
// Ordering :: ( LT | EQ | GT ) | ( -1 | 0 | 1 )
// compare :: a -> a -> Ordering var compare = function (a, b) { return a < b ? -1 : a > b ? 1 : 0; };
// mappendOrdering :: Ordering -> Ordering -> Ordering var mappendOrdering = function (a, b) { return a !== 0 ? a : b; };
// on :: (b -> b -> c) -> (a -> b) -> a -> a -> c var on = function (f, g) { return function (a, b) { return f(g(a), g(b)); }; };
// flip :: (a -> b -> c) -> b -> a -> c var flip = function (f) { return function (a, b) { return f.apply(null, [b, a]); }; };
// arrayCopy :: [a] -> [a] var arrayCopy = function (xs) { return xs.slice(0); };
// show :: a -> String var show = function (x) { return JSON.stringify(x, null, 2); };
// TEST var xs = ['Shanghai', 'Karachi', 'Beijing', 'Sao Paulo', 'Dhaka', 'Delhi', 'Lagos'];
var rs = [{ name: 'Shanghai', pop: 24.2 }, { name: 'Karachi', pop: 23.5 }, { name: 'Beijing', pop: 21.5 }, { name: 'Sao Paulo', pop: 24.2 }, { name: 'Dhaka', pop: 17.0 }, { name: 'Delhi', pop: 16.8 }, { name: 'Lagos', pop: 16.1 }];
// population :: Dictionary -> Num var population = function (x) { return x.pop; };
// length :: [a] -> Int var length = function (xs) { return xs.length; };
// toLower :: String -> String var toLower = function (s) { return s.toLowerCase(); };
// lengthThenAZ :: String -> String -> ( -1 | 0 | 1) var lengthThenAZ = function (a, b) { return mappendOrdering( on(compare, length)(a, b), on(compare, toLower)(a, b) ); };
// descLengthThenAZ :: String -> String -> ( -1 | 0 | 1) var descLengthThenAZ = function (a, b) { return mappendOrdering( on(flip(compare), length)(a, b), on(compare, toLower)(a, b) ); };
return show({ default: arrayCopy(xs) .sort(compare),
descendingDefault: arrayCopy(xs) .sort(flip(compare)),
byLengthThenAZ: arrayCopy(xs) .sort(lengthThenAZ),
byDescendingLengthThenZA: arrayCopy(xs) .sort(flip(lengthThenAZ)),
byDescendingLengthThenAZ: arrayCopy(xs) .sort(descLengthThenAZ),
byPopulation: arrayCopy(rs) .sort(on(compare, population)),
byDescendingPopulation: arrayCopy(rs) .sort(on(flip(compare), population)) });
})();</lang>
ES6
<lang JavaScript>(() => {
'use strict';
// GENERIC FUNCTIONS FOR COMPARISONS
// Ordering :: ( LT | EQ | GT ) | ( -1 | 0 | 1 ) // compare :: a -> a -> Ordering const compare = (a, b) => a < b ? -1 : (a > b ? 1 : 0);
// mappendOrdering :: Ordering -> Ordering -> Ordering const mappendOrdering = (a, b) => a !== 0 ? a : b;
// on :: (b -> b -> c) -> (a -> b) -> a -> a -> c const on = (f, g) => (a, b) => f(g(a), g(b));
// flip :: (a -> b -> c) -> b -> a -> c const flip = f => (a, b) => f.apply(null, [b, a]);
// arrayCopy :: [a] -> [a] const arrayCopy = (xs) => xs.slice(0);
// show :: a -> String const show = x => JSON.stringify(x, null, 2);
// TEST const xs = ['Shanghai', 'Karachi', 'Beijing', 'Sao Paulo', 'Dhaka', 'Delhi', 'Lagos'];
const rs = [{ name: 'Shanghai', pop: 24.2 }, { name: 'Karachi', pop: 23.5 }, { name: 'Beijing', pop: 21.5 }, { name: 'Sao Paulo', pop: 24.2 }, { name: 'Dhaka', pop: 17.0 }, { name: 'Delhi', pop: 16.8 }, { name: 'Lagos', pop: 16.1 }]
// population :: Dictionary -> Num const population = x => x.pop;
// length :: [a] -> Int const length = xs => xs.length;
// toLower :: String -> String const toLower = s => s.toLowerCase();
// lengthThenAZ :: String -> String -> ( -1 | 0 | 1) const lengthThenAZ = (a, b) => mappendOrdering( on(compare, length)(a, b), on(compare, toLower)(a, b) );
// descLengthThenAZ :: String -> String -> ( -1 | 0 | 1) const descLengthThenAZ = (a, b) => mappendOrdering( on(flip(compare), length)(a, b), on(compare, toLower)(a, b) );
return show({ default: arrayCopy(xs) .sort(compare),
descendingDefault: arrayCopy(xs) .sort(flip(compare)),
byLengthThenAZ: arrayCopy(xs) .sort(lengthThenAZ),
byDescendingLengthThenZA: arrayCopy(xs) .sort(flip(lengthThenAZ)),
byDescendingLengthThenAZ: arrayCopy(xs) .sort(descLengthThenAZ),
byPopulation: arrayCopy(rs) .sort(on(compare, population)),
byDescendingPopulation: arrayCopy(rs) .sort(on(flip(compare), population)) });
})();</lang>
- Output:
{ "default": [ "Beijing", "Delhi", "Dhaka", "Karachi", "Lagos", "Sao Paulo", "Shanghai" ], "descendingDefault": [ "Shanghai", "Sao Paulo", "Lagos", "Karachi", "Dhaka", "Delhi", "Beijing" ], "byLengthThenAZ": [ "Delhi", "Dhaka", "Lagos", "Beijing", "Karachi", "Shanghai", "Sao Paulo" ], "byDescendingLengthThenZA": [ "Sao Paulo", "Shanghai", "Karachi", "Beijing", "Lagos", "Dhaka", "Delhi" ], "byDescendingLengthThenAZ": [ "Sao Paulo", "Shanghai", "Beijing", "Karachi", "Delhi", "Dhaka", "Lagos" ], "byPopulation": [ { "name": "Lagos", "pop": 16.1 }, { "name": "Delhi", "pop": 16.8 }, { "name": "Dhaka", "pop": 17 }, { "name": "Beijing", "pop": 21.5 }, { "name": "Karachi", "pop": 23.5 }, { "name": "Shanghai", "pop": 24.2 }, { "name": "Sao Paulo", "pop": 24.2 } ], "byDescendingPopulation": [ { "name": "Shanghai", "pop": 24.2 }, { "name": "Sao Paulo", "pop": 24.2 }, { "name": "Karachi", "pop": 23.5 }, { "name": "Beijing", "pop": 21.5 }, { "name": "Dhaka", "pop": 17 }, { "name": "Delhi", "pop": 16.8 }, { "name": "Lagos", "pop": 16.1 } ] }
jq
The comparator, cmp, must have 0 arity, and may either be boolean or follow the negative/zero/positive convention.
