Assign dates from cell array to matrix elements

I have a file I'll call testdoc.csv. It's in the format below:

header
site, date, time, parm1, parm2
1098,2/23/2012,0:00,18.5,20.6
1098,2/23/2012,1:00,18.5,20.6
1098,2/23/2012,2:00,18.5,20.6
1098,2/23/2012,3:00,18.5,20.7

I want to extract the date and time and put it in a six column vector. The "time" in the file is HH:MM. The procedure I have adopted is

d = importdata('testdoc.csv')

d = 

        data: [4x4 double]
    textdata: {6x7 cell}

The textdata cell comes out like so

Columns 1 through 6

    'header'             []         []          []          []    
    'site'      ' date'        ' time'    ' parm1'    ' parm2'    
    '1098'      '2/23/2012'    '0:00'     ''          ''          
    '1098'      '2/23/2012'    '6:00'     ''          ''          
    '1098'      '2/23/2012'    '12:00'     ''          ''          
    '1098'      '2/23/2012'    '18:00'     ''          ''

I want to move the date and time to a matrix like

A = [day month year hours min sec]

so the first two rows in this case would be

2 23 2012 0 0 0
2 23 2012 6 0 0

Is there a way to re-assign the cell array to a matrix? I've tried

A = datevec([d.textdata{3:end,2} ' ' d.textdata{3:end,3}])

which works for a single element, but not for a vector of elements. I really want to avoid loops because of huge files.

Answer 1

[Andrei Bobrov]

d1 = d.textdata(3:end,2:3)
out = datevec(strcat(d1(:,1),d1(:,2)),'mm/dd/yyyyHH:MM');

[John Petersen]

Perfect. THANKYOU!

Answer 2

[Jan]

C = textdata(3:end, 2:3);
S = sprintf('%s ', transpose(C));
D = sscanf(S, '%d/%d/%d %d:%d');
D = transpose(reshape(D, 5, []));
D(:, 6) = 0;

[John Petersen]

>> S=sprintf('%s ',transpose(C));

Error using sprintf

Function is not defined for 'cell' inputs.