From: Beman Dawes (bdawes_at_[hidden])
Date: 2002-12-12 17:34:18
At 04:19 PM 12/12/2002, Peter Dimov wrote:
>From: "Beman Dawes" <bdawes_at_[hidden]>
>>
>> The alternative I considered was to try to map the system specific code
>> into a POSIX errno code. However, the macros seemed messy for C++.
>
>POSIX errno codes were what I meant, yes. On a POSIX implementation there
>would be no mapping. On Win32, you can use the POSIX-like C RTL
functions.
>
>I suppose you can introduce integral constant aliases for the <cerrno>
>macros but is it worth it? Aren't all caps identifiers starting with E
>reserved for the implementation, anyway?
Part of the problem is that there is no guarantee that the macros will be
available in <cerrno>. That's messy for Boost; we would have to provide
our own and configure accordingly. It might also be messy for
standardization; its seems to have a lot of trouble when multiple
committees start specifying the contents of the same header. The final
worry about a POSIX based scheme is that it might mislead users into
thinking every operating system can deliver exactly the POSIX errors and no
others. That isn't the case.
I'm not dead set against a POSIX based scheme, but it does seem a bit
messier that defining our own.
--Beman