>>> Joe Orton <email@example.com> 10/06/04 12:46 PM >>>
On Wed, Oct 06, 2004 at 10:12:33AM -0600, Jean-Jacques Clar wrote:
>> As far as not having a bug in the !HAS_WRITEV implementation,
>> I disagree.
>> The current implementation does not have a single chance of
>> being successful if there is more than 1 vector. This does
>> not make sense to me. Let assume the write part is successful,
>> the function will return apr_success but has completely failed
>> to his task.
>You're missing the point. It is valid for apr_file_writev() to return
>APR_SUCCESS and update *nbytes to less than the total number of bytes in
>the vector. The writev()-based implementation *may* do that. The
>non-writev-based implementation *will* do it, if passed a vector of more
>than one buffer. But the caller must cope with it either way.
I clearly understand "your" point, but could I disagree with it?
I have a problem with the *will* fail. Both option should
have a *may* be successful.
I don't think our views on that point are reconcilable,
but maybe others could share whatever they think
the current way is right or wrong.