fscanf question when ignoring header

Hello,

I have a text file that im having issues with. Ive been able to read my others files but this one for some reason isnt working properly when trying to us fscanf.

The file has 5 headers then data of 2k samples:

SAMPLES: 2000

BITSPERSAMPLE: 32

CHANNELS: 1

SAMPLERATE: 96000

NORMALIZED: FALSE

50

45

30

etc...till 2k values

My code: for convience lets say this file is file # 4.

formatSpec = '%*s %f'; %ignore strings and keep integers
myCell = {}; %create cell to put data into after uigetfile



[file,path] = uigetfile('*.rnd','MultiSelect', 'on');

            %creates file{1,X}
            
            %open file # 4
            pathname = file{1,4};
          
            %use current pathname to open file                    
            fileID = fopen(pathname,'r');
            
            %read file with formatSpec set to ignore strings and read only numbers
            myCell{1,1} = fscanf(fileID,formatSpec); 

Pretty straightforward. And my code for opening all of my other files that have different headers/footers works fine,and they all use this same setup for formatSpec and fscanf. But when opening this file it just returns [2000,32,1,96000] in myCell.

Answer 1

[Walter Roberson]

'%*s %f' does not mean to skip all strings and then look for a number. It means to skip one string and then look for a number. So it will skip 'NORMALIZED' and then try to read a number but will fail because FALSE is not a number.

[m j]

Smh.....Sorry, but thanks! Had misunderstood it,figured it out with:

fileID = fopen(name,'r');
%ignore first 5 lines
for i = 1:5
letsC=fgets(fileID);
end
%read file with formatSpec set to ignore strings and read only numbers
myCell{1,1} = fscanf(fileID,formatSpec); 

But is there anyway to remove a header/footer of file without knowing how long the header/footer is within the file? Or is there a way to remove all lines of file that contain strings/chars? And doing this will only leave data....

Just off top of head,maybe go through each line of file and test if line contains a char/string,if it does. fgets it??On cellphone currently so, apologies if this sounds dumb.

[Walter Roberson]

In the case where the header starts with a known string and ends with a known string then you can use textscan CommentStyle parameter to skip the content.

Otherwise you need to know something more about the pattern that indicates end of header.

A lot of the time it can make sense to read everything in as a single string and then manipulate it with regexp or regexprep.

[m j]

Thanks, ill look into making it into and single string a manipulating it. I would just like the freedom to be able to open up multiple files that have different headers/footers and extract the data,regardless if I know how they are set up.