import data from csv file

hi,

i would to import data from a file like attached . I have lot file and i would like extrat some parameter in order to plot variation between eatch cast

thanks

Answer 1

[Walter Roberson]

That file is not in csv format. It is also space delimited and uses two lines for each entry because the time has been folded down into the next line. There are fewer headers than there are columns because 'Sbeox0ML/LSbeox0Mm/Kg' is over two columns. You cannot use '/' to mark the end of column headings because you also have headers such as 'C1S/m'. The headers also require two lines.... Oh and I just noticed it isn't just the time that appears on the second line :(

I am working on code to import it.

[Walter Roberson]

fmt = ['%f%s%s%s', repmat('%f', 1, 19), '%s%*[^\n]\n%s', repmat('%f',1,11), '%s%*[^\n]\n'];

projectdir = 'blt-files';   %set as appropriate

dinfo = dir( fullfile(projectdir, '*.csv') );
nfiles = length(dinfo);

for K = nfiles : -1 : 1     %backwards increases efficiency
  thisfile = dinfo(K).name;
  filename = fullfile(projectdir, thisfile);
  [~, basename, ~] = fileparts(thisfile);
  filenames{K} = basename;

    fid = fopen(filename, 'rt');
    datacell = textscan(fid, fmt, 'HeaderLines', 10);
    fclose(fid);

    Bottle_Position{K} = datacell{1};

    %the date/time has to be pieced together from multiple parts over 2 lines
    Date_Time{K} = strcat(datacell{2},{' '}, datacell{3}, {' '}, datacell{4}, {' '}, datacell{25});

    Potemp090C{K} = datacell{5};
    Sal00{K} = datacell{6};
    Sigma_E00{K} = datacell{7};
    Potemp190C{K} = datacell{8};
    Sal11{K} = datacell{9};
    Sigma_E11{K} = datacell{10};
    Sbeox0ML{K} = datacell{11};
    LSbeox0Mm{K} = datacell{12};

    PrDM_avg{K} = datacell{13};
    PrDM_sdev{K} = datacell{26);

    T090C_avg{K} = datacell{14};
    T090C_sdev{K} = datacell{27};

    T190C_avg{K} = datacell{15};
    T190C_sdev{K} = datacell{28};

    C0S_avg{K} = datacell{16};
    C0S_sdev{K} = datacell{29};

    C1S_avg{K} = datacell{17};
    C1S_sdev{K} = datacell{30};

    Sbeox0V_avg{K} = datacell{18};
    Sbeox0V_sdev{K} = datacell{31};

    FlC_avg{K} = datacell{19};
    FlC_sdev{K} = datacell{32};

    CStarTr0_avg{K} = datacell{20};
    CStartTr0_sdev{K} = datacel{33};

    Upoly0_avg{K} = datacell{21};
    Upoly0_sdev{K} = datacell{34};

    Upoly1_avg{K} = datacell{22};
    Upoly1_sdev{K} = datacell{35};

    V1_avg{K} = datacell{23};
    V1_sdev{K} = datacell{36};

    units_avg{K} = datacell{24};    %should be all '(avg)'
    units_sdev{K} = datacell{37};    %should be all '(sdev)'

end

At the end of this, all of the *.csv files in the designated directory will be read in, and the contents placed into cell arrays, one cell entry for each file. The variables are named as closely as practical the same as the headings in the file.

In the file from PrDM to V1 there are two lines of data, one for avg and one for sdev. This has been handled in the code by creating two variables for each of the headings, one with a _avg suffix and the other with a _sdev suffix.

[Matthieu Labaste]

of course it's not a csv format ... it's .btl fil from seabird processing! and the format is really strange ....

[Walter Roberson]

The title of this Question asks about importing from csv files, and the file extension is .csv so everyone reading this question is going to assume that the file really is a Comma Separated Value file, but it isn't.

Answer 2

[Azzi Abdelmalek]

you can use csvread or xlsread

[ii,jj,kk]=xlsread('btl-file.csv')