Indexing a struct inside a struct

Hello!

I need to index a struct inside a struct, and dot indexing is not supported.

I'll explain better what I mean.

Each struct is a session (called dataA, corresponding for instance to the Race), and inside each session there is another struct, which is a stint, and inside the stint struct there are 1x98 structs (for instance, each is one lap). Inside each lap struct I have the variables I need (for instance, vCar, velocity vector).

For each stint, I want to go through all the 98 laps, and for each lap i want to go through the entire vCar vector and see where the vector has values that make no sense (for instance, when vCar(ii) > 500, add this index to a vector so at the end i have a vector of all the laps with errors in them).

The code is below (I'm sure there are also faster way to do this, but the main problem here is the indexing).

Thank you!

Fabrizio

for jj = 1:length(dataA.Stint)

for ii = 1:length(dataA.Stint{1, jj}.D_vCar) % ERROR, i want to call lap jj

if dataA.Stint{1, jj}.D_vCar(ii) > 10000 % ii is the length of the velocity vector

error_laps = [error_laps; jj]

break;

end

end

end

ERROR:

Dot indexing is not supported for variables of this type.

Error in dataanalysis (line 12)

for ii = 1:length(dataA.Stint{1, jj}.D_ETyreSlidingFLTotal)

[Stephen23]

@Fabrizio: please upload your data in a MAT file, by clicking the paperclip button.

Answer 1

[Catalytic]

If Stint is a struct array, as you say, then it should be indexed with parentheses (), not braces {},

dataA.Stint(jj)

Conversely, it would be better to store error_laps in a cell array (using braces) because you don't know what the length of the result will be. Overall, your code should probably be -

for jj = 1:numel(dataA.Stint)
    
  error_laps{jj}= find(dataA.Stint(jj).vCar>1000 ,1);  

end

[Steven Lord]

Conversely, it would be better to store error_laps in a cell array (using braces) because you don't know what the length of the result will be.

Or perhaps a table array with one variable containing the stint number and another the lap number, with whatever data variables you want for each stint / lap combination (velocity, race position, etc.)

If the data is time-based (how long since the start of the race the measurement was taken) a timetable array may also be an option.