Anagrams: Difference between revisions

When two or more words are composed of the same characters, but in a different order, they are called [[wp:Anagram|anagrams]].

 

Using the word list at   http://wiki.puzzlers.org/pub/wordlists/unixdict.txt,

<br>find the sets of words that share the same characters that contain the most words in them.

 

{{Related tasks/Word plays}}

 

=={{header|11l}}==

{{trans|Python}}

L(word) File(‘unixdict.txt’).read().split("\n")

 

V count = max(anagram.values().map(ana -> ana.len))

L(ana) anagram.values()

I ana.len == count

{{out}}

<pre>

 

=={{header|8th}}==

\

\ anagrams.8th

bye

;

=={{header|AArch64 Assembly}}==

{{works with|as|Raspberry Pi 3B version Buster 64 bits <br> or android 64 bits with application Termux }}

/* ARM assembly AARCH64 Raspberry PI 3B */

/* program anagram64.s */

/* for this file see task include a file in language AArch64 assembly */

.include "../includeARM64.inc"

<pre>

~/.../rosetta/asm1 $ anagram64

</pre>

=={{header|ABAP}}==

define update_progress.

call function 'SAPGUI_PROGRESS_INDICATOR'

return.

endif.

{{out}}

<pre>[ angel , angle , galen , glean , lange ]

 

=={{header|Ada}}==

 

with Ada.Containers.Indefinite_Ordered_Maps;

Iterate (Result, Put'Access);

Close (File);

{{out}}

<pre>

=={{header|ALGOL 68}}==

{{works with|ALGOL 68G|Any - tested with release 2.8.3.win32}} Uses the "read" PRAGMA of Algol 68 G to include the associative array code from the [[Associative_array/Iteration]] task.

# use the associative array in the Associate array/iteration task #

PR read "aArray.a68" PR

e := NEXT words

OD

{{out}}

<pre>

alger|glare|lager|large|regal

caret|carte|cater|crate|trace

</pre>

 

This is a rough translation of the J version, intermediate values are kept and verb trains are not used for clarity of data flow.

 

anagrams←{

tie←⍵ ⎕NTIE 0

({~' '∊¨(⊃/¯1↑[2]⍵)}ana)⌿ana ⋄ ⎕NUNTIE

}

On a unix system we can assume wget exists and can use it from dyalog to download the file.

 

 

'''Example:'''

⎕SH'wget http://wiki.puzzlers.org/pub/wordlists/unixdict.txt'

]display anagrams 'unixdict.txt'

'''Output:'''

<pre>

 

=={{header|AppleScript}}==

use scripting additions

 

on largestAnagramGroups(listOfWords)

script o

property wordList : listOfWords

end script

ignoring case

end repeat

set i to 1

if (thisText is not currentText) then

set currentText to thisText

set i to j

end if

end repeat

set {i, j} to thisRange

end repeat

end ignoring

end largestAnagramGroups

 

local wordFile, wordList

set wordList to paragraphs of (read wordFile as «class utf8»)

 

{{output}}

=={{header|ARM Assembly}}==

{{works with|as|Raspberry Pi <br> or android 32 bits with application Termux}}

/* ARM assembly Raspberry PI */

/* program anagram.s */

/***************************************************/

.include "../affichage.inc"

<pre>

bale able bela abel elba

=={{header|Arturo}}==

 

 

anagrams: #[]

 

loop select values anagrams 'x [5 =< size x] 'words ->

 

{{out}}

=={{header|AutoHotkey}}==

Following code should work for AHK 1.0.* and 1.1* versions:

Loop, Parse, Contents, % "`n", % "`r"

{ ; parsing each line of the file we just read

Else ; output only those sets of letters that scored the maximum amount of common words

Break

{{out}}

<pre>

 

=={{header|AWK}}==

# syntax: GAWK -f JUMBLEA.AWK UNIXDICT.TXT

{ for (i=1; i<=NF; i++) {

}

return(str)

{{out}}

<pre>

Alternatively, non-POSIX version:

{{works with|gawk}}

 

{ patsplit($0, chars, ".")

if (count[i] == countMax)

print substr(accum[i], 2)

 

 

DECLARE idx$ ASSOC STRING

FOR y = 0 TO x-1

IF MaxCount = AMOUNT(idx$(n$[y])) THEN PRINT n$[y], ": ", idx$(n$[y])

{{out}}

<pre>

</pre>

 

{{works with|BBC BASIC for Windows}}

sort% = FN_sortinit(0,0)

C% = LEN(word$)

CALL sort%, char&(0)

{{out}}

<pre>

=={{header|BQN}}==

 

⟨ "alger" "glare" "lager" "large" "regal" ⟩

⟨ "angel" "angle" "galen" "glean" "lange" ⟩

⟨ "caret" "carte" "cater" "crate" "trace" ⟩

⟨ "elan" "lane" "lean" "lena" "neal" ⟩

 

Assumes that <code>unixdict.txt</code> is in the same folder. The [[mlochbaum/BQN|JS implementation]] must be run in Node.js to have access to the filesystem.

This solution makes extensive use of Bracmat's computer algebra mechanisms. A trick is needed to handle words that are merely repetitions of a single letter, such as <code>iii</code>. That's why the variabe <code>sum</code> isn't initialised with <code>0</code>, but with a non-number, in this case the empty string. Also te correct handling of characters 0-9 needs a trick so that they are not numerically added: they are prepended with a non-digit, an <code>N</code> in this case. After completely traversing the word list, the program writes a file <code>product.txt</code> that can be visually inspected.

