How to print strings from a data text file that is in correlation to an indices text file?

Hi everyone, I was hoping someone could help me with this tricky concept. I am trying to write a program and I don't really know how to execute it. You will see in the text files I've attached, I have one called index.txt and that has an array from 0 - 7, each index has 4 array elements (looks like this 0 1 6 2 3 1 3 2 4 0 2 1 6 4 7 ..etc). Then in the data.txt file I have another array 0 - 7 but each index includes a hexidecimal value. There are 3 categories, it should make sense when you view the text files. I'm looking for this to happen:

1.) User selects a category (Category 1, 2, and 3)

2.) User can either select index from 0-7 or an index at random.

3.) The program prints 4 strings(since the indices each contain 4 numbers) from the data.txt that is correlated to index file.

I was thinking to read line by line searching for "First Index, Second Index or Third Index to accomplish #1, and then maybe put the strings in cell arrays. This is honestly really complex for me, so any pointers on how I can get started on this would be greatly appreciated!

Answer 1

[dpb]

Put the indices into a 3D array and just index --

indices(:,:,1)=[1 6 2 3; 3 2 4 0; 1 6 4 7; 0 1 4 7; ...
                0 5 6 1; 0 1 2 3; 0 5 6 2; 0 3 6 7];
indices(:,:,2)=[
                0 6 2 3; 3 2 4 0; 1 6 4 7; 0 1 4 7; ...
                0 5 6 1; 0 7 2 3; 0 5 6 2; 0 5 6 7];        % NB: presumed the last 4 0 5 6 7 ==>7 0 5 6 7 fixed typo
indices(:,:,3)=[
                4 6 2 3; 3 2 4 0; 1 6 2 7; 0 1 4 7; ...
                5 0 6 1; 0 1 2 3; 0 5 6 2; 0 5 6 0];

Do same thing with DATA array, then

icat=input('Enter category (1-3)');             % get the category
indx=input('Enter index (1-8)');                % and index (in MATLAB 1-based counting)
output=data(indices(indx,:,icat));              % retrieve data from icat plane, index row of lookup indices

[Akana Juliet]

@dpbThank you so much for your response! is there anyway of doing this by reading in text files and maybe put in a cell array? I have included shortened versions of my .txt files in my question here for simplicity. But the files are actually quite large. The format is the same just a lot more lines for both indices and data

[dpb]

Sure...just have to define how the files are constructed to be able to read them -- then you can manipulate the input as needed to reshape it.

I just showed the end result you need to be the most effective storage for the lookup table.

Inputting the extraneous index doesn't hurt anything but it's not needed but if you've already got the files created, then just delete [1:5:end] columns in the input arrays.

You could use cell arrays, but there's no real advantage to be gained that I see with additional overhead that a cell array takes in comparison to arrays of doubles (104 bytes/element overhead plus the data).

[dpb]

For your example index file, something like --

indices=readcell('index.txt');                      % read the file into cell array
indices(contains(indices,'Index'))=[];              % get rid of the text
indices=str2double(split(indices))+1;               % convert cell str to array numbers 1-8
indices(:,[1:5:end])=[];                            % remove superfluous column index 
indices=pagetranspose(reshape(indices.',4,[],3));   % create the 3D array based on 3 indices; four indices per

produces the above array in the needed 1:8 range.

>> indices
indices(:,:,1) =
     2     7     3     4
     4     3     5     1
     2     7     5     8
     1     2     5     8
     1     6     7     2
     1     2     3     4
     1     6     7     3
     1     4     7     8
indices(:,:,2) =
     1     7     3     4
     4     3     5     1
     2     7     5     8
     1     2     5     8
     1     6     7     2
     1     8     3     4
     1     6     7     3
     1     6     7     8
indices(:,:,3) =
     5     7     3     4
     4     3     5     1
     2     7     3     8
     1     2     5     8
     6     1     7     2
     1     2     3     4
     1     6     7     3
     1     6     7     1
>> 

The above lookup code can use this directly as written above; do similar rearrangement for the data array.

[Akana Juliet]

@dpb Thank you for helping me with this! I'm just confused on why the values this code stores is not the same as the text file values, I keep trying to figure out why it looks so different.

[dpb]

Two reasons --

1. I added "1" to all the values so the indexing is one-based as are MATLAB arrays -- you can't change lower bounds in ML to be 0-based;
2. I removed the extra index variable that you had in the text file ahead of each 4-vector because it isn't needed when arranged by index by row and plane -- you just index into each plane with the row number 1 thru 8 for the wanted plane 1-3 and retrieve the four indices into the data array associated.

If you subtract "1" from the array, then you'll be able to see the comparison more easily because then you'll have the 0-7 indices that are in the text file--but, you have to add one to them to use as array indices, so may as well just convert on input "in anger".

The first hand-written looked like

indices(:,:,1)=[1 6 2 3; 3 2 4 0; 1 6 4 7; 0 1 4 7; ...
                0 5 6 1; 0 1 2 3; 0 5 6 2; 0 3 6 7];

for the first plane -- you'll see [1 6 2 3] ==> [2 7 3 4] in the second and they're written out in 8x4 array display whereas I just edited in place keeping the two rows in writing them out explicitly before. They're the same array; if you display the above in command window w/o the trailing semicolon, it's an 8x4 array as well; the semicolons ensure that.

[dpb]

To show they're the same, I copied the code above for plane 1 to command line as I1 and

>> I1=[1 6 2 3; 3 2 4 0; 1 6 4 7; 0 1 4 7; ...
>> all(I1==indices(:,:,1),'all')                % they aren't the same as is, agreed...
ans =
  logical
   0
>> all(I1==indices(:,:,1)-1,'all')      % but they are the same when take off the "1"
ans =
  logical
   1
>>