|
Boost : |
Subject: Re: [boost] [range] Can iterator_range also model Random Access Range?
From: Nathan Ridge (zeratul976_at_[hidden])
Date: 2014-02-01 02:10:58
>>> Unfortunately, and respectfully, that is not at all clear from either
>>> docs, please reread range/reference/utilities/iterator_range.html and
>>> range/reference/utilities/sub_range.html. The latter defers to the
>>> former,
>>> and all the former states about range concepts is: "The intention of the
>>> iterator_range class is to encapsulate two iterators so they fulfill the
>>> Forward Range concept". And, obviously, just because some type fulfills
>>> the Forward Range concept does not mean that it fulfills refinements
>>> thereof. Additionally, there is no indication in the iterator_range doc
>>> if
>>> the underlying iterator models refinements of the Forward Iterator
>>> concept, than its corresponding iterator_range class also models
>>> analogous
>>> refinements of the Forward Range concept. A statement along the lines of
>>> the last sentence added to the iterator_range doc would go a long way
>>> towards clarifying this, else one is left with being forced to deduce
>>> this
>>> fact.
>>
>> I agree that the documentation is incomplete here. I would be happy
>> to review and commit a documentation patch.
>
> I've created a pull request at GitHub.
Thanks! I reviewed the PR on GitHub.
(Sorry for taking so long. The last time I committed a change to
Boost.Range, Boost was still using SVN. I had some process-related
catching up to do.)
Regards,
Nate
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk