Outputting cell vectors that contain specific information help?

I am trying to formulate a code that takes in input cell array of strings, evaluates the contents of each cell and outputs the cells that contain only uppercase letters

i.e.

cell =

  3×2 cell array

    {'ABC'}    {'Abc'}
    {'DEF'}    {'dEf'}
    {'GHI'}    {'ghi'}

ans = 
    ABC 
    DEF
    GHI

Here's my code so far:

function uppercase = all_caps(cell)
[rows, cols] = size(cell);
for ii = 1:rows
    for jj = 1:cols
            if contains(cell{ii, jj}, (['a', 'b','c','d','e','f','g','h','i','j','k','l','m','n','o','p','q','r','s','t','u','v','w','x','y','z']) == 0
                uppercase = [uppercase, cell{ii,jj}];
            end
        end
    end
end
fprintf('%s\n', uppercase)
end

How can I use MATLABS assortment of built-in functions to prevent me from typing out each individual letter? Additionally, this method doesn't even work.

My professor showed us this solution:

function all_caps(c_in)    
[rows, cols] = size(c_in);            
    for ii = 1:rows        
        for jj = 1:cols                        
            str = c_in{ii, jj};             
            if sum(str >= 'A' & str <= 'Z') == length(str)                
                fprintf('%s\n', str);           
            end        
        end     
    end 
end 

I am confused about his if statement. I have no clue what the inside is going on inside the sum() function. Please help.

I know these are "two separate questions," but they are about the same function and regard the same topic.

Answer 1

[Tommy]

You can use the colon operator to generate a character vector containing each lower case letter in the alphabet:

>> 'a':'z'
ans =
    'abcdefghijklmnopqrstuvwxyz'

Same with upper case letters:

>> 'A':'Z'
ans =
    'ABCDEFGHIJKLMNOPQRSTUVWXYZ'

Or this, if you want:

>> 'A':'z'
ans =
    'ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz'

See here for more info. You can use double() and char() to switch between number and corresponding character:

>> double('a')
ans =
    97
>> char(97)
ans =
    'a'

So, 'a':'z' is kind of like

>> char(double('a'):double('z'))
ans =
    'abcdefghijklmnopqrstuvwxyz'

When you compare characters,

>> 'a' < 'z'
ans =
  logical
   1

this is like

>> double('a') < double('z')
ans =
  logical
   1

or rather

>> 97 < 122
ans =
  logical
   1

In your professor's code:

str >= 'A' & str <= 'Z'

Here, each character within str is being compared to 'A' and 'Z'. The output is a logical array with the same size as str, containing a 1 where the corresponding character of str is between 'A' and 'Z' (i.e. somewhere within the array 'A':'Z').

>> 'GHSIEVBSK' >= 'A' & 'GHSIEVBSK' <= 'Z'
ans =
  1×9 logical array
   1   1   1   1   1   1   1   1   1

Here, str has all capital letters, so every element in the output is true.

>> 'GhSIeVBsK' >= 'A' & 'GhSIeVBsK' <= 'Z'
ans =
  1×9 logical array
   1   0   1   1   0   1   1   0   1

Here, some characters in str are lower-case. The corresponding elements in the output are false. With

if sum(str >= 'A' & str <= 'Z') == length(str)

your professor is checking that every element in the output is a 1. Also valid would be

if all(str >= 'A' & str <= 'Z')

Take another look at the documentation for contains.

>> contains('abc', 'a':'z')
ans =
  logical
   0

This returns 0 because the entirety of 'a':'z' is not found within 'abc'. To check if any element of 'a':'z' is within 'abc', you could use for example

>> contains('abc', cellstr(('a':'z')')')
ans =
  logical
   1

[James Metz]

Wow, this was super helpful! Thank you so much!

[James Metz]

One question though. What is the point of the additional apostrophies in the cell string?

contains('abc', cellstr(('a':'z')')')

^ ^

[Tommy]

Happy to help!

The first apostrophe is the difference between

>> cellstr(('a':'z')')'
ans =
  1×26 cell array
  Columns 1 through 13
    {'a'}    {'b'}    {'c'}    {'d'}    {'e'}    {'f'}    {'g'}    {'h'}    {'i'}    {'j'}    {'k'}    {'l'}    {'m'}
  Columns 14 through 26
    {'n'}    {'o'}    {'p'}    {'q'}    {'r'}    {'s'}    {'t'}    {'u'}    {'v'}    {'w'}    {'x'}    {'y'}    {'z'}

and

>> cellstr('a':'z')
ans =
  1×1 cell array
    {'abcdefghijklmnopqrstuvwxyz'}

based on how cellstr works.

Come to think of it, you don't need the second apostrophe.