How can I solve the problem 'Subscripted assignment dimension mismatch' in my looping progress?

I want to write vector to an array (matrix), but I get problem in the dimension of it. ntr is the number of trace. this is my code:

 for i=1:ntr                            % read traces into panel
   [seis(:,i),seqno(i),itr(i),irec(i),dt,offset,sdepth,selev,relev(i),...
            xs,ys,xr(i),yr(i),cdp(i)]=segyin(fid);
      if i==1
         cdp(i)                              % list cdp as progress flag
      end
      if (ixs == 0)
         xoff=[xoff;offset];                 % build offset vector
      else
         xoff = [xoff;10*(i-1)-1500];
      end

here is the error:

Subscripted assignment dimension mismatch.
Error in trpan (line 152)
      [seis(:,i),seqno(:,i),itr(i),irec(i),dt,offset,sdepth,selev,relev(i),...

Answer 1

[Walter Roberson]

In the error message, the first output is given as seis(nsamp,i), but in the code segment, the first output is given as seis(:,i) . The code you are running is not the same as the code you posted.

Note that we do not know anything about segyin() and we do not know how you initialized your variables.

[Kiswanto Ihsani]

i have revised the code. i just copy from my trial and error method.

this is the original function of segyin

function [trc,seqno,itr,irec,dt,offset,sdepth,selev,relev,...
      xs,ys,xr,yr,cdp]=segyin(fid)
% nsamp=length(trc);
%read the trace header

[stuff,count]=fread(fid,4,'int');

if(count<4)
  trc=nan;

          seqno=nan; itr=nan; irec=nan; dt=nan;
    offset=nan; sdepth=nan; selev=nan; relev=nan;
    xs=nan; ys=nan; xr=nan; yr=nan; cdp=nan;
    return;
  end
  seqno=stuff(1); irec=stuff(3); itr=stuff(4);
  stuff=fread(fid,4,'char');
  cdp=fread(fid,1,'int');
  stuff=fread(fid,4,'char');
  code=fread(fid,1,'short');
  stuff=fread(fid,6,'char');
  stuff=fread(fid,4,'int');
  offset=stuff(1);relev=stuff(2);selev=stuff(3);sdepth=stuff(4);
  stuff=fread(fid,16,'char');
  stuff=fread(fid,2,'short');
  stuff=fread(fid,4,'int');
  xs=stuff(1);ys=stuff(2);xr=stuff(3);yr=stuff(4);
  lenunit=fread(fid,1,'short');
  stuff=fread(fid,24,'char');
  stuff=fread(fid,2,'short');
  nsamp=stuff(1);dt=stuff(2)/1000000.;
  %Note SEGY stores dt in microseconds
  stuff=fread(fid,122,'char');
  %trace header complete
  %read in trace
  trc=fread(fid,nsamp,'float');

Answer 2

[Thorsten]

The error seems to be due to the first two arguemnts. Try

   [a,b,itr(i),irec(i),dt,offset,sdepth,selev,relev(i),...
            xs,ys,xr(i),yr(i),cdp(i)]=segyin(fid);

    whos a 
    whos b

What's the result? That may give you some idea of how to correct the code.

[Kiswanto Ihsani]

I got all of the parameters in segyin like this:

ntr = 7

Name      Size            Bytes  Class     Attributes

a         0x0                 0  double              

Name      Size            Bytes  Class     Attributes

b         1x1                 8  double

ans =

   101056512

but it produce the other error

Error using fread

Invalid size

trc=fread(fid,nsamp,'float')

I used 'float' as my precision to read binary data

[Walter Roberson]

nsamp is somehow invalid, such as if it is negative. You are reading it as 'short' which is a signed value.

[Kiswanto Ihsani]

i have give absolute value for nsamp, so it to be nsamp=abs(nsamp). as the result it will give positive value in all of the range. actually this code is an old code, it has build for more than 12 years ago, I am so tired to check it one by one. When one line is clear, there is a problem in the other line.

[Thorsten]

What's the value of nsamp? Could it be 0?

[Kiswanto Ihsani]

the value of nsamp is absolutely positive, based on the my previous experience, nsamp value more than 50.