bug#58283: 29.0.50; calendar-goto-day-of-year => wrong-type-argument numberp

[Eli Zaretskii]

> From: Stephen Berman <stephen.berman@gmx.net>
> Cc: pankaj@codeisgreat.org,  58283@debbugs.gnu.org
> Date: Tue, 04 Oct 2022 12:07:54 +0200
> 
> On Tue, 04 Oct 2022 12:58:38 +0300 Eli Zaretskii <eliz@gnu.org> wrote:
> 
> >> Cc: 58283@debbugs.gnu.org
> >> From: Stephen Berman <stephen.berman@gmx.net>
> >> Date: Tue, 04 Oct 2022 10:54:55 +0200
> >>
> >> On Tue, 04 Oct 2022 09:36:49 +0530 Pankaj Jangid <pankaj@codeisgreat.org> 
> >> wrote:
> >>
> >> > Steps:
> >> >
> >> > 1. emacs -Q
> >> > 2. M-x calendar RET g D RET RET
> >> >
> >> > =>
> >> >
> >> > Debugger entered--Lisp error: (wrong-type-argument numberp nil)
> >> >   abs(nil)
> >> >   #f(compiled-function (x) #<bytecode 0x1f4df5e08d503a2b>)(nil)
> >> >   calendar-read-sexp("Day number (+/- 1-%d)" #f(compiled-function (x)
> >> > #<bytecode 0x1f4df5e08d503a2b>) nil 365)
> >> >   byte-code("\300\301\302\303
> >> > \304\18\262\1#\211\211\305W\203\30\0\306\1!S\262\1\211\307\246\305U\205.\0\211\310\246\305U?\206.\0\211\311\246\305U\262\1\2037..."
> >> > [calendar-read-sexp "Year (>0)" #f(compiled-function (x) #<bytecode
> >> > -0x1433d194a6bf3cb2>) calendar-current-date 2 0 abs 4 100 400 366 365 
> >> > "Day
> >> > number (+/- 1-%d)" make-closure #f(compiled-function (x) #<bytecode
> >> > 0x79fd4d90378a2b>) nil] 7)
> >> >   call-interactively(calendar-goto-day-of-year nil nil)
> >> >   command-execute(calendar-goto-day-of-year)
> >>
> >> The value of the `default' argument in the calendar-read-sexp call here
> >> has to be a number instead of nil.
> >> But what number?  1, 365, (random 365), ...?
> >
> > I decided that today's day number is a good default.
> 
> Seems reasonable (though displaying that default value in the prompt
> ("Day number (+/- 1-365) (default 277): ") may make innocent users of
> the command wonder about it; though maybe that's not so bad...).

Let's see if users start complaining.  My guess is that no one ever
used the default anyway, until now.

> > Please see if the latest master solves this, and thanks.
> 
> Works for me.

Thanks, closing.