I have a cellArray data. Could you tell me how I can extract the first column of this cellArray data without depending any character and number?

cellArray =

    'A29G0202    465254.432       4066809.025'

    'A29G0008    499252.083        4042876.84'

    'A29G0014    494404.625       4047958.526'

 Name            Size            Bytes  Class     Attributes

 cellArray      22x1              ...    cell

%I want to extract first column of cellArray. (A29G0202, A29G0008, A29G0014)

[Jan]

This is the 7th question about this topic. Please calm down a little bit. Thanks.

Answer 1

[Andrei Bobrov]

cellArray ={'A29G0202    465254.432       4066809.025';
            'A29G0008    499252.083        4042876.84';
            'A29G0014    494404.625       4047958.526'};
a = regexp(cellArray,'^([a-zA-Z]*\d*)*','match');
out = cat(1,a{:});

[sermet]

if first column includes like p.1,p.2,p.3 what we should do?

Answer 2

[Jan]

No, the cell array you show us has one column only, as you can see by the size {22 x 1}. The elements of the cells contain the complete rows as a string and this is obviously a bad idea. So I suggest to improve the method you have used to create this cell array, such that you get something like this directly:

cellArray = { ...
  'A29G0202', '465254.432', '4066809.025'; ...
  'A29G0008', '499252.083',  '4042876.84'; ...
  ...
  }

Then the extraction is trivial:

cellArray(:, 1)

Perhaps something like this helps to create the cell array:

textscan(fid, '%s %s %s');

Answer 3

[dpb]

MATL
>> t=textscan(char(cellArray),'%s %*[^\n]');
>> reshape(char(t{:}),length(cellArray,[])
ans =
A29G0202
A29G0008
A29G0014
>>

Be easier if you would use

'CollectOutput', true

when you read the data to begin with.