Read csv file into two ouput variables - text & numbers

I have inherited a script that reads in a csv file, where the first column is text and the other columns are numbers. For example the data is similar to this:

hello          1       2       3        4
bye            5       6       7        8

and so forth...

my script includes the following code:

[summary,text] = xlsread(filename)

I understand from an answer related to another question I posted that it is not possible to use 'xlsread' as I am using matlab via linux (the person before me was using Windows) and csv is unsupported when using xlsread. Also, when I attempt to use 'csvread', it does not have the option to create two output variables (summary & text)

Does anyone know another method of loading in a csv file in a similar fashion to the code presented here?

[Michael]

Update

I have tried using csvread to load in each part separately like so:

numb = csvread(file,1:2,2:5); text = csvread(file,1:2,1);

However, I get the following message:

Error using dlmread (line 139)
Header lines must be integer-valued.

Any ideas? is there something I am missing?

Answer 1

[Guillaume]

Assuming you're on a 2013b or later, the simplest way to import this sort of file is with readtable. In your case:

t = readtable(filename, 'ReadVariableNames', false, 'ReadRowNames', true);

[Michael]

I am using 2012b unfortunately, although I may be able to change versions. Is there an equivalent function for 2012b?

[Michael]

nope...unfortunately the latest version of matlab I can use is 2013a

[Guillaume]

Then you will have to use textscan:

fid = fopen(filename, 'rt');
raw = textscan(fid, '%s %f %f %f %f'); %or whatever formatting your file is
fclose(fid);
rownames = raw(:, 1);
data = cell2mat(raw(:, 2:end));

Answer 2

[Stephen23]

Sadly csvread will only handle numeric data, and as your first column contains strings you will have to use another function, such as textscan . This code gives you an exact replacement of the xlsread functionality (for this usage case, not in general):

fid = fopen(filename,'rt');
C = textscan(fid,'%s%f%f%f%f', 'MultipleDelimsAsOne',true, 'CollectOutput',true);
fclose(fid);
text = C{1};
summary = C{2};

You should have a read through the textscan documentation, as there are many options that you might find useful.