Cannot import csv data with textscan

Hi, I'm trying to import several csv files to use the data. The problem is that xlsread is too slow and csvread, dlmread or importdata don't work. I suspect the encoding to be the source of the problem. So I'm trying to use textscan to do it.

My csv file is looking like that:

N°.  x[mm]  y[mm]  Rotation / x[°]
1  30.92221  3.66636  
2  30.92751  3.66634  -0.259299
3  30.93981  3.66649  0.696742
4  30.94510  3.66680  3.36023
5  30.95740  3.66706  1.24399
6  30.96970  3.66663  -2.04397
7  30.97500  3.66638  -2.66614
8  30.98029  3.66614  -2.61545
9  30.99259  3.66589  -1.17048
10  30.99789  3.66629  4.33688

1  31.10346  3.66539  -1.00677
2  30.99259  3.66589  -1.17048

So here i'm using the following code which I think is ok :

file = fopen('test.csv','rt');
data2 = textscan(file,'%f%f%f%f','Delimiter','\t','HeaderLines',1,'TreatAsEmpty',{'N°.','x[mm]','y[mm]','Rotation / x[°]'});

But it's returning

data2 =

    [0x1 double]    [0x1 double]    [0x1 double]    [0x1 double]

When I suppress the first line, it works. It appears that the parameter 'headerlines' doesn't work...

Any idea ?

Thank you !!

PS: I work with matlab R2009a

EDIT: The ° symbol appears to be the problem, does anyone know how to suppress it directly from matlab... ? Or any solution to bypass that??

Answer 1

[KSSV]

use xlsread

[num,txt,raw] = xlsread('test.csv') ;

[Thomas Hermelin]

Yes, I forgot to mention it, I'm already using it, but it takes 8 secs to import the data vs 0.1 with textscan. That's why I'm trying to use it. Thanks for your answer !