URL encoding

From Rosetta Code
Task
URL encoding
You are encouraged to solve this task according to the task description, using any language you may know.
Task

Provide a function or mechanism to convert a provided string into URL encoding representation.

In URL encoding, special characters, control characters and extended characters are converted into a percent symbol followed by a two digit hexadecimal code, So a space character encodes into %20 within the string.

For the purposes of this task, every character except 0-9, A-Z and a-z requires conversion, so the following characters all require conversion by default:

  • ASCII control codes (Character ranges 00-1F hex (0-31 decimal) and 7F (127 decimal).
  • ASCII symbols (Character ranges 32-47 decimal (20-2F hex))
  • ASCII symbols (Character ranges 58-64 decimal (3A-40 hex))
  • ASCII symbols (Character ranges 91-96 decimal (5B-60 hex))
  • ASCII symbols (Character ranges 123-126 decimal (7B-7E hex))
  • Extended characters with character codes of 128 decimal (80 hex) and above.


Example

The string "http://foo bar/" would be encoded as "http%3A%2F%2Ffoo%20bar%2F".


Variations
  • Lowercase escapes are legal, as in "http%3a%2f%2ffoo%20bar%2f".
  • Special characters have different encodings for different standards:
    • RFC 3986, Uniform Resource Identifier (URI): Generic Syntax, section 2.3, says to preserve "-._~".
    • HTML 5, section 4.10.22.5 URL-encoded form data, says to preserve "-._*", and to encode space " " to "+".
    • encodeURI function in Javascript will preserve "-._~" (RFC 3986) and ";,/?:@&=+$!*'()#".
Options

It is permissible to use an exception string (containing a set of symbols that do not need to be converted). However, this is an optional feature and is not a requirement of this task.


Related tasks



11l

F url_encode(s)
   V r = ‘’
   V buf = ‘’

   F flush_buf() // this function is needed because strings in 11l are UTF-16 encoded
      I @buf != ‘’
         V bytes = @buf.encode(‘utf-8’)
         L(b) bytes
            @r ‘’= ‘%’hex(b).zfill(2)
         @buf = ‘’

   L(c) s
      I c C (‘0’..‘9’, ‘a’..‘z’, ‘A’..‘Z’, ‘_’, ‘.’, ‘-’, ‘~’)
         flush_buf()
         r ‘’= c
      E
         buf ‘’= c

   flush_buf()
   R r

print(url_encode(‘http://foo bar/’))
print(url_encode(‘https://ru.wikipedia.org/wiki/Транспайлер’))
Output:
http%3A%2F%2Ffoo%20bar%2F
https%3A%2F%2Fru.wikipedia.org%2Fwiki%2F%D0%A2%D1%80%D0%B0%D0%BD%D1%81%D0%BF%D0%B0%D0%B9%D0%BB%D0%B5%D1%80

Action!

BYTE FUNC MustEncode(CHAR c CHAR ARRAY ex)
  BYTE i

  IF c>='a AND c<='z OR c>='A AND c<='Z OR c>='0 AND c<='9 THEN
    RETURN (0)
  FI
  IF ex(0)>0 THEN
    FOR i=1 TO ex(0)
    DO
      IF ex(i)=c THEN
        RETURN (0)
      FI
    OD
  FI
RETURN (1)

PROC Append(CHAR ARRAY s CHAR c)
  s(0)==+1
  s(s(0))=c
RETURN

PROC Encode(CHAR ARRAY in,out,ex BYTE spaceToPlus)
  CHAR ARRAY hex=['0 '1 '2 '3 '4 '5 '6 '7 '8 '9 'A 'B 'C 'D 'E 'F]
  BYTE i
  CHAR c

  out(0)=0
  FOR i=1 TO in(0)
  DO
    c=in(i)
    IF spaceToPlus=1 AND c=32 THEN
      Append(out,'+)
    ELSEIF MustEncode(c,ex) THEN
      Append(out,'%)
      Append(out,hex(c RSH 4))
      Append(out,hex(c&$0F))
    ELSE
      Append(out,c)
    FI
  OD
RETURN

PROC EncodeRaw(CHAR ARRAY in,out)
  Encode(in,out,"",0)
RETURN

PROC EncodeRFC3986(CHAR ARRAY in,out)
  Encode(in,out,"-._~",0)
RETURN

PROC EncodeHTML5(CHAR ARRAY in,out)
  Encode(in,out,"-._*",1)
RETURN

PROC PrintInv(CHAR ARRAY a)
  BYTE i

  IF a(0)>0 THEN
    FOR i=1 TO a(0)
    DO
      Put(a(i)%$80)
    OD
  FI
RETURN

PROC Test(CHAR ARRAY in)
  CHAR ARRAY out(256)

  PrintInv("input   ")
  PrintF(" %S%E",in)

  EncodeRaw(in,out)
  PrintInv("encoded ")
  PrintF(" %S%E",out)

  EncodeRFC3986(in,out)
  PrintInv("RFC 3986")
  PrintF(" %S%E",out)

  EncodeHTML5(in,out)
  PrintInv("HTML 5  ")
  PrintF(" %S%E%E",out)
RETURN

PROC Main()
  Test("http://foo bar/")
  Test("http://www.rosettacode.org/wiki/URL_encoding")
  Test("http://foo bar/*_-.html")
RETURN
Output:

Screenshot from Atari 8-bit computer

input    http://foo bar/
encoded  http%3A%2F%2Ffoo%20bar%2F
RFC 3986 http%3A%2F%2Ffoo%20bar%2F
HTML 5   http%3A%2F%2Ffoo+bar%2F

input    http://www.rosettacode.org/wiki/URL_encoding
encoded  http%3A%2F%2Fwww%2Erosettacode%2Eorg%2Fwiki%2FURL%5Fencoding
RFC 3986 http%3A%2F%2Fwww.rosettacode.org%2Fwiki%2FURL_encoding
HTML 5   http%3A%2F%2Fwww.rosettacode.org%2Fwiki%2FURL_encoding

input    http://foo bar/*_-.html
encoded  http%3A%2F%2Ffoo%20bar%2F%2A%5F%2D%2Ehtml
RFC 3986 http%3A%2F%2Ffoo%20bar%2F%2A_-.html
HTML 5   http%3A%2F%2Ffoo+bar%2F*_-.html

Ada

Library: AWS
with AWS.URL;
with Ada.Text_IO; use Ada.Text_IO;
procedure Encode is
   Normal : constant String := "http://foo bar/";
begin
   Put_Line (AWS.URL.Encode (Normal));
end Encode;
Output:
http%3A%2F%2Ffoo%20bar%2F

ALGOL 68

BEGIN
    # encodes the specified url - 0-9, A-Z and a-z are unchanged,            #
    # everything else is converted to %xx where xx are hex-digits            #
    PROC encode url = ( STRING url )STRING:
         IF url = "" THEN "" # empty string                                  #
         ELSE
            # non-empty string                                               #
            # ensure result will be big enough for a string of all encodable #
            # characters                                                     #
            STRING hex digits = "0123456789ABCDEF";
            [ 1 : ( ( UPB url - LWB url ) + 1 ) * 3 ]CHAR result;
            INT r pos := 0;
            FOR u pos FROM LWB url TO UPB url DO
                CHAR c = url[ u pos ];
                IF ( c >= "0" AND c <= "9" )
                OR ( c >= "A" AND c <= "Z" )
                OR ( c >= "a" AND c <= "z" )
                THEN
                    # no need to encode this character                       #
                    result[ r pos +:= 1 ] := c
                ELSE
                    # must encode                                            #
                    INT c code = ABS c;
                    result[ r pos +:= 1 ] := "%";
                    result[ r pos +:= 1 ] := hex digits[ ( c code OVER 16 ) + 1 ];
                    result[ r pos +:= 1 ] := hex digits[ ( c code  MOD 16 ) + 1 ]
                FI
            OD;
            result[ 1 : r pos ]
         FI; # encode url #
    # task test case                                                         #
    print( ( encode url( "http://foo bar/" ), newline ) )
END
Output:
http%3A%2F%2Ffoo%20bar%2F

