I just noticed last night that Nethack 3.6.7 has been released, even
though there has been no devteam announcement in this newsgroup.
Documentation says it is a bugfix release that contains also some
security fix, so apparently otherwise it behaves like 3.6.6.
https://www.nethack.org/

Just starting preparation for compiling it myself..

--
Jukka Lahtinen

On Sunday, February 19, 2023 at 4:38:03 AM UTC-8, Jukka Lahtinen wrote:
> I just noticed last night that Nethack 3.6.7 has been released, even
> though there has been no devteam announcement in this newsgroup.
> Documentation says it is a bugfix release that contains also some
> security fix, so apparently otherwise it behaves like 3.6.6.
> https://www.nethack.org/
>
> Just starting preparation for compiling it myself..

The release is genuine. Usually an announcement is posted
to rec.games.roguelike.announce with a request that the
moderator repost it here (or possibly cross-post it when
allowing it to appear in that group). I'm not sure why that didn't
happen this time. (I suspect nothing was sent to ...announce
rather than that the moderator is tardy with the repost.) An
announcement was posted to 'https://reddit.com/r/nethack'.

You're unlikely to notice any changes in game play from 3.6.6,
unless you're in the habit of assigning very long names to objects
or monsters. There was intended to be a limit of 63 characters
(going back many versions) but it wasn't being enforced. With
3.6.7 it is; too long names get truncated without any feedback.

There are a few other small changes, and the false positive for
malware in the Windows binary distribution reported by a tiny
subset of security checkers has been eliminated. (If I remember
rightly, 4 out of 62 of them didn't like the fact that nethack.exe
was using DLL code for the player's choice of keyboard handler.)

On Sunday, February 19, 2023 at 1:55:54 PM UTC-6, Pat Rankin wrote:
> On Sunday, February 19, 2023 at 4:38:03 AM UTC-8, Jukka Lahtinen wrote:
> > I just noticed last night that Nethack 3.6.7 has been released, even
> > though there has been no devteam announcement in this newsgroup.
> > Documentation says it is a bugfix release that contains also some
> > security fix, so apparently otherwise it behaves like 3.6.6.
> > https://www.nethack.org/
> >
> > Just starting preparation for compiling it myself..
> The release is genuine. Usually an announcement is posted
> to rec.games.roguelike.announce with a request that the
> moderator repost it here (or possibly cross-post it when
> allowing it to appear in that group). I'm not sure why that didn't
> happen this time. (I suspect nothing was sent to ...announce
> rather than that the moderator is tardy with the repost.) An
> announcement was posted to 'https://reddit.com/r/nethack'.
>
> You're unlikely to notice any changes in game play from 3.6.6,
> unless you're in the habit of assigning very long names to objects
> or monsters. There was intended to be a limit of 63 characters
> (going back many versions) but it wasn't being enforced. With
> 3.6.7 it is; too long names get truncated without any feedback.
>
> There are a few other small changes, and the false positive for
> malware in the Windows binary distribution reported by a tiny
> subset of security checkers has been eliminated. (If I remember
> rightly, 4 out of 62 of them didn't like the fact that nethack.exe
> was using DLL code for the player's choice of keyboard handler.)

I noticed the new moon started later in 3.6.7. It was only a matter
of an hour or two.

On 20.02.2023 07:29, Yosemite Sam wrote:
>> On Sunday, February 19, 2023 at 4:38:03 AM UTC-8, Jukka Lahtinen wrote:
>>> I just noticed last night that Nethack 3.6.7 has been released [...]
>
> I noticed the new moon started later in 3.6.7. It was only a matter
> of an hour or two.

I just had a peek into hacklib.c phase_of_the_moon(). Two observations;
the code seems unchanged, and the arithmetic seems to be of accuracy
'days' (not hours).

I would assume that differences of 'hours' (as you observed) might be
affected by the time zone configuration, since I think it's local time
based? (But currently I haven't examined it deeper.)

There's also an offset-discrepancy between reality and implementation;
in the past (in a concrete new moon instance) I observed 1 day offset.

See also: thread posted in re.games.roguelike.nethack dated 2022-11-07

See also: thread "Datediff script" in comp.unix.shell by castAway,
specifically (for example) some results from 2022.12.02 by castAway

See also: http://gridbug.de/random/moonphase.html

Janis

On Monday, February 20, 2023 at 5:22:17 AM UTC-6, Janis Papanagnou wrote:
> On 20.02.2023 07:29, Yosemite Sam wrote:
> >> On Sunday, February 19, 2023 at 4:38:03 AM UTC-8, Jukka Lahtinen wrote:
> >>> I just noticed last night that Nethack 3.6.7 has been released [...]
> >
> > I noticed the new moon started later in 3.6.7. It was only a matter
> > of an hour or two.
> I just had a peek into hacklib.c phase_of_the_moon(). Two observations;
> the code seems unchanged, and the arithmetic seems to be of accuracy
> 'days' (not hours).
>
> I would assume that differences of 'hours' (as you observed) might be
> affected by the time zone configuration, since I think it's local time
> based? (But currently I haven't examined it deeper.)
>
> There's also an offset-discrepancy between reality and implementation;
> in the past (in a concrete new moon instance) I observed 1 day offset.
>
> See also: thread posted in re.games.roguelike.nethack dated 2022-11-07
>
> See also: thread "Datediff script" in comp.unix.shell by castAway,
> specifically (for example) some results from 2022.12.02 by castAway
>
> See also: http://gridbug.de/random/moonphase.html
>
> Janis

It was probably the time zone configuration. There was about an hour of
difference, and I can imagine I haven't configured daylight savings time
into the nethackrc.

On 20.02.2023 16:26, Yosemite Sam wrote:
> On Monday, February 20, 2023 at 5:22:17 AM UTC-6, Janis Papanagnou wrote:
>> On 20.02.2023 07:29, Yosemite Sam wrote:
>>>
>>> I noticed the new moon started later in 3.6.7. It was only a matter
>>> of an hour or two.
>>
>> I would assume that differences of 'hours' (as you observed) might be
>> affected by the time zone configuration, since I think it's local time
>> based? (But currently I haven't examined it deeper.)
>
> It was probably the time zone configuration. There was about an hour of
> difference, and I can imagine I haven't configured daylight savings time
> into the nethackrc.

Hmm.. - can that be configured in Nethack? I thought it would be either
a system configuration (if you play locally) or the difference between
your local time and the server's local time (if you play remotely on a
server).

Janis

On Tuesday, February 21, 2023 at 3:12:16 AM UTC-6, Janis Papanagnou wrote:
> On 20.02.2023 16:26, Yosemite Sam wrote:
> > On Monday, February 20, 2023 at 5:22:17 AM UTC-6, Janis Papanagnou wrote:
> >> On 20.02.2023 07:29, Yosemite Sam wrote:
> >>>
> >>> I noticed the new moon started later in 3.6.7. It was only a matter
> >>> of an hour or two.
> >>
> >> I would assume that differences of 'hours' (as you observed) might be
> >> affected by the time zone configuration, since I think it's local time
> >> based? (But currently I haven't examined it deeper.)
> >
> > It was probably the time zone configuration. There was about an hour of
> > difference, and I can imagine I haven't configured daylight savings time
> > into the nethackrc.
> Hmm.. - can that be configured in Nethack? I thought it would be either
> a system configuration (if you play locally) or the difference between
> your local time and the server's local time (if you play remotely on a
> server).
>
> Janis

I think the fact that it is based on local time explains the phenomenon I
described. It is a matter of five or six hours difference between the two
machines in question.