please explain following encoding code at certain places where comment is present ?

function success = encode(msg,key)

num2add = 1000-length(msg); 
if num2add < 0, error('This message is too long to encode.'), end

newmsg = [msg,repmat(' ',1,num2add)];% whats happening here pls tell?
msgmat = dec2bin(newmsg)-48; 
[filen, pth]=uigetfile({'*.bmp';'*.tif';'*.jpg';'*.png'},'Choose Image To Encode.');

if isequal(filen,0) || isequal(pth,0)
    success = [];   return  % User cancelled.
end

pic1 = imread([pth filen]);  
B = pic1(:,:,1);   [piclngth, pichght] = size(B);
dim1 = piclngth-2;   dim2 = pichght-3; %why have we decreased 2 and 3 in dim ?   
keyb = key(end:-1:1);
rows = cumsum(double(key));  
columns = cumsum(double(keyb)); 
A = zeros(dim1,dim2);
A = crtmtrx(A,rows,columns,dim1,dim2,key);
idx = find(A==1); 

% please explain below for loop

for vv = 1:1000  
    for uu = 1:7
        if msgmat(vv,uu)==1;
           if rem(B(idx(uu+7*(vv-1))),2)==0
              B(idx(uu+7*(vv-1))) = B(idx(uu+7*(vv-1)))+1;
           end
        elseif rem(B(idx(uu+7*(vv-1))),2)==1
              B(idx(uu+7*(vv-1))) = B(idx(uu+7*(vv-1)))-1;
        end
    end
end

[dpb]

At least format the code so we can read it...

Answer 1

[Walter Roberson]

We don't know. There are an infinite number of different ways to encode something, and they can be completely arbitrary. The answer is just "Because the author chose to do it that way".

If the author had intended to implement some well-studied encoding mechanism, the author would have documented it in the comments. In the absence of author comments, any routine that does not always crash must be assumed to have been coded exactly the way the author wanted.