reading undelimited comma numbers

Apologise if this is an already resolved issue but I just cannot find the solution The lab. equipment spits out a .cti file which I try convert to .xlsx format I say try because I am unable to separate the comma delimited number so the rows looks like this:

-20.925926,119.3976

Any suggestion for an approach to read in the number, I have tried using

 num = xlsread(filename,sheet,xlRange)
num = xlsread(filename,sheet,xlRange,'basic')

none worked - num returns as an empty array

I am thinking I should try to read in as a string extract the number as a string and convert to numerics

Thanks, Philip

[Image Analyst]

Why do you say "undelimited" in the subject line? That row is delimited. What does "undelimited" mean to you?

[per isakson]

Can you upload a sample file (paper-clip-button)? I understand that cti-file is a text file, but that's about all. Matlab Interface User’s Guide doesn't help.

Answer 1

[per isakson]

Here is a file, which I think will extract the numerical data from your file. I tested with some data I found on the net.

Try

    >> [ var_list, num ] = CITIFILE()
    var_list =
       1.0e+09 *
        1.0000
        2.0000
        2.5000
        3.0000
    num = 
        [4x2 double]    [4x2 double]    [4x2 double]

where

    function    [ var_list, num ] = CITIFILE()
       str = fileread( 'CITIFILE.txt' );
       cac = regexp( str, '(?<=VAR_LIST_BEGIN).+?(?=VAR_LIST_END)', 'match' );
       var_list = str2num( cac{1} );
       cac = regexp( str, '(?<=\sBEGIN).+?(?=\sEND)', 'match' );
       num = cell( 1, length(cac) );
       for jj = 1 : length(cac)
          str = strrep( cac{jj}, '"', '' ); 
          num{jj} = str2num( str );
       end
    end

and &nbsp CITIFILE.txt &nbsp is copied from Agilent 8510 3-Term Frequency List Cal Set File and attached

[Philip]

The routine is quite powerful and efficient, thanks.

Regards,

Philip

Answer 2

[Kelly Kearney]

Trying to convert to an Excel format seems unnecessary and complicated... if you're planning to work with this data in Matlab only, I'd suggest skipping that step.

Can you show a few snippets of your file? In plain text. Are the comma-delimited portions of the file all one line, or are there a certain number of columns per row? Are all the data points in quoted pairs, or is that unique to line 16016?

[Philip]

Sure, the data file when open in note book looks like this

line 1 (in excel) to line 12 is text; looks like

CITIFILE A.01.01

"!Agilent Technologies,E8362B,MY43020295,A.06.03.05"

!Agilent E8362B: A.06.03.05

line 13 to line 16013 is data, no trouble reading these rows, looks like

10000000

10374375

10748750

2 lines of text at line 16014 and 5 then line 16016 on wards is shown below:

"-39.117367,-72.549469"

"-45.872181,76.079033"

"-46.924721,170.44943"

the qoutation marks shows up when the file is open using notepad

the first data line is missing using csvread and the first data value in the read is not read in, the read in data looks like

76.0790330000000

0

Thanks for the help this far,

Regards, Philip

[Kelly Kearney]

Okay, first off, this isn't a comma-delimited file... it's a text file with a few comma-delimited sections. So you're going to have to read it as such, parsing each section appropriately. A quick google search of the CITIFILE format suggests that the data portions of the file will be indicated by BEGIN and END keywords; is that the case in your file?