Apex

EncodingUtil.urlEncode('http://foo bar/', 'UTF-8')
http%3A%2F%2Ffoo+bar%2F

AppleScript

AST URL encode "http://foo bar/"
Output:
"http%3A%2F%2Ffoo%20bar%2F"

Applesoft BASIC

 100 URL$ = "http://foo bar/"
 110  GOSUB 140"URL ENCODE URL$ RETURNS R$
 120  PRINT R$;
 130  END
 140  LET R$ = ""
 150  LET L =  LEN (URL$)
 160  IF  NOT L THEN  RETURN
 170  LET H$ = "0123456789ABCDEF"
 180  FOR I = 1 TO L
 190      LET C$ =  MID$ (URL$,I,1)
 200      LET C =  ASC (C$)
 210      IF C <  ASC ("0") OR C >  ASC ("Z") + 32 OR C >  ASC ("9") AND C <  ASC ("A") OR C >  ASC ("Z") AND C <  ASC ("A") + 32 THEN H =  INT (C / 16):C$ = "%" +  MID$ (H$,H + 1,1) +  MID$ (H$,C - H * 16 + 1,1)
 220      LET R$ = R$ + C$
 230  NEXT I
 240  RETURN

Arturo

encoded: encode.url.slashes "http://foo bar/"
print encoded
Output:
http%3A%2F%2Ffoo+bar%2F

AutoHotkey

MsgBox, % UriEncode("http://foo bar/")

; Modified from http://goo.gl/0a0iJq
UriEncode(Uri, Reserved:="!#$&'()*+,/:;=?@[]") {
    Unreserved := "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-_.~"
	VarSetCapacity(Var, StrPut(Uri, "UTF-8"), 0)
	StrPut(Uri, &Var, "UTF-8")
	While (Code := NumGet(Var, A_Index - 1, "UChar")) {
		If InStr(Unreserved . Reserved, Chr(Code)) {
			Encoded .= Chr(Code)
        }
		Else {
			Encoded .= Format("%{:02X}", Code)
        }
    }
	Return Encoded 
}

AWK

BEGIN {
	for (i = 0; i <= 255; i++)
		ord[sprintf("%c", i)] = i
}

# Encode string with application/x-www-form-urlencoded escapes.
function escape(str,    c, len, res) {
	len = length(str)
	res = ""
	for (i = 1; i <= len; i++) {
		c = substr(str, i, 1);
		if (c ~ /[0-9A-Za-z]/)
		#if (c ~ /[-._*0-9A-Za-z]/)
			res = res c
		#else if (c == " ")
		#	res = res "+"
		else
			res = res "%" sprintf("%02X", ord[c])
	}
	return res
}

# Escape every line of input.
{ print escape($0) }

The array ord[] uses idea from Character codes#AWK.

To follow the rules for HTML 5, uncomment the two lines that convert " " to "+", and use the regular expression that preserves "-._*".

Bash

urlencode() {
    local LC_ALL=C # support unicode: loop bytes, not characters
    local c i n=${#1}
    for (( i=0; i<n; i++ )); do
        c="${1:i:1}"
        case "$c" in
            #' ') printf '+' ;; # HTML5 variant
            [[:alnum:].~_-]) printf '%s' "$c" ;;
            *) printf '%%%02x' "'$c" ;;
        esac
    done
}

urlencode "http://foo bar/"
Output:
http%3a%2f%2ffoo%20bar%2f

To produce upper-case hex codes, replace %02x with %02X

See also: BashFAQ/071

BBC BASIC

      PRINT FNurlencode("http://foo bar/")
      END
      
      DEF FNurlencode(url$)
      LOCAL c%, i%
      WHILE i% < LEN(url$)
        i% += 1
        c% = ASCMID$(url$, i%)
        IF c%<&30 OR c%>&7A OR c%>&39 AND c%<&41 OR c%>&5A AND c%<&61 THEN
          url$ = LEFT$(url$,i%-1) + "%" + RIGHT$("0"+STR$~c%,2) + MID$(url$,i%+1)
        ENDIF
      ENDWHILE
      = url$
Output:
http%3A%2F%2Ffoo%20bar%2F

Bracmat

(     ( encode
      =   encoded exceptions octet string
        .     !arg:(?exceptions.?string)
            & :?encoded
            & @( !string
               :   ?
                   ( %@?octet ?
                   &     !encoded
                         (   !octet
                           : ( ~<0:~>9
                             | ~<A:~>Z
                             | ~<a:~>z
                             )
                         |   @(!exceptions:? !octet ?)
                           & !octet
                         | "%" d2x$(asc$!octet)
                         )
                     : ?encoded
                   & ~
                   )
               )
          | str$!encoded
      )
    & out$"without exceptions:
"
    & out$(encode$(."http://foo bar/"))
    & out$(encode$(."mailto:Ivan"))
    & out$(encode$(."Aim <ivan.aim@email.com>"))
    & out$(encode$(."mailto:Irma"))
    & out$(encode$(."User <irma.user@mail.com>"))
    & out$(encode$(."http://foo.bar.com/~user-name/_subdir/*~.html"))
    & out$"
with RFC 3986 rules:
"
    & out$(encode$("-._~"."http://foo bar/"))
    & out$(encode$("-._~"."mailto:Ivan"))
    & out$(encode$("-._~"."Aim <ivan.aim@email.com>"))
    & out$(encode$("-._~"."mailto:Irma"))
    & out$(encode$("-._~"."User <irma.user@mail.com>"))
    & out$(encode$("-._~"."http://foo.bar.com/~user-name/_subdir/*~.html"))
);
Output:
without exceptions:

http%3A%2F%2Ffoo%20bar%2F
mailto%3AIvan
Aim%20%3Civan%2Eaim%40email%2Ecom%3E
mailto%3AIrma
User%20%3Cirma%2Euser%40mail%2Ecom%3E
http%3A%2F%2Ffoo%2Ebar%2Ecom%2F%7Euser%2Dname%2F%5Fsubdir%2F%2A%7E%2Ehtml

with RFC 3986 rules:

http%3A%2F%2Ffoo%20bar%2F
mailto%3AIvan
Aim%20%3Civan.aim%40email.com%3E
mailto%3AIrma
User%20%3Cirma.user%40mail.com%3E
http%3A%2F%2Ffoo.bar.com%2F~user-name%2F_subdir%2F%2A~.html

C

#include <stdio.h>
#include <ctype.h>
 
char rfc3986[256] = {0};
char html5[256] = {0};

/* caller responsible for memory */
void encode(const char *s, char *enc, char *tb)
{
	for (; *s; s++) {
		if (tb[*s]) sprintf(enc, "%c", tb[*s]);
		else        sprintf(enc, "%%%02X", *s);
		while (*++enc);
	}
}
 
int main()
{
	const char url[] = "http://foo bar/";
	char enc[(strlen(url) * 3) + 1];
 
	int i;
	for (i = 0; i < 256; i++) {
		rfc3986[i] = isalnum(i)||i == '~'||i == '-'||i == '.'||i == '_'
			? i : 0;
		html5[i] = isalnum(i)||i == '*'||i == '-'||i == '.'||i == '_'
			? i : (i == ' ') ? '+' : 0;
	}
 
	encode(url, enc, rfc3986);
	puts(enc);
 
	return 0;
}

C#

using System;

namespace URLEncode
{
    internal class Program
    {
        private static void Main(string[] args)
        {
            Console.WriteLine(Encode("http://foo bar/"));
        }

        private static string Encode(string uri)
        {
            return Uri.EscapeDataString(uri);
        }
    }
}
Output:
http%3A%2F%2Ffoo%20bar%2F