The program is not fast. (Minutes rather than seconds.)

& 1:?product

& whl

| out$!group

)

{{out}}

<pre> abel+able+bale+bela+elba

 

=={{header|C}}==

#include <stdlib.h>

#include <string.h>

fclose(f1);

return 0;

{{out}} (less than 1 second on old P500)

<pre>5:vile, veil, live, levi, evil,

</pre>

A much shorter version with no fancy data structures:

#include <stdlib.h>

#include <string.h>

close(fd);

return 0;

{{out}}

<pre>

 

=={{header|C sharp|C#}}==

using System.IO;

using System.Linq;

}

}

{{out}}

<pre>

 

=={{header|C++}}==

#include <fstream>

#include <string>

}

return 0;

{{out}}

abel, able, bale, bela, elba,

=={{header|Clojure}}==

Assume ''wordfile'' is the path of the local file containing the words. This code makes a map (''groups'') whose keys are sorted letters and values are lists of the key's anagrams. It then determines the length of the longest list, and prints out all the lists of that length.

 

(def groups

maxlength (count (first wordlists))]

(doseq [wordlist (take-while #(= (count %) maxlength) wordlists)]

 

(->> (slurp "http://wiki.puzzlers.org/pub/wordlists/unixdict.txt")

clojure.string/split-lines

;; ["evil" "levi" "live" "veil" "vile"]

;; ["abel" "able" "bale" "bela" "elba"])

 

=={{header|CLU}}==

anagrams = cluster is new, add, largest_size, sets

anagram_set = struct[letters: string, words: array[string]]

stream$putl(po, "")

end

{{out}}

<pre>Largest amount of anagrams per set: 5

Tested with GnuCOBOL 2.0. ALLWORDS output display trimmed for width.

 

*> wget http://wiki.puzzlers.org/pub/wordlists/unixdict.txt

*> or visit https://sourceforge.net/projects/souptonuts/files

.

 

 

{{out}}

 

=={{header|CoffeeScript}}==

 

show_large_anagram_sets = (word_lst) ->

req.end()

{{out}}

[ 'abel', 'able', 'bale', 'bela', 'elba' ]

[ 'alger', 'glare', 'lager', 'large', 'regal' ]

[ 'caret', 'carte', 'cater', 'crate', 'trace' ]

[ 'elan', 'lane', 'lean', 'lena', 'neal' ]

 

=={{header|Common Lisp}}==

{{libheader|DRAKMA}} to retrieve the wordlist.

(let ((words (drakma:http-request url :want-stream t))

(wordsets (make-hash-table :test 'equalp)))

else if (eql (car pair) maxcount)

do (push (cdr pair) maxwordsets)

Evalutating

(pprint wordsets)

{{out}}

<pre>(("vile" "veil" "live" "levi" "evil")

5</pre>

Another method, assuming file is local:

(with-open-file (stream file)

(loop with w = "" while w collect (setf w (read-line stream nil)))))

longest))

 

{{out}}

<pre>elba bela bale able abel

=={{header|Component Pascal}}==

BlackBox Component Builder

MODULE BbtAnagrams;

IMPORT StdLog,Files,Strings,Args;

END BbtAnagrams.

Execute:^Q BbtAnagrams.DoProcess unixdict.txt~<br/>

{{out}}

=={{header|Crystal}}==

{{trans|Ruby}}

 

response = HTTP::Client.get("http://wiki.puzzlers.org/pub/wordlists/unixdict.txt")

anagram.each_value { |ana| puts ana if ana.size >= count }

end

 

{{out}}

=={{header|D}}==

===Short Functional Version===

 

void main() {

immutable m = an.byValue.map!q{ a.length }.reduce!max;

writefln("%(%s\n%)", an.byValue.filter!(ws => ws.length == m));

{{out}}

<pre>["caret", "carte", "cater", "crate", "trace"]

===Faster Version===

Less safe, same output.

import std.stdio, std.algorithm, std.file, std.string;

 

immutable m = anags.byValue.map!q{ a.length }.reduce!max;

writefln("%(%-(%s %)\n%)", anags.byValue.filter!(ws => ws.length == m));

Runtime: about 0.06 seconds.

 

{{libheader| System.Classes}}

{{libheader| System.Diagnostics}}

program AnagramsTest;

 

end.

 

 

{{out}}

 

=={{header|E}}==

when (def wordText := <http://wiki.puzzlers.org/pub/wordlists/unixdict.txt> <- getText()) -> {

def words := wordText.split("\n")

println(anagramGroup.snapshot())

}

 

=={{header|EchoLisp}}==

For a change, we will use the french dictionary - '''(lib 'dico.fr)''' - delivered within EchoLisp.

(require 'struct)

(require 'hash)

(cdr h))

))

{{out}}

(length mots-français)

→ 209315

→ { alisen enlias enlisa ensila islaen islean laines lianes salien saline selina }

 

 

=={{header|Eiffel}}==

class

ANAGRAMS

 

end

{{out}}

<pre>

=={{header|Ela}}==

{{trans|Haskell}}

 

groupon f x y = f x == f y

let wix = groupBy (groupon fst) << sort $ zip (map sort words) words

let mxl = maximum $ map length wix

 

{{out}}<pre>["vile","veil","live","levi","evil"]

 

=={{header|Elena}}==

import system'calendar;

import system'io;

import extensions'routines;

import extensions'text;

 

extension op

auto dictionary := new Map<string,object>();

 

{

var key := word.normalized();

};

};

 

dictionary.Values

var end := now;

console.readChar()

{{out}}

<pre>

abel,able,bale,bela,elba

evil,levi,live,veil,vile

elan,lane,lean,lena,neal

emit,item,mite,time

amen,mane,mean,name

hare,hear,hera,rhea

keats,skate,stake,steak

</pre>

 

=={{header|Elixir}}==

def find(file) do

File.read!(file)

end

 

 

{{out}}

 

The same output, using <code>File.Stream!</code> to generate <code>tuples</code> containing the word and it's sorted value as <code>strings</code>.

|> Stream.map(&String.strip &1)

|> Enum.group_by(&String.codepoints(&1) |> Enum.sort)

|> Enum.max

|> elem(1)

 

{{out}}

=={{header|Erlang}}==

The function fetch/2 is used to solve [[Anagrams/Deranged_anagrams]]. Please keep backwards compatibility when editing. Or update the other module, too.

-compile(export_all).

 

get_value([], _, _, L) ->

L.

{{out}}

<pre>

 

=={{header|Euphoria}}==

 

function compare_keys(sequence a, sequence b)

puts(1,"\n")

end if

{{out}}

<pre>abel bela bale elba able

=={{header|F Sharp|F#}}==

Read the lines in the dictionary, group by the sorted letters in each word, find the length of the longest sets of anagrams, extract the longest sequences of words sharing the same letters (i.e. anagrams):

Note that it is necessary to convert the sorted letters in each word from sequences to arrays because the groupBy function uses the default comparison and sequences do not compare structurally (but arrays do in F#).

 

Takes 0.8s to return:

seq

[seq ["abel"; "able"; "bale"; "bela"; "elba"];

seq ["caret"; "carte"; "cater"; "crate"; "trace"];

seq ["elan"; "lane"; "lean"; "lena"; "neal"];

 

=={{header|Fantom}}==

{

// take given word and return a string rearranging characters in order

}

}

 

{{out}}

=={{header|Fortran}}==

This program:

module anagram_routines

!***************************************************************************************

!***************************************************************************************

end program main

 

{{out}}

=={{header|FBSL}}==

'''A little bit of cheating: literatim re-implementation of C solution in FBSL's Dynamic C layer.'''

 

DIM gtc = GetTickCount()

fclose(f1);

}

{{out}} (2.2GHz Intel Core2 Duo)

<pre>25104 words in dictionary max ana=5

 

=={{header|Factor}}==

[ [ natural-sort >string ] keep ] { } map>assoc sort-keys

[ [ first ] compare +eq+ = ] monotonic-split

{ "abel" "able" "bale" "bela" "elba" }

{ "caret" "carte" "cater" "crate" "trace" }

{ "elan" "lane" "lean" "lena" "neal" }

{ "evil" "levi" "live" "veil" "vile" }

 

=={{header|FreeBASIC}}==

 

Type IndexedWord

Print

Print "Press any key to quit"

 

{{out}}

 

=={{header|Frink}}==

d = new dict

for w = lines["http://wiki.puzzlers.org/pub/wordlists/unixdict.txt"]

i = i + 1

}

 

=={{header|FutureBasic}}==

 

 

 

next

end fn = result

 

end fn

 

end fn

 

 

Output:

<pre>

</pre>

 

 

 

 

 

 

 

 

 

 

 

 

<pre>

 

</pre>

 

=={{header|GAP}}==

local f, p, L, line, word, words, swords, res, cur, r;

words := [ ];

# [ "alger", "glare", "lager", "large", "regal" ],

# [ "elan", "lane", "lean", "lena", "neal" ],

 

=={{header|Go}}==

 

import (

func (b byteSlice) Len() int { return len(b) }

func (b byteSlice) Swap(i, j int) { b[i], b[j] = b[j], b[i] }

{{out}}

<pre>

=={{header|Groovy}}==

This program:

def groups = words.groupBy{ it.toList().sort() }

def bigGroupSize = groups.collect{ it.value.size() }.max()

def isBigAnagram = { it.value.size() == bigGroupSize }

{{out}}

<pre>

 

=={{header|Haskell}}==

 

groupon f x y = f x == f y

wix = groupBy (groupon fst) . sort $ zip (map sort words) words

mxl = maximum $ map length wix

{{out}}

["abel","able","bale","bela","elba"]

["caret","carte","cater","crate","trace"]

["alger","glare","lager","large","regal"]

["elan","lane","lean","lena","neal"]

 

and we can noticeably speed up the second stage sorting and grouping by packing the String lists of Chars to the Text type:

 

import Data.Ord (comparing)

import Data.Function (on)

mapM_

(print . fmap snd)

{{Out}}

<pre>["abel","able","bale","bela","elba"]

 

=={{header|Icon}} and {{header|Unicon}}==

every writeSet(!getLongestAnagramSets())

end

every (s := "") ||:= (find(c := !cset(w),w),c)

return s

Sample run:

<pre>->an <unixdict.txt

=={{header|J}}==

If the unixdict file has been retrieved and saved in the current directory (for example, using wget):

+-----+-----+-----+-----+-----+

|abel |able |bale |bela |elba |

+-----+-----+-----+-----+-----+

|evil |levi |live |veil |vile |

Explanation:

This reads in the dictionary and produces a list of boxes. Each box contains one line (one word) from the dictionary.

