array of Atoms

    Jan 06 2010 | 6:43 pm
    Dear all
    I want to create an array of atoms that I can reuse. I put a handle in my object structure and I assign the memory at instantiating time that way:
    x->array_out = (t_atom *)getbytes(cell_nb * 2 * sizeof(t_atom));
    but it does not seem to work, I hang Max all the time.
    Any help is welcome, I need the object to be Max4.6 compatible so new methods are forbidden ;-

    • Jan 06 2010 | 9:36 pm
      Are you sure cell_nb isn't too large for getbytes? Does it return a valid pointer, or 0?
    • Jan 06 2010 | 10:13 pm
      Dear Martin
      cell_nb is 32, so it is plenty within the range (512 byte only). I do seem to get a proper pointer but this code seem to be th problem.
      I'll continue to investigate, but this code worked fine before, so I am quite puzzled...
    • Jan 06 2010 | 10:18 pm
      ...tried running it in runtime under the debugger?
    • Jan 06 2010 | 10:29 pm
      tried it, and I made it work by doing this instead:
      strangely it did not accepted x->array_out[j] as a (t_atom *)
      I am quite surprised by this but it sorted my problem...
      thanks for your help
    • Jan 07 2010 | 4:25 pm
      I'm surprised the compiler didn't give you a warning with the first code.
      This code:
      is the same as this:
      but it is not the same as this:
      (note the &)
      Glad you got it figured out...
    • Jan 07 2010 | 4:46 pm
      Tim, I am so ashamed... such a midget mistake, thanks for pointing it !
    • Jan 07 2010 | 5:44 pm
      No need to be ashamed -- this kind of pointer snafu in the C language is part of the reason that so many other languages have been invented, and why C makes so many people run in the opposite direction. It snags us all from time to time...
    • Jan 07 2010 | 6:02 pm
      Thanks for the reassuring word ;-)
      Btw I did not get the error before because ext_obex.h was not included in my code. It was still compiling but the external was very unstable...