String interpolation (included)

From Rosetta Code
String interpolation (included)
You are encouraged to solve this task according to the task description, using any language you may know.

Basic Data Operation
This is a basic data operation. It represents a fundamental action on a basic data type.

You may see other such operations in the Basic Data Operations category, or:

Integer Operations
Arithmetic | Comparison

Boolean Operations
Bitwise | Logical

String Operations
Concatenation | Interpolation | Comparison | Matching

Memory Operations
Pointers & references | Addresses

Given a string and defined variables or values, string interpolation is the replacement of defined character sequences in the string by values or variable values.

For example, given an original string of "Mary had a X lamb.", a value of "big", and if the language replaces X in its interpolation routine, then the result of its interpolation would be the string "Mary had a big lamb".
(Languages usually include an infrequently used character or sequence of characters to indicate what is to be replaced such as "%", or "#" rather than "X").
The task is to
  1. Use your languages inbuilt string interpolation abilities to interpolate a string missing the text "little" which is held in a variable, to produce the output string "Mary had a little lamb".
  2. If possible, give links to further documentation on your languages string interpolation features.

Note: The task is not to create a string interpolation routine, but to show a language's built-in capability.


with Ada.Strings.Fixed, Ada.Text_IO;
use Ada.Strings, Ada.Text_IO;
procedure String_Replace is
Original : constant String := "Mary had a @[email protected] lamb.";
Tbr : constant String := "@[email protected]";
New_Str : constant String := "little";
Index : Natural := Fixed.Index (Original, Tbr);
Put_Line (Fixed.Replace_Slice (
Original, Index, Index + Tbr'Length - 1, New_Str));
end String_Replace;


Put_Line ("Mary had a " & New_Str & " lamb.");


const little = "little"
printf ("Mary had a %s lamb\n", little)
// alternatively
println ("Mary had a " + little + " lamb")

ALGOL 68[edit]

Translation of: C
Works with: ALGOL 68 version Revision 1 - no extensions to language used
Works with: ALGOL 68G version Any - tested with release 1.18.0-9h.tiny

strings are simply flex arrays of char. formats on the other hand take on some of the properties of procedures including the scoping rules.

# as a STRING #
STRING extra = "little";
printf(($"Mary had a "g" lamb."l$, extra));
# as a FORMAT #
FORMAT extraf = $"little"$;
printf($"Mary had a "f(extraf)" lamb."l$);
# or: use simply use STRING concatenation #
print(("Mary had a "+extra+" lamb.", new line))
Mary had a little lamb.
Mary had a little lamb.
Mary had a little lamb.


; Using the = operator
LIT = little
string = Mary had a %LIT% lamb.
; Using the := operator
LIT := "little"
string := "Mary had a" LIT " lamb."
MsgBox %string%

Documentation: Variables (see Storing values in variables and Retrieving the contents of variables)


String interpolation is usually done with functions sub() and gsub(). gawk has also gensub().

#!/usr/bin/awk -f
str="Mary had a # lamb."
gsub(/#/, "little", str)
print str

Batch File[edit]

@echo off
setlocal enabledelayedexpansion
call :interpolate %1 %2 res
echo %res%
goto :eof
set pat=%~1
set str=%~2
set %3=!pat:X=%str%!
goto :eof


>interpolate.cmd "Mary had a X lamb" little
Mary had a little lamb


Use pattern matching to find the part of the string up to and the part of the string following the magic X. Concatenate these parts with the string "little" in the middle.

@("Mary had a X lamb":?a X ?z) & str$(!a little !z)


Include the <stdio.h> header to use the functions of the printf family:

#include <stdio.h>
int main() {
const char *extra = "little";
printf("Mary had a %s lamb.\n", extra);
return 0;


#include <string>
#include <iostream>
int main( ) {
std::string original( "Mary had a X lamb." ) , toBeReplaced( "X" ) ,
replacement ( "little" ) ;
std::string newString = original.replace( original.find( "X" ) ,
toBeReplaced.length( ) , replacement ) ;
std::cout << "String after replacement: " << newString << " \n" ;
return 0 ;


