Natural sorting: Difference between revisions

{{task|Sorting Algorithms}}

:4. Sorting numeric portions of strings in numeric order. That is split the string into fields on numeric boundaries, then sort on each field, with the rightmost fields being the most significant, and numeric fields of integers treated as numbers.

:: Sort germanGerman eszett or scharfes S (ß)       as   ss

:: Sort ſ, LATIN SMALL LETTER LONG S     as   s

:: Sort ʒ, LATIN SMALL LETTER EZH           as   s

* Test each feature implemented separately with an ordered list of test strings from the &nbsp; '''Sample inputs''' &nbsp; section below, &nbsp; and make sure your naturally sorted output is in the same order as other language outputs such as &nbsp; <CODE> Python</CODE>.

# Ignoring leading spaces.

['ignore leading spaces: 2-2', ' ignore leading spaces: 2-1', ' ignore leading spaces: 2+0', ' ignore leading spaces: 2+1']

# Ignoring multiple adjacent spaces (mMAS).a.s)

['ignore m.a.sMAS spaces: 2-2', 'ignore m.a.s spaces: 2-1', 'ignore m.a.s spaces: 2+0', 'ignore m.a.s spaces: 2+1']

<langsyntaxhighlight lang="c">#include <stdio.h>

static const wchar_t *const tbl_accent[] = { /* copied from Perl6Raku code */

}</langsyntaxhighlight>output<syntaxhighlight lang="text">Sort flags: (collapse spaces)

100 niño</langsyntaxhighlight>

<langsyntaxhighlight lang="d">import std.stdio, std.string, std.algorithm, std.array, std.conv,

<langsyntaxhighlight lang="elixir">defmodule Natural do

|> Enum.each(fn {title, input} -> Natural.task(title, input) end)</langsyntaxhighlight>

Here, all previous worry over storage consumption is abandoned. An array of CHARACTER*66 is deemed long enough, and too bad over trailing spaces. An auxiliary array holds the length of each entry. The source texts are left alone, at the cost of having the "natural" compare process engage in a lot of cogitation. The plan for the comparison is to have subroutine ANOTHER yield two characters for comparison, one from each string being compared. Subroutine CRUSH deals with spaceish and character case simplification, and each scan can scoot ahead through whatever multiple spaces it encounters. Subroutine LIBRARIAN advances past antany initial spaceish stuff then detects articles at the start of a text and advances the initial scan past them while preparing the appropriate addendum for the end of the text, signified in TAIL, should a scan reach that far.

<langsyntaxhighlight lang="go">package main

}</langsyntaxhighlight>

<langsyntaxhighlight lang="j">require'strings regex'

natSor=: lines ;@/: norm&.>@lines</langsyntaxhighlight>

<langsyntaxhighlight lang="j">IgnoringLeadingSpaces=:0 :0

)</langsyntaxhighlight>

<langsyntaxhighlight lang="j"> natSor IgnoringLeadingSpaces

<langsyntaxhighlight lang="jq">def splitup:

sort_by(naturally);</langsyntaxhighlight>

<langsyntaxhighlight lang="jq">if type == "string" then "", . else natural_sort end</langsyntaxhighlight>

<langsyntaxhighlight lang="sh">jq -r -f Natural_sorting.jq Natural_sorting.json

]</langsyntaxhighlight>

<langsyntaxhighlight lang="perl">

<langsyntaxhighlight lang="perl">

=={{header|Perl 6Phix}}==

<langsyntaxhighlight PicoLisplang="picolisp">(de parseNatural (Str)

(link Word) ) ) ) ) ) ) ) )</langsyntaxhighlight>

<langsyntaxhighlight PicoLisplang="picolisp">: (parseNatural " ^MThe abc123Defß ^I Ghi ")

-> ("abc" 123 "defss" " " "ghi")</langsyntaxhighlight>

<langsyntaxhighlight PicoLisplang="picolisp">(de naturalSort (Lst)

(by parseNatural sort Lst) )</langsyntaxhighlight>

<langsyntaxhighlight PicoLisplang="picolisp">(de *TestData

(prinl) ) )</langsyntaxhighlight>

<langsyntaxhighlight lang="python"># -*- coding: utf-8 -*-

print 'Naturally sorted:'; print '\n'.join(ns(txt))</langsyntaxhighlight>

<langsyntaxhighlight lang="racket">

<langsyntaxhighlight lang="ruby">ar.sort_by{|str| str.downcase.gsub(/\Athe |\Aa |\Aan /, "").lstrip.gsub(/\s+/, " ")}</langsyntaxhighlight>

<langsyntaxhighlight lang="ruby">class NatSortString

<langsyntaxhighlight lang="ruby">tests =

<langsyntaxhighlight Scalalang="scala">object NaturalSorting {

}</langsyntaxhighlight>

<langsyntaxhighlight lang="scheme">

"# Numeric fields as numerics" "# TitleNumeric sortsfields as numerics - shows sorting from right")

Also for "Equivalent whitespace characters", Isorting believeincludes thenumeric Python output is incorrectfields, aswhich Isorts have described onfrom the discussion pageright.

{{trans|Perl 6Raku}}

<langsyntaxhighlight lang="ruby">class String {

}</langsyntaxhighlight>

<langsyntaxhighlight lang="ruby">var tests = [

}</langsyntaxhighlight>

<langsyntaxhighlight lang="tcl">package require Tcl 8.5

foo1000bar99baz9.txt}</langsyntaxhighlight>

<langsyntaxhighlight lang="zkl">fcn dsuSort(x,orig){ // decorate-sort-undecorate sort

}</langsyntaxhighlight>

<langsyntaxhighlight lang="zkl"># Ignoring leading spaces

dsuSort(ts1.apply("strip"),ts1).println();</langsyntaxhighlight>

<langsyntaxhighlight lang="zkl"># Ignoring multiple adjacent spaces (m.a.s)

dsuSort(ts2.apply('-(" ")),ts2).println();</langsyntaxhighlight>

<langsyntaxhighlight lang="zkl"># Equivalent whitespace characters

dsuSort(ts3.apply('-.fp1("\n\r\t\f\b\x0b ")),ts3).println();</langsyntaxhighlight>

<langsyntaxhighlight lang="zkl"># Case Indepenent sort

dsuSort(ts4.apply("toLower"),ts4).println();</langsyntaxhighlight>

<langsyntaxhighlight lang="zkl"># Numeric fields as numerics

}</langsyntaxhighlight>

<langsyntaxhighlight lang="zkl">ts5:=T("foo100bar99baz0.txt", "foo100bar10baz0.txt", "foo1000bar99baz10.txt",

fsort(x).println();</langsyntaxhighlight>