How to read arrays from files?

Hi. I inherited some code that I want to compile into an executable. It has a run() statement that brings in some data, and the compiler i guess doesn't want to compile run().

run(file.m);

File contents:

Headers = {...
    'a'  'b'  'c'  'd'  'e'  'f'  'g'  'h'  'i'  'j'  'k'  'l'  'm'  'n'  'o'  'p'  'q'  'r'  's'  't'  'u'  'v'  'w'  'x'  'y'  'z'  'aa'
    };

TowerData = {...
    'TwrBaseFX'  'TwrBaseFY'  'TwrBaseFZ'  'TwrBaseFR'  'TwrBaseMX'  'TwrBaseMY'  'TwrBaseMZ'  'TwrBaseMR'  1  0.745  8.475  4.300  0.022  21.925  0.600  25.000  1.000  0.200  1.000  0.650  0.400  0.600  0  0  0  0  0
    'TwrBaseFX'  'TwrBaseFY'  'TwrBaseFZ'  'TwrBaseFR'  'TwrBaseMX'  'TwrBaseMY'  'TwrBaseMZ'  'TwrBaseMR'  1  0.745  8.475  4.300  0.022  21.925  0.600  25.000  1.000  0.200  1.000  0.650  0.400  0.600  1  0.85  2.26  0.3  0.03
    'TwrS02FX'  'TwrS02FY'  'TwrS02FZ'  'TwrS02FR'  'TwrS02MX'  'TwrS02MY'  'TwrS02MZ'  'TwrS02MR'  1  8.475  16.205  4.300  0.018  21.925  0.600  25.000  1.000  0.200  1.000  0.650  0.400  0.600  0  0  0  0  0
    'TwrS03FX'  'TwrS03FY'  'TwrS03FZ'  'TwrS03FR'  'TwrS03MX'  'TwrS03MY'  'TwrS03MZ'  'TwrS03MR'  1  16.205  23.935  4.300  0.016  21.925  0.600  25.000  1.000  0.200  1.000  0.650  0.400  0.600  0  0  0  0  0
    'TwrS04FX'  'TwrS04FY'  'TwrS04FZ'  'TwrS04FR'  'TwrS04MX'  'TwrS04MY'  'TwrS04MZ'  'TwrS04MR'  2  23.935  31.665  4.122  0.015  25.912  0.600  25.000  1.000  0.200  1.000  0.650  0.400  0.600  0  0  0  0  0
    'TwrS05FX'  'TwrS05FY'  'TwrS05FZ'  'TwrS05FR'  'TwrS05MX'  'TwrS05MY'  'TwrS05MZ'  'TwrS05MR'  2  31.665  39.395  3.853  0.013  25.912  0.600  25.000  1.000  0.200  1.000  0.650  0.400  0.600  0  0  0  0  0
    'TwrS06FX'  'TwrS06FY'  'TwrS06FZ'  'TwrS06FR'  'TwrS06MX'  'TwrS06MY'  'TwrS06MZ'  'TwrS06MR'  2  39.395  47.125  3.585  0.012  25.912  0.600  25.000  1.000  0.200  1.000  0.650  0.400  0.600  0  0  0  0  0
    'TwrS07FX'  'TwrS07FY'  'TwrS07FZ'  'TwrS07FR'  'TwrS07MX'  'TwrS07MY'  'TwrS07MZ'  'TwrS07MR'  3  47.125  54.855  3.332  0.012  29.463  0.600  25.000  1.000  0.200  1.000  0.650  0.400  0.600  0  0  0  0  0
    'TwrS08FX'  'TwrS08FY'  'TwrS08FZ'  'TwrS08FR'  'TwrS08MX'  'TwrS08MY'  'TwrS08MZ'  'TwrS08MR'  3  54.855  62.585  3.112  0.011  29.463  0.600  25.000  1.000  0.200  1.000  0.650  0.400  0.600  0  0  0  0  0
    'TwrS09FX'  'TwrS09FY'  'TwrS09FZ'  'TwrS09FR'  'TwrS09MX'  'TwrS09MY'  'TwrS09MZ'  'TwrS09MR'  3  62.585  70.315  2.893  0.011  29.463  0.600  25.000  1.000  0.200  1.000  0.650  0.400  0.600  0  0  0  0  0
    'TwrTopFX'  'TwrTopFY'  'TwrTopFZ'  'TwrTopFR'  'TwrTopMX'  'TwrTopMY'  'TwrTopMZ'  'TwrTopMR'  3  70.315  78.045  2.674  0.011  29.463  0.600  25.000  1.000  0.200  1.000  0.650  0.400  0.600  0  0  0  0  0
    };

So I just want to replace the run() with a similar command to read these 2 arrays from a text file that loads Headers and TowerData the same as the Run() did.

I cannot put the data into the code to compile, because it can vary in length and content. I was playing with fscanf, but it seemed to mess up on the strings.

[Orion]

What is the error message that you get ?

Can you run this mfile alone ?

[David Pesetsky]

Yes the .m works on its own.

If i compile it, and run the exe, the RUN command no longer recognizes the .m script it's supposed to run as a valid script. Probably because no Matlab license is pulled when running a compiled executable, and RUN likely needs a licence.

[Orion]

Instead of a call to the run function, you could jsut call directly the mfile. You'll have the same result.

This should work with Matlab Compiler.

or another way : assuming your mfile is called Mydata.m

myMfile = @Mydata
feval(myMfile)

[David Pesetsky]

My syntax must be wrong. Trying not to use addpath, rather specify path...

TwrGeoFullFileName = 'D:\Users\PESETSDA\Documents\Documents\Systems\ComponentEnvelopes\FLEX5_and_Components\save\TestTwr.m';
myMfile = @TwrGeoFullFileName;
feval(myMfile);

Gives:

Error: File: ReactionsToStress_v05.m Line: 47 Column: 12 "TwrGeoFullFileName" was previously used as a variable, conflicting with its use here as the name of a function or command. See "How MATLAB Recognizes Command Syntax" in the MATLAB documentation for details.

[Walter Roberson]

You cannot take the handle of a script

[David Pesetsky]

What is the right way to execute the script, using a method the works after I compile into an executable?

[Walter Roberson]

There is no method in MATLAB Compiler to dynamically execute a script whose content the user may change. You need to parse anything like that as a data file and build the appropriate data structures yourself from the data.

[David Pesetsky]

Yes, that was what I was trying to do when I strated this thread. fscanf or textscan. The data is exactly above, it just loads those 2 arrays.

[Walter Roberson]

Does it need to be in that exact code-compatible format? A csv or tabbed columns would be easier to process.

[John BG]

would modifying the script and then use command evalin help?

[Walter Roberson]

evalin() cannot be used in compiled code to read outside code.

Remember, this question has to do with reading something user-alterable that is structured like code, at run-time, within a compiled application. The MATLAB Compiler is designed so that the parser is not present, so any call to eval() or evalin() or evalc() or run() is going to fail. All code must be known to MATLAB Compiler at the time of compiling. Anything that must be loaded at run-time according to user changes must be treated as a data file that has to be analyzed by code.

[David Pesetsky]

The original data above can be put in similar formats...it's basically a table with a header and labels on each row. So a mix of string and text. What's a good format, and method to read, such that I get the same 2 resulting arrays?

Answer 1

[Walter Roberson]

I suggest you use readtable() on a csv if you have R2013b or later.