Works with: C++11
// Variable argument template
#include <string>
#include <vector>
using std::string;
using std::vector;
template<typename S, typename... Args>
string interpolate( const S& orig , const Args&... args)
string out(orig);
// populate vector from argument list
auto va = {args...};
vector<string> v{va};
size_t i = 1;
for( string s: v)
string is = std::to_string(i);
string t = "{" + is + "}"; // "{1}", "{2}", ...
auto pos = out.find(t);
if ( pos != out.npos) // found token
out.erase(pos, t.length()); //erase token
out.insert( pos, s); // insert arg
i++; // next
catch( std::exception& e)
std::cerr << e.what() << std::endl;
} // for
return out;


This is called "composite formatting" in MSDN.

class Program
static void Main()
string extra = "little";
string formatted = $"Mary had a {extra} lamb.";


(let [little "little"]
(println (format "Mary had a %s lamb." little)))


Works with: OpenCOBOL
PROGRAM-ID. interpolation-included.
01 extra PIC X(6) VALUE "little".
DISPLAY FUNCTION SUBSTITUTE("Mary had a X lamb.", "X", extra)


As CoffeeScript, but the braces are optional if the expression to be interpolated is just a variable:

size = 'little'
console.log "Mary had a #size lamb."


size = 'little'
console.log "Mary had a #{size} lamb." # Mary had a little lamb.
console.log "Escaping: \#{}" # Escaping: #{}
console.log 'No #{ interpolation} with single quotes' # No #{ interpolation} with single quotes
# Multi-line strings and arbtrary expressions work: 20
console.log """
Multi-line strings and arbtrary expressions work: #{ 5 * 4 }


Common Lisp[edit]

(let ((extra "little"))
(format t "Mary had a ~A lamb.~%" extra))

More documentation on the FORMAT function.


void main() {
import std.stdio, std.string;
"Mary had a %s lamb.".format("little").writeln;
"Mary had a %2$s %1$s lamb.".format("little", "white").writeln;
Mary had a little lamb.
Mary had a white little lamb.

More documentation on the format() function.


program Project1;
Template : string;
Marker : string;
Description : string;
Value : integer;
Output : string;
// StringReplace can be used if you are definitely using strings
Template := 'Mary had a X lamb.';
Marker := 'X';
Description := 'little';
Output := StringReplace(Template, Marker, Description, [rfReplaceAll, rfIgnoreCase]);
// You could also use format to do the same thing.
Template := 'Mary had a %s lamb.';
Description := 'little';
Output := format(Template,[Description]);
// Unlike StringReplace, format is not restricted to strings.
Template := 'Mary had a %s lamb. It was worth $%d.';
Description := 'little';
Value := 20;
Output := format(Template,[Description, Value]);
Mary had a little lamb.
Mary had a little lamb.
Mary had a little lamb. It was worth $20.


PrintLn(Format('Mary had a %s lamb.', ['little']))
Mary had a little lamb.


Dyalect has a built-in string interpolation feature.

const lamb_size = "little"
print("Mary had a \(lamb_size) lamb.")


def adjective := "little"
`Mary had a $adjective lamb`

The `...` syntax in general may be used as a sort of syntax extension; string interpolation is just the default case. More information on E quasi-literals. (Note that this documentation may be somewhat out of date.)

The above code is equivalent to (expands into):

def adjective := "little"
simple__quasiParser.valueMaker("Mary had a ${0} lamb").substitute([adjective])

If an identifier precedes the opening `, then it replaces simple; the quasiParser may be an arbitrary user-defined object. In this way, E provides lightweight syntax for embedding other languages: XML, JSON, GUI layouts, regular expressions, etc.


format and printf use replacement directives to perform interpolation. See format specification in EchoLisp documentatiuon.