If "o" is an ordering, and if x and y are two entities for which "x o y" is defined, then "[x,y] | cmp" should return a number, or a boolean value.
As illustrated in the example, the comparator may be any jq filter, whether or not it is defined as a function. <lang jq>def quicksort(cmp):
if length < 2 then . # it is already sorted else .[0] as $pivot | reduce .[] as $x # state: [less, equal, greater] ( [ [], [], [] ]; # three empty arrays: if $x == $pivot then .[1] += [$x] # add x to equal else ([$x,$pivot]|cmp) as $order | if $order == 0 then .[1] += [$x] # ditto elif ($order|type) == "number" then if $order < 0 then .[0] += [$x] # add x to less else .[2] += [$x] # add x to greater end else ([$pivot,$x]|cmp) as $order2 | if $order and $order2 then .[1] += [$x] # add x to equal elif $order then .[0] += [$x] # add x to less else .[2] += [$x] # add x to greater end end end ) | (.[0] | quicksort(cmp) ) + .[1] + (.[2] | quicksort(cmp) ) end ;</lang>
Example: <lang jq># Sort by string length, breaking ties using ordinary string comparison. ["z", "yz", "ab", "c"]
| quicksort( (.[0]|length) > (.[1]|length) or ( (.[0]|length) == (.[1]|length) and .[0] < .[1] ) )
</lang>
- Output:
<lang jq>[
"ab", "yz", "c", "z"
]</lang>
Julia
My word list source is the opening sentence of Shelly's Frankenstein. <lang Julia> st = """You will rejoice to hear that no disaster has accompanied the commencement of an enterprise which you have regarded with such evil forebodings."""
wl = filter(x->length(x)>0, split(st, r"\W+"))
println("Original List:") for w in wl
println(" ", w)
end
wl = sort(wl, by=x->(-length(x), lowercase(x)))
println("\nSorted List:") for w in wl
println(" ", w)
end </lang>
- Output:
Original List: You will rejoice to hear that no disaster has accompanied the commencement of an enterprise which you have regarded with such evil forebodings Sorted List: commencement accompanied forebodings enterprise disaster regarded rejoice which evil have hear such that will with has the You you an no of to
Kotlin
A translation from Java, also showing the seamless interop between Java and Kotlin code.
<lang kotlin>import java.util.Arrays
fun main(args: Array<String>) {
val strings = arrayOf("Here", "are", "some", "sample", "strings", "to", "be", "sorted")
fun printArray(message: String, array: Array<String>) = with(array) { print("$message [") forEachIndexed { index, string -> print(if (index == lastIndex) string else "$string, ") } println("]") }
printArray("Unsorted:", strings)
Arrays.sort(strings) { first, second -> val lengthDifference = second.length - first.length if (lengthDifference == 0) first.compareTo(second, true) else lengthDifference }
printArray("Sorted:", strings)
}</lang>
- Output:
Unsorted: [Here, are, some, sample, strings, to, be, sorted] Sorted: [strings, sample, sorted, Here, some, are, be, to]
Lua
<lang lua>test = { "Here", "we", "have", "some", "sample", "strings", "to", "be", "sorted" }
function stringSorter(a, b) if string.len(a) == string.len(b) then return string.lower(a) < string.lower(b) end return string.len(a) > string.len(b) end table.sort(test, stringSorter)
-- print sorted table for k,v in pairs(test) do print(v) end</lang>
- Output:
strings sample sorted have Here some be to we
Mathematica
We define a new function to give true or false if two elements are in order. After that we can simply use the built-in Sort with an ordering function: <lang Mathematica>StringOrderQ[x_String, y_String] :=
If[StringLength[x] == StringLength[y], OrderedQ[{x, y}], StringLength[x] >StringLength[y] ]
words={"on","sunday","sander","sifted","and","sorted","sambaa","for","a","second"}; Sort[words,StringOrderQ]</lang> gives back:
{sambaa,sander,second,sifted,sorted,sunday,and,for,on,a}
Maxima
<lang maxima>strangeorderp(a, b) := slength(a) > slength(b) or (slength(a) = slength(b) and orderlessp(a, b))$ s: tokens("Lorem ipsum dolor sit amet consectetur adipiscing elit Sed non risus Suspendisse\
lectus tortor dignissim sit amet adipiscing nec ultricies sed dolor")$
sort(s, strangeorderp); ["Suspendisse", "consectetur", "adipiscing", "adipiscing", "dignissim", "ultricies", "lectus", "tortor", "Lorem", "dolor", "dolor", "ipsum", "risus", "amet", "amet", "elit", "Sed", "nec", "non", "sed", "sit", "sit"]</lang>
MAXScript
<lang maxscript>fn myCmp str1 str2 = (
