Use find to filter for results specified by user
1 view (last 30 days)
Show older comments
Petros Katsoulis
on 8 Nov 2019
Commented: Petros Katsoulis
on 14 Nov 2019
I have a table with some data:
clothes = table;
clothes.Type = ["Suit";"Shirt";"Dress"];
clothes.Colour = ["Blue";"Red";"White"];
clothes.Price = [300;50;400];
I want to write a filter that allows a user to define what they are interested in in each category and return the result. A simple way to do this would be:
% User defined input
filters = struct;
filters.Type = "Shirt";
filters.Colour = "Red";
filters.Price = 50;
% Generate result
f = find(strcmp(clothes.Type,filters.Type) ...
& strcmp(clothes.Colour,filters.Colour) ...
& clothes.Price == filters.Price);
output = clothes(f,:);
Suppose the user doesn't want to filter for the price and wants to see all results relevant to Red Shirts. Is there a way to write the code in such a way that when the user doesn't set any value to the filter (e.g. filters.Price = []), the find function will return all table entries with Red Shirts irrespectively of price? I thought about checking if it is empty then set the filter to contain all unique values of the relevant column of the table, but the real dataset is quite large and it's not very efficient.
Thank you.
0 Comments
Accepted Answer
Jan Studnicka
on 8 Nov 2019
Hi,
strcmp is slower then == and you can use == with strings. You can also save space and get better performance by using categorical arrays instead of strings when it makes sence.
3 Comments
Jan Studnicka
on 11 Nov 2019
I would do it this way:
function st = tfilters(t,userinput)
vn = t.Properties.VariableNames;
idxV = ones(size(t));
for i = 1:length(vn)
if ismember(vn{i},fields(userinput))
if isempty(userinput.(vn{i})) == false
idxV(:,i) = t{:,vn{i}} == userinput.(vn{i});
end
end
end
idx = all(idxV,2);
st = t(idx,:);
However, instead of structures I would suggest using OOP.
More Answers (0)
See Also
Categories
Find more on Matrices and Arrays in Help Center and File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!