Extract specific rows from a table, row names are dependent on user input.

Hello Everyone,

I have a function with two Inputs: table name and a string. The string usually contains information that would be used later in the function to locate spcfic values. Say you have a Table 5X5 and the rows are a,b,c,d,e and you want to find out which values in row c are smaller than 5, the input would be something like ( assuming function name is 'loc' and table name is 'cases'):

loc(cases,"cases.c<5")

The function returns a table that contains all the colmuns containing c values smaller than 5. However the table as a whole is still there, meaning that all the rows are still there, only the colmuns got filtered, and that's where the issue is: I need to automatically plot the values that I got, but the table itself has unecessary rows. So my Idea is to extract the rows/colmuns from the table that I already have and then use the values from a new table that basicly contains only the necessary values that i need for the plot.

Now I thought a good Idea for the new table would be something like this:

idx=cases.c<5;
Table_New=(idx,:)

but with this code I have two Problems: first of all I can't use the input value (cases.c<5) as an index. Even if I entred the idx line in the code, It's not working and I get the message 'Row Index exceeds table dimensions'

If you have an idea on how this can be realised or if you can help it would be much appreciated.

Answer 1

[Voss]

It's not clear exactly what you want to do, but you can refer to the following:

% create a 5x5 table with random integers between 1 and 10:
n = 5;
m = randi(10,n);
t = array2table(m);
t.Row = sprintfc('%c','a'+(0:n-1));
disp(t);

         m1    m2    m3    m4    m5
         __    __    __    __    __

    a    1     10     4    6     7 
    b    7      7     8    9     6 
    c    4      5    10    7     2 
    d    4      4     7    1     4 
    e    3      1     8    6     2 

% make a new table containing just the columns of t where the value in row
% c is less than 5:
new_t = t(:,t{strcmp(t.Row,'c'),:} < 5);
disp(new_t);

         m1    m5
         __    __

    a    1     7 
    b    7     6 
    c    4     2 
    d    4     4 
    e    3     2 

% make a new table containing just row c at the columns of t where the value in row
% c is less than 5:
idx = strcmp(t.Row,'c');
new_t = t(idx,t{idx,:} < 5);
disp(new_t);

         m1    m5
         __    __

    c    4     2 

[Peter Perkins]

Nicely done! Exactly what braces are for.