C9X and Time

[Paul Eggert]

   Date: Tue, 31 Aug 1999 17:23:18 +0100
   From: "Clive D.W. Feather" <clive at demon.net>

   The last WG14 meeting agreed, in principle, that a work item could
   be started on new time facilities for C....

   Where do we want to go from here ?

As seems clear from discussions on the tz mailing list, we need to
establish the scope of the new time facilities.  This can include:

  * Better time zone support (e.g. multiple time zones, a more
    realistic model of daylight saving time)
  * Better calendrical support (e.g. conversion to day numbers)
  * Better support for realtime applications (e.g. benchmarking)

Only the first item is on-topic for the tz mailing list; perhaps
a new mailing list should be established for the C0x time API.

Here is is some current work in this area, which can be the start of
the discussion.

D. J. Bernstein, libtai
ftp://koobera.math.uic.edu/www/libtai.html (1998-10-13)

C9x committee, Date and time <time.h> [proposed extensions]
section 7.23 of WG14/N843 committee draft (1998-08-03)
http://anubis.dkuug.dk/jtc1/sc22/open/n2794/n2794.txt

Paul Eggert, DRAFT Modernized <time.h> API for ISO C (1999-08-27)
http://www.twinsun.com/tz/timeapi.html

Markus Kuhn, Proposed new <time.h> for ISO C 9X (1999-04-02)
http://www.cl.cam.ac.uk/~mgk25/c-time/

David Tribble, Proposal for ISO C and C++ Extended-Range Time Type
(1998-07-14)
http://www.flash.net/~dtribble/text/c0xtime.htm