No BSD License  

Highlights from
EZread

from EZread by Paul Taylor
Easily read csv files containing both text and numeric columns

ezread(fname,varargin)
function out_struct = ezread(fname,varargin)
% Read a CSV file that contains both text and numeric columns, without
% having to manually specify the column type using %f, %s, etc.
%
%   Inputs:
%      1.  Filename
%      2.  (optional) file delimiter (e.g. ',')
%
%   Outputs
%      1.  A structure containing the data, one field per column
%
% This function expects the file to have one header row containing column 
% headers, and outputs a structure with one field per column in the file.  
% The fieldnames in the structure match the column headers.
%
% Paul Taylor
% The MathWorks Australia Pty Ltd
% 9/05/2006

% Set the file delimiter
if nargin == 2
    file_delim = varargin{1};
else
    file_delim = ',';
end

out_struct = [];

% Open the file
fid = fopen(fname,'r');

if fid == -1
    % could not find the file
    error('File not found')
end

% Read header line
header_str = fgetl(fid);

% Parse the header string into separate headers
header_cell = textscan(header_str,'%q','delimiter',file_delim);
header_cell = header_cell{1};

% Read the first line of data to determine the column data types
data = textscan(fid,'%s',length(header_cell),'delimiter',file_delim);
data = data{1};

% Check the contents of each column and contruct a column format specifier
% string
format_str = [];
for i = 1:length(data)
    % If str2num returns a numeric value, then the column in numeric,
    % otherwise if str2num returns empty, then the column is text
    if ~isnan(str2double(data{i}))
        col_format = '%f ';
    else
        col_format = '%q ';
    end
    format_str = [format_str col_format]; %#ok<AGROW>
end

% Move back to the start of the file
frewind(fid)

% Read the entire file using the new column format specifier string
data = textscan(fid,format_str,'delimiter',file_delim,'headerlines',1);

% Close the file
fclose(fid);

% Loop through each column of data and create a field in a structure for
% each column of data.  The fieldnames are based on the headers in the file
for i = 1:length(data)
    % Get the current header
    cur_hdr = header_cell{i};
    
    % Remove/replace standard non-compatibile parts of headers
    cur_hdr = strrep(cur_hdr,'"','');
    cur_hdr = strrep(cur_hdr,',','_');
    cur_hdr = strrep(cur_hdr,'#','_no');
    
    % Use the "genvarname" function to modify the header to handles any 
    % other parts of the header that would make the header an invalid
    % variable name
    cur_hdr = genvarname(cur_hdr);
    out_struct.(cur_hdr) = data{i};
end

Contact us at files@mathworks.com