[9fans] Some plan9 y2k problems

[9fans] Some plan9 y2k problems

[G.David]

>to guard against systems on which (-1)%7 == 1?

[checking K&R, yup the sign of % with
negative arguments is machine dependent]

How about this?

xtime.wday = (day + 4) % 7;
if(day < 0) {
	if(xtime.wday > 0)
		xtime.wday *= -1;
	xtime.wday += 7;
}

The reason I care is 7340036 fits easily in a vlong...

[9fans] Some plan9 y2k problems

[Andrew]

>
> I'm no bookmaker, but I'm tempted to take bets as to whether _any_
> aeroplane will in fact be in the air at midnight 1999/12/31.  My
> personal opinion is there won't, the airlines will be too shit-scared
> to risk it.


I've seen plenty of adverts and articles about a special Concorde flight.
I vaguely remember some German TV station giving away two tickets in a
competition.

The idea is that the participants celebrate the new year in Paris, fly
across the Atlantic and arrive in New York in time for a second
celebration. What a hoot it would be if Concorde never got off the runway
in Paris...

Andrew.

[9fans] Some plan9 y2k problems

[Lucio]

According to Dennis Ritchie:
>
> Of course, it's also my own birthday.  But maybe I'll
> have died in an aircrash on 1/1/00, so don't plan too
> far ahead.
>
I'm no bookmaker, but I'm tempted to take bets as to whether _any_
aeroplane will in fact be in the air at midnight 1999/12/31.  My
personal opinion is there won't, the airlines will be too shit-scared
to risk it.

Imagine the chaos, though, getting all those flight schedules back into
operation on the next day.

Bottom line, Dennis, I trust we'll live to wish you a happy birthday on
Unix's billionth secondversary.

++L

[9fans] Some plan9 y2k problems

[Dennis]

G. David Butler wrote:
>
> One more question, and sort of off topic --
>
> Does anyone know of any celebrations planned on Sep 8/9 2001
> for UNIX's 1,000,000,000 second anniversary?

Well, a couple of magazine reporters have asked me whether
999999999 seconds might be used as an EOF or sentinel or
other special value (I couldn't think of any such
applications, but what do I know)?

Of course, it's also my own birthday.  But maybe I'll
have died in an aircrash on 1/1/00, so don't plan too
far ahead.

	Dennis

[9fans] Some plan9 y2k problems

[G.David]

One more question, and sort of off topic --

Does anyone know of any celebrations planned on Sep 8/9 2001
for UNIX's 1,000,000,000 second anniversary?

:>

[9fans] Some plan9 y2k problems

[G.David]

This brings up another question, why is

xtime.wday = (day + 7340036L) % 7;

and not

xtime.wday = (day + 4) % 7;
if(xtime.wday < 0)
	xtime.wday += 7;

[9fans] Some plan9 y2k problems

[presotto]

Somewhere along the way we put it in libc, but when
I looked at the release version it was still everywhere.

------ forwarded message follows ------

>From cse.psu.edu!owner-9fans Wed Jan 13 01:09:14 EST 1999
Received: from plan9.bell-labs.com ([135.104.9.2]) by plan9; Wed Jan 13 01:09:14 EST 1999
Received: from cse.psu.edu ([130.203.3.50]) by plan9; Wed Jan 13 01:09:14 EST 1999
Received: from localhost (majordom@localhost)
	by cse.psu.edu (8.8.8/8.8.8) with SMTP id AAA24840;
	Wed, 13 Jan 1999 00:49:13 -0500 (EST)
Received: by claven.cse.psu.edu (bulk_mailer v1.5); Wed, 13 Jan 1999 00:48:58 -0500
Received: (from majordom@localhost)
	by cse.psu.edu (8.8.8/8.8.8) id AAA24802
	for 9fans-outgoing; Wed, 13 Jan 1999 00:48:50 -0500 (EST)
X-Authentication-Warning: claven.cse.psu.edu: majordom set sender to owner-9fans using -f
Received: from bio.cse.psu.edu (galapagos.cse.psu.edu [130.203.12.17])
	by cse.psu.edu (8.8.8/8.8.8) with SMTP id AAA24798
	for <9fans@cse.psu.edu>; Wed, 13 Jan 1999 00:48:47 -0500 (EST)
Received: (qmail 5800 invoked by uid 991); 13 Jan 1999 05:48:46 -0000
Message-ID: <19990113054846.5799.qmail@g.bio.cse.psu.edu>
Date: 13 Jan 1999 00:48:46 -0500
To: 9fans@cse.psu.edu
Subject: Re: [9fans] Some plan9 y2k problems
In-reply-to: Your message of "Tue, 12 Jan 1999 15:42:42 EST."
             <199901122043.PAA16347@cse.psu.edu>
Date: Wed, 13 Jan 1999 00:48:46 -0500
From: Scott Schwartz <schwartz@bio.cse.psu.edu>
Sender: owner-9fans@cse.psu.edu
Reply-To: 9fans@cse.psu.edu
Precedence: bulk

presotto@plan9.bell-labs.com writes:
| /sys/src/cmd/ftpfs/tm2sec.c:
| /sys/src/9/*/devrtc.c:
| ...
| There are probably other versions of this.

Good grief... isn't that what libraries are for?

"cutting and pasting considered harmful"

[9fans] Some plan9 y2k problems

[Scott]

presotto@plan9.bell-labs.com writes:
| /sys/src/cmd/ftpfs/tm2sec.c:
| /sys/src/9/*/devrtc.c:
| ...
| There are probably other versions of this.

Good grief... isn't that what libraries are for?

"cutting and pasting considered harmful"

[9fans] Some plan9 y2k problems

[presotto]

In case anyone cares...

They all involve screw ups in figuring out leap years.
The ctime.c one is stupidly calls dysize with the year
minus 1900 instead of the year.  Change it like the
following.

/sys/src/libc/9sys/ctime.c:

% diff /n/dump/1998/0513/sys/src/libc/9sys/ctime.c ctime.c
137c137
< 		for(d1 = 70; day >= dysize(d1); d1++)
---
> 		for(d1 = 1970; day >= dysize(d1); d1++)
140c140
< 		for (d1 = 70; day < 0; d1--)
---
> 		for (d1 = 1970; day < 0; d1--)
142c142
< 	xtime.year = d1;
---
> 	xtime.year = d1-1900;

/sys/src/cmd/ftpfs/tm2sec.c:
/sys/src/9/*/devrtc.c:

yrsize() computes the leap year incorrectly.  It should look like:

static int *
yrsize(int y)
{
	if((y%4) == 0 && ((y%100) != 0 || (y%400) == 0))
		return ldmsize;
	else
		return dmsize;
}

There are probably other versions of this.  I know I fixed it in
several places.  Good luck.