MATLAB: Remove rows in a csv file

csv csvread

Hello,

Like many others, I have csv file where the first rows (in my case 10 rows) containing text (or are empty), and hence I haven't find a way to figure how to import the file. I have tested other solutions, like (<http://www.mathworks.com/matlabcentral/answers/22710-help-removing-columns-from-csv-file)>, but that didn't solve the problem. When open the csv file in Excel, the beginning looks like this: (It looks very fuzzy, sorry for that).

Model,"WE7000" BlockNumber,"1" TraceName,"CH3","CH4", BlockSize,288154,288154, Date,08/11/25,08/11/25, Time,10:07:43,10:07:43, VUnit,V,V, HResolution,5.000000e-006,5.000000e-006,

block0 -4.999995E+000,-3.230750E+002,-3.685000E+001 -4.999990E+000,-3.230750E+002,-3.685000E+001 -4.999985E+000,3.104250E+002,-3.677500E+002 -4.999980E+000,3.104250E+002,-3.677500E+002 -4.999975E+000,-8.191750E+002,-8.189500E+002

Best Answer

I have extracted the first 12 rows from the original file into a separate one, and attached that. I hope that will give a view how the the original file looks like.

Previous, I have been succesful with using dlmread with other kind of files (i.e dat), where the first rows have been pure text, i.e using this kind of command:

m=dlmread('Test_20130404_0006.dat','/',13,0); or m=dlmread('Test_20130404_0006.dat','',13,0);

But when I´m doing that with my original csv file I get a quite common error message:

Mismatch between file and format string. Trouble reading number from file (row 2u, field 1u) ==>

Related Solutions

MATLAB: Import and format CSV file

"bank format for point(.)," and "converts the period (.) to a comma (,) code to import" &nbsp I don't fully understand. The screen-clip does it show a piece of the csv-file?

Your csv-file uses for

list separator - I guess ','
decimal separator - I guess '.'
thousand separator - ????

Try

    >> out = cssm('cssm.txt')
    out =
       1.0e+05 *
        7.3456    0.0005    0.0005    0.0005    0.0005    0.0206
        7.3459    0.0005    0.0005    0.0005    0.0005    0.0277
        7.3462    0.0005    0.0005    0.0005    0.0005    0.0229
        7.3465    0.0005    0.0005    0.0005    0.0005    0.0526
        7.3468    0.0005    0.0005    0.0005    0.0005    0.0301
        7.3478    0.0005    0.0005    0.0005    0.0005    0.0333

where

    function    out = cssm( filespec ) 
        fid = fopen( filespec );
        cac = textscan( fid, '%s%s%f%f%f%f%f'                   ...
                      , 'Delimiter', ',', 'CollectOutput', true );
        fclose( fid );
        str = cell2mat(cac{1});
        sdn = datenum( str, 'mm/dd/yyyyHH:MM');
        out = cat( 2, sdn, cac{2} );    
    end

and where cssm.txt contains

    03/01/2011,00:00,50.38,51.62,50.38,50.91,2061
    04/01/2011,00:00,51.42,51.42,49.81,50.21,2771
    05/01/2011,00:00,49.60,50.82,49.60,49.81,2287
    06/01/2011,00:00,49.95,50.05,48.35,48.60,5259
    07/01/2011,00:00,48.60,48.91,47.90,48.16,3012
    10/01/2011,00:00,48.47,48.89,47.57,47.63,3333

(thanks to OCR)

MATLAB: How to load read only the two number columns skipping headerlines using dlmread

In versions before R2015a or so (we are not sure exactly which version), it was not possible to use dlmread() for any file that contained text anywhere, even if you were passing parameters that asked for the lines to be skipped.

In newer versions, dlmread() can be used with files that contain text in the header lines that you skip, but not text anywhere else (or possibly text in leading columns you ask to skip might also be okay.)

Your file cannot be handled by dlmread at all, because it contains the text character '«' after numeric columns to be read.

You should be switching to using textscan with 'HeaderLines', 104, and with format '%f%f%*s'

Best Answer

Related Solutions

MATLAB: Import and format CSV file

MATLAB: How to load read only the two number columns skipping headerlines using dlmread

Related Question