C++

using Qt 4.6 as a library

#include <QByteArray>
#include <iostream>

int main( ) {
   QByteArray text ( "http://foo bar/" ) ;
   QByteArray encoded( text.toPercentEncoding( ) ) ;
   std::cout << encoded.data( ) << '\n' ;
   return 0 ;
}
Output:
http%3A%2F%2Ffoo%20bar%2F

Clojure

Using Java's URLEncoder:

(import 'java.net.URLEncoder)
(URLEncoder/encode "http://foo bar/" "UTF-8")
Output:
"http%3A%2F%2Ffoo+bar%2F"

ColdFusion

Common Lisp

(defun needs-encoding-p (char)
  (not (digit-char-p char 36)))

(defun encode-char (char)
  (format nil "%~2,'0X" (char-code char)))

(defun url-encode (url)
  (apply #'concatenate 'string
         (map 'list (lambda (char)
                      (if (needs-encoding-p char)
                          (encode-char char)
                          (string char)))
              url)))

(url-encode "http://foo bar/")
Output:
"http%3A%2F%2Ffoo%20bar%2F"

Crystal

The standard library URI class provides methods for both the RFC 3986 and HTML 5 standards. The RFC 3986 method defaults to replacing space characters with %20 but will replace them with + instead if the optional parameter space_to_plus is true. The HTML 5 method has the opposite default behavior.

require "uri"

puts URI.encode("http://foo bar/")
puts URI.encode("http://foo bar/", space_to_plus: true)
puts URI.encode_www_form("http://foo bar/")
puts URI.encode_www_form("http://foo bar/", space_to_plus: false)
Output:
http://foo%20bar/
http://foo+bar/
http%3A%2F%2Ffoo+bar%2F
http%3A%2F%2Ffoo%20bar%2F

D

import std.stdio, std.uri;

void main() {
    writeln(encodeComponent("http://foo bar/"));
}
http%3A%2F%2Ffoo%20bar%2F

Delphi

Works with: Delphi version 6.0


function EncodeURL(URL: string): string;
var I: integer;
begin
Result:='';
for I:=1 to Length(URL) do
 if URL[I] in ['0'..'9', 'A'..'Z', 'a'..'z'] then Result:=Result+URL[I]
 else Result:=Result+'%'+IntToHex(byte(URL[I]),2);
end;

procedure EncodeAndShowURL(Memo: TMemo; URL: string);
var ES: string;
begin
Memo.Lines.Add('Unencoded URL: '+URL);
ES:=EncodeURL(URL);
Memo.Lines.Add('Encoded URL:   '+ES);
Memo.Lines.Add('');
end;

procedure ShowEncodedURLs(Memo: TMemo);
begin
EncodeAndShowURL(Memo,'http://foo bar/');
EncodeAndShowURL(Memo,'https://rosettacode.org/wiki/URL_encoding');
EncodeAndShowURL(Memo,'https://en.wikipedia.org/wiki/Pikes_Peak_granite');
end;
Output:
Unencoded URL: http://foo bar/
Encoded URL:   http%3A%2F%2Ffoo%20bar%2F

Unencoded URL: https://rosettacode.org/wiki/URL_encoding
Encoded URL:   https%3A%2F%2Frosettacode%2Eorg%2Fwiki%2FURL%5Fencoding

Unencoded URL: https://en.wikipedia.org/wiki/Pikes_Peak_granite
Encoded URL:   https%3A%2F%2Fen%2Ewikipedia%2Eorg%2Fwiki%2FPikes%5FPeak%5Fgranite

Elapsed Time: 11.734 ms.


Elixir

iex(1)> URI.encode("http://foo bar/", &URI.char_unreserved?/1)
"http%3A%2F%2Ffoo%20bar%2F"

Erlang

Built in, http_uri:encode/1 accepts lists and binary:

1> http_uri:encode("http://foo bar/").
"http%3A%2F%2Ffoo%20bar%2F"

Unicode

If you are URL encoding unicode data though http_uri:encode/1 will produce incorrect results:

1> http_uri:encode("étanchéité d une terrasse").
"étanchéité%20d%20une%20terrasse"

You should use the built-in (non-documented) edoc_lib:escape_uri/1 instead:

1> edoc_lib:escape_uri("étanchéité d une terrasse").                    
"%c3%a9tanch%c3%a9it%c3%a9%20d%20une%20terrasse"

And for binary you will need to take care and use unicode:characters_to_{list,binary}/1:

1> unicode:characters_to_binary(edoc_lib:escape_uri(unicode:characters_to_list(<<"étanchéité d une terrasse"/utf8>>))).
<<"%c3%a9tanch%c3%a9it%c3%a9%20d%20une%20terrasse">>

F#

open System

[<EntryPoint>]
let main args =
    printfn "%s" (Uri.EscapeDataString(args.[0]))
    0
Output:
>URLencoding.exe "http://foo bar/"
http%3A%2F%2Ffoo%20bar%2F

Factor

Factor's built-in URL encoder doesn't encode : or /. However, we can write our own predicate quotation that tells (url-encode) what characters to exclude.

USING: combinators.short-circuit unicode urls.encoding.private ;

: my-url-encode ( str -- encoded )
    [ { [ alpha? ] [ "-._~" member? ] } 1|| ] (url-encode) ;

"http://foo bar/" my-url-encode print
Output:
http%3A%2F%2Ffoo%20bar%2F

Free Pascal

function urlEncode(data: string): AnsiString;
var
  ch: AnsiChar;
begin
  Result := '';
  for ch in data do begin
    if ((Ord(ch) < 65) or (Ord(ch) > 90)) and ((Ord(ch) < 97) or (Ord(ch) > 122)) then begin
      Result := Result + '%' + IntToHex(Ord(ch), 2);
    end else
      Result := Result + ch;
  end;
end;


FreeBASIC

Translation of: Liberty BASIC
Dim Shared As String lookUp(256)
For cadena As Integer = 0 To 256
    lookUp(cadena) = "%" + Hex(cadena)
Next cadena

Function string2url(cadena As String) As String
    Dim As String cadTemp, cadDevu
    For j As Integer = 1 To Len(cadena)
        cadTemp = Mid(cadena, j, 1)
        If Instr( "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz", cadTemp) Then
            cadDevu &= cadTemp 
        Else
            cadDevu &= lookUp(Asc(cadTemp))
        End If
    Next j
    Return cadDevu
End Function

Dim As String URL = "http://foo bar/"

Print "Supplied URL '"; URL; "'"
Print "URL encoding '"; string2url(URL); "'"
Sleep
Output:
Supplied URL 'http://foo bar/'
URL encoding 'http%3A%2F%2Ffoo%20bar%2F'

Frink

Frink has a built-in URLEncode[string, encoding="UTF-8"] function that correctly encodes strings (including high Unicode characters) for inclusion in a URL:

println[URLEncode["http://foo bar/"]]
Output:
http%3A%2F%2Ffoo+bar%2F


FutureBasic

In addition to the generic alphanumeric character set used in the demo code below, FB offers several special character sets for URL encoding:

fn CharacterSetURLFragmentAllowedSet
fn CharacterSetURLHostAllowedSet
fn CharacterSetURLPasswordAllowedSet
fn CharacterSetURLPathAllowedSet
fn CharacterSetURLQueryAllowedSet
fn CharacterSetURLUserAllowedSet

