Re: Tcl C API documentation?

In article <m21yj0d6ju(dot)fsf(at)kuiper(dot)rlent(dot)pnet>,
Roland Roberts <roland(at)rlenter(dot)com> wrote:
>>>>>> "Neil" == Neil Madden <nem00u(at)cs(dot)nott(dot)ac(dot)uk> writes:
>
> Neil> If it is a widely available database system, there is a high
> Neil> likely-hood that there exists a Tcl extension out there
> Neil> which will do most if not all of what you want. I don't do
> Neil> much database work, so I can't comment on any particularly
> Neil> useful extensions.
>
>Actually, it is PostgreSQL and there is a pgtclsh provided. So my
>exercise is, well, an exercise. While I can read the man pages on
>each of the calls used, I can't quite put together the rationale for
>*why* things are done the way they are.
>
>libpgtcl.so doesn't use the Tcl object calls, but that's a pretty
>minor thing to change. It also is not really set up the way I would
>have preferred which is to make is usable via "package require", but
>that's a nit...
.
.
.
I'll say a few words about Postgres (PG).

While Great Bridge's demise might lead some to expect
stagnation for PG, it's far healthier than that. Look,
for example, at Bruce Momjian's new book. In any case,
there's a lot of life left in PG.

PG has a history of friendship and co-operation with
Tcl. Other languages have received more attention
lately. This is a good time, though, I think, for
those with an interest in pgtclsh to work up patches
to implement the (Tcl) object interface, stubs, and
proper [packag]ing.

While we're on the subject, I'd sure appreciate it if
someone at www.postgresql.org would fix up the hyper-
link anchors with meaningful "ALT ="-s so image-free
browsers can make better use of the site.
--

Cameron Laird <Cameron(at)Lairds(dot)com>
Business: http://www.Phaseit.net
Personal: http://starbase.neosoft.com/~claird/home.html