;; format uses %a or ~a as replacement directive
(format "Mary had a ~a lamb" "little")
"Mary had a little lamb"
(format "Mary had a %a lamb" "little")
"Mary had a little lamb"


STD.Str.FindReplace('Mary had a X Lamb', 'X','little');


ELENA 4.x :

import extensions;
public program()
var s := "little";
console.printLineFormatted("Mary had a {0} lamb.",s).readChar()


Elixir borrows Ruby's #{...} interpolation syntax.

x = "little"
IO.puts "Mary had a #{x} lamb"


7> S1 = "Mary had a ~s lamb".
8> S2 = lists:flatten( io_lib:format(S1, ["big"]) ).
9> S2.
"Mary had a big lamb"


constant lambType = "little"
sequence s
s = sprintf("Mary had a %s lamb.",{lambType})

See sprintf, printf



let lambType = "little"
printfn "Mary had a %s lamb." lambType


USE: formatting 
SYMBOL: little
"little" little set
little get "Mary had a %s lamb" sprintf

I tried to be as specific as possible here. The challenge says to use a variable so that is what I used. It could have been done more cleanly using a CONSTANT.

USE: formatting 
CONSTANT: little "little"
little "Mary had a %s lamb" sprintf


'VBA/Python programmer's approach. I'm just a junior Falconeer but this code seems falconic

/* created by Aykayayciti Earl Lamont Montgomery
April 9th, 2018 */

size = "little"
> @ "Mary had a $size lamb"
// line 1: use of the = operator
// line 2: use of the @ and $ operator
Mary had a little lamb
[Finished in 0.2s]


Interpolating a variable value into a string is done by using a $ prefix on the variable name within a string. For example:

fansh> x := "little"
fansh> echo ("Mary had a $x lamb")
Mary had a little lamb

Documentation at: Fantom website


program interpolate
write (*,*) trim(inter("Mary had a X lamb.","X","little"))
elemental function inter(string,place,ins) result(new)
character(len=*), intent(in) :: string,place,ins
character(len=len(string)+max(0,len(ins)-len(place))) :: new
integer :: idx
idx = index(string,place)
if ( idx == 0 ) then
new = string
new = string(1:idx-1)//ins//string(idx+len(place):len(string))
end if
end function inter
end program interpolate


FreeBASIC has a complex Print statement which, amongst other things, enables variables to be embedded in the string to be printed.

It is also possible to use C library functions such as printf or sprintf, which allow more conventional string interpolation, as easily as if they were part of FB itself:

' FB 1.05.0 Win64
#Include "crt/" '' header needed for printf
Dim x As String = "big"
Print "Mary had a "; x; " lamb" '' FB's native Print statement
x = "little"
printf("Mary also had a %s lamb", x)
Mary had a big lamb
Mary also had a little lamb


x = "little"
println["Mary had a $x lamb."]


X = 'little'
println( "Mary had a $X lamb." )


Click this link to run this code

Public Sub Main()
Print Subst("Mary had a &1 lamb", "little")


Mary had a little lamb


This kind of string interpolation is indeed a strong feature in Gastona. We add one more indirection in the sample just to ilustrate it.

<how> //little
<what> //has a @<how> lamb
<main> //Mary @<what>
Mary has a little lamb



package main
import (
func main() {
str := "Mary had a %s lamb"
txt := "little"
out := fmt.Sprintf(str, txt)


def adj = 'little'
assert 'Mary had a little lamb.' == "Mary had a ${adj} lamb."


No such facilities are defined in Haskell 98, but the base package distributed with GHC provides a printf function.

import Text.Printf
main = printf "Mary had a %s lamb\n" "little"


Further documentation on HicEst string interpolation function EDIT()

CHARACTER original="Mary had a X lamb", little = "little", output_string*100
output_string = original
EDIT(Text=output_string, Right='X', RePLaceby=little)

Icon and Unicon[edit]

Icon and Unicon are descended from a line of languages with a wealth of string manipulation capabilities. See The Icon Programming Language, 3rd Edition; Griswold and Griswold; Chapter 3 String Scanning

  s2 := "humongous"
s3 := "little"
s1 := "Mary had a humongous lamb."
s1 ?:= tab(find(s2)) || (=s2,s3) || tab(0) # replaces the first instance of s2 with s3
while s1 ?:= tab(find(s2)) || (=s2,s3) || tab(0) # replaces all instances of s2 with s3, equivalent to replace

Note the strings library includes convenient procedures for string replacement such as replace(s1,s2,s3) which replaces all occurrences of s2 in s1 with s3 and replacem(s1,s2,s3,...) which replaces multiple pairs.