Users can also create custom character strings with:

fn CharacterSetWithCharactersInString( CFStringRef string ) = CFCharacterSetRef

include "NSLog.incl"

local fn PercentEncodeURLString( urlStr as CFStringRef ) as CFStringRef
  CFStringRef encodedStr = fn StringByAddingPercentEncodingWithAllowedCharacters( urlStr, fn CharacterSetAlphanumericSet )
end fn = encodedStr

NSLog( @"%@", fn PercentEncodeURLString( @"http://foo bar/" ) )
NSLog( @"%@", fn PercentEncodeURLString( @"http://www.rosettacode.org/wiki/URL_encoding" ) )

HandleEvents
Output:
http%3A%2F%2Ffoo%20bar%2F
http%3A%2F%2Fwww%2Erosettacode%2Eorg%2Fwiki%2FURL%5Fencoding


Go

package main

import (
    "fmt"
    "net/url"
)

func main() {
    fmt.Println(url.QueryEscape("http://foo bar/"))
}
Output:
http%3A%2F%2Ffoo+bar%2F


Groovy

def normal = "http://foo bar/"
def encoded = URLEncoder.encode(normal, "utf-8")
println encoded
Output:
http%3A%2F%2Ffoo+bar%2F

Haskell

import qualified Data.Char as Char
import Text.Printf

encode :: Char -> String
encode c
  | c == ' ' = "+"
  | Char.isAlphaNum c || c `elem` "-._~" = [c]
  | otherwise = printf "%%%02X" c

urlEncode :: String -> String
urlEncode = concatMap encode

main :: IO ()
main = putStrLn $ urlEncode "http://foo bar/"
Output:
http%3A%2F%2Ffoo+bar%2F

Icon and Unicon

link hexcvt

procedure main()
write("text    = ",image(u := "http://foo bar/"))
write("encoded = ",image(ue := encodeURL(u)))
end

procedure encodeURL(s)           #: encode data for inclusion in a URL/URI
static en                           
initial {                           # build lookup table for everything
   en := table()
   every en[c := !string(~(&digits++&letters))] := "%"||hexstring(ord(c),2)
   every /en[c := !string(&cset)] := c
   }

every (c := "") ||:= en[!s]         # re-encode everything
return c   
end

hexcvt provides hexstring

Output:
text    = "http://foo bar/"
encoded = "http%3A%2F%2Ffoo%20bar%2F"

J

J has a urlencode in the gethttp package, but this task requires that all non-alphanumeric characters be encoded.

Here's an implementation that does that:

require'strings convert'
urlencode=: rplc&((#~2|_1 47 57 64 90 96 122 I.i.@#)a.;"_1'%',.hfd i.#a.)

Example use:

   urlencode 'http://foo bar/'
http%3A%2F%2Ffoo%20bar%2F

Java

Java includes the URLEncoder and URLDecoder classes for this specific task.

import java.net.URLEncoder;
import java.nio.charset.StandardCharsets;
URLEncoder.encode("http://foo bar/", StandardCharsets.UTF_8)

Alternately, you could implement this with a basic for-loop.

String encode(String string) {
    StringBuilder encoded = new StringBuilder();
    for (char character : string.toCharArray()) {
        switch (character) {
            /* rfc3986 and html5 */
            case '-', '.', '_', '~', '*' -> encoded.append(character);
            case ' ' -> encoded.append('+');
            default -> {
                if (alphanumeric(character))
                    encoded.append(character);
                else {
                    encoded.append("%");
                    encoded.append("%02x".formatted((int) character));
                }
            }
        }
    }
    return encoded.toString();
}

boolean alphanumeric(char character) {
    return (character >= 'A' && character <= 'Z')
        || (character >= 'a' && character <= 'z')
        || (character >= '0' && character <= '9');
}
http%3a%2f%2ffoo+bar%2f

JavaScript

Confusingly, there are 3 different URI encoding functions in JavaScript: escape(), encodeURI(), and encodeURIComponent(). Each of them encodes a different set of characters. See this article and this article for more information and comparisons.

var normal = 'http://foo/bar/';
var encoded = encodeURIComponent(normal);

jq

Works with: jq version 1.4

jq has a built-in function, @uri, for "percent-encoding". It preserves the characters that RFC 3968 mandates be preserved, but also preserves the following five characters: !'()*

To address the task requirement, therefore, we can first use @uri and then convert the exceptional characters. (For versions of jq with regex support, this could be done using gsub, but here we perform the translation directly.)

Note that @uri also converts multibyte characters to multi-triplets, e.g.

"á" | @uri
produces: "%C3%A1"
def url_encode:
   # The helper function checks whether the input corresponds to one of the characters: !'()*
   def recode: . as $c | [33,39,40,41,42] | index($c);
   def hex:   if . < 10 then 48 + . else  55 + . end;
   @uri
   | explode
   # 37 ==> "%", 50 ==> "2"
   | map( if recode then (37, 50, ((. - 32) | hex)) else . end )
   | implode;

Examples:

"http://foo bar/" | @uri

produces: "http%3A%2F%2Ffoo%20bar%2F"

"http://foo bar/" | @uri == url_encode
produces: true

To contrast the difference between "@uri" and "url_encode", we compare the characters that are unaltered:

[range(0;1024) | [.] | implode | if @uri == . then . else empty end] | join(null)

produces: "!'()*-.0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ_abcdefghijklmnopqrstuvwxyz~"

[range(0;1024) | [.] | implode | if url_encode == . then . else empty end] | join(null)

produces: "-.0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ_abcdefghijklmnopqrstuvwxyz~"

Julia

 
//version 1.0.1
import HTTP.URIs: escapeuri

dcd = "http://foo bar/"
enc = escapeuri(dcd)

println(dcd, " => ", enc)
Output:
http://foo bar/ => http%3A%2F%2Ffoo%20bar%2F

Kotlin

// version 1.1.2

import java.net.URLEncoder

fun main(args: Array<String>) {
    val url = "http://foo bar/"
    println(URLEncoder.encode(url, "utf-8")) // note: encodes space to + not %20
}
Output:
http%3A%2F%2Ffoo+bar%2F

Ksh

url_encode()
{
        printf "%(url)q\n" "$*"
}


url_encode "http://foo bar/"
url_encode "https://ru.wikipedia.org/wiki/Транспайлер"
url_encode "google.com/search?q=`Abdu'l-Bahá"
Output:
http%3A%2F%2Ffoo%20bar%2F
https%3A%2F%2Fru.wikipedia.org%2Fwiki%2F%D0%A2%D1%80%D0%B0%D0%BD%D1%81%D0%BF%D0%B0%D0%B9%D0%BB%D0%B5%D1%80
google.com%2Fsearch%3Fq%3D%60Abdu%27l-Bah%C3%A1


langur

val .urlEncode = f(.s) replace(
    .s, re/[^A-Za-z0-9]/,
    f(.s2) join "", map f $"%\.b:X02;", s2b .s2,
)

writeln .urlEncode("https://some website.com/")
Output:
https%3A%2F%2Fsome%20website%2Ecom%2F

Lasso

bytes('http://foo bar/') -> encodeurl

-> http%3A%2F%2Ffoo%20bar%2F

Liberty BASIC

    dim lookUp$( 256)

    for i =0 to 256
        lookUp$( i) ="%" +dechex$( i)
    next i

    string$ ="http://foo bar/"

    print "Supplied string '"; string$; "'"
    print "As URL          '"; url$( string$); "'"

    end

function url$( i$)
    for j =1 to len( i$)
        c$ =mid$( i$, j, 1)
        if instr( "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz", c$) then
            url$ =url$ +c$ 
        else
            url$ =url$ +lookUp$( asc( c$))
        end if
    next j
end function
 Supplied string 'http://foo bar/'
 As URL 'http%3A%2F%2Ffoo%20bar%2F'

Lingo

Lingo implements old-school URL encoding (with spaces encoded as "+") out of the box:

put urlencode("http://foo bar/")
-- "http%3a%2f%2ffoo+bar%2f"

For RFC 3986 URL encoding (i.e. without the "+" fuss) a custom function is needed - which might call the above function and then just replace all "+" with "%20".

