Character codes

Given a character value in your language, print its code (could be ASCII code, Unicode code, or whatever your language uses). For example, the character 'a' (lowercase letter A) has a code of 97 in ASCII (as well as Unicode, as ASCII forms the beginning of Unicode). Conversely, given a code, print out the corresponding character.

Character codes
You are encouraged to solve this task according to the task description, using any language you may know.

ABAP

In ABAP you must first cast the character to a byte field and back to a number in order to get its ASCII value. <lang ABAP>report zcharcode data: c value 'A', n type i. field-symbols <n> type x.

assign c to <n> casting. move <n> to n. write: c, '=', n left-justified. </lang> Output:

A = 65

ActionScript

In ActionScript, you cannot take the character code of a character directly. Instead you must create a string and call charCodeAt with the character's position in the string as a parameter. <lang ActionScipt>trace(String.fromCharCode(97)); //prints 'a' trace("a".charCodeAt(0));//prints '97'</lang>

Ada

<lang ada>with Ada.Text_IO; use Ada.Text_IO;

procedure Char_Code is begin

  Put_Line (Character'Val (97) & " =" & Integer'Image (Character'Pos ('a')));

end Char_Code;</lang> The predefined language attributes S'Pos and S'Val for every discrete subtype, and Character is such a type, yield the position of a value and value by its position correspondingly. Sample output.

a = 97

ALGOL 68

In ALGOL 68 the format $g$ is type aware, hence the type conversion operators abs & repr are used to set the type. <lang algol68>main:(

 printf(($gl$, ABS "a")); # for ASCII this prints "+97" EBCDIC prints "+129" #
 printf(($gl$, REPR 97))  # for ASCII this prints "a"; EBCDIC prints "/" #

)</lang> Character conversions may be available in the standard prelude so that when a foreign tape is mounted, the characters will be converted transparently as the tape's records are read. <lang algol68>FILE tape; INT errno = open(tape, "/dev/tape1", stand out channel) make conv(tape, ebcdic conv); FOR record DO getf(tape, ( ~ )) OD; ~ # etc ... #</lang> Every channel has an associated standard character conversion that can be determined using the stand conv query routine and then the conversion applied to a particular file/tape. eg. <lang algol68> make conv(tape, stand conv(stand out channel))</lang>

AutoHotkey

<lang AutoHotkey>MsgBox % Chr(97) MsgBox % Asc("a")</lang>

AWK

AWK has not built-in way to convert a character into ASCII (or whatever) code; but a function that does so can be easily built using an associative array (where the keys are the characters). The opposite can be done using printf (or sprintf) with %c

<lang awk>function ord(c) {

 return chmap[c]

} BEGIN {

 for(i=0; i < 256; i++) {
   chmap[sprintf("%c", i)] = i
 }
 print ord("a"), ord("b")
 printf "%c %c\n", 97, 98
 s = sprintf("%c%c", 97, 98)
 print s

}</lang>

BASIC

Works with: QuickBasic version 4.5

<lang qbasic>charCode = 97 char = "a" PRINT CHR$(charCode) 'prints a PRINT ASC(char) 'prints 97</lang>

Befunge

<lang befunge>"A". 88*1+,</lang>

C

char is already an integer type in C, and it gets automatically promoted to int. So you can use a character where you would otherwise use an integer. Conversely, you can use an integer where you would normally use a character, except you may need to cast it, as char is smaller.

<lang c>#include <stdio.h>

int main() {

 printf("%d\n", 'a'); /* prints "97" */
 printf("%c\n", 97); /* prints "a"; we don't have to cast because printf is type agnostic */
 return 0;

}</lang>

C++

char is already an integer type in C++, and it gets automatically promoted to int. So you can use a character where you would otherwise use an integer. Conversely, you can use an integer where you would normally use a character, except you may need to cast it, as char is smaller.

