Re: [Orbit-python-list] ORBit-Python Roadmap

[Jason T. <ta...@li...>]

> I'd like to vote for pickling object references as a 0.4.0 feature, to be
> implemented in the first round of that.

That would make a reasonable 0.4.0 feature.  

> Task taken here; will be showing progress over the weekend.

Great.  Make the test cases simple and focused.  For example, one
client/server to test sequences, with a series of functions to test the
various cases:

interface SequenceTest {
	typedef sequence<long> seq;

	seq get_sequence(in long n);
	void set_sequence(in seq s);
	seq pass_sequence(in seq s);
};

The client should loop over each function individually 25000 or so
times.  My hope is that we can expand on these tests to identify memory
leaks and help in profiling.

> Can we target 0.3.0 for closing on June 1st, and release on the next week?

We can target all we want. :)

2 weeks might be enough, but then it might not.  I'll want to play it by
ear.  Remember I just committed a lot of hefty changes to the tree that
has the potential to break a lot of things.  If things are wishy washy
to begin with, I'll probably need more time to sort out the problems.

On the other hand, maybe I was in the zone last night and wrote solid,
bug free code.  Pfft, yeah right. :)

Jason.