LiveCode

urlEncode("http://foo bar/")
-- http%3A%2F%2Ffoo+bar%2F

Lua

function encodeChar(chr)
	return string.format("%%%X",string.byte(chr))
end

function encodeString(str)
	local output, t = string.gsub(str,"[^%w]",encodeChar)
	return output
end

-- will print "http%3A%2F%2Ffoo%20bar%2F"
print(encodeString("http://foo bar/"))

M2000 Interpreter

Module Checkit {
            Function decodeUrl$(a$) {
                  DIM a$()
                  a$()=Piece$(a$, "%")
                  if len(a$())=1 then =str$(a$):exit
                  k=each(a$(),2)
                  acc$=str$(a$(0))
                  While k {
                  acc$+=str$(Chr$(Eval("0x"+left$(a$(k^),2)))+Mid$(a$(k^),3))
                  }
                  =string$(acc$ as utf8dec)
            }
            Group Parse$ {
                 all$, c=1
                 tc$=""           
                 Enum UrlType {None=0, RFC3986, HTML5}
                 variation
                 TypeData=("","-._~","-._*")
                 Function Next {                 
                            .tc$<=mid$(.all$,.c,1) 
                             .c++
                             =.tc$<>""
                  }
                  Value {
                        =.tc$
                  }
                  Function DecodeOne$ {
                        if .tc$="" then exit
                        if .tc$ ~"[A-Za-z0-9]" then =.tc$ : exit
                        If .tc$=" " Then =if$(.variation=.HTML5->"+","%20") :exit
                        if instr(.TypeData#val$(.variation),.tc$)>0 then =.tc$ :exit
                        ="%"+hex$(asc(.tc$), 1)
                  }
                  Function Decode$ {
                        acc$=""
                        .c<=1
                        While .Next()  {
                               acc$+=.DecodeOne$()
                        }
                        =acc$
                  }
                  Set () {
                        \\ using optional argument
                        var=.None
                        Read a$, ? var
                        a$=chr$(string$(a$ as utf8enc))
                        .variation<=var
                        .all$<=a$
                        .c<=1
                  }
            }
            \\ MAIN
            Parse$()="http://foo bar/"
            Print Quote$(Parse.Decode$())
            Parse.variation=Parse.HTML5
            Print Quote$(Parse.Decode$())
            Parse.variation=Parse.RFC3986
            Print Quote$(Parse.Decode$())
            Parse$(Parse.RFC3986) ={mailto:"Irma User" <irma.user@mail.com>}
            Print Quote$(Parse.Decode$())
            Parse$(Parse.RFC3986) ={http://foo.bar.com/~user-name/_subdir/*~.html}
            m=each(Parse.UrlType)
            while m {
                 Parse.variation=eval(m)
                 Print Quote$(Parse.Decode$())
                 Print decodeUrl$(Parse.Decode$())
            }            
}
CheckIt
"http%3A%2F%2Ffoo%20bar%2F"
"http%3A%2F%2Ffoo+bar%2F"
"mailto%3A%22Irma%20User%22%20%3Cirma.user%40mail.com%3E"
"http%3A%2F%2Ffoo%2Ebar%2Ecom%2F%7Euser%2Dname%2F%5Fsubdir%2F%2A%7E%2Ehtml"
"http%3A%2F%2Ffoo.bar.com%2F~user-name%2F_subdir%2F%2A~.html"
"http%3A%2F%2Ffoo.bar.com%2F%7Euser-name%2F_subdir%2F*%7E.html"

Maple

URL:-Escape("http://foo bar/");
Output:
"http%3A%2F%2Ffoo%20bar%2F"

Mathematica/Wolfram Language

URLEncoding[url_] := 
 StringReplace[url, 
  x : Except[
     Join[CharacterRange["0", "9"], CharacterRange["a", "z"], 
      CharacterRange["A", "Z"]]] :> 
   StringJoin[("%" ~~ #) & /@ 
     IntegerString[ToCharacterCode[x, "UTF8"], 16]]]

Example use:

URLEncoding["http://foo bar/"]
Output:
http%3a%2f%2ffoo%20bar%2f

MATLAB / Octave

function u = urlencoding(s)
	u = '';
	for k = 1:length(s),
		if isalnum(s(k))
			u(end+1) = s(k);
		else
			u=[u,'%',dec2hex(s(k)+0)];
		end; 	
	end
end

Usage:

octave:3> urlencoding('http://foo bar/')
ans = http%3A%2F%2Ffoo%20bar%2F

NetRexx

/* NetRexx */
options replace format comments java crossref symbols nobinary

/* -------------------------------------------------------------------------- */
  testcase()
  say
  say 'RFC3986'
  testcase('RFC3986')
  say
  say 'HTML5'
  testcase('HTML5')
  say
  return

/* -------------------------------------------------------------------------- */
method encode(url, varn) public static

  variation = varn.upper 
  opts = ''
  opts['RFC3986'] = '-._~'
  opts['HTML5']   = '-._*'

  rp = ''
  loop while url.length > 0
    parse url tc +1 url
    select
      when tc.datatype('A') then do
        rp = rp || tc
        end
      when tc == ' ' then do
        if variation = 'HTML5' then
          rp = rp || '+'
        else
          rp = rp || '%' || tc.c2x
        end
      otherwise do
        if opts[variation].pos(tc) > 0 then do
          rp = rp || tc
          end
        else do
          rp = rp || '%' || tc.c2x
          end
        end
      end
    end

  return rp

/* -------------------------------------------------------------------------- */
method testcase(variation = '') public static

  url = [ -
      'http://foo bar/' -
    , 'mailto:"Ivan Aim" <ivan.aim@email.com>' -
    , 'mailto:"Irma User" <irma.user@mail.com>' -
    , 'http://foo.bar.com/~user-name/_subdir/*~.html' -
    ]

  loop i_ = 0 to url.length - 1
    say url[i_]
    say encode(url[i_], variation)
    end i_

  return
Output:
http://foo bar/ 
http%3A%2F%2Ffoo%20bar%2F 
mailto:"Ivan Aim" <ivan.aim@email.com> 
mailto%3A%22Ivan%20Aim%22%20%3Civan%2Eaim%40email%2Ecom%3E 
mailto:"Irma User" <irma.user@mail.com> 
mailto%3A%22Irma%20User%22%20%3Cirma%2Euser%40mail%2Ecom%3E 
http://foo.bar.com/~user-name/_subdir/*~.html 
http%3A%2F%2Ffoo%2Ebar%2Ecom%2F%7Euser%2Dname%2F%5Fsubdir%2F%2A%7E%2Ehtml 
 
RFC3986 
http://foo bar/ 
http%3A%2F%2Ffoo%20bar%2F 
mailto:"Ivan Aim" <ivan.aim@email.com> 
mailto%3A%22Ivan%20Aim%22%20%3Civan.aim%40email.com%3E 
mailto:"Irma User" <irma.user@mail.com> 
mailto%3A%22Irma%20User%22%20%3Cirma.user%40mail.com%3E 
http://foo.bar.com/~user-name/_subdir/*~.html 
http%3A%2F%2Ffoo.bar.com%2F~user-name%2F_subdir%2F%2A~.html 
 
HTML5 
http://foo bar/ 
http%3A%2F%2Ffoo+bar%2F 
mailto:"Ivan Aim" <ivan.aim@email.com> 
mailto%3A%22Ivan+Aim%22+%3Civan.aim%40email.com%3E 
mailto:"Irma User" <irma.user@mail.com> 
mailto%3A%22Irma+User%22+%3Cirma.user%40mail.com%3E 
http://foo.bar.com/~user-name/_subdir/*~.html 
http%3A%2F%2Ffoo.bar.com%2F%7Euser-name%2F_subdir%2F*%7E.html

