MATLAB Answers

0

Splitting one column into multiple columns

Asked by Gopika Rajan on 16 Aug 2019
Latest activity Commented on by Walter Roberson
on 17 Aug 2019
Hello,
I am writing a script to split a singe cell to 9 columns. Here is the sample data. I want to be able to do some operations on these columns. I have tried reshape but it doesn't seem to work. Also the actual dataset consists of unknown number of rows.
'0.17 1.7285 0.001763792 1.000977 56651.41 0.000017652 DHT11, OK, 35.0,'
'0.34 1.6992 0.001733896 1.000977 57629.88 0.000017352 DHT11, OK, 34.0,'
'0.51 1.6602 0.001694037 0.996094 58700.00 0.000017036 DHT11, OK, 34.0,'
'0.68 1.6650 0.001699019 1.000977 58814.96 0.000017002 DHT11, OK, 34.0,'
'0.85 1.7187 0.001753827 0.996094 56695.46 0.000017638 DHT11, OK, 34.0,'
'1.02 1.6943 0.001728914 0.996094 57513.83 0.000017387 DHT11, OK, 34.0,'
'1.19 1.6016 0.001634247 0.996094 60851.21 0.000016434 DHT11, OK, 34.0,'
Any help would be very useful.
Thanks
Gopika

  1 Comment

Is it correct that the input is a cell array of character vectors?
Is there a reason that you are not using readtable() on the data?

Sign in to comment.

4 Answers

Answer by Andrei Bobrov
on 16 Aug 2019

input = {'0.17 1.7285 0.001763792 1.000977 56651.41 0.000017652 DHT11, OK, 35.0,';
'0.34 1.6992 0.001733896 1.000977 57629.88 0.000017352 DHT11, OK, 34.0,';
'0.51 1.6602 0.001694037 0.996094 58700.00 0.000017036 DHT11, OK, 34.0,';
'0.68 1.6650 0.001699019 1.000977 58814.96 0.000017002 DHT11, OK, 34.0,';
'0.85 1.7187 0.001753827 0.996094 56695.46 0.000017638 DHT11, OK, 34.0,';
'1.02 1.6943 0.001728914 0.996094 57513.83 0.000017387 DHT11, OK, 34.0,';
'1.19 1.6016 0.001634247 0.996094 60851.21 0.000016434 DHT11, OK, 34.0,'};
v = regexp(input,'\d*(\.\d*)?|\w*\d*','match');
out = cat(1,v{:});
lo=~cellfun(@(x)isletter(x(1)),out);
out(lo) = cellfun(@str2double,out(lo),'un',0);
T = cell2table(out);

  3 Comments

Thanks. I am getting the following error. I am quite new to MATLAB so I am not sure what this means.
Error using regexp
All cells must be char row vectors.
Thanks
If you could post an example of the input you are using;
The above solutions expect the input to be in rows and in a cell format.
Please attach your 'input' as mat-file.
On my laptop:
>> input = {'0.17 1.7285 0.001763792 1.000977 56651.41 0.000017652 DHT11, OK, 35.0,';
'0.34 1.6992 0.001733896 1.000977 57629.88 0.000017352 DHT11, OK, 34.0,';
'0.51 1.6602 0.001694037 0.996094 58700.00 0.000017036 DHT11, OK, 34.0,';
'0.68 1.6650 0.001699019 1.000977 58814.96 0.000017002 DHT11, OK, 34.0,';
'0.85 1.7187 0.001753827 0.996094 56695.46 0.000017638 DHT11, OK, 34.0,';
'1.02 1.6943 0.001728914 0.996094 57513.83 0.000017387 DHT11, OK, 34.0,';
'1.19 1.6016 0.001634247 0.996094 60851.21 0.000016434 DHT11, OK, 34.0,'};
v = regexp(input,'\d*(\.\d*)?|\w*\d*','match');
out = cat(1,v{:});
lo=~cellfun(@(x)isletter(x(1)),out);
out(lo) = cellfun(@str2double,out(lo),'un',0);
T = cell2table(out)
T =
7×9 table
out1 out2 out3 out4 out5 out6 out7 out8 out9
____ ______ _________ _______ _____ __________ _______ ____ ____
0.17 1.7285 0.0017638 1.001 56651 1.7652e-05 'DHT11' 'OK' 35
0.34 1.6992 0.0017339 1.001 57630 1.7352e-05 'DHT11' 'OK' 34
0.51 1.6602 0.001694 0.99609 58700 1.7036e-05 'DHT11' 'OK' 34
0.68 1.665 0.001699 1.001 58815 1.7002e-05 'DHT11' 'OK' 34
0.85 1.7187 0.0017538 0.99609 56695 1.7638e-05 'DHT11' 'OK' 34
1.02 1.6943 0.0017289 0.99609 57514 1.7387e-05 'DHT11' 'OK' 34
1.19 1.6016 0.0016342 0.99609 60851 1.6434e-05 'DHT11' 'OK' 34
>>