The strings and printf scripts are part of the base library.

   require 'printf'
'Mary had a %s lamb.' sprintf <'little'
Mary had a little lamb.
require 'strings'
('%s';'little') stringreplace 'Mary had a %s lamb.'
Mary had a little lamb.
'Mary had a %s lamb.' rplc '%s';'little'
Mary had a little lamb.


The comments in these library files give brief descriptions of their contents, and you can browse them using open:

   open'strings printf'

Alternatively, both strings and printf have various web pages describing them, and printf has a lab demonstrating its use (from J's IDE's menu, go Studio -> Labs... and then look in the System category).

That said, note that in recent versions of J, strings is no longer a separate script but part of the core library.


String original = "Mary had a X lamb";
String little = "little";
String replaced = original.replace("X", little); //does not change the original String
System.out.printf("Mary had a %s lamb.", little);
String formatted = String.format("Mary had a %s lamb.", little);


var original = "Mary had a X lamb";
var little = "little";
var replaced = original.replace("X", little); //does not change the original string


// ECMAScript 6
var X = "little";
var replaced = `Mary had a ${X} lamb`;


"little" as $x
| "Mary had a \($x) lamb"
Any valid jq expression (including a pipeline) can appear between the interpolating parentheses, e.g.:
$ jq -M -n -r '"Jürgen" as $x | "The string \"\($x)\" has \($x|length) codepoints."'
The string "Jürgen" has 6 codepoints.

Documentation: String interpolation


X = "little"
"Mary had a $X lamb"


// version 1.0.6
fun main(args: Array<String>) {
val s = "little"
// String interpolation using a simple variable
println("Mary had a $s lamb")
// String interpolation using an expression (need to wrap it in braces)
println("Mary had a ${s.toUpperCase()} lamb")
// However if a simple variable is immediately followed by a letter, digit or underscore
// it must be treated as an expression
println("Mary had a ${s}r lamb") // not $sr
Mary had a little lamb
Mary had a LITTLE lamb
Mary had a littler lamb


Lasso doesn't really have built-in string interpolation, but you can use the built-in email mail-merge capability:

email_merge("Mary had a #adjective# lamb", map("token"="little", "adjective"=""), null, 'plain')
Mary had a little lamb


Livecode has a merge function for interpolation

local str="little"
put merge("Mary had a [[str]] lamb.")
-- Mary had a little lamb.


Variable names[edit]

There is no default support for automatic interpolation of variables names being used as placeholders within a string. However, interpolation is easily emulated by using the [string.gsub] function:

str = string.gsub( "Mary had a X lamb.", "X", "little" )
print( str )

Literal characters[edit]

Interpolation of literal characters escape sequences does occur within a string:

print "Mary had a \n lamb"    -- The \n is interpreted as an escape sequence for a newline

M2000 Interpreter[edit]

module checkit {
m$=format$("Mary had a {0} lamb.", size$)
Print m$
Const RightJustify=1
\\ format$(string_expression) process escape codes
Report RightJustify, format$(format$("Mary had a {0} {1} lamb.\r\n We use {0} for size, and {1} for color\r\n", size$, "wh"+"ite"))
\\ we can use { } for multi line string
Report RightJustify, format$({Mary had a {0} {1} lamb.
We use {0} for size, and {1} for color
}, size$, "wh"+"ite")


Extra = "little";
StringReplace["Mary had a X lamb.", {"X" -> Extra}]
->"Mary had a little lamb."


printf(true, "Mary had a ~a lamb", "little");


<doc><h2>String interpolation, in Neko</h2>
<p><a href="">NekoVM String Library</a></p>

var sprintf = $loader.loadprim("[email protected]", 2)
$print(sprintf("Mary had a %s lamb\n", "little"))
prompt$ nekoc string-interpolation.neko
prompt$ neko string-interpolation.n
Mary had a little lamb


Nemerle has a few ways to accomplish this. It provides an implementation of printf(), $ interpolation within the print() method, and the most general use is $ interpolation within $ quoted strings.

using System;
using System.Console;
using Nemerle.IO; // contains printf() and print()
module Stringy
Main() : void
def extra = "little";
printf("Mary had a %s lamb.\n", extra);
print("Mary had a $extra lamb.\n");
WriteLine($"Mary had a $extra lamb.");


The Built In Functions (BIFs) of NetRexx can be employed to manipulate strings quite successfully but for more flexible string interpolation a function package like Java's MessageFormat should be used.

