On 02/19/2011 04:42 PM, Leo Butler wrote:
> >
> < Ray:
> <
> < I understand that Leo's branch is different than the official maxima project.
> <
> < My main reason for responding initially was to try to determine if the problem
> < was with GCL or if the problem was with Leo's CL-PPCRE code which GCL chokes
> < on. I went through the process of building maxima from CVS against GCL
> < successfully so I think GCL seems to work where some people were saying GCL
> < didn't work when doing the same thing. Keep in mind that I may not be using
> < the same version of GCL as those people were using, which is why I gave my
> < version of GCL as release 2.6.8pre (which chokes on CL-PPCRE code downloaded
> < via "git").
> <
> < Reporting it to Camm Maguire should be done if there is consensus that there
> < is a problem with GCL, but maybe Leo should actually be the one to report it
> < since it is his code that has the problem.
>
> Paul,
> I think that it would be easier to try to compile cl-ppcre with gcl, if
> your intent is to troubleshoot that end of things.
>
> However, I think that cl-ppcre simply uses features of common lisp that
> gcl does not support. Rather than try to shoehorn cl-ppcre into an ill
> fitting gcl shoe, I've been working on extending the nregex code so
> that more modern lisps can use cl-ppcre+build-index while gcl can use
> nregex+build-index. This seems quite feasible--and with a few tweaks,
> the nregex regexes can be quite fast in gcl.
>
> Leo
>
Leo:
Troubleshooting cl-ppcre would be quite an adventure for me because I'd
have to spend a lot of time learning lisp (which is on my list of things
to do --- someday). I like learning programming languages, but I've not
yet had the time to tackle lisp because of some other things I've been
attempting.
Do you think Camm would be interested in extending GCL to handle the
features found in more modern lisps or is that unlikely. I ask because
of all the lisps I've tried with maxima, GCL is the fastest and seems to
work properly for anything I have enough mathematical expertise to use.
Paul