Ask for Data Extraction

Hello everyone, hope you have a good day today, i want to ask for help

I have a matriks around 1000x3, and the examples of some rows look like this

1 A1 Apple

1 A2 Apricot

1 A3 Avocado

2 B1 Banana

2 B2 Blackberry

2 B3 Blueberry

3 C1 Coconut

3 C2 Cherry

3 C3 Cranberry

4 D1 Dragonfruit

4 D2 Durian

4 D3 Date

I want to extract the data from column 2 and 3 based on the same value on column 1, but i can't find the solution, here is the idea of my script, sorry for the basic examples because i'm a beginner

data = fopen('dataexample.txt');
firstcolumn = data(:,1);

Var1 = data(firstcolumn == 1, :);
Var2 = data(firstcolumn == 2, :);
Var3 = data(firstcolumn == 3, :);

anyone can help me with this problem will be greatly appreciated. is it possible to generate a new file (in this case a txt file) for each same values in column 1? thank you for the attention, may you all always be healthy :)

Answer 1

[Walter Roberson]

T = readtable('dataexample.txt', 'readvariablenames', false);
G = findgroups(T{:,1});
grouped = splitapply(@(var1,var2,var3) {table(var1, var2, var3)}, T, G);

for K = 1 : numel(grouped)
    thistable = grouped{K};
    groupid = thistable{1,1};
    filename = sprintf('example_grouped_%d.txt', groupid);
    writetable(thistable, filename, 'writevariablenames', false);
end

[Blaze Dexter]

any suggestions why i got the readtable error message? (Undefined function 'readtable' for input arguments of type 'char'.)

[Walter Roberson]

You did not post a readtable error message?

Also you did not indicate which MATLAB release you are using.

[Blaze Dexter]

sorry, i'm using r2012b

[Walter Roberson]

Then you need a completely different approach.

fid = fopen('dataexample.txt', 'r');
datacell = textscan(fid, '%f%s%s');
fclose(fid);
C1 = datacell{1}; C2 = datacell{2}; C3 = datacell{3};

group_ids = unique(C1, 'stable');

for K = 1:numel(group_ids)
    groupid = group_ids(K);
    idx = find(C1 == groupid);
    filename = sprintf('example_grouped_%d.txt', groupid);
    fid = fopen(filename, 'wt');
    for row = idx(:).'      %force row vector
       fprintf('%g %s %s\n', C1(row), C2{row}, C3{row});
    end
    fclose(fid);
end

[Blaze Dexter]

ah i see, the readtables is not in r2012b, thank you so much

Answer 2

[Ive J]

t = readtable('text.txt');
t.Properties.VariableNames = "col" + (1:size(t, 2)); % set col names
12×3 table

    col1     col2          col3      
    ____    ______    _______________

     1      {'A1'}    {'Apple'      }
     1      {'A2'}    {'Apricot'    }
     1      {'A3'}    {'Avocado'    }
     2      {'B1'}    {'Banana'     }
     2      {'B2'}    {'Blackberry' }
     2      {'B3'}    {'Blueberry'  }
     3      {'C1'}    {'Coconut'    }
     3      {'C2'}    {'Cherry'     }
     3      {'C3'}    {'Cranberry'  }
     4      {'D1'}    {'Dragonfruit'}
     4      {'D2'}    {'Durian'     }
     4      {'D3'}    {'Date'       }

The question for you would be, what do you wanna exactly do with data itself? You simply can loop over col1, or more efficiently use groupfilter

ucol1 = unique(t.col1, 'stable');
Vars = ({});
for i = 1:numel(ucol1)
    Vars{i, 1} = t{t.col1 == ucol1(i), 2:end};
end