[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: difference of cell arrays behavior between octave-3.0.5 and octave-3
From: |
Sergei Steshenko |
Subject: |
Re: difference of cell arrays behavior between octave-3.0.5 and octave-3.4.2 |
Date: |
Mon, 1 Aug 2011 02:35:50 -0700 (PDT) |
--- On Mon, 8/1/11, Sergei Steshenko <address@hidden> wrote:
> From: Sergei Steshenko <address@hidden>
> Subject: difference of cell arrays behavior between octave-3.0.5 and
> octave-3.4.2
> To: address@hidden
> Date: Monday, August 1, 2011, 2:31 AM
> In octave-3.0.5 I have:
>
> "
> octave:1> foo{1}(:) = []
> foo =
>
> {
> [1,1] = [](0x0)
> }
>
> octave:2> foo{2}(:) = [2;3]
> foo =
>
> {
> [1,1] = [](0x0)
> [1,2] =
>
> 2 3
>
> }
>
> octave:3>
> ".
>
> In octave-3.4.2 I have:
>
> "
> foo{1}(:) = []
> foo =
> {
> [1,1] = [](0x0)
> }
> octave:2> foo{2}(:) = [2;3]
> error: A(I) = X: X must have the same size as I
> octave:2>
> ".
>
>
> Why is it so that in octave-3.4.2 I have an error ?
>
> Why at all size checking kicks in - "foo{2}(:)" doesn't yet
> have size,
> it should be created according to RHS.
>
> Which of the two behaviors is nominally correct (Matlab) ?
>
> I've discovered this by running my code which works under
> octave-3.0.5
> and which stopped working under octave-3.4.2.
>
> Thanks,
> Sergei.
>
>
>
Probably not an issue. I have packages from octave-3.0.5; when I try
with these packages removed, the small test case works the same way as
in octave-3.0.5.
Thanks,
Sergei.