In this case, the output operator << is overloaded to handle integer (outputs the decimal representation) and character (outputs just the character) types differently, so we need to cast it in both cases. <lang cpp>#include <iostream>

int main() {

 std::cout << (int)'a' << std::endl; // prints "97"
 std::cout << (char)97 << std::endl; // prints "a"
 return 0;

}</lang>

C#

C# represents strings and characters internally as Unicode, so casting a char to an int returns its Unicode character encoding. <lang csharp>using System;

namespace RosettaCode.CharacterCode {

   class Program
   {
       static void Main(string[] args)
       {
           Console.WriteLine((int) 'a');   //Prints "97"
           Console.WriteLine((char) 97);   //Prints "a"
       }
   }

}</lang>

Clojure

<lang clojure>(print (int \a)) ; prints "97" (print (char 97)) ; prints \a

Unicode is also available, as Clojure uses the underlying java Strings & chars

(print (int \π))  ; prints 960 (print (char 960)) ; prints \π</lang>

Common Lisp

<lang lisp>(princ (char-code #\a)) ; prints "97" (princ (code-char 97)) ; prints "a"</lang>

D

Could be treated like C, but since the standard string type is UTF-8, let's be verbose. <lang d>import std.stdio, std.utf;

void main() {

 string test = "a";
 size_t index = 0;   
 // get four-byte utf32 value for index 0
 // this returns dchar, so cast it to numeric
 writefln(cast(uint) test.decode(index));
 // index has moved to next character position in input
 assert(index == 1);

}</lang>

E

<lang e>? 'a'.asInteger()

1. value: 97

? <import:java.lang.makeCharacter>.asChar(97)

1. value: 'a'</lang>

Erlang

In Erlang, lists and strings are the same, only the representation changes. Thus: <lang erlang>1> F = fun([X]) -> X end.

1. Fun<erl_eval.6.13229925>

2> F("a"). 97</lang>

If entered manually, one can also get ASCII codes by prefixing characters with $: <lang erlang>3> $a. 97</lang>

Unicode is fully supported since release R13A only.

F#

<lang fsharp>let c = 'A' let n = 65 printfn "%d" (int c) printfn "%c" (char n)</lang>

Output

65
A

Factor

<lang factor>CHAR: katakana-letter-a . "ア" first .

12450 1string print</lang>

FALSE

<lang false>'A." "65,</lang>

Forth

As with C, characters are just integers on the stack which are treated as ASCII. <lang forth>char a dup . \ 97 emit \ a</lang>

Fortran

Functions ACHAR and IACHAR specifically work with the ASCII character set, while the results of CHAR and ICHAR will depend on the default character set being used. <lang fortran>WRITE(*,*) ACHAR(97), IACHAR("a") WRITE(*,*) CHAR(97), ICHAR("a")</lang>

GAP

<lang gap># Code must be in 0 .. 255. CHAR_INT(65);

1. 'A'

INT_CHAR('Z');

1. 90</lang>

Go

In Go, a character literal is simply an integer constant of the character code: <lang go>fmt.Println('a') // prints "97" fmt.Println('π') // prints "960"</lang>

To obtain the character code of the first Unicode character in a UTF-8 string: <lang go>rune, _ := utf8.DecodeRuneInString("π") fmt.Println(rune) // prints "960"</lang>

To print the character represented by a character code, we can create a string of that character (by converting the character code to string, it encodes the character in UTF-8): <lang go>fmt.Println(string(97)) // prints "a" fmt.Println(string(960)) // prints "π"</lang>

Golfscript

To convert a number to a string, we use the array to string coercion. <lang golfscript>97[]++p</lang>

To convert a string to a number, we have a many options, of which the simplest and shortest are: <lang golfscript>'a')\;p 'a'(\;p 'a'0=p 'a'{}/p</lang>

Groovy

Groovy does not have a character literal at all, so one-character strings have to be coerced to char. Groovy printf (like Java, but unlike C) is not type-agnostic, so the cast or coercion from char to int is also required. The reverse direction is considerably simpler. <lang groovy>printf ("%d\n", ('a' as char) as int) printf ("%c\n", 97)</lang>

Output:

97
a

Haskell

<lang haskell>import Data.Char

main = do

 print (ord 'a') -- prints "97"
 print (chr 97) -- prints "'a'"
 print (ord 'π') -- prints "960"
 print (chr 960) -- prints "'\960'"</lang>

HicEst

<lang hicest>WRITE(Messagebox) ICHAR('a'), CHAR(97)</lang>

Icon and Unicon

<lang Icon>procedure main(arglist) if *arglist > 0 then L := arglist else L := [97, "a"]

every x := !L do

  write(x, " ==> ", char(integer(x)) | ord(x) )  # char produces a character, ord produces a number

end</lang> Icon and Unicon do not currently support double byte character sets. Sample output:

97 ==> a
a ==> 97

J

<lang j> 4 u: 97 98 99 9786 abc☺

  3 u: 7 u: 'abc☺'

97 98 99 9786</lang>

Java

char is already an integer type in Java, and it gets automatically promoted to int. So you can use a character where you would otherwise use an integer. Conversely, you can use an integer where you would normally use a character, except you may need to cast it, as char is smaller.

