Matlab app designer confuses the end written inside matrix index with the end for if and for loop

Hi all

I got a piece of my code that is causing function nesting error and the root cause is that matlab is not understanding the end written in Matrix(end)

is not related to the loop or function end and has messed up

exactly where I haveTtot(end) , this end is confused

for fn=1:numel(files)
    
    ff= xlsread(strcat(files{fn},'.xlsx'));
    filename=files{fn};
    t = ff(:,1);
    resol=t(2)-t(1);
    motion6t=ff;
  
    mulnames=readcell('Ms.xlsx')
    mask = strcmp(filename, mulnames(:,1));
    multip = cell2mat(mulnames(mask, 2));
    
    
    
    resol=t(2)-t(1)
    if fn==1
        motiontot=[];
        for m=1:multip
            if m==1
               Ttot=t.';
            else
               Ttot= [Ttot,t.'+Ttot(end)+resol];
            end
        end
        
    else
        for m=1:multip
            Ttot= [Ttot,t.'+Ttot(end)+resol];
        end
    end
    for m=1:multip
        motiontot=[motiontot;motion6t];
    end
         
end

[Mehmed Saad]

what is the exact error message.

[farzad]

What error would you get if you misplace an end expression??

Matlab will take the tail of that misplacement to the next end and the error will depend on the number of loops and functions. I think the focus of the question is confusing the end inside the paranthesis. When I comment that line, everything is fine

[Mehmed Saad]

is this your error?

What is the value of Ttot when your loop run for the first time? because if f1 is not equal to 1, the else condition will run.

[Stephen23]

@farzad: please show us the complete error message. This means all of the red text.

[Mehmed Saad]

@farzad: you are asking a lot of questions in community and sometimes your behaviour is very rude with contributors.

[farzad]

@Mehmed: First I thank you for your contribution. But I don't understand your rant. This is the first time I have such a complaint here while have not even felt where the problem is. I may have asked a lot of questions but my interactions with you as far as I remember has not been in all of my questions, so it remains a question for me, why before letting me answer your question, you jumped to judge my behavior , where have I been very rude ( and not even just rude ) ? that' a very strange judgement. I might be frustrated with matlab error not you, by the way.

and sorry , it had not to be f1 , but fn, but still the same problem

When I hover on the word end in Tot(end) , Matlab should not match that end with a previous if or for expression. it just shouldn't. I get some sort of the error message you mention but in the grey part of app designer code. only due to the misreasing of end expression, the next chain of end words assigned to irrelevant loops and this causes the pushbutton end get lost and the code give me some error like

Illegal use of reserved keyword "methods". 

which is in the grey part of app designer

[Walter Roberson]

The error you indicate does not occur in the section of code you posted. If you were to attach the full code, we might be able to figure out what is going on.

[farzad]

The full code is really long dear Walter. and I couldn't share it. I was assuming this error is known or has some known root cause. I am checking al the paranethsis and intdentations under the push button

[Walter Roberson]

The typical root cause for these kinds of messages is a missing ' or " or ]

Double-check that you do not accidentally have a ‘ (right quote, U+2018) instead of a ' (apostrphone, U+0027)

[farzad]

Dear Walter, I checked, and it seems that whenever I remove this block , the whole code is ok. the only suspicious ' sign, was t.' in the summation equation with Ttot. but when I removed it, I still have the same error. I had some variables with the warning : the variable spans among more functions. I resolved it as well, but still I have the same problem :

for fn=1:numel(files)
    
    ff= xlsread(strcat(files{fn},'.xlsx'));
    fname=files{fn};
    t = ff(:,1);
    
    resol=t(2)-t(1);
    motion6t=ff;
    
    cd(myperc)
    mulnames=readcell('Multipliers.xlsx');
    mask = strcmp(fname, mulnames(:,1));
    mu = cell2mat(mulnames(mask, 2));
    
    
    
    resol=t(2)-t(1);
    if fn==1
        motiontot=[];
        for m=1:mu
            if m==1
               mTtot=t;
            else
               mTtot= [mTtot ,t.'+mTtot(end)+resol];
            end
        end
        
    else
        for m=1:mu
            mTtot= [mTtot ,t.'+mTtot(end)+resol];
        end
    end
    for m=1:mu
        motiontot=[motiontot;motion6t];
    end

[farzad]

Ok this is the complete code, and trying this on workspace I get error on mTtot:

Array indices must be positive integers or logical values.

