TEXTSCAN - .csv - mutliple columns to a vector/double/matrix

--

data.csv

t1    1     5     [4 5 8]    [3 4 5]

--

How can I use textscan to import the [4 5 8] and the [3 4 5] so that they are converted to 1x3 doubles? As though it were entered at the prompt as:

a = [4 5 8]

fid = fopen('data.csv')
data = textscan(fid,...
     '%s %f %f [%f %*f %f] [%f %*f %f]')

Note that I'm trying to skip the middle values (5 and 4) using the *.

Doesn't seem to work! I've tried a number of variations on this and I always get weird results.

Answer 1

[Oleg Komarov]

I don't get strange results but expected behaviour of textscan.

cell2mat(data(:,2:end))

or when importing:

data = textscan(fid, '%s%f%f[%f%*f%f][%f%*f%f]','CollectOutput',1);
data = data{2};