Import not consistent table data

Hi! Sorry again for a (maybe) naif question, but I have been struggling all day trying to figure out how I can import a not consistent .txt dataset. I show you in figure what I mean:

From that, I just need to extract the column with inside the blu limiters. I tried to set import option and get the column where those numbers are in by the following code:

opts = detectImportOptions(sample_txt_file);
disp([opts.VariableNames' opts.VariableTypes']); 
opts = setvartype(opts,{'x1100071'},'char'); 
disp([opts.VariableNames' opts.VariableTypes']);
values = readtable(sample_txt_file,opts);
times = values{:,4};

As result, I got 4 table columns and then I extracted the last one, the one with numbers I need.

Checking on how I could change 'char' values into a numeric array I coded:

S = sprintf('%s',times{:});
D = sscanf(S, '%f');

So I got an array containing number values of all those from column 4 but still don't understand how it works or if there are more suitable ways to extract data from that dataset.

Hope you could help me and thank you again :)

[Mathieu NOE]

hello

can you provide your txt file ?

[Luca Castrogiovanni]

I attach here the sample.zip. Thank you :)

[Mathieu NOE]

do you want separate blocks of data or everything in one block ?

Answer 1

[Mathieu NOE]

hello again

so this first simple code will extract the data as one single block

if you need to separate the data in individual blocks as organized in your file , that needs a bit of extra code but shoudn't be very difficult :

str = readlines('samples.txt');% read the file:
C = regexp(str,' MODfromPPD:: PARA:'); % find lines that contains that string
ind = find(~cellfun('isempty',C)); % indice of non empty cells
data_lines = split(str(ind)); % split columns of selected rows
your_data = str2double(data_lines(:,7)); % your data are in the 7th column of data_lines

[Mathieu NOE]

in the mean time , this is the code extended that puts data in blocks like in the txt file

one block of data is stored in a cell - you have now a cell array of dimensions 4000 x 1 containing your 4000 blocks of data

hope it helps

str = readlines('samples.txt');% read the file:
C = regexp(str,' MODfromPPD:: PARA:'); % find lines that contains that string
ind = find(~cellfun('isempty',C)); % indice of non empty cells
data_lines = split(str(ind)); % split columns of selected rows
your_data = str2double(data_lines(:,7));

% split in blocks 
ind_blocks = str2double(data_lines(:,6)); % 
ind_first_line = find(ind_blocks<2); % find start lines correponding to value = 1 

for ck = 1:numel(ind_first_line)-1
    block{ck} =  your_data(ind_first_line(ck):ind_first_line(ck+1)-1,1);
end

% add the last block 
block{ck+1} = your_data(ind_first_line(end):end,1);

[Luca Castrogiovanni]

Yes, it worked! I just needed not splitted so the first code was good, but I thank you for you detailed and helpful answer. It was so kind of you.

I think I should spend more time on how to import different datatypes but your explanation helped me much! :)

[Mathieu NOE]

yes , it takes some time to figure out how to deal with the multiple file formats , but after some time you go faster as you get more experienced.

also this forum is a good place to ask and learn from other's Q & A