/* NetRexx */
options replace format comments java crossref savelog symbols
import java.text.MessageFormat
import java.text.FieldPosition
method useBif public static
st = "Mary had a %1$ lamb."
si = 'little'
say st.changestr('%1$', si)
method useMessageFormat public static
result = StringBuffer('')
args = Object [ -
Object Integer(7), -
Object Date(), -
Object 'a disturbance in the Force' -
msgfmt = MessageFormat('At {1, time} on {1, date}, there was {2} on planet {0, number, integer}.')
result = msgfmt.format(args, result, FieldPosition(0))
say result
Mary had a little lamb.
At 5:43:05 PM on Aug 22, 2011, there was a disturbance in the Force on planet 7.


import strutils
var str = "little"
echo "Mary had a $# lamb" % [str]
# doesn't need an array for one substitution, but use an array for multiple substitutions
import strformat
var str: string = "little"
echo fmt"Mary had a {str} lamb"
echo &"Mary had a {str} lamb"


The OCaml standard library provides the module Printf:

let extra = "little" in
Printf.printf "Mary had a %s lamb." extra


In a String all expressions between #{...} will be evaluated.

main: func {
X := "little"
"Mary had a #{X} lamb" println()


String interpolation is unidiomatic in Oz. Instead, "virtual strings" are used. Virtual strings are tuples of printable values and are supported by many library functions.

X = "little"
{System.showInfo "Mary had a "#X#" lamb"}


The Pari library has string interpolation, which extends C's:

string_interpolate(GEN n)
pari_printf("The value was: %Ps.\n", n);
GEN s = pari_sprintf("Storing %Ps in a string", n);
Works with: PARI/GP version version 2.4.4 and above

GP can also interpolate strings:

s=Strprintf("The value was: %Ps", 1<<20);
printf("The value was: %Ps", 1<<20);


$extra = "little";
print "Mary had a $extra lamb.\n";
printf "Mary had a %s lamb.\n", $extra;

Perl 6[edit]

my $extra = "little";
say "Mary had a $extra lamb"; # variable interpolation
say "Mary had a { $extra } lamb"; # expression interpolation
printf "Mary had a %s lamb.\n", $extra; # standard printf
say $extra.fmt("Mary had a %s lamb"); # inside-out printf
my @lambs = <Jimmy Bobby Tommy>;
say Q :array { $$$ The lambs are called @lambs[]\\\.} # only @-sigiled containers are interpolated


string size = "little"
string s = sprintf("Mary had a %s lamb.",{size})
"Mary had a little lamb."


$extra = 'little';
echo "Mary had a $extra lamb.\n";
printf("Mary had a %s lamb.\n", $extra);


(let Extra "little"
(prinl (text "Mary had a @1 lamb." Extra)) )


*process or(!) source xref attributes;
sit: Proc Options(main);
* Test string replacement
* 02.08.2013 Walter Pachl

Dcl s Char(50) Var Init('Mary had a &X lamb. It is &X');
Put Edit(repl(s,'little','&X'))(Skip,A);
repl: Proc(str,new,old) Returns(Char(50) Var);
* ooREXX has CHANGESTR(old,str,new[,count])
* repl follows, however, the translate "philosophy"
* translate(str,new,old) when old and new are just a character each
* and replaces all occurrences of old in str by new

Dcl str Char(*) Var;
Dcl (new,old) Char(*);
Dcl (res,tmp) Char(50) Var init('');
Dcl p Bin Fixed(31);
tmp=str; /* copy the input string */
Do Until(p=0);
p=index(tmp,old); /* position of old in tmp */
If p>0 Then Do; /* found */
res=res!!left(tmp,p-1)!!new; /* append new to current result*/
tmp=substr(tmp,p+length(old)); /* prepare rest of input */
res=res!!tmp; /* final append */
Mary had a little lamb. It is little


Using the format (-f) operator:

$extra = "little"
"Mary had a {0} lamb." -f $extra

Using format string with the WriteLine static method

$extra = "little"
[console]::writeline("Mary had a {0} lamb.", $extra)

Using the format method of the string type

$extra = "little"
[string]::Format("Mary had a {0} lamb.", $extra)

Note: numeric and date/time formats can be specified with {index:formatString} (i.e. {0:###,###})


Extra = little,
format('Mary had a ~w lamb.', [Extra]), % display result
format(atom(Atom), 'Mary had a ~w lamb.', [Extra]). % ... or store it a variable

Using library(func) for SWI-Prolog:

Extra = little,
Atom = 'Mary had a ~w lamb' $ Extra.

Using library(interpolate) for SWI-Prolog:

Extra = little,
Atom = 'Mary had a $Extra lamb'.


The function ReplaceString() is built-in and can have both constants and variables as parameters.

ReplaceString("Mary had a X lamb.","X","little")

Implemented in a program context

; String variable can be defined by appending .s to its name during definition or by appending and using $ as a part of its name.
Define txt$, txtvar.s="little"
;Load datasegment into variable txt$
Restore Mary
Read.s txt$
; Replace X with "little" and store result in txt$
OpenConsole(): Print(txt$)
Data.s "Mary had a X lamb."


Python has more than one inbuilt way of accomplishing the task. The methods have different capabilities that are not stretched by this small task

Using the % string interpolation operator:

>>> original = 'Mary had a %s lamb.'
>>> extra = 'little'
>>> original % extra
'Mary had a little lamb.'

Using the .format method of strings:

>>> original = 'Mary had a {extra} lamb.'
>>> extra = 'little'
>>> original.format(**locals())
'Mary had a little lamb.'

Using the format method, but replace by an expressions position as an argument to the format method call instead of by name:

>>> original = 'Mary had a {0} lamb.'
>>> extra = 'little'
>>> original.format(extra)
'Mary had a little lamb.'

Using the Template class of the string module:

>>> from string import Template
>>> original = Template('Mary had a $extra lamb.')
>>> extra = 'little'
>>> original.substitute(**locals())
'Mary had a little lamb.'

Using the new f-strings string literal available from Python 3.6:

>>> extra = 'little'
>>> f'Mary had a {extra} lamb.'
'Mary had a little lamb.'


See the documentation on fprintf for more information on string interpolation in Racket.

#lang racket
(format "Mary had a ~a lamb" "little")


str: "Mary had a <%size%> lamb"
size: "little"
build-markup str
;REBOL3 also has the REWORD function
str: "Mary had a $size lamb"
reword str [size "little"]


Interpolation does not occur in literal strings, neither within   singlequote   or   doublequote   enclosures.
However, interpolation can be emulated using the   changestr   function:

/*REXX program to demonstrate string interpolation (string replacement).*/
/*the string to be replaced is */
replace = "little" /*usually a unique character(s) */
/*string and is case sensative.*/
original1 = "Mary had a X lamb."
new1 = changestr('X', original1, replace)
say 'original1 =' original1
say 'replaced =' new1
original2 = "Mary had a % lamb."
new2 = changestr('%', original2, replace)
say 'original2 =' original2
say 'replaced =' new2
original3 = "Mary had a $$$ lamb."
new3 = changestr('$$$',original3,replace)
say 'original3 =' original3
say 'replaced3 =' new3
original4 = "Mary had a someKindOf lamb."
new3 = changestr('someKindOf', original4, "little")
say 'original4 =' original4
say 'replaced4 =' new3
/*stick a fork in it, we're done.*/

Some older REXXes don't have a   changestr   BIF,   so one is included here   ──►   CHANGESTR.REX.


original1 = Mary had a X lamb.
replaced  = Mary had a little lamb.

original2 = Mary had a % lamb.
replaced  = Mary had a little lamb.

original3 = Mary had a $$$ lamb.
replaced3 = Mary had a little lamb.

original4 = Mary had a someKindOf lamb.
replaced4 = Mary had a little lamb.


aString =substr("Mary had a X lamb.", "X", "little")
see aString + nl


irb(main):001:0> extra = 'little'
=> "little"
irb(main):002:0> "Mary had a #{extra} lamb."
=> "Mary had a little lamb."
irb(main):003:0> "Mary had a %s lamb." % extra
=> "Mary had a little lamb."