case of ( (str1.count < str2.count): 1 (str1.count > str2.count): -1 default:( -- String compare is case sensitive, name compare isn't. Hence... str1 = str1 as name str2 = str2 as name case of ( (str1 > str2): 1 (str1 < str2): -1 default: 0 ) ) )
)
strList = #("Here", "are", "some", "sample", "strings", "to", "be", "sorted") qSort strList myCmp print strList</lang>
Nemerle
<lang Nemerle>using System.Console;
module CustomSort {
Main() : void { def strings1 = ["these", "are", "strings", "of", "different", "length"]; def strings2 = ["apple", "House", "chewy", "Salty", "rises", "Later"]; WriteLine(strings1.Sort((x, y) => y.Length.CompareTo(x.Length))); WriteLine(strings2.Sort((x, y) => x.CompareTo(y))) }
}</lang>
- Output:
[different, strings, length, these, are, of] [apple, chewy, House, Later, rises, Salty]
NetRexx
<lang NetRexx>/* NetRexx */ options replace format comments java crossref symbols nobinary
-- ============================================================================= class RSortCustomComparator public
-- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ method main(args = String[]) public static
sample = [String 'Here', 'are', 'some', 'sample', 'strings', 'to', 'be', 'sorted'] say displayArray(sample) Arrays.sort(sample, LengthComparator()) say displayArray(sample) return
method displayArray(harry = String[]) constant
disp = loop elmt over harry disp = disp','elmt end elmt return '['disp.substr(2)']' -- trim leading comma
-- ============================================================================= class RSortCustomComparator.LengthComparator implements Comparator
method compare(lft = Object, rgt = Object) public binary returns int
cRes = int if lft <= String, rgt <= String then do cRes = (String rgt).length - (String lft).length if cRes == 0 then cRes = (String lft).compareToIgnoreCase(String rgt) end else signal IllegalArgumentException('Arguments must be Strings') return cRes
</lang>
- Output:
[Here,are,some,sample,strings,to,be,sorted] [strings,sample,sorted,Here,some,are,be,to]
Nial
<lang nial>sort fork [=[tally first,tally last],up, >= [tally first,tally last]] ['Here', 'are', 'some', 'sample', 'strings', 'to', 'be', 'sorted'] =+-------+------+------+----+----+---+--+--+ =|strings|sample|sorted|Here|some|are|be|to| =+-------+------+------+----+----+---+--+--+</lang>
Nim
<lang nim>import strutils, algorithm
var strings = "here are Some sample strings to be sorted".split(' ')
strings.sort(proc (x,y: string): int =
cmp(y.len, x.len))
echo strings</lang>
Objeck
<lang objeck>use Collection;
class Test {
function : Main(args : String[]) ~ Nil { v := CreateHolders(["Here", "are", "some", "sample", "strings", "to", "be", "sorted"]); "unsorted: "->Print(); Show(v); v->Sort(); "sorted: "->Print(); Show(v); } function : CreateHolders(strings : String[]) ~ CompareVector { vector := CompareVector->New(); each(i : strings) { vector->AddBack(StringHolder->New(strings[i])); }; return vector; } function : Show(v : CompareVector) ~ Nil { each(i : v) { s := v->Get(i)->As(StringHolder); s->ToString()->Print(); if(i + 1 < v->Size()) { ','->Print(); }; }; '\n'->Print(); }
}
class StringHolder implements Compare {
@s : String; New(s : String) { @s := s; } method : public : Compare(c : Compare) ~ Int { h := c->As(StringHolder); r := h->ToString(); size := r->Size() - @s->Size(); if(size = 0) { size := @s->ToUpper()->Compare(r->ToUpper()); }; return size; } method : public : HashID() ~ Int { return @s->HashID(); } method : public : ToString() ~ String { return @s; }
}</lang>
unsorted: Here,are,some,sample,strings,to,be,sorted sorted: strings,sample,sorted,Here,some,are,be,to
Objective-C
Using blocks: <lang objc>#import <Foundation/Foundation.h>
- define esign(X) (((X)>0)?1:(((X)<0)?-1:0))
int main() {
@autoreleasepool {
NSMutableArray *arr = [NSMutableArray arrayWithArray: [@"this is a set of strings to sort" componentsSeparatedByString: @" "] ];
[arr sortUsingComparator: ^NSComparisonResult(id obj1, id obj2){ NSComparisonResult l = esign((int)([obj1 length] - [obj2 length])); return l ? -l // reverse the ordering : [obj1 caseInsensitiveCompare: obj2]; }];
for( NSString *str in arr ) { NSLog(@"%@", str); }
} return EXIT_SUCCESS;
}</lang>
<lang objc>#import <Foundation/Foundation.h>
@interface NSString (CustomComp) - (NSComparisonResult)my_compare: (id)obj; @end
- define esign(X) (((X)>0)?1:(((X)<0)?-1:0))
@implementation NSString (CustomComp) - (NSComparisonResult)my_compare: (id)obj {