NewLISP

;; simple encoder
;; (source http://www.newlisp.org/index.cgi?page=Code_Snippets)
(define (url-encode str) 
  (replace {([^a-zA-Z0-9])} str (format "%%%2X" (char $1)) 0))

(url-encode "http://foo bar/")

Nim

import cgi

echo encodeUrl("http://foo/bar/")
Output:
http%3A%2F%2Ffoo%2Fbar%2F

Oberon-2

Works with: oo2c
MODULE URLEncoding;
IMPORT
  Out := NPCT:Console,
  ADT:StringBuffer, 
  URI := URI:String;
VAR
  encodedUrl: StringBuffer.StringBuffer;
BEGIN
  encodedUrl := NEW(StringBuffer.StringBuffer,512);
  URI.AppendEscaped("http://foo bar/","",encodedUrl);
  Out.String(encodedUrl.ToString());Out.Ln
END URLEncoding.
Output:
http%3A%2F%2Ffoo%20bar%2F

Objeck

use FastCgi;

bundle Default {
  class UrlEncode {
    function : Main(args : String[]) ~ Nil {
      url := "http://foo bar/";
      UrlUtility->Encode(url)->PrintLine();
    }
  }
}

Objective-C

NSString *normal = @"http://foo bar/";
NSString *encoded = [normal stringByAddingPercentEscapesUsingEncoding:NSUTF8StringEncoding];
NSLog(@"%@", encoded);

The Core Foundation function CFURLCreateStringByAddingPercentEscapes() provides more options.

Works with: Mac OS X version 10.9+
Works with: iOS version 7+
NSString *normal = @"http://foo bar/";
NSString *encoded = [normal stringByAddingPercentEncodingWithAllowedCharacters:[NSCharacterSet alphanumericCharacterSet]];
NSLog(@"%@", encoded);

For encoding for various parts of the URL, the allowed character sets [NSCharacterSet URLUserAllowedCharacterSet], [NSCharacterSet URLPasswordAllowedCharacterSet], [NSCharacterSet URLHostAllowedCharacterSet], [NSCharacterSet URLPathAllowedCharacterSet], [NSCharacterSet URLQueryAllowedCharacterSet], or [NSCharacterSet URLFragmentAllowedCharacterSet] are provided.

OCaml

Using the library ocamlnet from the interactive loop:

$ ocaml
# #use "topfind";;
# #require "netstring";;

# Netencoding.Url.encode "http://foo bar/" ;;
- : string = "http%3A%2F%2Ffoo+bar%2F"

ooRexx

The solution shown at Rexx (version 1) is a valid ooRexx program.

version 2 uses constructs not supported by ooRexx:
url.=; and $ as variable symbol

Perl

sub urlencode {
    my $s = shift;
    $s =~ s/([^-A-Za-z0-9_.!~*'() ])/sprintf("%%%02X", ord($1))/eg;
    $s =~ tr/ /+/;
    return $s;
}

print urlencode('http://foo bar/')."\n";
Output:
http%3A%2F%2Ffoo+bar%2F
use URI::Escape;

my $s = 'http://foo/bar/';
print uri_escape($s);

Use standard CGI module:

use 5.10.0;
use CGI;

my $s = 'http://foo/bar/';
say $s = CGI::escape($s);
say $s = CGI::unescape($s);

Phix

--
-- demo\rosetta\encode_url.exw
-- ===========================
--
with javascript_semantics
function nib(integer b)
    return b+iff(b<=9?'0':'A'-10)
end function

function encode_url(string s, string exclusions="", integer spaceplus=0)
string res = ""
    for i=1 to length(s) do
        integer ch = s[i]
        if ch=' ' and spaceplus then
            ch = '+'
        elsif not find(ch,exclusions)
          and (ch<'0'
           or (ch>'9' and ch<'A')
           or (ch>'Z' and ch<'a')
           or  ch>'z') then
            res &= '%'
            res &= nib(floor(ch/#10))
            ch = nib(and_bits(ch,#0F))
        end if
        res &= ch
    end for
    return res
end function

printf(1,"%s\n",{encode_url("http://foo bar/")})
{} = wait_key()
Output:
http%3A%2F%2Ffoo%20bar%2F

PHP

<?php
$s = 'http://foo/bar/';
$s = rawurlencode($s);
?>

There is also urlencode(), which also encodes spaces as "+" signs

PicoLisp

(de urlEncodeTooMuch (Str)
   (pack
      (mapcar
         '((C)
            (if (or (>= "9" C "0") (>= "Z" (uppc C) "A"))
               C
               (list '% (hex (char C))) ) )
         (chop Str) ) ) )

Test:

: (urlEncodeTooMuch "http://foo bar/")
-> "http%3A%2F%2Ffoo%20bar%2F"

Pike

Protocols.HTTP.uri_encode( "http://foo bar/" );
Output:
http%3A%2F%2Ffoo%20bar%2F

Powershell

[uri]::EscapeDataString('http://foo bar/')

http%3A%2F%2Ffoo%20bar%2F

PureBasic

URL$ = URLEncoder("http://foo bar/")

Python

import urllib
s = 'http://foo/bar/'
s = urllib.quote(s)

There is also urllib.quote_plus(), which also encodes spaces as "+" signs

R

R has a built-in

URLencode("http://foo bar/")

function, but it doesn't fully follow RFC guidelines, so we have to use another R package to accomplish the task:

library(RCurl)
curlEscape("http://foo bar/")

Racket

#lang racket
(require net/uri-codec)
(uri-encode "http://foo bar/")

Raku

(formerly Perl 6)

my $url = 'http://foo bar/';

say $url.subst(/<-alnum>/, *.ord.fmt("%%%02X"), :g);
Output:
http%3A%2F%2Ffoo%20bar%2F

REALbasic

Using the built-in encoding method, which doesn't permit exceptions:

  Dim URL As String = "http://foo bar/"
  URL = EncodeURLComponent(URL)
  Print(URL)

With optional exceptions. A "ParamArray" is an array of zero or more additional arguments passed by the caller:

Function URLEncode(Data As String, ParamArray Exceptions() As String) As String
  Dim buf As String
  For i As Integer = 1 To Data.Len
    Dim char As String = Data.Mid(i, 1)
    Select Case Asc(char)
    Case 48 To 57, 65 To 90, 97 To 122, 45, 46, 95
      buf = buf + char
    Else
      If Exceptions.IndexOf(char) > -1 Then
        buf = buf + char
      Else
        buf = buf + "%" + Left(Hex(Asc(char)) + "00", 2)
      End If
    End Select
  Next
  Return buf
End Function

 'usage
 Dim s As String = URLEncode("http://foo bar/") ' no exceptions
 Dim t As String = URLEncode("http://foo bar/", "!", "?", ",") ' with exceptions

REXX

version 1

/* Rexx */
do
  call testcase
  say
  say RFC3986
  call testcase RFC3986
  say
  say HTML5
  call testcase HTML5
  say
  return
end
exit

