How do I delete a row in a table containing certain text?

Hi, I am very new to MATLAB and I have been reading all the help and I do not understand where my code is wrong. I need to simply delete a row of my table 'indices' which contains the text 'Totals:'.

Below is an excerpt of my table 'indices' (originally 18x9) and I want to delete the row with the Total:, i.e the last row (or row 18).

CalendarYear Premium

'2017' 35216284

'2018' 36432973

'2019' 37895599

'Total:' 428730614

I tired several things in order to find the row number or the location where the word Total is without success:

1.

>> isequal(indices.CalendarYear,'Total:')

ans =

  logical

   0

2.

 >> strcmp(indices(:,1),'Total:')

ans =

  logical

   0  

3. This gives me a logical output but then I do not know how to continue using it

   >> strfind(indices.CalendarYear,'Total:')

ans =

  18×1 cell array

    {0×0 double}
    .........
    {[       1]}   %finally a positive answer

none of these attempts gives me the row number 18 which I wold like to delete form my table.

What am I missing?

Thank you in advance,

ni7

Answer 1

[Jan]

You are almost there.

isequal(indices.CalendarYear, 'Total:')

This tests, if indices.CalenderYear is equal to the char vector 'Total:'. This cannot be equal.

strcmp(indices(:,1), 'Total:')

This is almost working. Use braces instead

strcmp(indices{:,1}, 'Total:')
% Or to get the index:
find(strcmp(indices{:,1}, 'Total:'))

Or equivalently (I prefer this!):

find(strcmp(indices.CalendarYear, 'Total:'))

The command:

strfind(indices.CalendarYear, 'Total:')

searchs, where the part 'Total:' appears, not if the string equals this exactly. But as long as you do not have a 'not Total:' anywhere, this would work also:

find(~cellfun('isempty', strfind(indices.CalendarYear, 'Total:')))  % [TYPO FIXED]

[EDITED] Use contains in modern Matlab versions:

find(contains(indices.CalendarYear, 'Total:'))

[Nina Nikolova]

Thank you, @Jan, for the educational answer. I still need to understand better the concept of the different brackets and braces. The solution worked perfect!

[Stefan Grandl]

Perfect answer, thanks!

Just a remark: in Jan's last answer there is a final parenthesis missing at the end. It should be:

find(~cellfun('isempty', strfind(indices.CalendarYear, 'Total:')))

Answer 2

[Luna]

Hi Nina,

You can use this code below:

myTable = table({'2017','2018','2019','Total:'}',[35216284,36432973,37895599,428730614]','VariableNames',{'CalendarYear','Premium'});
reducedTable = myTable(~contains(myTable.CalendarYear,'Total:'),:);

[Luna]

Yes, both indexing or the real row number can be used to delete the row.

In this case:

rowNumber = find(contains(myTable.CalendarYear,'Total:'));
myTable(rowNumber,:) = []; % this deletes that row from myTable itself without creating a new table

[Jan]

@Luna: Sorry, my comment was off track. I've overseen the detail of "deleting". +1 for your efficient solution.

[Luna]

Thanks @Jan :)

[Nina Nikolova]

Dear @Luna, thank you for showing another efficient way, this is all so beneficial for my learning!

[Luna]

Your welcome :)

Answer 3

[madhan ravi]

Another possibility using regexp but not as effective as the above two answers:

k=cellfun(@(x)regexp(x,'Total:'),T.CalendarYear,'un',0) % T is your table
T(~cellfun(@any,k),:)