Path: csiph.com!eternal-september.org!feeder.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: Keith Thompson <kst-u@mib.org>
Newsgroups: comp.lang.c
Subject: Re: What about my last date functions? there are some error?
Date: Tue, 08 Nov 2016 12:05:29 -0800
Organization: None to speak of
Lines: 31
Message-ID: <lnh97hrb3q.fsf@kst-u.example.com>
References: <gh932ctlqkebuct208fqbbc7rdak1b2c1o@4ax.com> <87d1i69r5w.fsf@bsb.me.uk> <nvsd9k$nd4$1@dont-email.me> <871sym9dpv.fsf@bsb.me.uk> <nvstum$jt5$1@dont-email.me> <87bmxp978j.fsf@bsb.me.uk> <nvt73j$o68$1@dont-email.me>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Injection-Info: mx02.eternal-september.org; posting-host="e43f89eff67778d2e2548cd041b513fe"; logging-data="31840"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX18wqfHqndtCzNLuoNtx4OiC"
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.1 (gnu/linux)
Cancel-Lock: sha1:GDRMErjAg5RlKyh5NlJQorcXa5s= sha1:S7MjNQUcwAjsIEHZC1Amg8xo7NA=
Xref: csiph.com comp.lang.c:96123

Richard Heathfield <rjh@cpax.org.uk> writes:
> On 08/11/16 18:06, Ben Bacarisse wrote:
> <snip>
>> I suspect your preferred code will not go that far
>> out on exactly the same systems that have a limited range of time_t
>> values.
>
> You're right, of course. Nevertheless, it's still a win over time_t. 
> Let's assume time_t uses a long int under the bonnet, and let's further 
> assume that it encodes seconds (a not uncommon implementation, as we 
> both know). On a 32-bit-long-int system, that gives us a range of 49710 
> days, which is 136+ years. Julian dates, on the same system, can cover 
> well over 11,000,000 years.
>
> I have no particular need for /millions/ of years, at least not yet, but 
> I find the ability to stretch back through the second millennium (and 
> even into the first) to be useful from time to time.

Desktop Unix-like systems these days tend to make time_t a 64-bit signed
integer.

On the other hand, 32-bit systems still exist.  They can still be
updated to use a 64-bit time_t, and I expect that's what will happen
between now and 2038, but so far I haven't heard of a system with 32-bit
long int and 64-bit time_t.

-- 
Keith Thompson (The_Other_Keith) kst-u@mib.org  <http://www.ghoti.net/~kst>
Working, but not speaking, for JetHead Development, Inc.
"We must do something.  This is something.  Therefore, we must do this."
    -- Antony Jay and Jonathan Lynn, "Yes Minister"