This groups the words into rows where anagram equivalents appear in the same row. In other words, creates a copy of the original list where the characters contained in each box have been sorted. Then it organizes the contents of the original list in rows, with each new row keyed by the values in the new list.

This selects rows whose last element is not an empty box.<br>

(In the previous step we created an array of rows of boxes. The short rows were automatically padded with empty boxes so that all rows would be the same length.)

The key to this algorithm is the sorting of the characters in each word from the dictionary. The line <tt>Arrays.sort(chars);</tt> sorts all of the letters in the word in ascending order using a built-in [[quicksort]], so all of the words in the first group in the result end up under the key "aegln" in the anagrams map.

{{works with|Java|1.5+}}

import java.io.*;

import java.util.*;

System.out.println(ana);

}

{{works with|Java|1.8+}}

import java.io.*;

import java.util.*;

;

}

{{out}}

[angel, angle, galen, glean, lange]

===ES5===

{{Works with|Node.js}}

var words = fs.readFileSync('unixdict.txt', 'UTF-8').split('\n');

 

}

}

 

{{Out}}

 

Alternative using reduce:

var dictionary = fs.readFileSync('unixdict.txt', 'UTF-8').split('\n');

 

keysSortedByFrequency.slice(0, 10).forEach(function (key) {

console.log(sortedDict[key].join(' '));

 

 

Using JavaScript for Automation

(A JavaScriptCore interpreter on macOS with an Automation library).

'use strict';

 

// MAIN ---

return main();

{{Out}}

<pre>[

 

=={{header|jq}}==

(reduce .[] as $word (

{table: {}, max: 0}; # state

# The task:

split("\n") | anagrams

{{Out}}

$ jq -M -s -c -R -f anagrams.jq unixdict.txt

["abel","able","bale","bela","elba"]

["elan","lane","lean","lena","neal"]

["evil","levi","live","veil","vile"]

 

=={{header|Jsish}}==

From Javascript, nodejs entry.

var datafile = 'unixdict.txt';

if (console.args[0] == '-more' && Interp.conf('maxArrayList') > 500000)

evil levi live veil vile

=!EXPECTEND!=

 

{{out}}

=={{header|Julia}}==

{{works with|Julia|1.6}}

wordlist = open(readlines, download(url))

 

end

 

 

{{out}}

 

=={{header|K}}==

 

=={{header|Kotlin}}==

{{trans|Java}}

import java.io.InputStreamReader

import java.net.URL

.filter { it.size == count }

.forEach { println(it) }

 

{{out}}

 

=={{header|Lasso}}==

anagrams = map,

words = include_url('http://wiki.puzzlers.org/pub/wordlists/unixdict.txt')->split('\n'),

 

#findings -> join('<br />\n')

{{out}}

<pre>abel, able, bale, bela, elba

 

=={{header|Liberty BASIC}}==

open "unixdict.txt" for input as #1

while not(eof(#1))

sorted$=sorted$+chrSort$(chr)

next

 

=={{header|LiveCode}}==

LiveCode could definitely use a sort characters command. As it is this code converts the letters into items and then sorts that. I wrote a merge sort for characters, but the conversion to items, built-in-sort, conversion back to string is about 10% faster, and certainly easier to write.

 

put mostCommonAnagrams(url "http://wiki.puzzlers.org/pub/wordlists/unixdict.txt")

end mouseUp

replace comma with empty in X

return X

{{out}}

<pre>abel,able,bale,bela,elba

=={{header|Lua}}==

Lua's core library is very small and does not include built-in network functionality. If a networking library were imported, the local file in the following script could be replaced with the remote dictionary file.

local bytes = {word:byte(1, -1)}

table.sort(bytes)

print('') -- Finish with a newline.

end

{{out}}

<pre>abel able bale bela elba

 

=={{header|M4}}==

changequote(`[',`]')

define([for],

_max

for([x],1,_n,[ifelse(_get([count],x),_max,[_get([list],x)

 

Memory limitations keep this program from working on the full-sized dictionary.

The convert call discards the hashes, which have done their job, and leaves us with a list L of anagram sets.

Finally, we just note the size of the largest sets of anagrams, and pick those off.

words := HTTP:-Get( "http://wiki.puzzlers.org/pub/wordlists/unixdict.txt" )[2]: # ignore errors

use StringTools, ListTools in

m := max( map( nops, L ) ); # what is the largest set?

A := select( s -> evalb( nops( s ) = m ), L ); # get the maximal sets of anagrams

The result of running this code is

A := [{"abel", "able", "bale", "bela", "elba"}, {"angel", "angle", "galen",

"glean", "lange"}, {"alger", "glare", "lager", "large", "regal"}, {"evil",

"levi", "live", "veil", "vile"}, {"caret", "carte", "cater", "crate", "trace"}

, {"elan", "lane", "lean", "lena", "neal"}];

 

=={{header|Mathematica}}/{{header|Wolfram Language}}==

Download the dictionary, split the lines, split the word in characters and sort them. Now sort by those words, and find sequences of equal 'letter-hashes'. Return the longest sequences:

text={#,StringJoin@@Sort[Characters[#]]}&/@list;

text=SortBy[text,#[[2]]&];

splits=Split[text,#1[[2]]==#2[[2]]&][[All,All,1]];

maxlen=Max[Length/@splits];

gives back:

