Reading character and numeric data

Hi,

I have the following dataset:

'9 b 4 13'

'b4 c0 93 94'

'10 4 8 b'

'8 10 8 d'

'18 17 e 17'

'5 9 4 9'

'11 9 9 16'

I need to specify the first two data entries (seperated by space) in each row, which can be a combination of numeric and characters, with varying size lengths into seperare variables. So for example:

row1=['9';'b']

row2=['b4';'c0']

row3=['10';'4']

I have tried many different avenues with limited success. Can someone please advise?

Answer 1

[Walter Roberson]

sscanf() cannot be used to do this in a single call. The %s format element can be used to a string, and %*s to skip a string; however, if you have multiple strings read in the same call then sscanf() will concatenate them together in the output.

You are better off using

textscan(OneRow, '%s%s%*s%*s')

where OneRow is a row of your input.

Neither textscan nor sscanf can accept cell arrays of strings; if that is what you have then you can do them all at one time by using

regexp(CellArray, '(\S+)\s+(\S+).*', 'match')

This would return a cell array, each entry of which was a cell array per line, with (if possible) the two matched strings, with the rest of each line ignored.

A version simpler to code the pattern for is

regexp(CellArray, '\s+', 'split')

This would also return a cell array, each entry of which was a cell array per line, with as many strings as there were per input line; you would then code to ignore the entries you did not need.