/* -------------------------------------------------------------------------- */
encode:
procedure
do
  parse arg url, varn .
  parse upper var varn variation 
  drop RFC3986 HTML5
  opts. = ''
  opts.RFC3986 = '-._~'
  opts.HTML5   = '-._*'

  rp = ''
  do while length(url) > 0
    parse var url tc +1 url
    select
      when datatype(tc, 'A') then do
        rp = rp || tc
        end
      when tc == ' ' then do
        if variation = HTML5 then
          rp = rp || '+'
        else
          rp = rp || '%' || c2x(tc)
        end
      otherwise do
        if pos(tc, opts.variation) > 0 then do
          rp = rp || tc
          end
        else do
          rp = rp || '%' || c2x(tc)
          end
        end
      end
    end

  return rp
end
exit

/* -------------------------------------------------------------------------- */
testcase:
procedure
do
  parse arg variation
  X = 0
  url. = ''
  X = X + 1; url.0 = X; url.X = 'http://foo bar/'
  X = X + 1; url.0 = X; url.X = 'mailto:"Ivan Aim" <ivan.aim@email.com>'
  X = X + 1; url.0 = X; url.X = 'mailto:"Irma User" <irma.user@mail.com>'
  X = X + 1; url.0 = X; url.X = 'http://foo.bar.com/~user-name/_subdir/*~.html'

  do i_ = 1 to url.0
    say url.i_
    say encode(url.i_, variation)
    end i_

  return
end
Output:
http://foo bar/
http%3A%2F%2Ffoo%20bar%2F
mailto:"Ivan Aim" <ivan.aim@email.com>
mailto%3A%22Ivan%20Aim%22%20%3Civan%2Eaim%40email%2Ecom%3E
mailto:"Irma User" <irma.user@mail.com>
mailto%3A%22Irma%20User%22%20%3Cirma%2Euser%40mail%2Ecom%3E
http://foo.bar.com/~user-name/_subdir/*~.html
http%3A%2F%2Ffoo%2Ebar%2Ecom%2F%7Euser%2Dname%2F%5Fsubdir%2F%2A%7E%2Ehtml

RFC3986
http://foo bar/
http%3A%2F%2Ffoo%20bar%2F
mailto:"Ivan Aim" <ivan.aim@email.com>
mailto%3A%22Ivan%20Aim%22%20%3Civan.aim%40email.com%3E
mailto:"Irma User" <irma.user@mail.com>
mailto%3A%22Irma%20User%22%20%3Cirma.user%40mail.com%3E
http://foo.bar.com/~user-name/_subdir/*~.html
http%3A%2F%2Ffoo.bar.com%2F~user-name%2F_subdir%2F%2A~.html

HTML5
http://foo bar/
http%3A%2F%2Ffoo+bar%2F
mailto:"Ivan Aim" <ivan.aim@email.com>
mailto%3A%22Ivan+Aim%22+%3Civan.aim%40email.com%3E
mailto:"Irma User" <irma.user@mail.com>
mailto%3A%22Irma+User%22+%3Cirma.user%40mail.com%3E
http://foo.bar.com/~user-name/_subdir/*~.html
http%3A%2F%2Ffoo.bar.com%2F%7Euser-name%2F_subdir%2F*%7E.html

version 2

/*REXX program encodes a  URL  text,    blanks ──► +,    preserves  -._*    and   -._~  */
url.=;                              url.1= 'http://foo bar/'
                                    url.2= 'mailto:"Ivan Aim" <ivan.aim@email.com>'
                                    url.3= 'mailto:"Irma User" <irma.user@mail.com>'
                                    url.4= 'http://foo.bar.com/~user-name/_subdir/*~.html'
     do j=1  while url.j\=='';  say
     say '  original: '             url.j
     say '   encoded: '   URLencode(url.j)
     end   /*j*/
exit                                             /*stick a fork in it,  we're all done. */
/*──────────────────────────────────────────────────────────────────────────────────────*/
URLencode: procedure; parse arg $,,z;        t1= '-._~'              /*get args, null Z.*/
           skip=0;                           t2= '-._*'
                    do k=1  for length($);   _=substr($, k, 1)       /*get a character. */
                    if skip\==0  then do;    skip=skip-1             /*skip t1 or t2 ?  */
                                             iterate                 /*skip a character.*/
                                      end
                       select
                       when datatype(_, 'A')    then z=z || _        /*is alphanumeric ?*/
                       when _==' '              then z=z'+'          /*is it  a  blank ?*/
                       when substr($, k, 4)==t1 |,                   /*is it  t1 or t2 ?*/
                            substr($, k, 4)==t2   then do;  skip=3   /*skip 3 characters*/
                                                            z=z || substr($, k, 4)
                                                       end
                       otherwise   z=z'%'c2x(_)                      /*special character*/
                       end   /*select*/
                    end      /*k*/
           return z

output   when using the default input:

  original:  http://foo bar/
   encoded:  http%3A%2F%2Ffoo+bar%2F

  original:  mailto:"Ivan Aim" <ivan.aim@email.com>
   encoded:  mailto%3A%22Ivan+Aim%22+%3Civan%2Eaim%40email%2Ecom%3E

  original:  mailto:"Irma User" <irma.user@mail.com>
   encoded:  mailto%3A%22Irma+User%22+%3Cirma%2Euser%40mail%2Ecom%3E

  original:  http://foo.bar.com/~user-name/_subdir/*~.html
   encoded:  http%3A%2F%2Ffoo%2Ebar%2Ecom%2F%7Euser%2Dname%2F%5Fsubdir%2F%2A%7E%2Ehtml

Ruby

CGI.escape encodes all characters except '-.0-9A-Z_a-z'.

require 'cgi'
puts CGI.escape("http://foo bar/").gsub("+", "%20")
# => "http%3A%2F%2Ffoo%20bar%2F"

Programs should not call URI.escape (alias URI.encode), because it fails to encode some characters. URI.escape is obsolete since Ruby 1.9.2.

URI.encode_www_form_component is a new method from Ruby 1.9.2. It obeys HTML 5 and encodes all characters except '-.0-9A-Z_a-z' and '*'.

Works with: Ruby version 1.9.2
require 'uri'
puts URI.encode_www_form_component("http://foo bar/").gsub("+", "%20")
# => "http%3A%2F%2Ffoo%20bar%2F"

Run BASIC

urlIn$ = "http://foo bar/"

for i = 1 to len(urlIn$)
  a$ = mid$(urlIn$,i,1)
  if (a$ >= "0" and a$ <= "9") _
  or (a$ >= "A" and a$ <= "Z") _
  or (a$ >= "a" and a$ <= "z") then url$ = url$ + a$ else url$ = url$ + "%"+dechex$(asc(a$))
next i
print urlIn$;" -> ";url$
http://foo bar/ -> http%3A%2F%2Ffoo%20bar%2F

Rust

const INPUT: &str = "http://foo bar/";
const MAX_CHAR_VAL: u32 = std::char::MAX as u32;
 
fn main() {
    let mut buff = [0; 4];
    println!("{}", INPUT.chars()
        .map(|ch| {
            match ch as u32 {
                0 ..= 47 | 58 ..= 64 | 91 ..= 96 | 123 ..= MAX_CHAR_VAL => {
                    ch.encode_utf8(&mut buff);
                    buff[0..ch.len_utf8()].iter().map(|&byte| format!("%{:X}", byte)).collect::<String>()
                },
                _ => ch.to_string(),
            }
        })
        .collect::<String>()
    );
}
Output:
http%3A%2F%2Ffoo%20bar%2F

Scala

Library: Scala
import java.net.{URLDecoder, URLEncoder}
import scala.compat.Platform.currentTime

object UrlCoded extends App {
  val original = """http://foo bar/"""
  val encoded: String = URLEncoder.encode(original, "UTF-8")

  assert(encoded == "http%3A%2F%2Ffoo+bar%2F", s"Original: $original not properly encoded: $encoded")

  val percentEncoding = encoded.replace("+", "%20")
  assert(percentEncoding == "http%3A%2F%2Ffoo%20bar%2F", s"Original: $original not properly percent-encoded: $percentEncoding")

  assert(URLDecoder.decode(encoded, "UTF-8") == URLDecoder.decode(percentEncoding, "UTF-8"))

  println(s"Successfully completed without errors. [total ${currentTime - executionStart} ms]")
}

Seed7

The library encoding.s7i defines functions to handle URL respectively percent encoding. The function toPercentEncoded encodes every character except 0-9, A-Z, a-z and the characters '-', '.', '_', '~'. The function toUrlEncoded works like toPercentEncoded and additionally encodes a space with '+'. Both functions work for byte sequences (characters beyond '\255\' raise the exception RANGE_ERROR).

