function y = flip(x,dim)

%FLIP Flip the order of elements

%   Y = FLIP(X) returns a vector Y with the same dimensions as X, but with the

%   order of elements flipped. If X is a matrix, each column is flipped 

%   vertically. For N-D arrays, FLIP(X) operates on the first

%   nonsingleton dimension.

%

%   FLIP(X,DIM) works along the dimension DIM.

%

%   For example, FLIP(X) where

%

%       X = 1 4  produces  3 6

%           2 5            2 5

%           3 6            1 4

%

%

%   Class support for input X:

%      float: double, single

%      integers: uint8, int8, uint16, int16, uint32, int32, uint64, int64

%      char, logical

%

%   See also FLIPLR, FLIPUD, ROT90, PERMUTE.

%   Copyright 1984-2018 The MathWorks, Inc.

% Argument parsing

if nargin == 1 

    dim = find(size(x)~=1,1,'first'); % Find leading singleton dimensions

    if isempty(dim) % scalar case

        dim = 2;

    end

else 

    dim = matlab.internal.math.getdimarg(dim); 

end 

dimsize = size(x,dim); 

if dimsize <= 1 && (isreal(x) || ~isnumeric(x)) 

    % No-op.

    y = x; 

else 

    % Create the index that will transform x.

    v(1:max(dim,ndims(x))) = {':'}; 

    % Flip dimension dim.

    v{dim} = dimsize:-1:1; 

    % Index with v.

    y = x(v{:}); 

end