[boost] Re: any_container?

From: Dylan Cuthbert (dylan_at_[hidden])
Date: 2002-09-06 00:59:26

Hi again,

Looks like you've had a good play with it, I presume it supports any types
now rather than the limited set you had before?  Sorry, for the lack of
response in the meantime, I got really busy with a deadline (deadlines in
the gaming industry are particularly disastrous).

I'll have some time free over the next couple of weeks to look over the code
and see if I can use it (regardless of whether boost deems it useable or
not), hopefully some other people on this list will also "get it", ie. what
this is trying to do... no, it isn't trying to be perl/python etc. which are
external libs and cannot be relied on to be available, its a simple generic
interface for access to "any" based totally generic structures

It seems like a very good extension to the any classes to me, but I'm used
to having the need for this kind of code.

Regards

---------------------------------
Q-Games, Dylan Cuthbert.
http://www.q-games.com
P2P internet radio - http://www.peercast.org

"Geoff Leyland" <geoff.leyland_at_[hidden]> wrote in message
news:A81DC312-C0DA-11D6-8E17-003065F9F514_at_epfl.ch...
> Hi,
>
> I wrote to the list some time ago about a container for holding "any"
> (or at least for the moment, my cheap equivalent to any).  At one point
> I said I'd put something up for people to look at by the end of the
> week.  So, it's next month, and I've finally tidied it up a bit, and you
> can get it to have a look at it.
>
> It's at http://leniwww.epfl.ch/~geoff/rts_boost.tgz
>
> I've compiled in under CW 8.2 and an apple-modified gcc 3.1 under osx
> 10.2.  There's a CW project and makefile in rts_boost/rts/build.
>
> What is it?  It's containers for a thing like an any, to enable the use
> of "run time structures".  There's maps (like structures) and vectors.
> So you can go
>
> whatever_map m;
>
> m["a[5].b.c.d"] = 3.0;
> m["a"][6]["b.c.d"] = "hello";
>
> and so on.  It can also read such a container from a simple text file
> (using a very cheap and nasty parser I hand-wrote):
>
> a =
> {
>    b = [1 "hello" false 3.0]
>    c.d = "a name"
> }
>
> and so on...
>
> I find it very useful, and there was a little interest from others in
> the list.  On the other hand, it's not really finished or up to boost
> standard or anything like that.  I tried to be careful about exception
> safety in the container code, but otherwise there's lots of things to
> fix.  Clearly the contained type should be "any" (dynamic_any would be
> better) or even parameterised, the parser "needs work" and so on.
>
> If there's any interest and any ideas, I'll have a go at making it
> better.
>
> Cheers,
> Geoff
>
> --
> Geoff Leyland
> Laboratoire d'energetique industrielle
> LENI-DGM-EPFL, CH-1015, Lausanne, Switzerland
> Phone: [phone], Fax: [phone]
>
> _______________________________________________
> Unsubscribe & other changes:
http://lists.boost.org/mailman/listinfo.cgi/boost
>