NSComparisonResult l = esign((int)([self length] - [obj length])); return l ? -l // reverse the ordering : [self caseInsensitiveCompare: obj];
} @end
int main() {
@autoreleasepool {
NSMutableArray *arr = [NSMutableArray arrayWithArray: [@"this is a set of strings to sort" componentsSeparatedByString: @" "] ];
[arr sortUsingSelector: @selector(my_compare:)];
for ( NSString *str in arr ) { NSLog(@"%@", str); }
} return EXIT_SUCCESS;
}</lang>
This example can also be written using sort descriptors:
<lang objc>#import <Foundation/Foundation.h>
int main() {
@autoreleasepool {
NSArray *strings = [@"Here are some sample strings to be sorted" componentsSeparatedByString:@" "];
NSSortDescriptor *sd1 = [[NSSortDescriptor alloc] initWithKey:@"length" ascending:NO]; NSSortDescriptor *sd2 = [[NSSortDescriptor alloc] initWithKey:@"lowercaseString" ascending:YES];
NSArray *sorted = [strings sortedArrayUsingDescriptors:@[sd1, sd2]]; NSLog(@"%@", sorted);
}
return 0;
}</lang>
OCaml
<lang ocaml>let mycmp s1 s2 =
if String.length s1 <> String.length s2 then compare (String.length s2) (String.length s1) else String.compare (String.lowercase s1) (String.lowercase s2)</lang>
List: <lang ocaml># let strings = ["Here"; "are"; "some"; "sample"; "strings"; "to"; "be"; "sorted"];; val strings : string list =
["Here"; "are"; "some"; "sample"; "strings"; "to"; "be"; "sorted"]
- List.sort mycmp strings;;
- : string list = ["strings"; "sample"; "sorted"; "Here"; "some"; "are"; "be"; "to"]</lang>
Array: <lang ocaml># let strings = [|"Here"; "are"; "some"; "sample"; "strings"; "to"; "be"; "sorted"|];; val strings : string array =
[|"Here"; "are"; "some"; "sample"; "strings"; "to"; "be"; "sorted"|]
- Array.sort mycmp strings;;
- : unit = ()
- strings;;
- : string array = [|"strings"; "sample"; "sorted"; "Here"; "some"; "are"; "be"; "to"|]</lang>
Oforth
<lang Oforth>String method: customCmp(s)
s size self size > ifTrue: [ true return ] s size self size < ifTrue: [ false return ] s toUpper self toUpper <= ;
["this", "is", "a", "set", "of", "strings", "to", "sort", "This", "Is", "A", "Set", "Of", "Strings", "To", "Sort"] sortWith(#customCmp) println</lang>
- Output:
[Strings, strings, Sort, sort, this, This, Set, set, is, Is, of, Of, To, to, A, a]
ooRexx
<lang ooRexx>A=.array~of('The seven deadly sins','Pride','avarice','Wrath','envy','gluttony','sloth','Lust') say 'Sorted in order of descending length, and in ascending lexicographic order' say A~sortWith(.DescLengthAscLexical~new)~makeString
- class DescLengthAscLexical mixinclass Comparator
- method compare
use strict arg left, right if left~length==right~length
then return left~caselessCompareTo(right) else return right~length-left~length</lang>
- Output:
Sorted in order of descending length, and in ascending lexicographic order The seven deadly sins gluttony avarice Pride sloth Wrath envy Lust
Oz
<lang oz>declare
fun {LexicographicLessThan Xs Ys} for X in {Map Xs Char.toLower} Y in {Map Ys Char.toLower} return:Return default:{Length Xs}<{Length Ys} do if X < Y then {Return true} end end end fun {LessThan Xs Ys} {Length Xs} > {Length Ys} orelse {Length Xs} == {Length Ys} andthen {LexicographicLessThan Xs Ys} end Strings = ["Here" "are" "some" "sample" "strings" "to" "be" "sorted"]
in
{ForAll {Sort Strings LessThan} System.showInfo}</lang>
PARI/GP
<lang parigp>cmp(a,b)=if(#a<#b,1,if(#a>#b,-1,lex(a,b))); vecsort(v,cmp)</lang>
Pascal
See at http://rosettacode.org/wiki/Sorting_algorithms/Merge_sort#improvement struct/record with myText implementing this task too
Perl
<lang perl>sub mycmp { length $b <=> length $a || lc $a cmp lc $b }
my @strings = ("Here", "are", "some", "sample", "strings", "to", "be", "sorted"); my @sorted = sort mycmp @strings;</lang>
Or inline: <lang perl>my @strings = qw/here are some sample strings to be sorted/; my @sorted = sort {length $b <=> length $a || lc $a cmp lc $b} @strings</lang>
Faster with a Schwartzian transform: <lang perl>my @strings = qw/here are some sample strings to be sorted/; my @sorted = map { $_->[0] }
sort { $a->[1] <=> $b->[1] || $a->[2] cmp $b->[2] } map { [ $_, length, lc ] } @strings;</lang>
Perl 6
<lang perl6>my @strings = <Here are some sample strings to be sorted>; my @sorted_strings = sort { $^a.chars <=> $^b.chars or $^a.lc cmp $^b.lc }, @strings; .say for @sorted_strings;</lang> This behavior is triggered by use of an arity 2 sort routine.
If instead the function you feed to sort
is of arity 1, it will do the Schwartzian transform for you, automatically sorting numeric fields numerically, and strings fields stringily:
<lang perl6>my @sorted_strings = sort -> $x { [ $x.chars, $x.lc ] }, @strings;</lang>
Phix
Copy of Euphoria <lang Phix>function my_compare(sequence a, sequence b)
if length(a)!=length(b) then return -compare(length(a),length(b)) else return compare(lower(a),lower(b)) end if
end function ?custom_sort(routine_id("my_compare"),{"Here", "are", "some", "sample", "strings", "to", "be", "sorted"})</lang>
- Output:
{"strings","sample","sorted","Here","some","are","be","to"}
PHP
<lang php><?php function mycmp($s1, $s2) {
if ($d = strlen($s2) - strlen($s1)) return $d; return strcasecmp($s1, $s2);
}
$strings = array("Here", "are", "some", "sample", "strings", "to", "be", "sorted"); usort($strings, "mycmp"); ?></lang>
PicoLisp
By default, the sort function in PicoLisp returns an ascending list (of any type). To get a result in descending order, the "greater than" function can be supplied <lang PicoLisp>: (sort '("def" "abc" "ghi") >) -> ("ghi" "def" "abc")</lang> or simply the result reversed (which is, btw, the most efficient way) <lang PicoLisp>: (flip (sort '("def" "abc" "ghi"))) -> ("ghi" "def" "abc")</lang>
PL/I
Platform: WIN <lang pli>MRGEPKG: package exports(MERGESORT,MERGE,RMERGE);
DCL (T(4)) CHAR(20) VAR; /* scratch space of length N/2 */
MERGE: PROCEDURE (A,LA,B,LB,C,CMPFN);
DECLARE (A(*),B(*),C(*)) CHAR(*) VAR; DECLARE (LA,LB) FIXED BIN(31) NONASGN; DECLARE (I,J,K) FIXED BIN(31); DECLARE CMPFN ENTRY( NONASGN CHAR(*) VAR, NONASGN CHAR(*) VAR) RETURNS (FIXED bin(31)); I=1; J=1; K=1; DO WHILE ((I <= LA) & (J <= LB)); IF CMPFN(A(I),B(J)) <= 0 THEN DO; C(K)=A(I); K=K+1; I=I+1; END; ELSE DO; C(K)=B(J); K=K+1; J=J+1; END; END; DO WHILE (I <= LA); C(K)=A(I); I=I+1; K=K+1; END; return;
END MERGE;
MERGESORT: PROCEDURE (A,N,CMPFN) RECURSIVE ;
DECLARE (A(*)) CHAR(*) VAR; DECLARE N FIXED BINARY(31) NONASGN; DECLARE CMPFN ENTRY( NONASGN CHAR(*) VAR, NONASGN CHAR(*) VAR) RETURNS (FIXED bin(31)); DECLARE (M,I) FIXED BINARY; DECLARE AMP1(N) CHAR(20) VAR BASED(P); DECLARE P POINTER;
IF (N=1) THEN RETURN; M = trunc((N+1)/2); IF M > 1 THEN CALL MERGESORT(A,M,CMPFN); P=ADDR(A(M+1)); IF (N-M > 1) THEN CALL MERGESORT(AMP1,N-M,CMPFN); IF CMPFN(A(M),AMP1(1)) <= 0 THEN RETURN; DO I=1 to M; T(I)=A(I); END; CALL MERGE(T,M,AMP1,N-M,A,CMPFN);
END MERGESORT;
RMERGE: PROC OPTIONS(MAIN); DCL I FIXED BIN(31); DCL A(8) CHAR(20) VAR INIT("this","is","a","set","of","strings","to","sort");
MyCMP: PROCEDURE(A,B) RETURNS (FIXED BIN(31));
DCL (A,B) CHAR(*) VAR NONASGN; DCL (I,J) FIXED BIN(31);
I = length(trim(A)); J = length(trim(B)); IF I < J THEN RETURN(+1); IF I > J THEN RETURN(-1); IF lowercase(A) < lowercase(B) THEN RETURN(-1); IF lowercase(A) > lowercase(B) THEN RETURN(+1); RETURN (0);
END MyCMP;
CALL MERGESORT(A,8,MyCMP); DO I=1 TO 8;
put edit (I,A(I)) (F(5),X(2),A(10)) skip;
END;
put skip; END RMERGE;</lang>
Pop11
<lang pop11>lvars ls = ['Here' 'are' 'some' 'sample' 'strings' 'to' 'be' 'sorted']; define compare(s1, s2); lvars k = length(s2) - length(s1); if k < 0 then
return(true);
elseif k > 0 then
return(false);
else
return (alphabefore(uppertolower(s1), uppertolower(s2)));
endif; enddefine;
syssort(ls, compare) -> ls;
NOTE: The definition of compare can also be written thus: define compare(s1, s2);
lvars l1 = length(s1), l2 = length(s2); l1 > l2 or (l1 == l2 and alphabefore(uppertolower(s1), uppertolower(s2)))
enddefine;</lang>
PowerBASIC
<lang powerbasic>FUNCTION Sorter(p1 AS STRING, p2 AS STRING) AS LONG
'if p1 should be first, returns -1 'if p2 should be first, returns 1 ' if they're equal, returns 0 IF LEN(p1) > LEN(p2) THEN FUNCTION = -1 ELSEIF LEN(p2) > LEN(p1) THEN FUNCTION = 1 ELSEIF UCASE$(p1) > UCASE$(p2) THEN 'if we get here, they're of equal length, 'so now we're doing a "normal" string comparison FUNCTION = -1 ELSEIF UCASE$(p2) > UCASE$(p1) THEN FUNCTION = 1 ELSE FUNCTION = 0 END IF
END FUNCTION
FUNCTION PBMAIN()
DIM x(7) AS STRING ARRAY ASSIGN x() = "Here", "are", "some", "sample", "strings", "to", "be", "sorted"
'pb's built-in sorting; "USING" tells it to use our custom comparator ARRAY SORT x(), USING Sorter()
END FUNCTION</lang>
PowerShell
The Sort-Object
cmdlet accepts script blocks as arguments as well as multiple criteria after which to sort.
