[elephant-devel] Re: Common lisp transition issues?

[Robert L. Read]

Hi, Ian, and welcome.  It sounds like you will be able to make some nice
contributions.
I am the maintainer, so I am the main person to coordinate with.  I will
be out of email range
today, but we can discuss it more tomorrow.  The original authors are on
the "devel" list,
and we can re-open the licensing conversation with them.

There are some known PPC problems; Andrew Blumberg and others are
working on that,
and I'm sure they would welcome your help.

Can you give me the link about elephant being down?  I don't know what
that means
and would like to find it and repair it.  As far as I know, elephant has
not been down,
and I know of no reason why you shouldn't be able to sign up to the
list.

On Tue, 2006-01-17 at 16:03 -0500, Ian Eslick wrote:

> Aside from the common-lisp.net pointer to elephant being down (.org 
> works) I have been unable to sign up to the developers list and have 
> some patches to make the Allegro bdb version of elephant on Mac OS X 
> more robust (at least in my particular setup) that I would like to submit.
> 
> I'm planning to use elephant for a large data collection effort and 
> expect I'll need/desire to contribute to the system as I proceed in 
> using it for this application so wanted to find out who are the key 
> people to coordinate with for checking in code and get on the list so I 
> can stay up to date with development.
> 
> I'm using the recent 0.4.0 release with version 4.3 of sleepycat under 
> Allegro 7.0 on a PPC G4 PowerBook running Mac OS X 10.4.  The approach 
> to serialization of strings in 0.4.0 didn't work for me.  I suspect it's 
> a byte ordering problem in the unicode encoding.  On my PPC under 
> allegro the native-to-string failed to work (returned all '???' - a good 
> indication of byte order problems) with the default copy in serializing 
> 2-byte characters.  Also in my version of allegro strings are stored 
> internall as 1-bit entities but the :ics features is still set so :ics 
> does not categorically imply 2-byte unicode.  Not sure if this is a 
> Franz feature or a bug.  I also added a little bit the build process to 
> support .dylib builds under Mac OS X and some vendor variances in 
> *features* (Allegro doesn't provide the :darwin feature, only :macosx).  
> etc, etc.  I have little patches for all this.
> 
> My setup passes all tests on bdb except migration because I haven't 
> installed Postgres yet.  I'll do that at some point and perhaps check 
> out MySQL.
> 
> I also want to discuss the LLGPL license vs. the GPL license for systems 
> based on the code.  Now that we can backend to different SQL platforms 
> there it would be nice to have more flexibility on using the elephant 
> lisp code which LLGPL would provide.  I saw a discussion in November on 
> this but notice that the license hasn't been update.
> 
> Many thanks,
> Ian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.common-lisp.net/pipermail/elephant-devel/attachments/20060118/236c3bd1/attachment.html>