Boost logo

Boost :

From: Daryle Walker (darylew_at_[hidden])
Date: 2002-03-24 03:54:02


on 3/20/02 7:07 PM, Beman Dawes at bdawes_at_[hidden] wrote:

> At 03:55 PM 3/20/2002, Daryle Walker wrote:
>
>>> For consistency, shouldn't the forward declaration header be named
>>> <boost/io/iostate_fwd.hpp> and placed in the boost/io subdirectory?
>>
>> No. The <boost/io_fwd.hpp> header is for everything in the boost::io
>> namespace. Future libraries that go in here would change this forwarding
>> header too. See the math sub-library for a current example. The new
>> format library should go in next to my stuff. (I should contact format's
>> developer about rearranging his headers.) I should post my namespace,
>> forwarding header, and sub-library philosophy.
>
> Yes, that might be a help. I think the confusion is in the name
> io_fwd.hpp.

I posted an explanation right before the switch away from Yahoo. I'll
re-post it under "Re(post): My Namespace, Forwarding Header, and Sub-library
Philosophy Post".

> The _fwd could signify either two things:
>
> (1) The header includes only forward declarations. Like the
> standard's <iosfwd>
>
> (2) The header just forwards to other headers. For example,
> <boost/smart_ptr.hpp> contains only:
>
> #include <boost/scoped_ptr.hpp>
> #include <boost/scoped_array.hpp>
> #include <boost/shared_ptr.hpp>
> #include <boost/shared_array.hpp>
>
> Personally, I'd like to reserve appending "_fwd" for case (1), but would
> also like to hear other's opinions.
>
> In which sense are you using "_fwd"?

I'm using it as in case [1].

> Aside from that one header name question, is the library ready to be
> committed to CVS?

Almost; I need to come up with test cases for the combined saver classes.

-- 
Daryle Walker
Mac, Internet, and Video Game Junkie
darylew AT mac DOT com

Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk