How should I import dataset separately from a long text file?

Hi,

I have a dataset like this (attached). Please let me know if there is a way to import all the "tables" individually as a separate dataset. Basically, the data loops start after the summary with a string "table x", then some parameters. After parameters the data starts with the variable names.

I also want to ignore the summary of the dataset at the beginning.

delimeter: tab

[Mathieu NOE]

hello

you need a dedicated parser for that , but a "real" data file would be more efficient to work with

all the best

[Md Redwanul Islam]

Thank you for your comment.

Here is an example real dataset. The first Table 1 is the summary.

[Mathieu NOE]

hello again

to be sure, what you need is those tables ?

like

Time [s] V+ [V] V- [V] I1 [A] P1 [uC/cm2] I2 [A] P2 [uC/cm2] I3 [A] P3 [uC/cm2] D1 [nm] D2 [nm] D3 [nm] CH3 [V]

0.000000e+000 7.820238e-001 0.000000e+000 5.293780e-008 -4.931280e+001 5.293780e-008 -4.931280e+001 0.000000e+000 -2.672872e+001 0.000000e+000 0.000000e+000 0.000000e+000 0.000000e+000

2.500000e-004 2.778735e+000 0.000000e+000 5.542453e-008 -4.917735e+001 5.542453e-008 -4.917735e+001 0.000000e+000 -2.672872e+001 -2.207564e-003 -2.207564e-003 0.000000e+000 0.000000e+000

5.000000e-004 4.777116e+000 0.000000e+000 6.000530e-008 -4.903306e+001 6.000530e-008 -4.903306e+001 0.000000e+000 -2.672872e+001 -1.466427e-002 -1.466427e-002 0.000000e+000 0.000000e+000

7.500000e-004 6.777614e+000 0.000000e+000 6.009513e-008 -4.888294e+001 6.009513e-008 -4.888294e+001 0.000000e+000 -2.672872e+001 -1.587351e-002 -1.587351e-002 0.000000e+000 0.000000e+000

1.000000e-003 8.768961e+000 0.000000e+000 5.680956e-008 -4.873681e+001 5.680956e-008 -4.873681e+001 0.000000e+000 -2.672872e+001 -2.400754e-002 -2.400754e-002 0.000000e+000 0.000000e+000

1.250000e-003 1.077756e+001 0.000000e+000 5.808965e-008 -4.859318e+001 5.808965e-008 -4.859318e+001 0.000000e+000 -2.672872e+001 -3.077039e-002 -3.077039e-002 0.000000e+000 0.000000e+000

1.500000e-003 1.276941e+001 0.000000e+000 5.580522e-008 -4.845081e+001 5.580522e-008 -4.845081e+001 0.000000e+000 -2.672872e+001 -3.635528e-002 -3.635528e-002 0.000000e+000 0.000000e+000

1.750000e-003 1.476892e+001 0.000000e+000 6.166729e-008 -4.830397e+001 6.166729e-008 -4.830397e+001 0.000000e+000 -2.672872e+001 -3.822271e-002 -3.822271e-002 0.000000e+000 0.000000e+000

2.000000e-003 1.676584e+001 0.000000e+000 5.885770e-008 -4.815332e+001 5.885770e-008 -4.815332e+001 0.000000e+000 -2.672872e+001 -4.513578e-002 -4.513578e-002 0.000000e+000 0.000000e+000

2.250000e-003 1.876169e+001 0.000000e+000 5.659342e-008 -4.800900e+001 5.659342e-008 -4.800900e+001 0.000000e+000 -2.672872e+001 -4.782756e-002 -4.782756e-002 0.000000e+000 0.000000e+000

[Md Redwanul Islam]

Hi,

Yes. so, I have these separate tables of dataset in one large text file. I wanted to load these table dataset separately e.g. table-1, table-2, ..., table-n. Then it would be easier for me to plot them, like i can say plot column2 and column4 of table-n.

Answer 1

[Mathieu NOE]

hello

the demo txt files has two tables so this is how the code generates the labels and data (as cell arrays)

label =

1×13 cell array

Columns 1 through 7

{'Time [s]'} {'→V+ [V]'} {'→V- [V]'} {'→I1 [A]'} {'→P1 [uC/cm2]'} {'→I2 [A]'} {'→P2 [uC/cm2]'}

Columns 8 through 13

{'→I3 [A]'} {'→P3 [uC/cm2]'} {'→D1 [nm]'} {'→D2 [nm]'} {'→D3 [nm]'} {'→CH3 [V]'}

data =

1×2 cell array

{10×13 double} {10×13 double}

code

fid = fopen('data.txt');
tline = fgetl(fid);

% init variables
n = 0;
flag =0;

while ischar(tline)
    
    if findstr(tline,'Time [s]')     % extract data only for lines below 'Time [s]'
        
        headerline = tline;
        n = n+1;
%         disp(n)  % debug only
%         disp(tline) % debug only
        flag =1;
        k = 0;
    end
    
     tline = fgetl(fid);
     
     if flag == 1 
        
        out = tline;
        
        if length(out) <2
            flag =0;
        else
            k = k+1;
            data{n}(k,:) = str2num(out);
        end
        
    end
    


    
end
fclose(fid);

% labels
in2 = strfind(headerline,']');
label{1} = headerline(1:in2(1));
for ci = 2:length(in2)
    label{ci} = headerline(in2(ci-1)+1:in2(ci));
end

% display in command window for info

label
data

[Md Redwanul Islam]

Hi,

Works perfectly! Thank you so much for the help.

[Mathieu NOE]

Glad it matches your expectations !

do you mind accepting my answer ?

tx !