An alternative is faster, but requires version 7 (for <code>Gather</code>):

maxlen = Max[Length /@ splits];

 

Or using build-in functions for sorting and gathering elements in lists it can be implimented as:

Also, Mathematica's own word list is available; replacing the list definition with <code>list = WordData[];</code> and forcing <code>maxlen</code> to 5 yields instead this result:

 

 

Also if using Mathematica 10 it gets really concise:

 

=={{header|Maxima}}==

while stringp(s: readline(file)) do L: cons(s, L), close(file), L)$

 

["angel", "angle", "galen", "glean", "lange"],

["caret", "carte", "cater", "crate", "trace"],

 

=={{header|MUMPS}}==

Set file="unixdict.txt"

Open file:"r" Use file

Quit

 

<pre>

The anagrams with the most variations:

===Java&ndash;Like===

{{trans|Java}}

options replace format comments java crossref symbols nobinary

 

 

return

{{out}}

<pre>

===Rexx&ndash;Like===

Implemented with more NetRexx idioms such as indexed strings, <tt>PARSE</tt> and the NetRexx &quot;built&ndash;in functions&quot;.

options replace format comments java crossref symbols nobinary

 

 

Return

{{out}}

<pre>

 

=={{header|NewLisp}}==

;;; Get the words as a list, splitting at newline

(setq data

;;; Print out only groups of more than 4 words

(map println (filter (fn(x) (> (length x) 4)) (group-by-key)))

{{out}}

<pre>

 

=={{header|Nim}}==

import tables, strutils, algorithm

 

 

main()

{{out}}

<pre>

=={{header|Oberon-2}}==

Oxford Oberon-2

MODULE Anagrams;

IMPORT Files,Out,In,Strings;

DoProcess("unixdict.txt");

END Anagrams.

{{out}}

<pre>

 

=={{header|Objeck}}==

use Collection;

 

}

}

{{out}}

<pre>[abel,able,bale,bela,elba]

 

=={{header|OCaml}}==

let l = ref [] in

let n = String.length str in

( List.iter (Printf.printf " %s") lw;

print_newline () )

 

=={{header|Oforth}}==

 

import: collect

import: quicksort

filter( #[ second size m == ] )

apply ( #[ second .cr ] )

 

{{out}}

Two versions of this, using different collection classes.

===Version 1: Directory of arrays===

-- This assumes you've already downloaded the following file and placed it

-- in the current directory: http://wiki.puzzlers.org/pub/wordlists/unixdict.txt

say letters":" list~makestring("l", ", ")

end

===Version 2: Using the relation class===

This version appears to be the fastest.

-- This assumes you've already downloaded the following file and placed it

-- in the current directory: http://wiki.puzzlers.org/pub/wordlists/unixdict.txt

say letters":" words~makestring("l", ", ")

end

Timings taken on my laptop:

<pre>

 

=={{header|Oz}}==

%% Helper function

fun {ReadLines Filename}

%% Display result (make sure strings are shown as string, not as number lists)

{Inspector.object configureEntry(widgetShowStrings true)}

 

=={{header|Pascal}}==

 

// assumes a local file

AnagramList[i].Destroy;

 

{{out}}

<pre>

 

=={{header|Perl}}==

 

my @words = split "\n", do { local( @ARGV, $/ ) = ( 'unixdict.txt' ); <> };

for my $ana (values %anagram) {

print "@$ana\n" if @$ana == $count;

If we calculate <code>$max</code>, then we don't need the CPAN module:

$max > @$_ or $max = @$_ for values %anagram;

{{out}}

alger glare lager large regal

=={{header|Phix}}==

copied from Euphoria and cleaned up slightly

<span style="color: #004080;">integer</span> <span style="color: #000000;">fn</span> <span style="color: #0000FF;">=</span> <span style="color: #7060A8;">open</span><span style="color: #0000FF;">(</span><span style="color: #008000;">"demo/unixdict.txt"</span><span style="color: #0000FF;">,</span><span style="color: #008000;">"r"</span><span style="color: #0000FF;">)</span>

<span style="color: #004080;">sequence</span> <span style="color: #000000;">words</span> <span style="color: #0000FF;">=</span> <span style="color: #0000FF;">{},</span> <span style="color: #000000;">anagrams</span> <span style="color: #0000FF;">=</span> <span style="color: #0000FF;">{},</span> <span style="color: #000000;">last</span><span style="color: #0000FF;">=</span><span style="color: #008000;">""</span><span style="color: #0000FF;">,</span> <span style="color: #000000;">letters</span>

<span style="color: #008080;">end</span> <span style="color: #008080;">if</span>

<span style="color: #008080;">end</span> <span style="color: #008080;">for</span>

{{out}}

<pre>

 

=={{header|Phixmonti}}==

 

"unixdict.txt" "r" fopen var f

len for

get len maxlen == if ? else drop endif

 

Other solution

 

 

( )

len for

get len maxlen == if ? else drop endif

 

{{out}}<pre>["abel", "able", "bale", "bela", "elba"]

 

=={{header|PHP}}==

$words = explode("\n", file_get_contents('http://wiki.puzzlers.org/pub/wordlists/unixdict.txt'));

foreach ($words as $word) {

if (count($ana) == $best)

print_r($ana);

 

=={{header|Picat}}==

Using foreach loop:

Dict = new_map(),

foreach(Line in read_file_lines("unixdict.txt"))

println(Value)

end,

 

[alger,glare,lager,large,regal]

[evil,levi,live,veil,vile]

 

Same idea, but shorter version by (mis)using list comprehensions.

