how to read columns with different length correctly in txt file?

I have a simple array saved in test.txt file:

[1 2 3; 
 4 5 6;
   2  ]

I used importdata:

>> x=importdata('test.txt')

x =

       1     2     3
       4     5     6
       2   NaN   NaN

I also tried dlmread:

>> x=dlmread('test.txt')

x =

       1     2     3
       4     5     6
       2     0     0

>>

Both are not what I want. I want 2nd column to keep 3 elements.

What's the best way to do it?

Thanks,

Jean

[Jean]

The array is

[1 2 3;

4 5 6;

2 ]

2 in the last row is in the middle column.

[Oleg Komarov]

If there's no delimiter before the 2 in the third row I'm afraid it's not possible.

[Walter Roberson]

It is possible, but only if the columns are fixed width.

[Jean]

Thanks, Walter. If the columns are fixed width, how will you do it?

Answer 1

[Walter Roberson]

For fixed width content:

fopen() the file.

Loop:

fgetl() a line. If ~ischar() the line, break out of the loop. Pad the line with blanks to a multiple of the field width (if need be). Call that Lin. Then,

L_by_field = reshape(Lin, FieldWidth, []) .';
blankfield = all(L_by_field == ' ',2);
L_by_field(blankfield,end-2:end) = 'NaN';
fieldvalues = str2double(cellstr(L_by_field));

Store the values, then let the loop continue.

When the loop finishes, fclose() the file.

[Jean]

Walter, thanks. It works on the simple example I made, but doesn't work if I have data with different length. It gotta have a better way. Thanks anyway.

Jean