How to use correctly a string from a variable into a for loop??

Hi all,

Firstly thank you for your time!

I'm working in a matlab script for data analysis and plotting from a csv file. I'm trying to use a variable name into a for loop to get the data on a specific table column but I can't get it right...

A simplified version of my code is:

CSV_Import=readtable('Data.csv');
Ch_sLap ='distancelap';
Channels={'Ch_sLap'}
sLap = CSV_Import.(Ch_sLap); 
for i=1:length(Channels)
    if exist(char(Channels(i)))
       char(Channels(i)) = CSV_Import.((fprintf('%s\n', Channels{i})));
    end
end

What I want to do is:

• Import a csv file with some data arranged in columns with headers into a table
• Find the column which is named "distancelap" on that table
• That column header (distancelap) is coming from an array of strings that I've created previously
• Then, if I find the column using this: sLap = CSV_Import.(Ch_sLap); is working fine
• But instead of that, if I use the for loop (In case I had a long Channels array) I can't make it to work because it takes "Ch_sLap" instead of its value which is "distancelap"

I can do it without the loop and just put one by one all the variables of the Channels array but I don't like it at all...

What do you think???

Thank you!!

[Walter Roberson]

http://www.mathworks.com/matlabcentral/answers/304528-tutorial-why-variables-should-not-be-named-dynamically-eval

[Jonathan Fernandez]

Great Walter than you for the help.

I knew that eval shouldn´t be used, but not that we shouldn´t create dynamic variables.

Then I'll use array indexing or struct (I like the struc more because I can put headers) but then, the question still is the same, how can I access that data??

I mean, if I have everything in columns, I still have that column named "distancelap" which is a string inside Ch_sLap. How can I go straight to the 'distancelap' column inside the array using a for loop?? (I will then put it in a structure).

Thank you!

[Walter Roberson]

You do not have an array, you have a table. And those table variables do not have newlines in their names.

Channels_struct = table2struct(CSV_Import(:,Channels));

This does not do checking to be sure everything in Channels is in CSV_Import; you would get an error if there were a problem.

Or if you prefer for some reason:

Channels_struct = struct();
known_channels = find( ismember(Channels, CSV_Import.Properties.Variablenames) );
for K = known_channels
    Channels_struct.(Channels{K}) = CSV_Import.(Channels{K});
end

This filters down to only the Channels that are in the table; it also illustrates copying one at a time, which might be of interest to you for other reasons.

[Jonathan Fernandez]

Great Walter, thank you!!

[Jonathan Fernandez]

I've been checking what you proposed Walter and it still gives me the same problem. I may not explained myself correctly:

In the Channels table one of the column is named "distancelap".

That name ("distancelap") is inside the Ch_sLap variable, I mean:

Ch_sLap = 'distancelap'

What I want is to find the 'distancelap' column in the table from a command. If I use this:

Channels_struct = table2struct(CSV_Import(:,Channels));

it tries to find Ch_sLap in the table, but what I want to find is 'distancelap' (The string placed inside Ch_sLap).

I hope not to be a pain...

Thank you!!

Answer 1

[Rajani Mishra]

By writing below mentioned code you are creating a string variable with value as 'Ch_sLap' and storing in cell array Channels. Hence, it takes column name as 'Ch_sLap'.

Channels={'Ch_sLap'}

I created a table mentioned in the first example on this page and tried your code. I changed "Channels" array into the following way (variable 'Ch_sLap' has value 'Age') :

Channels={Ch_sLap}
for i=1:length(Channels)
    myTable.(Channels{i})
end

Above code was able to print values present in 'Age' column.