|
Boost : |
From: Eric Niebler (eric_at_[hidden])
Date: 2005-05-18 17:02:21
Rob Stewart wrote:
> From: "Eric Niebler" <eric_at_[hidden]>
>
>>Rob Stewart wrote:
>
> Perhaps there's time to improve the proposed
> interface, too?
It's too late for TR1 but not for C++0X.
> Maybe a note clarifying that while spaces are significant in a
> Perl or, for that matter, a dynamic xpressive RE, they aren't
> significant in a static xpressive RE other than in literals.
But this is obvious, and I don't think pointing out the obvious makes
for better documentation.
> What about "beg" and "end?" I realize they aren't reusing the
> proposed terminology, but they avoid the "sequence/buffer/input"
> issue.
Beginning of what? End of what? The line? The word? The
sequence/input/buffer? beg/end are not specific enough, nor are they
more memorable than bos/eos, IMO. Besides, "end" is too common to make
it a namespace scoped constant, and "beg" is a word of its own with a
meaning distinct than "begin". Nope, bos/eos are it.
-- Eric Niebler Boost Consulting www.boost-consulting.com
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk