On Sun, 2012-04-15 at 19:45 -0400, Hans-Christoph Steiner wrote:
On Apr 13, 2012, at 3:44 AM, Roman Haefeli wrote:
On Thu, 2012-04-12 at 17:52 +0200, Roman Haefeli wrote:
On Wed, 2012-04-11 at 22:49 -0400, Hans-Christoph Steiner wrote:
It would affect vanilla if you have the libdir loader running in it. Its related to that.
I think I experienced once also problem with "maximum object loading depth 1000 reached" and this was on Pd-vanilla without libdir. So probably there is still a problem outside of libdir?
I'll post again, when I figured out how to exactly reproduce the situation.
This also triggers the same error, although this is with a binary external and with no libdir involved:
[declare -stdpath extra/iemnet]
[tcpclient] <- works fine [iemnet/tcplient] <- triggers error from subject
Why is that?
Roman
The "depth 1000" error is completely separate from libdir.
Yes, that is why I said in the above example there is _no_ libdir loader involved.
It is caused by when you try to load an object, and Pd can load the files, but can't find a setup() function mapped to the classname.
Aha, thanks for the explanation.
But why is it, that [iemnet/tcpclient] works fine without having the above [declare] statement in the patch, but fails with "maximum depth reached" when there is a such statement? Sounds like a bug in [declare], but is it really?
Roman