Re: next release?

From: Erick Gallesio <eg_at_kaolin.unice.fr>
Date: Wed, 06 Dec 1995 14:50:37 +0000

> Grant Edwards writes
.....
>
> Right now when I try to load the file (which worked under 2.x) I get a
> never-ending string of warnings about "&" being undefined while it appears
> to recurse infinitely (not sure about the recusion -- but it uses up so much
> memory/CPU that it's hard to get it killed).

I have made the assumption that I can delete it since I have never documented
it. This assumptions is probably the good one when you have a rather
exhaustive
documentation it is surely not realist when the the most common answer of the
author of the package is "Hey, look at the demos code if you want
documentation". Consequently, & has been put back in the relase (I will try to
not forget to document it :-)

> I did get a smaller program to run without errors, but some of the Tk widget
> behaviors have changed, so it doesn't work correctly.
>
This is Tk fault, not mine ;-)

> It looks like switching over to STk 3.0 is going to be much harder than I
> expected.

It is hard in effect. Bindings are the most difficult problem. However, I
think that current solution (especially for closure callback) is better than
previous hack (especially closure callback). I was not reluctant to add
incompabilities in this version, since I knew that Tk already induced some
incompabilities, and that, consequently, some pieces of code must be
rewritten.
I will try to avoid them for next release.

                -- Erick
Received on Wed Dec 06 1995 - 14:55:30 CET