  • string_spec.rb describes interpolation using #{....} in double-quoted strings.
  • Core API describes printf-style interpolation by String#% and Kernel#sprintf.

Run BASIC[edit]

a$ = Mary had a X lamb."
a$ = word$(a$,1,"X")+"little"+word$(a$,2,"X")


Rust has very powerful string interpolation. Documentation here.

fn main() {
println!("Mary had a {} lamb", "little");
// You can specify order
println!("{1} had a {0} lamb", "little", "Mary");
// Or named arguments if you prefer
println!("{name} had a {adj} lamb", adj="little", name="Mary");


Library: Scala
object StringInterpolation extends App {
import util.matching.Regex._
val size = "little"
{ // Method I (preferred)
// Scala 2.10.0 supports direct string interpolation
// by putting "s" at the beginning of the string.
println("V2.10+  : " + s"Mary had a $size lamb,")
{ // Method II
// Pre Scala 2.10 indirect use of Java Class Formatter
val originalFormatter = "Mary had a %s lamb,"
println("V2.10- 1: " + originalFormatter format size)
// Above mentioned is Scala's postfix notation and equivalent for:
println("V2.10- 2: " + originalFormatter.format(size))
// Also possible
printf(("V2.10- 3: " + originalFormatter + '\n').format(size))
// All will be expanded to
print(("V2.10- 3: " + originalFormatter + '\n').format(size))
print((new java.util.Formatter).format("V2.10- 4: " + originalFormatter + '\n', size))
{ // Method III
// Regular expressions, only for demonstration
val extractor = """\$\{([^}]+)\}""".r
println((extractor.replaceAllIn("Regex 1: Mary had a ${x} lamb,", "snow white")))
// RegEx freaking
def interpolate(text: String, vars: (String, String)*) =
_ match { case Groups(v) => vars.toMap.getOrElse(v, "" /*in case nothing provided*/ ) })
println(interpolate("Regex 2: ${who} had a ${size} ${pet}, ${unknown}",
("pet", "lamb"), ("size", "fat"), ("size", "humongous"), ("who", "Mary")))
{ // Method IV, not recommended.
// Standard API method, search argument (1st ones) supposed to be a regular expression
println("Replace1: " + "Mary had a ${x} lamb".replaceAll("""\$\{x\}""", size))
// Standard API method, literally, on regular expression
println("Replace2: " + "Mary had a ${x} lamb".replaceAllLiterally("${x}", size))
{ // Method IV, not recommended.
println("Split  : " + "Mary had a ${x} lamb.".split("""\$\{([^}]+)\}""").mkString(size))



# Usage example: . interpolate "Mary has a X lamb" "quite annoying"
echo "$1" | sed "s/ X / $2 /g"


$ include "seed7_05.s7i";
const proc: main is func
const string: original is "Mary had a X lamb";
const string: little is "little";
var string: replaced is "";
replaced := replace(original, "X", little);
end func;
Mary had a little lamb


var extra = 'little';
say "Mary had a #{extra} lamb";


say ("Mary had a %s lamb" % 'little');

See: documentation


Every statement in SNOBOL can is a subset of pattern replacement having a subject (s1 in this case), object (s2), and replacement (s3).

        s1 = "Mary had a humongous lamb."
s2 = "humongous"
s3 = "little"
s1 s2 = s3

See The SNOBOL4 Programming Language; Griswold, Poage, Polonsky; Chapter 2 Pattern Matching


See printf in Stata help.

: printf("Mary had a %s lamb.\n", "little")
Mary had a little lamb.


let extra = "little"
println("Mary had a \(extra) lamb.")