To encode Unicode characters it is necessary to convert them to UTF-8 with striToUtf8 before.
$ include "seed7_05.s7i";
  include "encoding.s7i";

const proc: main is func
  begin
    writeln(toPercentEncoded("http://foo bar/"));
    writeln(toUrlEncoded("http://foo bar/"));
  end func;
Output:
http%3A%2F%2Ffoo%20bar%2F
http%3A%2F%2Ffoo+bar%2F

Sidef

Translation of: Perl
func urlencode(str) {
    str.gsub!(%r"([^-A-Za-z0-9_.!~*'() ])", {|a| "%%%02X" % a.ord});
    str.gsub!(' ', '+');
    return str;
}

say urlencode('http://foo bar/');
Output:
http%3A%2F%2Ffoo+bar%2F

Tcl

# Encode all except "unreserved" characters; use UTF-8 for extended chars.
# See http://tools.ietf.org/html/rfc3986 §2.4 and §2.5
proc urlEncode {str} {
    set uStr [encoding convertto utf-8 $str]
    set chRE {[^-A-Za-z0-9._~\n]};		# Newline is special case!
    set replacement {%[format "%02X" [scan "\\\0" "%c"]]}
    return [string map {"\n" "%0A"} [subst [regsub -all $chRE $uStr $replacement]]]
}

Demonstrating:

puts [urlEncode "http://foo bar/"]
Output:
http%3A%2F%2Ffoo%20bar%2F%E2%82%AC

TUSCRIPT

$$ MODE TUSCRIPT
text="http://foo bar/"
BUILD S_TABLE spez_char="::>/:</::<%:"
spez_char=STRINGS (text,spez_char)
LOOP/CLEAR c=spez_char
c=ENCODE(c,hex),c=concat("%",c),spez_char=APPEND(spez_char,c)
ENDLOOP
url_encoded=SUBSTITUTE(text,spez_char,0,0,spez_char)
print "text:    ", text
PRINT "encoded: ", url_encoded
Output:
text:    http://foo bar/
encoded: http%3A%2F%2Ffoo%20bar%2F

UNIX Shell

Works with: bash

unfortunately ksh does not support "'$c" syntax

function urlencode
{
	typeset decoded=$1 encoded= rest= c=
	typeset rest2= bug='rest2=${rest}'

	if [[ -z ${BASH_VERSION} ]]; then
		# bug /usr/bin/sh HP-UX 11.00
		typeset _decoded='xyz%26xyz'
		rest="${_decoded#?}"
		c="${_decoded%%${rest}}"
		if (( ${#c} != 1 )); then
			typeset qm='????????????????????????????????????????????????????????????????????????'
			typeset bug='(( ${#rest} > 0 )) && typeset -L${#rest} rest2="${qm}" || rest2=${rest}'
		fi
	fi

	rest="${decoded#?}"
	eval ${bug}
	c="${decoded%%${rest2}}"
	decoded="${rest}"

	while [[ -n ${c} ]]; do
		case ${c} in
		[-a-zA-z0-9.])
			;;
		' ')
			c='+'
			;;
		*)
			c=$(printf "%%%02X" "'$c")
			;;
		esac

		encoded="${encoded}${c}"

		rest="${decoded#?}"
		eval ${bug}
		c="${decoded%%${rest2}}"
		decoded="${rest}"
	done

	if [[ -n ${BASH_VERSION:-} ]]; then
		\echo -E "${encoded}"
	else
		print -r -- "${encoded}"
	fi
}

VBScript

Function UrlEncode(url)
	For i = 1 To Len(url)
		n = Asc(Mid(url,i,1))
		If (n >= 48 And n <= 57) Or (n >= 65 And n <= 90) _
			Or (n >= 97 And n <= 122) Then
			UrlEncode = UrlEncode & Mid(url,i,1)
		Else
			ChrHex = Hex(Asc(Mid(url,i,1)))
                        For j = 0 to (Len(ChrHex) / 2) - 1
			    UrlEncode = UrlEncode & "%" & Mid(ChrHex,(2*j) + 1,2)
                        Next
		End If
	Next
End Function

WScript.Echo UrlEncode("http://foo baré/")
Output:
http%3A%2F%2Ffoo%20bar%C3%A9%2F

V (Vlang)

import net.urllib
fn main() {
	println(urllib.query_escape("http://foo bar/"))
}
Output:
http%3A%2F%2Ffoo+bar%2F

Wren

Library: Wren-fmt
import "./fmt" for Fmt

var urlEncode = Fn.new { |url|
    var res = ""
    for (b in url.bytes) {
        if ((b >= 48 && b <= 57) || (b >= 65 && b <= 90) || (b >= 97 && b <= 122)) {
            res = res + String.fromByte(b)
        } else {
            res = res + Fmt.swrite("\%$2X", b)
        }
    }
    return res
}

var urls = [
    "http://foo bar/",
    "mailto:\"Ivan Aim\" <ivan.aim@email.com>",
    "mailto:\"Irma User\" <irma.user@mail.com>",
    "http://foo.bar.com/~user-name/_subdir/*~.html"
]
for (url in urls) System.print(urlEncode.call(url))
Output:
http%3A%2F%2Ffoo%20bar%2F
mailto%3A%22Ivan%20Aim%22%20%3Civan%2Eaim%40email%2Ecom%3E
mailto%3A%22Irma%20User%22%20%3Cirma%2Euser%40mail%2Ecom%3E
http%3A%2F%2Ffoo%2Ebar%2Ecom%2F%7Euser%2Dname%2F%5Fsubdir%2F%2A%7E%2Ehtml

XPL0

code Text=12;
string 0;               \use zero-terminated strings

func Encode(S0);        \Encode URL string and return its address
char S0;
char HD, S1(80);        \BEWARE: very temporary string space returned
int  C, I, J;
[HD:= "0123456789ABCDEF"; \hex digits
I:= 0;  J:= 0;
repeat  C:= S0(I);  I:= I+1;
        if C>=^0 & C<=^9 ! C>=^A & C<=^Z ! C>=^a & C<=^z ! C=0
        then    [S1(J):= C;  J:= J+1]           \simply pass char to S1
        else    [S1(J):= ^%;  J:= J+1;          \encode char into S1
                 S1(J):= HD(C>>4);  J:= J+1;
                 S1(J):= HD(C&$0F); J:= J+1;
                ];
until   C=0;
return S1;
];

Text(0, Encode("http://foo bar/"))
Output:
http%3A%2F%2Ffoo%20bar%2F

Yabasic

Translation of: Phix
sub encode_url$(s$, exclusions$, spaceplus)
    local res$, i, ch$
	
    for i=1 to len(s$)
        ch$ = mid$(s$, i, 1)
        if ch$ = " " and spaceplus then
            ch$ = "+"
        elsif not instr(esclusions$, ch$) and (ch$ < "0" or (ch$ > "9" and ch$ < "A") or (ch$ > "Z" and ch$ < "a") or  ch$ > "z") then
            res$ = res$ + "%"
            ch$ = upper$(hex$(asc(ch$)))
        end if
        res$ = res$ + ch$
    next i
    return res$
end sub

print encode_url$("http://foo bar/")

zkl

Using lib cURL:

var CURL=Import("zklCurl");
CURL.urlEncode("http://foo bar/") //--> "http%3A%2F%2Ffoo%20bar%2F"