In this case, the println method is overloaded to handle integer (outputs the decimal representation) and character (outputs just the character) types differently, so we need to cast it in both cases. <lang java>public class Foo {

   public static void main(String[] args) {
       System.out.println((int)'a'); // prints "97"
       System.out.println((char)97); // prints "a"
   }

}</lang>

Java characters support Unicode: <lang java>public class Bar {

   public static void main(String[] args) {
       System.out.println((int)'π'); // prints "960"
       System.out.println((char)960); // prints "π"
   }

}</lang>

JavaScript

Here character is just a string of length 1 <lang javascript>document.write('a'.charCodeAt(0)); // prints "97" document.write(String.fromCharCode(97)); // prints "a"</lang>

Joy

<lang joy>'a ord. 97 chr.</lang>

Liberty BASIC

<lang lb>charCode = 97 char$ = "a" print chr$(charCode) 'prints a print asc(char$) 'prints 97</lang>

Logo

Logo characters are words of length 1. <lang logo>print ascii "a  ; 97 print char 97  ; a</lang>

Logtalk

<lang logtalk>|?- char_code(Char, 97), write(Char). a Char = a yes</lang> <lang logtalk>|?- char_code(a, Code), write(Code). 97 Code = 97 yes</lang>

Lua

<lang lua>print(string.byte(io.read()))</lang>

Mathematica

Use the FromCharacterCode and ToCharacterCode functions: <lang Mathematica>ToCharacterCode["abcd"] FromCharacterCode[{97}]</lang>

Result:

{97, 98, 99, 100}

"a"

MATLAB

There are two built-in function that perform these tasks. To convert from a number to a character use: <lang MATLAB>character = char(asciiNumber)</lang>

To convert from a character to its corresponding ascii character use: <lang MATLAB>asciiNumber = double(character)</lang>

or if you need this number as an integer not a double use: <lang MATLAB>asciiNumber = uint16(character) asciiNumber = uint32(character) asciiNumber = uint64(character)</lang>

Sample Usage: <lang MATLAB>>> char(87)

ans =

W

>> double('W')

ans =

   87

>> uint16('W')

ans =

    87</lang>

Metafont

Metafont handles only ASCII (even though codes beyond 127 can be given and used as real ASCII codes)

<lang metafont>message "enter a letter: "; string a; a := readstring; message decimal (ASCII a); % writes the decimal number of the first character

                          % of the string a

message "enter a number: "; num := scantokens readstring; message char num;  % num can be anything between 0 and 255; what will be seen

                   % on output depends on the encoding used by the "terminal"; e.g.
                   % any code beyond 127 when UTF-8 encoding is in use will give
                   % a bad encoding; e.g. to see correctly an "è", we should write

message char10;  % (this add a newline...) message char hex"c3" & char hex"a8";  % since C3 A8 is the UTF-8 encoding for "è" end</lang>

Modula-3

The built in functions ORD and VAL work on characters, among other things. <lang modula3>ORD('a') (* Returns 97 *) VAL(97, CHAR); (* Returns 'a' *)</lang>

MUMPS

<lang MUMPS>WRITE $ASCII("M") WRITE $CHAR(77)</lang>

Objeck

<lang objeck> 'a'->As(Int)->PrintLine(); 97->As(Char)->PrintLine(); </lang>

OCaml

<lang ocaml>Printf.printf "%d\n" (int_of_char 'a'); (* prints "97" *) Printf.printf "%c\n" (char_of_int 97); (* prints "a" *)</lang>

Oz

Characters in Oz are the same as integers in the range 0-255 (ISO 8859-1 encoding). To print a number as a character, we need to use it as a string (i.e. a list of integers from 0 to 255): <lang oz>{System.show &a}  %% prints "97" {System.showInfo [97]}  %% prints "a"</lang>

PARI/GP

<lang>print(Vecsmall("a")[1]); print(Strchr([72, 101, 108, 108, 111, 44, 32, 119, 111, 114, 108, 100, 33]))</lang>

Pascal

<lang pascal>writeln(ord('a')); writeln(chr(97));</lang>

Perl

Here character is just a string of length 1 <lang perl>print ord('a'), "\n"; # prints "97" print chr(97), "\n"; # prints "a"</lang>

Perl 6

As Perl 5.

PHP

Here character is just a string of length 1 <lang php>echo ord('a'), "\n"; // prints "97" echo chr(97), "\n"; // prints "a"</lang>

PicoLisp

<lang PicoLisp>: (char "a") -> 97

(char "字")

-> 23383

(char 23383)

-> "字"

(chop "文字")

-> ("文" "字")

(mapcar char @)

-> (25991 23383)</lang>

PL/I

<lang PL/I> declare 1 u union,

         2 c character (1),
         2 i fixed binary (8) unsigned;

c = 'a'; put skip list (i); /* prints 97 */ i = 97; put skip list (c); /* prints 'a' */ </lang>

PowerShell

PowerShell does not allow for character literals directly, so to get a character one first needs to convert a single-character string to a char: <lang powershell>$char = [char] 'a'</lang> Then a simple cast to int yields the character code: <lang powershell>$charcode = [int] $char # => 97</lang> This also works with Unicode: <lang powershell>[int] [char] '☺' # => 9786</lang> For converting an integral character code into the actual character, a cast to char suffices: <lang powershell>[char] 97 # a [char] 9786 # ☺</lang>

PureBasic

PureBasic allows compiling code so that it will use either Ascii or a Unicode (UCS-2) encoding for representing its string content. It also allows for the source code that is being compiled to be in either Ascii or UTF-8 encoding. A one-character string is used here to hold the character and a numerical character type is used to hold the character code. The character type is either one or two bytes in size, depending on whether compiling for Ascii or Unicode respectively. <lang PureBasic>If OpenConsole()

 ;Results are the same when compiled for Ascii or Unicode
 charCode.c = 97
 Char.s = "a"
 PrintN(Chr(charCode))   ;prints a
 PrintN(Str(Asc(Char)))  ;prints 97

 Print(#CRLF$ + #CRLF$ + "Press ENTER to exit")
 Input()
 CloseConsole()

EndIf</lang>

This version should be compiled with Unicode setting and the source code to be encoded using UTF-8. <lang PureBasic>If OpenConsole()

 ;UTF-8 encoding compiled for Unicode (UCS-2)
 charCode.c = 960 
 Char.s = "π"            
 PrintN(Chr(charCode))   ;prints π
 PrintN(Str(Asc(Char)))  ;prints 960

 Print(#CRLF$ + #CRLF$ + "Press ENTER to exit")
 Input()
 CloseConsole()

EndIf</lang>

Python

2.x

Here character is just a string of length 1

8-bit characters: <lang python>print ord('a') # prints "97" print chr(97) # prints "a"</lang>

Unicode characters: <lang python>print ord(u'π') # prints "960" print unichr(960) # prints "π"</lang>

3.x

Here character is just a string of length 1 <lang python>print(ord('a')) # prints "97" print(ord('π')) # prints "960" print(chr(97)) # prints "a" print(chr(960)) # prints "π"</lang>

R

<lang R>ascii <- as.integer(charToRaw("hello world")); ascii text <- rawToChar(as.raw(ascii)); text</lang>

REXX

REXX supports handling of characters with built-in functions, whether it be hexadecimal, binary (bits), or decimal codes. <lang rexx> yyy='c' /*assign a lowercase c to YYY.*/ yyy='34'x /*assign hexadecimal 34 to YYY.*/ yyy=x2c(34) /* (same as above) */ yyy='00110100'b /* (same as above) */ yyy=d2c(97) /*assign decimal code 97 to YYY.*/

say y /*displays the value of YYY. */ say c2x(yyy) /*displays the value of YYY in hexadecimal. */ say c2d(yyy) /*displays the value of YYY in decimal. */ say x2b(c2x(yyy)) /*displays the value of YYY in binary (bit string). */ </lang>

Ruby

1.8

In Ruby 1.8 characters are usually represented directly as their integer character code. Ruby has a syntax for "character literal" which evaluates directly to the integer code: ?a evaluates to the integer 97. Subscripting a string also gives just the integer code for the character. <lang ruby>> ?a => 97 > "a"[0] => 97 > 97.chr => "a"</lang>

1.9

In Ruby 1.9 characters are represented as length-1 strings; same as in Python. The previous "character literal" syntax ?a is now the same as "a". Subscripting a string also gives a length-1 string. There is now an "ord" method of strings to convert a character into its integer code.

