apr-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Brad Nicholes" <BNICHO...@novell.com>
Subject Re: [STATUS] (apr) Wed Feb 5 23:45:28 EST 2003
Date Thu, 06 Feb 2003 15:32:51 GMT
It is there because I need it there to remind me that there is work
(although minor) to be done.  As a result of the previous discussion, I
talked to our C runtime engineers and asked them to update our SDK so
that it is in compliance.  They have done it but that version of the SDK
is not publically available yet.  Until it is, I can't change the APR
headers.  So it is here as a reminder.

Brad Nicholes
Senior Software Engineer
Novell, Inc., the leading provider of Net business solutions
http://www.novell.com 

>>> Craig Rodrigues <rodrigc@attbi.com> Wednesday, February 05, 2003
9:57:11 PM >>>
On Wed, Feb 05, 2003 at 11:45:28PM -0500, Rodent of Unusual Size
wrote:
>         apr.hnw               (READDIR_IS_THREAD_SAFE, ENUM_BITFIELD,

>                               _POSIX_THREAD_SAFE_FUNCTIONS (?))
                                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Why is this still in this notice?  I mentioned on this list that it is
wrong
for an application to declare this constant, since this
is a manifest POSIX constant that is defined by including <unistd.h>.
See:
http://www.opengroup.org/onlinepubs/007904975/basedefs/unistd.h.html 

In a follow-up post, Brad Nicholes at Novell mentioned that 
it is all OK to unconditionally define this constant on Novell.
If someone who works at Novell says it is OK to do this on
Novell, then why is this still an issue?

Mime
View raw message