How to read specific lines from a text file and store them in a matrix?

Hello everyone,

I want to create a matrix from the attached text file that only includes the lines after the one that reads "Interval: 0.02 sec". Basically, I want a matrix that includes the headers of the columns and all the numeric data from the file, but none of the information preceding it.

Also, would it be possible to use the same code when reading information from a file with a different number or rows and columns?

Thank you!

Answer 1

[Stephen23]

This code should work regardless of the number of rows or columns:

opt = {'Headerlines',3, 'CollectOutput',true};
fid = fopen('Example.txt','rt');
tgt = '';
while ~strncmpi(tgt,'Target',6)
    tgt = fgetl(fid);
end
itm = textscan(fid,'Item%d%s','Delimiter','=');
fmt = repmat('%f',1,1+numel(itm{1}));
dat = textscan(fid,fmt,opt{:});
fclose(fid);
dat = dat{1};

giving:

>> size(dat)
ans =
       12111          17
>> dat
dat =
        12110      -66.174      -69.167      -77.607      -74.388      -66.369       -69.19      -77.367      -73.986      -2.6341    -0.046144      0.34017      -0.7492        -1.97      -2.3501      -2.3501      -2.3501
        12109      -66.174      -69.167      -77.607      -74.388      -66.369       -69.19      -77.367      -73.986      -2.6341    -0.046144      0.34017      -0.7492        -1.97      -2.3501      -2.3501      -2.3501
        12108      -68.547      -72.838       -79.49      -77.035      -68.078      -72.151      -78.954      -75.901      -2.8352     -0.24927      -1.2125      -4.7041        -1.97        -1.97      -2.3501      -2.3501
        12107      -68.547      -72.838       -79.49      -77.035      -68.078      -72.151      -78.954      -75.901      -2.8352     -0.24927      -1.2125      -4.7041        -1.97        -1.97      -2.3501      -2.3501
        12106      -71.372      -75.556      -80.754      -79.204       -70.49      -74.884      -80.715      -78.217      -3.1016      -3.7407      -1.6364      -2.9571        -1.97        -1.97        -1.97        -1.97
        12105      -71.372      -75.556      -80.754      -79.204       -70.49      -74.884      -80.715      -78.217      -3.1016      -3.7407      -1.6364      -2.9571        -1.97        -1.97        -1.97        -1.97
        12104      -73.434      -76.856      -83.268      -80.121      -72.997      -76.536      -82.741      -80.363      -3.6008     -0.69239      -1.8817       -4.322        -1.97        -1.97        -1.97        -1.97
        12103      -73.434      -76.856      -83.268      -80.121      -72.997      -76.536      -82.741      -80.363      -3.6008     -0.69239      -1.8817       -4.322        -1.97        -1.97        -1.97        -1.97
        12102      -75.472      -78.823      -85.593      -82.589      -74.971      -78.008      -84.924      -82.487      0.42387      -1.7753      -2.9945      -1.1647        -1.97        -1.97        -1.97        -1.97
        12101      -75.472      -78.823      -85.593      -82.589      -74.971      -78.008      -84.924      -82.487      0.42387      -1.7753      -2.9945      -1.1647        -1.97        -1.97        -1.97        -1.97
   ... lots more rows here
            6            0            0            0            0            0            0            0            0            0            0            0            0            0            0            0            0
            5            0            0            0            0            0            0            0            0            0            0            0            0            0            0            0            0
            4            0            0            0            0            0            0            0            0            0            0            0            0            0            0            0            0
            3            0            0            0            0            0            0            0            0            0            0            0            0            0            0            0            0
            2            0            0            0            0            0            0            0            0            0            0            0            0            0            0            0            0
            1            0            0            0            0            0            0            0            0            0            0            0            0            0            0            0            0
            0            0            0            0            0            0            0            0            0            0            0            0            0            0            0            0            0

And some of the header information:

>> tgt
tgt =
Target: 0,31
>> itm{1}
ans =
           1
           2
           3
           4
           5
           6
           7
           8
           9
          10
          11
          12
          13
          14
          15
          16
>> itm{2}
ans = 
    '0,31:1.03 SPEED MEASURED 1 [rpm]'
    '0,32:1.03 SPEED MEASURED 1 [rpm]'
    '0,33:1.03 SPEED MEASURED 1 [rpm]'
    '0,34:1.03 SPEED MEASURED 1 [rpm]'
    '0,31:1.04 MOTOR SPEED [rpm]'
    '0,32:1.04 MOTOR SPEED [rpm]'
    '0,33:1.04 MOTOR SPEED [rpm]'
    '0,34:1.04 MOTOR SPEED [rpm]'
    '0,31:1.08 MOTOR TORQUE [%]'
    '0,32:1.08 MOTOR TORQUE [%]'
    '0,33:1.08 MOTOR TORQUE [%]'
    '0,34:1.08 MOTOR TORQUE [%]'
    '0,31:25.04 TORQUE REF B [%]'
    '0,32:25.04 TORQUE REF B [%]'
    '0,33:25.04 TORQUE REF B [%]'
    '0,34:25.04 TORQUE REF B [%]'

[Heidi Mäkitalo]

Thank you so much, it works perfectly!

Another question that I've been wondering about today is as follows, and it's going to be hard for me to even explain exactly what I want. Basically, I always want to plot four consecutive columns in the same figure. In the attached file for example, I want all the different speeds (items 1-4) in one figure, all the motor speeds in another, etc.

However, the data isn't always in the same order as it is in the attached file. Sometimes, for example, the torque may be before the speed in the file. Sometimes not everything is measured - like maybe the motor current isn't relevant in some experiment.

Is there any way to always plot four consecutive columns in the same figure (maybe with a for-loop?) and do this for all the columns? The number of columns is always 4*n + 1 where n is the number of recorded values (and the 1 is result from the leftmost column that records the intervals). Could I also in this loop somehow name the axels, make a legend and give the figure a title, maybe with the help of the header information? I know that this is a lot so it's totally okay if you don't have time to help me.

[Stephen23]

"However, the data isn't always in the same order as it is in the attached file"

Convert the numeric array to a table:

https://www.mathworks.com/help/matlab/ref/array2table.html

then you can access the "variables" by their names (and the order on the files becomes irrelevant).

[Heidi Mäkitalo]

If I convert the array to a table, do the headers/names "follow" through? And if yes, how can I access them? What I'd like to do with them, if possible, is use an if-function to see if a header contains a specific word or words (for ex. motor speed), and if the function comes back as True, plot a figure.

[Stephen23]

"If I convert the array to a table, do the headers/names "follow" through?"

A numeric array does not store any header information (it is simply an array of numbers), so where exactly would this header information be coming from?

Generating a table from an numeric array requires that either you specify the headers yourself, or use whatever ones that the conversion function invents itself (unlikely to be very useful to you).

Note that you could generate the headers from cell array itm's contents (which is why I showed it to you in my answer), you will of course need to replace/remove invalid characters and ensure that each name is unique, etc., according to the documentation on tables (e.g. by reading the link I gave you in my previous comment).

"...use an if-function to see if a header contains a specific word or words (for ex. motor speed), and if the function comes back as True, plot a figure."

Once you have the data in a table T you can get the "variable" names using:

T.Properties.VariableNames

and use any string comparison function, e.g. strcmp.

[Heidi Mäkitalo]

Right, of course, the headers are in the cell array, not the numeric array. Thanks for all the help, really, but I'm still left wondering how exactly I can generate the headers from the cell array itm's contents. Sorry for being slow!

[Stephen23]

"...how exactly I can generate the headers from the cell array itm's contents"

First read the documentation for array2table where it specifies how to specify the table's "variable" names:

https://www.mathworks.com/help/matlab/ref/array2table.html#btx3z8_-1-VariableNames

Then read the specification of valid variable names:

https://www.mathworks.com/help/matlab/matlab_prog/variable-names.html

Are the contents of itm valid variable names? No, the headers contain many invalid characters, and are also not unique. So you will need to remove/replace those invalid characters, and make them unique. You could use any string manipulation tools/functions of your choice, e.g. regexprep, strrep, genvarname, etc., whatever you prefer. Read the documentation and experiment!

[Heidi Mäkitalo]

Okay, I figured it out! Thanks a lot man! :-)