<lang ruby>> "a".ord => 97 > 97.chr => "a"</lang>

Sather

<lang sather>class MAIN is

 main is
   #OUT + 'a'.int + "\n"; -- or
   #OUT + 'a'.ascii_int + "\n";
   #OUT + CHAR::from_ascii_int(97) + "\n";
 end;

end;</lang>

Scala

Scala supports unicode characters, but each character is UTF-16, so there is not a 1-to-1 relationship for supplementary character sets.

Without worrying about supplemental character sets:

<lang scala>scala> 'a' toInt res9: Int = 97

scala> 97 toChar res10: Char = a</lang>

Worrying about supplemental character sets, we need to test the "next" character as well:

<lang scala>def charToInt(c: Char, next: Char): Option[Int] = (c, next) match {

 case _ if (c.isHighSurrogate && next.isLowSurrogate) => Some(java.lang.Character.toCodePoint(c, next))
 case _ if (c.isLowSurrogate) => None
 case _ => Some(c.toInt)

}

def intToChars(n: Int): Array[Char] = java.lang.Character.toChars(n)</lang>

Scheme

<lang scheme>(display (char->integer #\a)) (newline) ; prints "97" (display (integer->char 97)) (newline) ; prints "a"</lang>

Seed7

<lang seed7>writeln(ord('a')); writeln(chr(97));</lang>

Slate

<lang slate>$a code. 97 as: String Character.</lang>

Smalltalk

<lang smalltalk>($a asInteger) displayNl. "output 97" (Character value: 97) displayNl. "output a"</lang>

SNOBOL4

Snobol implementations may or may not have built-in char( ) and ord ( ) or asc( ). These are based on examples in the Snobol4+ tutorial and work with the native (1-byte) charset.

<lang SNOBOL4> define('chr(n)') :(chr_end) chr &alphabet tab(n) len(1) . chr :s(return)f(freturn) chr_end

       define('asc(str)c') :(asc_end)

asc str len(1) . c

       &alphabet break(c) @asc :s(return)f(freturn)

asc_end

• # Test and display

       output = char(65) ;* Built-in
       output = chr(65)
       output = asc('A')

end</lang>

Output:

A
A
65

Standard ML

<lang sml>print (Int.toString (ord #"a") ^ "\n"); (* prints "97" *) print (Char.toString (chr 97) ^ "\n"); (* prints "a" *)</lang>

Tcl

<lang tcl># ASCII puts [scan "a" %c]  ;# ==> 97 puts [format %c 97]  ;# ==> a

1. Unicode is the same

puts [scan "π" %c]  ;# ==> 960 puts [format %c 960] ;# ==> π</lang>

TI-89 BASIC

The TI-89 uses an 8-bit charset/encoding which is similar to ISO-8859-1, but with more mathematical symbols and Greek letters. At least codes 14-31, 128-160, 180 differ. The ASCII region is unmodified. (TODO: Give a complete list.)

The TI Connect X desktop software converts between this unique character set and Unicode characters, though sometimes in a consistent but inappropriate fashion.

The below program will display the character and code for any key pressed. Some keys do not correspond to characters and have codes greater than 255. The portion of the program actually implementing the task is marked with a line of “©”s.

<lang ti89b>Prgm

 Local k, s
 ClrIO
 Loop
   Disp "Press a key, or ON to exit."
   getKey() © clear buffer
   0 → k : While k = 0 : getKey() → k : EndWhile
   ClrIO
   If k ≥ 256 Then
     Disp "Not a character."
     Disp "Code: " & string(k)
   Else

     char(k) → s                           ©
     © char() and ord() are inverses.      ©
     Disp "Character: " & s                ©
     Disp "Code: " & string(ord(s))        ©

   EndIf
 EndLoop

EndPrgm</lang>

Trith

Characters are Unicode code points, so the solution is the same for Unicode characters as it is for ASCII characters: <lang trith>"a" ord print 97 chr print</lang> <lang trith>"π" ord print 960 chr print</lang>

TUSCRIPT

<lang tuscript> $$ MODE TUSCRIPT SET character ="a", code=DECODE (character,byte) PRINT character,"=",code </lang> Output:

a=97

Visual Basic .NET

<lang vbnet>Console.WriteLine(Chr(97)) 'Prints a Console.WriteLine(Asc("a")) 'Prints 97</lang>

Ursala

Character code functions are not built in but easily defined as reifications of the character table. <lang Ursala>#import std

1. import nat

chr = -: num characters asc = -:@rlXS num characters

1. cast %cnX

test = (chr97,asc`a)</lang> output:

(`a,97)