Error in test (line 47)
            mTtot= [mTtot ,t.'+mTtot(end)+resol];

mydir=uigetdir()
content = dir('*.xlsx'); 
content([content.isdir]) = [];   % remove directories
files = cell(size(content)); 
% Extract the filename without the extentions
for fi = 1:numel(files)
    [~, files{fi}] = fileparts(content(fi).name); 
end


for fn=1:numel(files)
    
    ff= xlsread(strcat(files{fn},'.xlsx'));
    fname=files{fn};
    t = ff(:,1);
    
    resol=t(2)-t(1);
    motion6t=ff;
   
    mulnames=readcell('Ms.xlsx');
    mask = strcmp(fname, ulnames(:,1));
    mu = cell2mat(mulnames(mask, 2));
    
    
    mTtot=[];
    resol=t(2)-t(1);
    if fn==1
        motiontot=[];
        for m=1:mu
            if m==1
               mTtot=t;
            else
               mTtot= [mTtot ,t.'+mTtot(end)+resol];
            end
        end
        
    else
        for m=1:mu
            mTtot= [mTtot ,t.'+mTtot(end)+resol];
        end
    end
    for m=1:mu
        motiontot=[motiontot;motion6t];
    end


        
    maxlim=max(max(motion6t(:,2:7)));
    minlim=min(min(motion6t(:,2:7)));
    directory=pwd;
end

Answer 1

[Walter Roberson]

    mTtot=[]
    size(mTtot)
    
    mTtot =
     []
ans =
     0     0

with it being empty, end as an index would be 0. mTtot(end) would be mTtot(0) which is not permitted.

mTtot= [mTtot ,t.'+mTtot(end)+resol];

Perhaps you are wanting mTtot(end) to be empty instead of an error when mTtot is empty. But if that were the case, then you would have the situation

mTtot = [[], t.'+[]+resol]

and anything + [] is [], so you would be dealing with

 mTtot = [[], []]

which would leave mTtot empy, but I suspect you are trying to append to it in that statement,.

[farzad]

Thank you very much for the answer Walter. well my aim is to append the time series one after the other. this was working somewhere else in my code, but since my code has become super complex, I tried to understand it here.

so in the first step , you have the first time series , let's say 35 seconds with 0.01 seconds of resolution, if the second time series is 15 seconds, by appending , mTtot should be 35+15 = 50 seconds but to append it to an array, I need the above algorithm, maybe it needs imrpovements

[farzad]

when mTtot =[] ; we are in the first step of the loop, and in that step, meaning

if m==1
    mTtot= t.'
end

[farzad]

am I doing a simple mistake or there is a better way to do this to not encounter this error ?

[Walter Roberson]

You missed the case of fn not equal to 1.

[farzad]

I'm not sure , cause I have put an else after that if

[Walter Roberson]

Yes, but suppose that fn is not 1 when mTtot is empty. You go to the else part, and in that else part you do not have the if m==1 protection.

[farzad]

you are right ! I tried that as bellow, and I got :

Error :

Error using horzcat
Dimensions of arrays being concatenated are not consistent.

Error in test (line 50)
               mTtot= [mTtot ,t.'+mTtot(end)+resol];

if fn==1
        motiontot=[];
        for m=1:mu
            if m==1
               mTtot=t;
            else
               mTtot= [mTtot ,t.'+mTtot(end)+resol];
            end
        end
        
    else
        for m=1:mu
            if m==1
               mTtot=t;
            else
               mTtot= [mTtot ,t.'+mTtot(end)+resol];
            end
        end
    end

[Walter Roberson]

if fn==1
        motiontot=[];
        for m=1:mu
            if m==1
               mTtot = t(:).';
            else
               mTtot= [mTtot, t(:).'+mTtot(end)+resol];
            end
        end
        
    else
        for m=1:mu
            if m==1
               mTtot = t(:).';
            else
               mTtot= [mTtot, t(:).'+mTtot(end)+resol];
            end
        end
    end

[farzad]

Thank you so much !!! this works perfectly

[farzad]

dear Walter, after going ahead with my code, and adding other lines, not relevant to this loop, I mean : they should not affect it at all , now I see that Matlab does not pass through this loop and does not generate mTtot at all. that confusion of end expressions still exists though, but I hope it's just a visualization problem of app designer. the loop works fine itself in matlab base, but now I don't knw how to resolve it

[farzad]

if this problem lasts, my code won't work.

[Walter Roberson]

Perhaps you could use the editor "Smart Indent" feature on your code; by following the indentation levels you might be able to figure out why it thinks you have an extra end statement.

[farzad]

Thank you so much !!! that was actually the problem, I had put all the functions on the zero columns under the button

[farzad]

dear Walter, I wanted to avoid using end in brackets. So I tried

if fn==1
    motiontot=[];
    for m=1:mu
        if m==1
            Ttot = t(:).';
            Tsize=size(Ttot) 
        else
            Ttot= [Ttot, t(:).'+Ttot(Tsize)+resol];
            Tsize=size(Ttot)
        end
    end
    
else
    for m=1:mu
        if m==1
            
            Ttot = t(:).'; % I am not if sure it should be Ttot= [Ttot, t(:).'+Ttot(Tsize)+resol]; ?
            
            Tsize=size(Ttot)
        else
            Ttot= [Ttot, t(:).'+Ttot(Tsize)+resol];
            Tsize=size(Ttot)
        end
    end
end

I have a doubt, that when fn > 1 , I think Ttot should be already assigned with some values. but in matlab I get error. more than that, if the total number of fn is bigger than 2 I get a dimension mismatch problem

[Walter Roberson]

Tsize = numel(Ttot);

[farzad]

fantastic !! thank you