How can I export only specific data from a text file to a matrix?

I'm need to import only the numeric data of the attached .txt file into a 9x8 matrix and I'm not sure how to do this. Any help is greatly appreciated!

[Adam Danz]

I see several sections of numeric data, one of which is a 9x8 matrix. Where does the 2x2 come from?

Also do all of the text files follow this exact same format? For example is ":Solar_Radio_Flux" always on the 8th line?

[dpb]

What Adam said...

But, it's generally simpler to read what section of a file contains the info of interest and then just throw out what you don't need than it is to specifically hunt and peck through a file reading a bit here and there.

But, it certainly isn't at all clear what you want to retrieve from this file given the question wording...

[Caleb Sanders]

My bad! I'm not sure where I got 2x2 from either lol, it was late in the day when I posted. I'm looking to grab that 9X8 matrix and import into MATLAB. And yes Adam, I believe the text files follow this exact same format.

Answer 1

[Jan]

With some guessing, that you want the part behind ':Solar_Radio_Flux:' :

Key = ':Solar_Radio_Flux:';
Str = fileread(FileName);
C   = strsplit(Str, newline);
index = find(strncmp(C, Key, numel(Key)));

Block    = C(index+3:index+11);
BlockStr = sprintf('%s ', Block{:});  % Horizontal concatenation
% Data     = sscanf(BlockStr, '%g ', [9, 8]);  [EDITED:]
Data     = sscanf(BlockStr, '%g ', [8, 9])';   % Thanks Adam

This can be expanded e.g. to read the 2 header lines also or to accept any block of numbers until the next comment line.

[Caleb Sanders]

Thanks so much! Time for me to try and figure out what this means:)

[Adam Danz]

That last line should be

Data     = sscanf(BlockStr, '%g ', [8, 9])';

Answer 2

[Adam Danz]

rows = 11:19; % NUMERIC ROWS IN TEXT FILE TO IMPORT
nColumns = 8; % NUMBER OF COLUMNS OF TEXT
file = 'Testdata.txt'; % THE FULL PATH WOULD BE BETTER

opts = delimitedTextImportOptions('NumVariables',nColumns,'DataLines',rows([1,end]),'Delimiter', " ", ...
    'VariableTypes',repmat({'double'},1,nColumns),'ConsecutiveDelimitersRule','join',...
    'LeadingDelimitersRule','ignore');
Testdata = readtable(file, opts);
Testdata = table2array(Testdata);

Result

Testdata =
         245          14          12          11          -1          -1          -1          -1
         410          25          28          26          -1          -1          -1          -1
         610          34          -1          34          -1          -1          -1          -1
        1415          45          49          55          -1          -1          -1          -1
        2695         104          -1          68          -1          -1          -1          -1
        2800          -1          -1          -1          68          68          -1          -1
        4995         107         113         126          -1          -1          -1          -1
        8800         223         230         218          -1          -1          -1          -1
       15400         519         549         496          -1          -1          -1          -1

[Caleb Sanders]

Thanks for this, this was the easiest to understand!

[Adam Danz]

No problem! The main difference between this solution and Jan's is that this one only reads in the matrix data while Jan's reads in the whole file and then hones in on the matrix data.

[madhan ravi]

Last line is simply Testdata{:,:}

[Adam Danz]

@madhan ravi, it's the same. That's the only line in the table2array function. Matlab's import data tool uses table2array() which is why I ended up using it, too. But I agree that Testdata{:,:} is more direct.

[Jeremy Hughes]

Or if you have R2019a

Testdata = readtable(file, opts);

Testdata = table2array(Testdata);

can just be

A = readmatrix(file,opts)