From Wikipedia, the free encyclopedia
Jump to navigation Jump to search

In computer programming, append is the operation for concatenating linked lists or arrays in some high-level programming languages.


Append originates in the Lisp programming language. The append procedure takes zero or more (linked) lists as arguments, and returns the concatenation of these lists.

(append '(1 2 3) '(a b) '() '(6))
;Output: (1 2 3 a b 6)

Since the append procedure must completely copy all of its arguments except the last, both its time and space complexity are O(n) for a list of elements. It may thus be a source of inefficiency if used injudiciously in code.

The nconc procedure (called append! in Scheme) performs the same function as append, but destructively: it alters the cdr of each argument (save the last), pointing it to the next list.


Append can easily be defined recursively in terms of cons. The following is a simple implementation in Scheme, for two arguments only:

(define append
  (lambda (ls1 ls2)
    (if (null? ls1)
      (cons (car ls1) (append (cdr ls1) ls2)))))

Append can also be implemented using fold-right:

(define append
   (lambda (a b)
      (fold-right cons b a)))

Other languages[edit]

Following Lisp, other high-level languages which feature linked lists as primitive data structures have adopted an append. Haskell uses the ++ operator to append lists. OCaml uses the @ operator to append lists.

Other languages use the + or ++ symbols for nondestructive string/list/array concatenation.


The logic programming language Prolog features a built-in append predicate, which can be implemented as follows:

append([X|Xs],Ys,[X|Zs]) :-

This predicate can be used for appending, but also for picking lists apart. Calling

 ?- append(L,R,[1,2,3]).

yields the solutions:

L = [], R = [1, 2, 3] ;
L = [1], R = [2, 3] ;
L = [1, 2], R = [3] ;
L = [1, 2, 3], R = []


This right-fold, from Hughes (1989:5-6), has the same semantics (by example) as the Scheme implementation above, for two arguments.

append a b = reduce cons b a

Where reduce is Miranda's name for fold, and cons constructs a list from two values or lists.

For example,

append [1,2] [3,4] = reduce cons [3,4] [1,2]
    = (reduce cons [3,4]) (cons 1 (cons 2 nil))
    = cons 1 (cons 2 [3,4]))
        (replacing cons by cons and nil by [3,4])
    = [1,2,3,4]


This right-fold has the same effect as the Scheme implementation above:

append :: [a] -> [a] -> [a]
append xs ys = foldr (:) ys xs

This is essentially a reimplementation of Haskell's ++ operator.


In Perl, the push function is equivalent to the append method, and can be used in the following way.

my @list;
push @list, 1;
push @list, 2, 3;

The end result is a list containing [1, 2, 3]

The unshift function appends to the front of a list, rather than the end

my @list;
unshift @list, 1;
unshift @list, 2, 3;

The end result is a list containing [2, 3, 1]

When opening a file, use the ">>" mode to append rather than over write.

open(my $fh, '>>', "/some/file.txt");
print $fh "Some new text\n";
close $fh;

Note that when opening and closing file handles, one should always check the return value.


In Python, use the list method "extend" or the infix operators + and += to append lists.

l = [1, 2]
l.extend([3, 4, 5])
print l + [6, 7]

After executing this code, l is a list containing [1, 2, 3, 4, 5], while the output generated is the list [1, 2, 3, 4, 5, 6, 7].

Do not confuse with the list method "append", which adds a single element to a list:

l = [1, 2]

Here, the result is a list containing [1, 2, 3].


In Bash the append redirect is the usage of ">>" for adding a stream to something, like in the following series of shell commands:

echo Hello world! >text; echo Goodbye world! >>text; cat text

The stream "Goodbye world!" is added to the text file written in the first command. The ";" implies the execution of the given commands in order not simultaneously. So, the final content of the text file is:

Hello world!
Goodbye world!