MATLAB: Is textscan ever better than readtable

data importMATLABreadtabletextscan

I have some legacy code that reads in data from a text file. The author seems to avoid using readtable() in favour of using textscan() to get a cell array of strings and then converting the strings to the correct format afterwards. This seems like an awkward way of doing things, and takes a long time for big files so my questions are:

Is there any obvious reason to do this? Is textscan somehow more flexible/robust than readtable?
Is readtable optimised for reading data in a specified format? (i.e. faster than reading a string and converting)

Best Answer

readtable internally calls textscan, but does a lot of work before to automatically detect the format of the file and after to split the data into variables of the correct type. So, a properly designed call to textscan and direct conversion to a table is always going to be faster than going through readtable.

What little you may lose in speed (file i/o is probably dominant anyway, so processing speed may not be critical), you make up for it by the flexibility of readtable. readtable is simply textscan on steroid (and it gets better with each release) so unless it is demonstrably slower I would always use it.

Note that early readtable wasn't as good at the autodetection. As said it gradually improved with each release since R2013b. The introduction of the import options in R2016b really makes it powerful.

Related Solutions

MATLAB: How to read and create column vectors from this txt file (attached)

A fixedWidthImportOptions object can help here altho the heading row doesn't quite line up to make it as trivial as could otherwise be...

opt=fixedWidthImportOptions('NumVariables',10);                 % create base import object w 10 variables
opt.VariableWidths=[8 10*ones(1,9)];                            % set field width to match file
opt.DataLines=[4 inf];                                          % data starts row 4
tout=readtable('output_nostri.txt',opt);                        % import data; will be cellstr arrays
for i=2:10                                                      % convert the numeric data
  tout.(tout.Properties.VariableNames{i})=str2double(strrep(tout{:,i},',','.'));
end
tout.Var1=datetime(tout.Var1,'InputFormat','MMM-yyyy');         % and the time

I presume local settings will handle the month name abbreviations; the above fails on all that aren't english translations here; I would presume that comes from a locale setting for other places but I've never had the opportunity to test to know if that is so or not.

Also leaves variable names Var1 thru Var10; the column headings in line two don't exactly align with the data columns so to read it would have to do a second read of that line...

MATLAB: Readtable defaulting to wrong date format

Look at the "Read Foreign-Language Dates from Text File" example in the documentation for the readtable function. When you don't tell readtable the format of your data, it tries to determine the format on its own. If it doesn't quite get it right, you can use the 'Format' option to tell it exactly how to read the data.

That particular example uses the "%{fmt}D" specifier as shown in the "Nonnumeric fields" section of the documentation of the formatSpec input argument to the textscan function.

Best Answer

Related Solutions

MATLAB: How to read and create column vectors from this txt file (attached)

MATLAB: Readtable defaulting to wrong date format

Related Question