M = new_map(),

_ = [_:W in read_file_lines("unixdict.txt"),S=sort(W),M.put(S,M.get(S,"")++[W])],

X = max([V.len : _K=V in M]),

println(maxLen=X),

 

[[evil,levi,live,veil,vile],[abel,able,bale,bela,elba],[caret,carte,cater,crate,trace],[angel,angle,galen,glean,lange],[elan,lane,lean,lena,neal],[alger,glare,lager,large,regal]]</pre>

 

=={{header|PicoLisp}}==

A straight-forward implementation using 'group' takes 48 seconds on a 1.7 GHz Pentium:

(by length sort

(by '((L) (sort (copy L))) group

Using a binary tree with the 'idx' function, it takes only 0.42 seconds on the same machine, a factor of 100 faster:

(in "unixdict.txt"

(while (line)

(push (car @) Word)

(set Key (list Word)) ) ) ) )

{{out}}

<pre>-> (("vile" "veil" "live" "levi" "evil") ("trace" "crate" "cater" "carte" "caret

 

=={{header|PL/I}}==

 

word_test: proc options (main);

end is_anagram;

 

{{out}}

<pre>

 

=={{header|Pointless}}==

readFileLines("unixdict.txt")

|> reduce(logWord, {})

getMax(groups) =

groups |> filter(g => length(g) == maxLength)

 

{{out}}

=={{header|PowerShell}}==

{{works with|PowerShell|2}}

$words = -split ($c.DownloadString('http://wiki.puzzlers.org/pub/wordlists/unixdict.txt'))

$top_anagrams = $words `

| Select-Object -First 1

 

{{out}}

<pre>abel, able, bale, bela, elba

evil, levi, live, veil, vile</pre>

Another way with more .Net methods is quite a different style, but drops the runtime from 2 minutes to 1.5 seconds:

 

$uri = 'http://wiki.puzzlers.org/pub/wordlists/unixdict.txt'

[string]::join('', $entry.Value)

}

 

=={{header|Processing}}==

 

void setup() {

}

}

 

{{out}}

=={{header|Prolog}}==

{{works with|SWI-Prolog|5.10.0}}

 

anagrams:-

length(V1, L1),

length(V2, L2),

The result is

<pre>[abel,able,bale,bela,elba]

=={{header|PureBasic}}==

{{works with|PureBasic|4.4}}

OpenConsole()

PrintN("Press any key"): Repeat: Until Inkey() <> ""

EndIf

{{out}}

<pre>evil, levi, live, veil, vile

===Python 3.X Using defaultdict===

Python 3.2 shell input (IDLE)

>>> from collections import defaultdict

>>> words = urllib.request.urlopen('http://wiki.puzzlers.org/pub/wordlists/unixdict.txt').read().split()

>>> for ana in anagram.values():

if len(ana) >= count:

 

===Python 2.7 version===

Python 2.7 shell input (IDLE)

>>> from collections import defaultdict

>>> words = urllib.urlopen('http://wiki.puzzlers.org/pub/wordlists/unixdict.txt').read().split()

>>> count

5

 

===Python: Using groupby===

{{trans|Haskell}}

{{works with|Python|2.6}} sort and then group using groupby()

>>> words = urllib.urlopen('http://wiki.puzzlers.org/pub/wordlists/unixdict.txt').read().split()

>>> len(words)

>>> count

5

 

 

Or, disaggregating, speeding up a bit by avoiding the slightly expensive use of ''sorted'' as a key, updating for Python 3, and using a local ''unixdict.txt'':

{{Works with|Python|3.7}}

 

from os.path import expanduser

# MAIN ---

if __name__ == '__main__':

{{Out}}

<pre>caret carte cater crate creat creta react recta trace

 

=={{header|QB64}}==

$CHECKING:OFF

' Warning: Keep the above line commented out until you know your newly edited code works.

IF i < Finish THEN QSort i, Finish

END SUB

 

'''2nd solution (by Steve McNeill):'''

$CHECKING:OFF

SCREEN _NEWIMAGE(640, 480, 32)

LOOP UNTIL gap = 1 AND swapped = 0

END SUB

 

'''Output:'''

LOOPER: 7134 executions from start to finish, in one second.

Note, this is including disk access for new data each time.

caret, trace, crate, carte, cater

bale, abel, able, elba, bela

 

=={{header|Quackery}}==

 

[] swap witheach

[ dup sort

else drop ]

drop cr ]

 

{{out}}

 

=={{header|R}}==

word_group <- sapply(

strsplit(words, split=""), # this will split all words to single letters...