<lang powershell>$list = "Here", "are", "some", "sample", "strings", "to", "be", "sorted"
$list | Sort-Object {-$_.Length},{$_}</lang>
The negated string length is the first sort criterion, the second is the string itself, resulting in descending length and ascending lexicographic order.
Prolog
Works with SWI-Prolog. <lang Prolog>rosetta_sort :- L = ["Here", "are", "some", "sample", "strings", "to", "be", "sorted" ], predsort(my_comp, L, L1), writeln('Input list :'), maplist(my_write, L), nl,nl, writeln('Sorted list :'), maplist(my_write, L1).
my_comp(Comp, W1, W2) :-
length(W1,L1),
length(W2, L2),
( L1 < L2 -> Comp = '>'
; L1 > L2 -> Comp = '<'
; compare(Comp, W1, W2)).
my_write(W) :- format('~s ', [W]). </lang>
- Output:
?- rosetta_sort. Input list : Here are some sample strings to be sorted Sorted list : strings sample sorted Here some are be to true.
Python
Using a key function is usually more efficient than a comparator. We can take advantage of the fact that tuples are ordered first by the first element, then by the second, etc., to perform a sort on multiple criteria. <lang python>strings = "here are Some sample strings to be sorted".split()
def mykey(x):
return -len(x), x.upper()
print sorted(strings, key=mykey)</lang>
- Output:
<lang python>['strings', 'sample', 'sorted', 'here', 'Some', 'are', 'be', 'to']</lang>
Alternative method using cmp
To technically comply with this task, we can also use an actual comparator (cmp) function which will be called every time members of the original list are to be compared. Note that this feature is worse than using the key argument and has been removed from Python 3, so should no longer be used in new code. <lang python>def mycmp(s1, s2):
return cmp(len(s2), len(s1)) or cmp(s1.upper(), s2.upper())
print sorted(strings, cmp=mycmp)</lang>
R
<lang R>v = c("Here", "are", "some", "sample", "strings", "to", "be", "sorted") print(v[order(-nchar(v), tolower(v))])</lang>
Racket
<lang Racket>
- lang racket
- Using a combination of the two comparisons
(define (sort1 words)
(sort words (λ(x y) (define xl (string-length x)) (define yl (string-length y)) (or (> xl yl) (and (= xl yl) (string-ci<? x y))))))
(sort1 '("Some" "pile" "of" "words"))
- -> '("words" "pile" "Some" "of")
- Doing two sorts, relying on `sort's stability
(define (sort2 words)
(sort (sort words string-ci<?) > #:key string-length))
(sort2 '("Some" "pile" "of" "words"))
- -> '("words" "pile" "Some" "of")
</lang>
REXX
<lang rexx>/*REXX program sorts a (stemmed) array using the merge-sort method. */ /* using mycmp function for the sort order */ /**********************************************************************
- mergesort taken from REXX (adapted for ooRexx (and all other REXXes))
- 28.07.2013 Walter Pachl
- /
Call gena /* generate the array elements. */ Call showa 'before sort' /* show the before array elements.*/ Call mergeSort highitem /* invoke the merge sort for array*/ Call showa ' after sort' /* show the after array elements.*/ Exit /* stick a fork in it, we're done.*/
/*---------------------------------GENa subroutine-------------------*/ gena:
a.= /* assign default value for a stem*/ a.1='---The seven deadly sins---'/* everybody: pick your favorite.*/ a.2='===========================' a.3='pride' a.4='avarice' a.5='wrath' a.6='envy' a.7='gluttony' a.8='sloth' a.9='lust' Do highitem=1 While a.highitem\== /*find number of entries */ End highitem=highitem-1 /* adjust highitem by -1. */ Return
/*---------------------------------MERGETOa subroutine---------------*/ mergetoa: Procedure Expose a. !.
Parse Arg l,n Select When n==1 Then Nop When n==2 Then Do h=l+1 If mycmp(a.l,a.h)=1 Then Do _=a.h a.h=a.l a.l=_ End End Otherwise Do m=n%2 Call mergeToa l+m,n-m Call mergeTo! l,m,1 i=1 j=l+m Do k=l While k<j If j==l+n|mycmp(!.i,a.j)<>1 Then Do a.k=!.i i=i+1 End Else Do a.k=a.j j=j+1 End End End End Return
/*---------------------------------MERGESORT subroutine--------------*/ mergesort: Procedure Expose a.
Call mergeToa 1,arg(1) Return
/*---------------------------------MERGETO! subroutine---------------*/ mergeto!: Procedure Expose a. !.
Parse Arg l,n,_ Select When n==1 Then !._=a.l When n==2 Then Do h=l+1 q=1+_ If mycmp(a.l,a.h)=1 Then Do q=_ _=q+1 End !._=a.l !.q=a.h Return End Otherwise Do m=n%2 Call mergeToa l,m Call mergeTo! l+m,n-m,m+_ i=l j=m+_ Do k=_ While k<j If j==n+_|mycmp(a.i,!.j)<>1 Then Do !.k=a.i i=i+1 End Else Do !.k=!.j j=j+1 End End End End Return
/*---------------------------------SHOWa subroutine------------------*/ showa:
widthh=length(highitem) /* maximum the width of any line.*/ Do j=1 For highitem Say 'element' right(j,widthh) arg(1)':' a.j End Say copies('-',60) /* show a separator line (fence).*/ Return
mycmp: Procedure /**********************************************************************
- shorter string considered higher
- when lengths are equal: caseless 'Z' considered higher than 'X' etc.
