Camm Maguire :
> Greetings! If there is any ansi feature currently implemented in gcl
> 2.7.0 which you would like backported to gcl 2.6.2 before we release
> it, please let me know and I will consider. In general we are too
> close to release now for such items, but we have still mysterious
> windows problems outstanding, which may delay the release long enough
> for a simple backport to be feasible.
>
No, I don't think that any backports are required at present.
I hope that ANSI compliance will be the primary goal for
gcl 2.7.X and ultimately 2.8.0 (3.0.0 ?)
> Take care,
>
> "Vadim V. Zhytnikov" <vvzhy@mail.ru> writes:
>
>
>>Raymond Toy :
>>
>>
>>>>>>>>"Vadim" == Vadim V Zhytnikov <vvzhy@users.sourceforge.net> writes:
>>>
>>> Vadim> - ;; Hmm, gcl 2.5.1 does appear to have
>>>handler-case. Ick. Vadim> + ;; GCL 2.6.2
>>>has handler-case but not quite ANSI yet. Vadim>
>>>(let ((result
>>> Vadim> #-gcl
>>> Vadim> (handler-case Vadim>
>>>(catch 'errorsw
>>> Vadim> ($float ($realpart (meval* ',expr))))
>>> Vadim> (arithmetic-error () t))
>>> Vadim> - #+gcl
>>> Vadim> - (catch 'errorsw
>>> Vadim> + #+gcl
>>> Vadim> + (conditions::handler-case
>>>Vadim> + (catch 'errorsw
>>> Vadim> ($float ($realpart (meval* ',expr))))
>>> Vadim> + (conditions::internal-error () t))
>>> Vadim> ))
>>>Does this mean we no longer support versions of gcl earlier than
>>>2.6.2? Is that what we want?
>>
>>I suppose so.
>>
>>
>>>FWIW, I still use gcl 2.4.0 on my Solaris box. Mostly because I
>>>haven't been able to build more recent ones. I think my Linux box
>>>also uses 2.5.x or so, because I can't build later ones.
>>>Ray
>>>
>>
>>As far as I remember we decided to use ANSI GCL for Maxima
>>so 2.4.0 is certainly out of date. I'm not 100% sure about
>>Solaris but I don't expect any troubles with 2.6.2
>>(now still 2.6.1 CVS) on Linux.
>>
--
Vadim V. Zhytnikov
<vvzhy@mail.ru>
<vvzhy@netorn.ru>