Re: [boost] Filesystem attributes

From: Jan Langer (jan_at_[hidden])
Date: 2002-03-03 12:00:57

On Sun, 3 Mar 2002, Dietmar Kuehl wrote:
>> but the key should be the same like the one being passed to remove. IMO
>> string, not in the meaning of pathname but of an unique identifier.
>
>Unless we stick only to the currently existing function taking a
>C-string we can overload it as well for the opaque type returned from
>directory iterator's 'operator*()' which also acts as a unique
>identifier of a file.

i think thats a possible solution. this opaque type should be impl.
defined and act nearly like a string.

so the posix impl. can typedef it to string, because the name is the
only attribute related information one usually gets during iteration
(except the inode number and a unusable type information)

-- 
jan langer ... jan_at_[hidden]
"pi ist genau drei"