"angel, angle, galen, glean, lange" "alger, glare, lager, large, regal"

aeln eilv

 

=={{header|Racket}}==

#lang racket

 

 

(get-maxes (hash-words (get-lines "http://wiki.puzzlers.org/pub/wordlists/unixdict.txt")))

{{out}}

<pre>

 

{{works with|Rakudo|2016.08}}

my $max = @anagrams».elems.max;

 

 

{{out}}

 

{{works with|Rakudo|2016.08}}

'unixdict.txt'.IO.words # load words from file

.classify(*.comb.sort.join) # group by common anagram

.classify(*.value.elems) # group by number of anagrams in a group

.max(*.key).value # get the group with highest number of anagrams

 

=={{header|RapidQ}}==

dim x as integer, y as integer

dim SortX as integer

End

 

{{out}}

<pre>

 

=={{header|Rascal}}==

 

list[str] OrderedRep(str word){

longest = max([size(group) | group <- range(AnagramMap)]);

return [AnagramMap[rep]| rep <- AnagramMap, size(AnagramMap[rep]) == longest];

Returns:

{"glean","galen","lange","angle","angel"},

{"glare","lager","regal","large","alger"},

{"able","bale","abel","bela","elba"},

{"levi","live","vile","evil","veil"}

 

=={{header|Red}}==

 

m: make map! [] 25000

]

foreach v values-of m [ if maxx = length? v [print v] ]

{{out}}

<pre>abel able bale bela elba

This version doesn't assume that the dictionary is in alphabetical order, &nbsp; nor does it assume the

<br>words are in any specific case &nbsp; (lower/upper/mixed).

iFID= 'unixdict.txt' /*the dictionary input File IDentifier.*/

$=; !.=; ww=0; uw=0; most=0 /*initialize a bunch of REXX variables.*/

/*reassemble word with sorted letters. */

return @.a || @.b || @.c || @.d || @.e || @.f||@.g||@.h||@.i||@.j||@.k||@.l||@.m||,

Programming note: &nbsp; the long (wide) assignment for &nbsp; &nbsp; '''return @.a||'''... &nbsp; &nbsp; could've been coded as an elegant &nbsp; '''do''' &nbsp; loop instead of hardcoding 26 letters,<br>but since the dictionary (word list) is rather large, a rather expaciated method was used for speed.

 

===version 1.2, optimized===

This optimized version eliminates the &nbsp; '''sortA''' &nbsp; subroutine and puts that subroutine's code in-line.

iFID= 'unixdict.txt' /*the dictionary input File IDentifier.*/

$=; !.=; ww=0; uw=0; most=0 /*initialize a bunch of REXX variables.*/

/*reassemble word with sorted letters. */

return @.a || @.b || @.c || @.d || @.e || @.f||@.g||@.h||@.i||@.j||@.k||@.l||@.m||,

{{out|output|text=&nbsp; is the same as REXX version 1.1}}

 

===annotated version using &nbsp; PARSE===

(This algorithm actually utilizes a &nbsp; ''bin'' &nbsp; sort, &nbsp; one bin for each Latin letter.)

upper u /*fast method to uppercase a variable. */

/*another: u = translate(u) */

/*Note: the ? is prefixed to the letter to avoid */

/*collisions with other REXX one-character variables.*/

{{out|output|:}}

<pre>

 

===annotated version using a &nbsp; DO &nbsp; loop===

upper u /*fast method to uppercase a variable. */

L=length(u) /*get the length of the word (in bytes)*/

_.?n||_.?o||_.?p||_.?q||_.?r||_.?s||_.?t||_.?u||_.?v||_.?w||_.?x||_.?y||_.?z

 

{{out|output|:}}

<pre>

 

===version 2===

* 07.08.2013 Walter Pachl

* sorta for word compression courtesy Gerard Schildberger,

End

Return c.a||c.b||c.c||c.d||c.e||c.f||c.g||c.h||c.i||c.j||c.k||c.l||,

{{out}}

<pre>

 

=={{header|Ring}}==

# Project : Anagrams

 

end

return cnt

Output:

<pre>

 

=={{header|Ruby}}==

 

anagram = Hash.new {|hash, key| hash[key] = []} # map sorted chars to anagrams

p ana

end

{{out}}

<pre>

 

Short version (with lexical ordered result).

 

anagrams = open('http://wiki.puzzlers.org/pub/wordlists/unixdict.txt'){|f| f.read.split.group_by{|w| w.each_char.sort} }

anagrams.values.group_by(&:size).max.last.each{|group| puts group.join(", ") }

{{Out}}

<pre>

 

=={{header|Run BASIC}}==

mem$ = "CREATE TABLE anti(gram,ordr);

CREATE INDEX ord ON anti(ordr)"

print

next i

<pre>

abel able bale bela elba

Unicode is hard so the solution depends on what you consider to be an anagram: two strings that have the same bytes, the same codepoints, or the same graphemes. The first two are easily accomplished in Rust proper, but the latter requires an external library. Graphemes are probably the most correct way, but it is also the least efficient since graphemes are variable size and thus require a heap allocation per grapheme.

 

use std::fs::File;

use std::io::{BufRead,BufReader};

}

}

{{out}}

<pre>

If we assume an ASCII string, we can map each character to a prime number and multiply these together to create a number which uniquely maps to each anagram.

 

use std::path::Path;

use std::io::{self, BufRead, BufReader};

}

Ok(map.into_iter().map(|(_, entry)| entry).collect())

 

=={{header|Scala}}==

val vls = src.getLines.toList.groupBy(_.sorted).values

val max = vls.map(_.size).max

{{out}}

<pre>

----

Another take:

.fromURL("http://wiki.puzzlers.org/pub/wordlists/unixdict.txt").getLines.toList

.groupBy(_.sorted).values

.groupBy(_.size).maxBy(_._1)._2

.map(_.mkString("\t"))

{{out}}

<pre>

Uses two SRFI libraries: SRFI 125 for hash tables and SRFI 132 for sorting.

