Re[2]: Re: Re: new print function in interim release 660

On Fri, 25 Jun 2004 08:05:13 -0700 Paul Prescod <[email protected]> wrote:

> Mark Hahn wrote:
> 
> >...
> > It will stay untested until we try to write real code with it and shake out
> > the bugs.  When Lenard tried to write real code he found many bugs and
> > helped test it a lot.  If you were to write real code it would help test it
> > a lot also.
> > 
> > If you were to write real code I would promise to give bugs you find very
> > high priority.  Instead of working on the bug list every few days like now,
> > I would work on your bugs as soon as you find them.
> 
> It isn't the bugs that are the problem. Its the shifting design. One day 
> you would have a ton of code treating print as a keyword and then the 
> next day it would be a function. It's great that Prothon is in a mutable 
> stage right now but you an't expect people to write stuff and then have 
> to rewrite it every week until all of the basics are worked out.
> 
In my case it is the bugs. As new features are added and changed old bugs
can reappear. Also some bugs only appear with large programs. These could be stack
or garbage collector related. They are very time consuming to isolate. I think
the memory management module should be given a good workout with a C program
written for the purpose. And there are still some problems involving generators.
Until the interpreter is more stable - it does not regularly crash - it is too
early to start anything major.

Lenard Lindstrom
<[email protected]>