Index exceeds matrix dimensions

Hi. I am trying to export some data to a database and I am getting following error message

Index exceeds matrix dimensions.

Error in data2postgresql (line 135)
                exportData = {data{k,4}, data{k,2}, data{k,1},
                data{k,6}, data{k,7}};

for this code

          %%Export data to the database
          switch databaseExport
              case 0 % no database export
              case 1 % export to database
                  for k = 1:nRows
                  % sort the imported data and move it into exportData
                  exportData = {data{k,4}, data{k,2}, data{k,1}, data{k,6}, data{k,7}};
                  % insert data to the appropriate table and columns of the database
                  colNames = {'"A"', '"B"', '"C"', '"D"', '"E"'};
                  fastinsert(conn, exportTable, colNames, exportData);
                  end
              otherwise % no export
          end

Some info: nRows is the length of data{:,1} and data is a 1x7 2074487 Bytes cell and exportData is a 1x5 691766 Bytes cell with following entries for k=1, I guess, '001' / 1 / 86399x1 double / [] / []

Do you have any idea why I get a wrong dimension error and how I could fix it? Thank you very much for your help. Best regards

Answer 1

[Guillaume]

Well, either data has less than 7 columns or less than nRows rows.

Before the for loop insert these two lines:

assert(nRows <= size(data, 1), 'data has less than nRows rows');
assert(size(data, 2) >= 7, 'data has less than 7 columns');

or put a breakpoint on the exportData line and check the size of data manually.

[GuidingDrifter]

Thank you. It says: 'data has less than nRows rows'

If I change the loop to

for k = 1:size(data)

I just get 1 entry in my database of '001' / 1 / 0 / [] / [] instead of entries in the magnitude of data{:,1} which has around 80000+ lines. The end result should therefore be a lot of rows in the database.

for k = 1:size(data{:,1})

produces the same dimension error.

[Guillaume]

It should be

for k =1:size(data, 1)

I'm not completely clear on the size and structure of your data. It sounds like you're mixing up the size of the container (data, a cell array) and the size of the content of the container.

What do

size(data)
size(data{1})
class(data{1})

return?

[GuidingDrifter]

Thank you for helping me and taking your time. I have changed the code to

for k = 1:size(data,1)
% sort the imported data and move it into exportData
exportData = {data{:,4}, data{:,2}, data{k,1}, data{:,6}, data{:,7}};

as only data{:,1} has more than one row.

K>> size(data)

ans =

       1     7

K>> size(data{1})

ans =

       86399           1

K>> class(data{1}) 

ans =

double

[Guillaume]

Ok, the cell array, the container has only one row and 7 columns. The 1st element of the container, a double vector, has only 1 column but 86399 rows. What about the other 6 elements of data. Have they all got 86399 rows?

What is it you're trying to put in exportdata?

Maybe what you meant to do was:

for k = 1:numel(data{1})
    exportdata = {data{1}(k), data{2}(k), data{3}(k), data{4}(k), data{5}(k), data{6}(k), data{7}(k)};
    %...
 end

[GuidingDrifter]

Thanks! Merci beaucoup! Yes, numel is what I was looking for. I do have a lot of entries now.

The first element of the container is the only one that has 80000+ rows. The other elements consist of just 1 row.