- Result: 1 B consider higher than A
- -1 A consider higher than B
- 0 A==B (caseless)
- /
Parse Upper Arg A,B A=strip(A) B=strip(B) I = length(A) J = length(B) Select When I << J THEN res=1 When I >> J THEN res=-1 When A >> B THEN res=1 When A << B THEN res=-1 Otherwise res=0 End RETURN res</lang>
- Output:
element 1 before sort: ---The seven deadly sins--- element 2 before sort: =========================== element 3 before sort: pride element 4 before sort: avarice element 5 before sort: wrath element 6 before sort: envy element 7 before sort: gluttony element 8 before sort: sloth element 9 before sort: lust ------------------------------------------------------------ element 1 after sort: ---The seven deadly sins--- element 2 after sort: =========================== element 3 after sort: gluttony element 4 after sort: avarice element 5 after sort: pride element 6 after sort: sloth element 7 after sort: wrath element 8 after sort: envy element 9 after sort: lust ------------------------------------------------------------
Ruby
Since Ruby 1.8.6 Enumerables have a "sort_by" method, taking a key block, which is more efficient than a comparator. We can take advantage of the fact that Arrays are ordered first by the first element, then by the second, etc., to perform a sort on multiple criteria.
<lang ruby>words = %w(Here are some sample strings to be sorted) p words.sort_by {|word| [-word.size, word.downcase]}</lang>
To technically comply with this task, we can also use an actual comparator block which will be called every time members of the original list are to be compared. <lang ruby>p words.sort {|a, b| d = b.size <=> a.size
d != 0 ? d : a.upcase <=> b.upcase}</lang>
Rust
<lang rust> fn main() {
let mut words = ["Here", "are", "some", "sample", "strings", "to", "be", "sorted"]; words.sort_by(|l, r| if l.len() == r.len() { l.cmp(&r) } else { r.len().cmp(&l.len()) }); println!("{:?}", words);
} </lang>
Sather
<lang sather>class MAIN is
custom_comp(a, b:STR):BOOL is l ::= a.length - b.length; if l = 0 then return a.lower < b.lower; end; return l > 0; end;
main is s:ARRAY{STR} := |"this", "is", "an", "array", "of", "strings", "to", "sort"|; s.insertion_sort_by(bind(custom_comp(_,_))); loop #OUT + s.elt! + "\n"; end; end;
end;</lang>
Scala
<lang scala>List("Here", "are", "some", "sample", "strings", "to", "be", "sorted").sortWith{(a,b) =>
val cmp=a.size-b.size (if (cmp==0) -a.compareTo(b) else cmp) > 0
}</lang>
- Output:
List(strings, sample, sorted, Here, some, are, be, to)
Scheme
<lang scheme>(use srfi-13);;Syntax for module inclusion depends on implementation,
- a sort function may be predefined, or available through srfi 95
(define (mypred? a b)
(let ((len-a (string-length a))
(len-b (string-length b)))
(if (= len-a len-b)
(string>? (string-downcase b) (string-downcase a))
(> len-a len-b))))
(sort '("sorted" "here" "strings" "sample" "Some" "are" "be" "to") mypred?) </lang>
- Output:
<lang scheme>("strings" "sample" "sorted" "here" "Some" "are" "be" "to")</lang>
An alternative solution:
<lang Scheme>(define strings '(
"This" "Is" "A" "Set" "Of" "Strings" "To" "Sort" "duplicated" "this" "is" "a" "set" "of" "strings" "to" "sort" "duplicated"))
(sort strings (lambda two (define sizes (map string-length two)) (if (apply = sizes) (apply string-ci<? two) (apply > sizes)))))
</lang>
- Output:
(duplicated duplicated Strings strings Sort sort This this Set set Is is Of of To to A a)
Sidef
<lang ruby>func mycmp(a, b) { (b.len <=> a.len) || (a.lc <=> b.lc) }; var strings = %w(Here are some sample strings to be sorted); var sorted = strings.sort(mycmp);</lang>
Slate
<lang slate>define: #words -> #('here' 'are' 'some' 'sample' 'strings' 'to' 'sort' 'since' 'this' 'exercise' 'is' 'not' 'really' 'all' 'that' 'dumb' '(sorry)'). words sortBy: [| :first :second | (first lexicographicallyCompare: second) isNegative]</lang>
Smalltalk
<lang smalltalk>#('here' 'are' 'some' 'sample' 'strings' 'to' 'sort' 'since' 'this' 'exercise' 'is' 'not' 'really' 'all' 'that' 'dumb' '(sorry)' ) asSortedCollection
sortBlock: [:first :second | (second size = first size) ifFalse: [second size < first size] ifTrue: [first < second]]</lang>
Standard ML
List:
<lang sml>fun mygt (s1, s2) =
if size s1 <> size s2 then size s2 > size s1 else String.map Char.toLower s1 > String.map Char.toLower s2</lang>
<lang sml>- val strings = ["Here", "are", "some", "sample", "strings", "to", "be", "sorted"]; val strings = ["Here","are","some","sample","strings","to","be","sorted"]
: string list
- ListMergeSort.sort mygt strings; val it = ["strings","sample","sorted","Here","some","are","be","to"]
