Improving the compactness of code

Hi all,

I am working on four string arrays on which I performed some structuring. I am new to Matlab so I am very happy that the code finally works. However, I would like to improve the code by using loops or statements to improve the compactness. For each of the four data sets the last three operations are performed on all arrays, therefore I would like to use a structure that can reduce the number of lines and perform a operation that simplifies to one operation.

I tried using a for-loop but this did not work out since I didnt know how to state the i/j/k for the different data sets. Could anyone explain me how to approach such a problem?

In the code below you can see the operations that are performed on each dataset:

% Importing and sorting population data
pop = tsv2cell('Population.tsv');
pop = sortrows(pop(2:end,:),1);
pop(:,[2:4,11:end]) = [];                         % Delete column 2,3,4
pop = erase(pop,[" e", " b"]);                    % Delete non-numerical attributes   
pop = standardizeMissing(pop,':');                % Find invalid data    
pop = rmmissing(pop);                             % Remove rows with invalid data

% Importing and sorting GDP data
gdp = tsv2cell('GDP.tsv');
gdp = sortrows(gdp(2:end,:),1); 
gdp(:,[2,3,4,11:end]) = [];                            
gdp = erase(gdp,[" e", " b"]); 
gdp = standardizeMissing(gdp,':');
gdp = rmmissing(gdp);
                             
% Importing and sorting passenger data
fp = tsv2cell('Flightpassengers.tsv');
fp(38:end,:) = [];                                     
fp(:,[2:5]) = [];
fp = erase(fp,[" e", " b"]); 
fp = sortrows(fp(2:end,:),1);
fp = standardizeMissing(fp,':');
fp = rmmissing(fp);


%Importing and sorting QoL data
lr = tsv2cell('LifeRating.tsv');
lr(3:36,8) = lr(38:end,7);                             
lr(37:end,:) = [];
lr(:,[2:6,9:end]) = [];
lr = sortrows(lr(2:end,:),1);
lr = erase(lr,[" e", " b"]); 
lr = standardizeMissing(lr,':');
lr = rmmissing(lr);

Answer 1

[Turlough Hughes]

You could put them in a structure as follows:

s.pop=pop; s.gdp=gdp; s.fp=fp; s.lr=lr; 

and then to run the loop you just iterate through each field in the structure, s:

f=fieldnames(s);
for c=1:length(f)
    s.(f{c})=erase(s.(f{c}),[" e", " b"]); 
    s.(f{c})=standardizeMissing(s.(f{c}),':');
    s.(f{c})=rmmissing(s.(f{c}));
end

This should be fine but i'm not certain as I don't have your data, if it doesn't could you attach the datasets in the original question.

[Wouter Wizard]

Thanks for replying Turlough!

I tried to implement the code, but the result should be 27 rows while the output is 34 rows. I add the code in this document

[Turlough Hughes]

Can you upload the .tsv files?

[Wouter Wizard]

I have placed them in a ZIP since tsv is not supported. Thanks for your help!