How to read some specific columns from xls file with matlab

Hello,

i'm a new user of Matlab, and i would like to know how I can choose some specific columns

For example :

                B          C       D         E
              a480.1    *c450*    a488    *c555*
                0.5      *0.8*     0.6     *0.84*

In this case i need to read C and E.

Thank you very much

H

Answer 1

[Azzi Abdelmalek]

data1=xlsread('file.xlsx','C:C');
data2=xlsread('file.xlsx','E:E');
data=[data1 data2]

[Sabrina]

Thanks Azzi, the problem is that I ve tried this method, but it took me a lot of time, that's why I'm trying to find the simple way

Answer 2

[dpb]

Unless there are a very large number of columns being ignored, far simpler to just read the whole spreadsheet then eliminate the columns not needed.

In this case since there's a mess of alphanumeric and numeric data, probably best is to use the RAW cell array result directly --

[~,~,dat]=xlsread('filename.xls');
dat=[dat(:,2) dat(:,4)];

[Sabrina]

Exactly in my case I have 172 columns thank you I will try it

[Image Analyst]

I think this is the best approach unless you want to use ActiveX, which will be more complicated, but possibly faster if you have a large worksheet that you need only a small amount of data from (since ActiveX won't read unneeded cells like xlsread() does).

[dpb]

If the columns you want are close to each other you can try to minimize the extra input by defining the block to enclose the columns you wish as first and last. I'd guess w/o trying it that the extra overhead in xlsread of handling the block will be as much or more as that in reading the full file.

Certainly the option of reading the file multiple times is surely going to be the slowest option I'd think without question.

So, "it depends" I suspect, altho I still think the fastest will likely be the read it all into memory, then operate on that.

[Sabrina]

waw it worked, thank you so much :-) dpb !! of course I have to put all the colmuns I need but it's more easy and helpful than my first script, which is too long

thanks a lot

 H