 

(import (scheme base)

(scheme char)

(map (lambda (grp) (list-sort string<? grp))

(largest-groups (read-groups)))))

 

{{out}}

 

=={{header|Seed7}}==

include "gethttp.s7i";

include "strifile.s7i";

var integer: maxLength is 0;

begin

while hasNext(dictFile) do

readln(dictFile, word);

end if;

end for;

 

{{out}}

 

=={{header|SETL}}==

anagrams := {};

while not eof(h) loop

end loop;

return A;

{{out}}

<pre>{abel able bale bela elba}

 

=={{header|Sidef}}==

file.open_r(\var fh, \var err) ->

|| die "Can't open file `#{file}' for reading: #{err}\n";

}

 

{{out}}

<pre>alger glare lager large regal

 

=={{header|Simula}}==

$ cim -m64 anagrams-hashmap.sim

;

 

END

{{out}}

<pre>

 

=={{header|Smalltalk}}==

dict:= Dictionary new.

list do: [:val|

add: val.

].

Documentation:

<pre>

{{works with|Smalltalk/X}}

instead of asking for the strings, read the file:

'unixdict.txt' asFilename

readingLinesDo:[:eachWord |

sortBySelector:#size)

reverse

{{out}}

<pre>

...</pre>

not sure if getting the dictionary via http is part of the task; if so, replace the file-reading with:

 

=={{header|SNOBOL4}}==

{{works with|Macro Spitbol}}

Note: unixdict.txt is passed in locally via STDIN. Newlines must be converted for Win/DOS environment.

define('sortw(str)a,i,j') :(sortw_end)

sortw a = array(size(str))

L3 j = j + 1; key = kv<j,1>; val = kv<j,2> :f(end)

output = eq(countw(val),max) key ': ' val :(L3)

{{out}}

<pre>abel: abel able bale bela elba

 

=={{header|Stata}}==

mata

a=st_sdata(.,.)

reshape wide v1, i(k) j(group) string

drop k

 

'''Output'''

 

=={{header|SuperCollider}}==

var text, words, sorted, dict = IdentityDictionary.new, findMax;

File.use("unixdict.txt".resolveRelative, "r", { |f| text = f.readAllString });

};

findMax.(dict)

 

Answers:

 

=={{header|Swift}}==

{{works with|Swift 2.0}}

 

 

let wordsURL = NSURL(string: "http://wiki.puzzlers.org/pub/wordlists/unixdict.txt")!

print("set \(i): \(thislist.sort())")

}

 

{{out}}

 

=={{header|Tcl}}==

package require http

 

puts $anagrams($key)

}

{{out}}

<pre>evil levi live veil vile

 

=={{header|Transd}}==

 

 

MainModule: {

_start: (λ

(with fs FileStream() words String()

(textin fs words)

( -|

)

))

<pre>

[[abel, able, bale, bela, elba],

 

=={{header|TUSCRIPT}}==

requestdata = REQUEST ("http://wiki.puzzlers.org/pub/wordlists/unixdict.txt")

 

PRINT cs," ",f,": ",a

ENDLOOP

{{out}}

<pre>

Process substitutions eliminate the need for command pipelines.

 

local host=$1

local uri=$2

done

 

 

{{output}}

The algorithm is to group the words together that are made from the same unordered lists of letters, then collect the groups together that have the same number of words in

them, and then show the collection associated with the highest number.

 

#show+

 

{{out}}

<pre>

 

=={{header|VBA}}==

Option Explicit

 

If (mini < j) Then Call SortTwoDimArray(myArr, mini, j, Colonne)

If (i < Maxi) Then Call SortTwoDimArray(myArr, i, Maxi, Colonne)

{{out}}

<pre>25104 words, in the dictionary

 

Time to go : 2,464844 seconds.</pre>

 

=={{header|Vedit macro language}}==

 

The word list is expected to be in the same directory as the script.

 

Repeat(ALL) {

Ins_Char(#8, OVERWRITE)

}

{{out}}

<pre>

evil levi live veil vile

</pre>

 

=={{header|Visual Basic .NET}}==

Imports System.Collections.ObjectModel

 

End Function

 

{{out}}

<PRE>

[EILV] evil, levi, live, veil, vile

</PRE>

 

=={{header|Wren}}==

{{libheader|Wren-sort}}

 

var words = File.read("unixdict.txt").split("\n").map { |w| w.trim() }

for (key in wordMap.keys) {

if (wordMap[key].count == most) System.print(wordMap[key])

 

{{out}}

 

=={{header|Yabasic}}==

maxw = 0 : c = 0 : dimens(c)

i = 0

d(j,p) = c

end if

 

=={{header|zkl}}==

// blob to dictionary: key is word "fuzzed", values are anagram words

.pump(Void,T(fcn(w,d){

"%d:%s: %s".fmt(v.len(),zz.strip(),

v.apply("strip").concat(","))

{{out}}

<pre>

</pre>

In the case where it is desirable to get the dictionary from the web, use this code:

var ZC=Import("zklCurl");

unixdict:=ZC().get(URL); //--> T(Data,bytes of header, bytes of trailer)

unixdict=unixdict[0].del(0,unixdict[1]); // remove HTTP header

 

{{omit from|6502 Assembly|unixdict.txt is much larger than the CPU's address space.}}

{{omit from|8080 Assembly|See 6502 Assembly.}}

{{omit from|Z80 Assembly|See 6502 Assembly.}}