Sign in to comment.


Answer by Walter Roberson
on 16 Aug 2019

input = {'0.17 1.7285 0.001763792 1.000977 56651.41 0.000017652 DHT11, OK, 35.0,';
'0.34 1.6992 0.001733896 1.000977 57629.88 0.000017352 DHT11, OK, 34.0,';
'0.51 1.6602 0.001694037 0.996094 58700.00 0.000017036 DHT11, OK, 34.0,';
'0.68 1.6650 0.001699019 1.000977 58814.96 0.000017002 DHT11, OK, 34.0,';
'0.85 1.7187 0.001753827 0.996094 56695.46 0.000017638 DHT11, OK, 34.0,';
'1.02 1.6943 0.001728914 0.996094 57513.83 0.000017387 DHT11, OK, 34.0,';
'1.19 1.6016 0.001634247 0.996094 60851.21 0.000016434 DHT11, OK, 34.0,'};
t = textscan(strjoin(input, '\n'), '%f%f%f%f%f%f%s%s%f,'); %this might be what you are looking for directly
as_table = cell2table([num2cell(cell2mat(t(1:6))),t{7},t{8}, num2cell(t{9})]); %if you want a table

  0 Comments

Sign in to comment.


Answer by Banister on 16 Aug 2019

Not the most efficient, but,
input = {'0.17 1.7285 0.001763792 1.000977 56651.41 0.000017652 DHT11, OK, 35.0,';
'0.34 1.6992 0.001733896 1.000977 57629.88 0.000017352 DHT11, OK, 34.0,';
'0.51 1.6602 0.001694037 0.996094 58700.00 0.000017036 DHT11, OK, 34.0,';
'0.68 1.6650 0.001699019 1.000977 58814.96 0.000017002 DHT11, OK, 34.0,';
'0.85 1.7187 0.001753827 0.996094 56695.46 0.000017638 DHT11, OK, 34.0,';
'1.02 1.6943 0.001728914 0.996094 57513.83 0.000017387 DHT11, OK, 34.0,';
'1.19 1.6016 0.001634247 0.996094 60851.21 0.000016434 DHT11, OK, 34.0,'}
output = {}
for i = 1:size(input,1)
[b c] = regexp(input(i,1),'\s*|,\s*|,*','match','split')
output = [output; c{:,1}]
end

  1 Comment

Is the original as posted a txt file or .csv? What command have you used to get the data into the workspace?

Sign in to comment.


Answer by Gopika Rajan on 17 Aug 2019

Here is the script i have been working on. The first five rows have een removed.
fid = fopen('LE.txt', 'r') ; % Open source file.
fgetl(fid) ; % Read/discard line.
fgetl(fid) ; % Read/discard line.
fgetl(fid) ; % Read/discard line.
fgetl(fid) ; % Read/discard line.
fgetl(fid) ; % Read/discard line.
buffer = fread(fid, Inf) ; % Read rest of the file.
fclose(fid);
fid = fopen('LE_truncated.txt', 'w'); ; % Open destination file.
fwrite(fid, buffer) ; % Save to file.
fclose(fid) ;
File = importdata('LE_truncated.txt');
file1 = File(:,1);
Data = (file1.textdata);
input = {Data}; % This is where it shows the error ( Error using regexp,
All cells must be char row vectors)
v = regexp(input,'\d*(\.\d*)?|\w*\d*','match');
out = cat(1,v{:});
lo=~cellfun(@(x)isletter(x(1)),out);
out(lo) = cellfun(@str2double,out(lo),'un',0);
T = cell2table(out);

  1 Comment

Don't do that. Use readtable() with 'HeaderLines' option, or textscan() with 'HeaderLines' option.

Sign in to comment.