String interpolation is a fundamental operation of the Tcl language itself, and is carried out in a "double-quoted" program strings as well as bare-words. Thus, interpolation of the string from a variable is carried out with the $ syntax and the string result of a command is interpolated with the […] syntax.

set size "little"
puts "Mary had a $size lamb."
proc RandomWord {args} {
lindex $args [expr {int(rand()*[llength $args])}]
puts "Mary had a [RandomWord little big] lamb."

When more sophisticated control is required the format command can be used, which is very similar to the standard C library's sprintf function:

puts [format "Mary had a %s %s." [RandomWord little big] [RandomWord lamb piglet calf]]

A third approach is to use string map.

set s "Mary had a @[email protected] lamb."
puts [string map {@[email protected] "miniscule"} $s]

Tcl also supports variable variable names. Even more powerful is nested interpolation with the subst command.

set grandpa {$daddy}; set grandma \$mommy
set daddy myself; set mommy {lil' bro}
set fun1 \[string\ to
set fun2 lower
set lower middle
set middle upper
set fun3 {aNd]}
puts [subst "$grandpa $fun1$[subst $$fun2] $fun3 $grandma"]


sentence_old="Mary had a X lamb."
DATA little
DATA big
sentence_new=SUBSTITUTE (sentence_old,":X:",0,0,values)
PRINT sentence_old
PRINT sentence_new
Mary had a X lamb.
Mary had a little lamb. 

UNIX Shell[edit]

Works with: Bourne Shell

Within the Unix shell, interpolation only occurs within doublequotes. Strings enclosed in singlequotes will not be subject to interpolation. Note that within the shell, a string may be used bare. If this is done each word within the string is treated separately, and any variable references or escape sequences will be substituted for their values:

echo Mary had a $extra lamb.
echo "Mary had a $extra lamb."
printf "Mary had a %s lamb.\n" $extra

A parameter substitution, like $extra or ${extra}, interpolates its value into some string. This happens outside quotes or inside "double quotes". The other form of interpolation is printf(1) with %s.

The shell has more forms of parameter substitution, like ${tea:?no tea}. Your shell's manual explains those. For the original Bourne Shell, sh(1) manual explains those.

C Shell[edit]

set extra='little'
echo Mary had a $extra lamb.
echo "Mary had a $extra lamb."
printf "Mary had a %s lamb.\n" $extra

C Shell has $extra and ${extra}. There are also modifiers, like $file:t; csh(1) manual explains those.


Expressions like this

-[foo-[ x ]-bar]-

evaluate to a list of character strings beginning with foo and ending with bar, where foo and bar are literal text (possibly multiple lines) and x is any expression evaluating to a list of character strings. Using a dot like this

-[foo-[. f ]-bar]-

makes it a function returning a list of character strings consisting of the output from the function f bracketed by the literal text foo and bar. In this task, the identity function, ~&, is used for f.

x = <'little'>
main = -[Mary had a -[. ~& ]- lamb.]- x

These operators are parsed like parentheses.

Mary had a little lamb.


Here are 2 examples:

With Replace

debug.print replace("Mary had a X lamb","X",a)  'prints Mary had a little lamb


With Interpolation function

Sub Main()
    debug.print Format("Mary had a {0} lamb",a)
End Sub

Public Function Format(ParamArray arr() As Variant) As String
    Dim i As Long, temp As String
    temp = CStr(arr(0))
    For i = 1 To UBound(arr)
        temp = Replace(temp, "{" & i - 1 & "}", CStr(arr(i)))
    Format = temp
End Function

CStr UBound ParamArray


// The @INTERPOLATE verb processes a string with imbedded blocks of code. The code
// blocks are parsed and evaluated. Any results are converted to a string, which
// is then inserted into the output string, replacing the code and braces.
// example: @INTERPOLATE "text{ @IF (x > y) then:{x} else:{y} }more text "
@VAR v = "little";
@SAY (@INTERPOLATE "Mary had a { v } lamb");
// output: Mary had a litle lamb

Visual Basic .NET[edit]

Dim name as String = "J. Doe"
Dim balance as Double = 123.45
Dim prompt as String = String.Format("Hello {0}, your balance is {1}.", name, balance)


"Mary had a X lamb.".replace("X","big")

Generates a new string. For more info, refer to manual in the downloads section of zkl page