: string list</lang>
Array:
<lang sml>fun mycmp (s1, s2) =
if size s1 <> size s2 then Int.compare (size s2, size s1) else String.compare (String.map Char.toLower s1, String.map Char.toLower s2)</lang>
<lang sml>- val strings = Array.fromList ["Here", "are", "some", "sample", "strings", "to", "be", "sorted"]; val strings = [|"Here","are","some","sample","strings","to","be","sorted"|]
: string array
- ArrayQSort.sort mycmp strings; val it = () : unit - strings; val it = [|"strings","sample","sorted","Here","some","are","be","to"|]
: string array</lang>
Swift
<lang swift>import Foundation
var list = ["this",
"is", "a", "set", "of", "strings", "to", "sort", "This", "Is", "A", "Set", "Of", "Strings", "To", "Sort"]
list.sortInPlace {lhs, rhs in
let lhsCount = lhs.characters.count let rhsCount = rhs.characters.count let result = rhsCount - lhsCount if result == 0 { return lhs.lowercaseString > rhs.lowercaseString } return lhsCount > rhsCount
}</lang>
<lang swift>import Foundation
var list = ["this",
"is", "a", "set", "of", "strings", "to", "sort", "This", "Is", "A", "Set", "Of", "Strings", "To", "Sort"]
sort(&list) {lhs, rhs in
let lhsCount = count(lhs) let rhsCount = count(rhs) let result = rhsCount - lhsCount if result == 0 { return lhs.lowercaseString > rhs.lowercaseString } return lhsCount > rhsCount
}</lang>
Tcl
<lang tcl>proc sorter {a b} {
set la [string length $a] set lb [string length $b] if {$la < $lb} { return 1 } elseif {$la > $lb} { return -1 } return [string compare [string tolower $a] [string tolower $b]]
}
set strings {here are Some sample strings to be sorted} lsort -command sorter $strings ;# ==> strings sample sorted here Some are be to</lang>
TUSCRIPT
<lang tuscript> $$ MODE TUSCRIPT setofstrings="this is a set of strings to sort This Is A Set Of Strings To Sort" unsorted=SPLIT (setofstrings,": :") PRINT "1. setofstrings unsorted" index="" LOOP l=unsorted PRINT l length=LENGTH (l),index=APPEND(index,length) ENDLOOP index =DIGIT_INDEX (index) sorted=INDEX_SORT (unsorted,index) PRINT "2. setofstrings sorted"
- {sorted}
</lang>
- Output:
1. setofstrings unsorted this is a set of strings to sort This Is A Set Of Strings To Sort 2. setofstrings sorted a A is of to Is Of To set Set this sort This Sort strings Strings
Ursala
A standard library function, psort, takes a list of binary relational predicates and returns a function that uses them in order of decreasing priority to perform a sort. The less or equal length predicate (leql) and lexically less or equal predicate (lleq) are also standard library functions. This task is therefore easily dispatched as shown.
<lang Ursala>#import std
- show+
data = <'this','is','a','list','of','strings','to','be','sorted'>
example = psort<not leql,lleq+ ~* ~&K31K30piK26 letters> data</lang> The lleq library function is case sensitive, so it is composed with a function to convert the words to lower case on the fly (without destructively modifying them) in order to meet the task requirement of case insensitivity.
- Output:
strings sorted list this be is of to a
Visual Basic .NET
<lang vbnet>Imports System
Module Sorting_Using_a_Custom_Comparator
Function CustomComparator(ByVal x As String, ByVal y As String) As Integer Dim result As Integer result = y.Length - x.Length If result = 0 Then result = String.Compare(x, y, True) End If Return result End Function
Sub Main() Dim strings As String() = {"test", "Zoom", "strings", "a"}
Array.Sort(strings, New Comparison(Of String)(AddressOf CustomComparator)) End Sub
End Module</lang>
zkl
<lang zkl>s:=T("Cat","apple","Adam","zero","Xmas","quit","Level","add","Actor","base","butter"); r:=s.sort(fcn(a,b){ an,bn := a.len(),b.len();
if(an==bn)(a.toLower() < b.toLower()) else (an > bn)
}); r.pump(Console.println);</lang>
- Output:
butter Actor apple Level Adam base quit Xmas zero add Cat
- Programming Tasks
- Sorting
- BBC BASIC/Omit
- Ada
- AutoHotkey
- AWK
- Babel
- Burlesque
- C
- C sharp
- C++
- Ceylon
- Clean
- Clojure
- Common Lisp
- D
- Delphi
- E
- EGL
- Elixir
- Erlang
- Euphoria
- F Sharp
- Factor
- Fantom
- Fortran
- FreeBASIC
- FunL
- Go
- Groovy
- Haskell
- Icon
- Unicon
- J
- Java
- JavaScript
- Jq
- Julia
- Kotlin
- Lua
- Mathematica
- Maxima
- MAXScript
- Nemerle
- NetRexx
- Nial
- Nim
- Objeck
- Objective-C
- OCaml
- Oforth
- OoRexx
- Oz
- PARI/GP
- Pascal
- Perl
- Perl 6
- Phix
- PHP
- PicoLisp
- PL/I
- Pop11
- PowerBASIC
- PowerShell
- Prolog
- Python
- R
- Racket
- REXX
- Ruby
- Rust
- Sather
- Scala
- Scheme
- Sidef
- Slate
- Smalltalk
- Standard ML
- Swift
- Tcl
- TUSCRIPT
- Ursala
- Visual Basic .NET
- Zkl
- GUISS/Omit