Re: [boost] Re: Any interest for a parser class?

From: David B. Held (dheld_at_[hidden])
Date: 2002-01-18 16:43:16

-----Original Message-----
From: Beman Dawes <bdawes_at_[hidden]>
To: boost_at_[hidden] <boost_at_[hidden]>; boost_at_[hidden]
<boost_at_[hidden]>
Date: Friday, January 18, 2002 03:13 PM
Subject: Re: [boost] Re: Any interest for a parser class?

>I'd personally rather see the effort now going into this discussion go
>instead into finishing Sprint and submitting it to Boost for formal review.

Actually, the library is "Spirit", as in "ephemeral", not "fast".  I hope
nobody takes advantage of the obvious pun. ;)

>It seems a distraction from the main show to spend so much time
>discussing one possible use.  It is clear that Sprint is of great interest
>to many Boosters.  And interesting to hear from those who think it's
>inappropriate for some uses, and why.

I guess I didn't see this thread so much as being about parsing as
about what a CLA library should do.  Would it be constructive to
discuss the desired requirements for a CLA library, since it has
come up several times, and people seem to want one quite a bit?

>But let's think about moving on now.  Once a release version has
>been available for awhile, we can ask users what they use it for
>and why or why not.  But if someone doesn't like Sprint for some
>particular use, we aren't going to make them use it against their will.
>
>Worry about those of us who do want to use Sprint, not those of us
>who don't.

I agree Spirit shouldn't be forced on anybody.  On the other hand,
it looks like there are divergent views on how a CLA library ought
to be implemented.  Just as nobody wants to see code duplication
by including both MPL and Loki::typelist in their entirety, I would
imagine that nobody wants two competing CLA library
candidates, so it seems like it would be nice to decide what
people want/expect from a CLA library, and maybe that will
help decide whether there is a preferred approach.

Dave