Secify what is the end of a line using textscan?

I have a .dat file with varied spacing between a series of numbers. I want to be able to jump to any one of these numbers using textscan with its 'headerlines' feature. The problem is, I need to specify that each separate number should be a new line. That way if is use 'headerlines',2 in textscan it will skip the first two numbers and go from there.

My .dat file starts like this

      1      55      63       78  45.6  34.6  45.7  23.5  23.7  34.5  23.6 ...

how can I make textscan see line one as 1... line two as 55... line three as 63... line four as 78 etc? I know about the 'EndOfLine' feature and have tried everything with that but it's still not giving me what I need.

Thanks

-Kyle

Answer 1

[Walter Roberson]

I would not do it that way. What I would do is something like:

fid =  fopen(YourFileName, 'rt');

wantpos = 3;  %want 3rd number
fmt = [repmat('%*f', 1, wantpos-1), '%f'];

datacell = textscan(fid, fmt, 1);

wantednum = datacell{1};

fclose(fid);

This can be used even if you have fewer numbers per line than "wantpos": it will read as many lines as it needs to in order to get the proper number.

Afterwards the file will be positioned right after the number that was read.

[kschau]

Trying to implement your solution but I cant get it to output anything but NaN. Here is my code

fid=fopen('dummy.dat');

wantpos=3;

fmt = [repmat('%*f', 1, wantpos-1), '%f'];

datacell = textscan(fid, fmt, 1);

U = datacell{1}

Thanks for the help!

[Walter Roberson]

When I use the string you gave, it works for me. Is it certain that the data is separated by spaces? And why did you remove the 'rt' from the fopen() ? Try this:

fid = fopen('dummy.dat', 'r');
data = fread(fid, 20, '*uint8');
fclose(fid)

and then disp(data) and show us what it indicates