bug#69454: Not possible to insert an empty vtable

[Eli Zaretskii]

> Date: Mon, 11 Mar 2024 14:57:20 -0500
> Cc: 69454@debbugs.gnu.org
> From: Adam Porter <adam@alphapapa.net>
> 
> Hi Eli,
> 
>  > P.S. Adam, I took the liberty of adding you to this discussion, since
>  > you seem lately to be interested in vtable.
> 
> Thanks for adding me.  Indeed, I've found vtable to be very useful in my 
> new listen.el package.
> 
> On 3/9/24 02:54, Eli Zaretskii wrote:
> >> Date: Wed, 28 Feb 2024 15:29:11 +0100
> >> From: Eric Marsden <eric.marsden@risk-engineering.org>
> >>
> >> Hello,
> >>
> >> The following generates an error. It seems to me that it would be
> >> preferable to insert the header line and show zero rows for the vtable.
> >>
> >>      (require 'vtable)
> >>      (make-vtable :columns '("tweedle" "dum") :objects (list))
> >>
> >> Debugger entered--Lisp error: (wrong-number-of-arguments #<subr max> 0)
> >>     max()
> >>     apply(max nil)
> >>     seq-max(nil)
> > 
> > I'm not sure we want to support zero-size vtables.  A better error
> > message would be nice, though.  What do others think?
> 
> I tend to agree with Eric that it would be helpful if vtable could 
> handle having an empty objects collection value to insert, because it 
> saves the application from having to wrap the rather large `make-vtable' 
> form in a `when' block, like here:
> 
> https://github.com/alphapapa/listen.el/blob/e9ea67350cf3b6cd870561c5e52d4b5255b04d34/listen-queue.el#L135
> 
> Also, it's possible that, after inserting a vtable, the collection of 
> objects may be modified so that the collection is empty--then if the the 
> vtable is reverted, it should be able to handle the case of the 
> collection being empty.
> 
> AFAICT there's not much the application could do to avoid errors in that 
> case, other than working outside of vtable's revert API and calling the 
> function that tested the collection and conditionally inserted the 
> vtable in the first place--in which case the vtable revert API would 
> seem useless.
> 
> So IMO, when inserting or reverting a vtable, vtable ought to check 
> whether the collection is empty; and if so, handle it gracefully, 
> meaning that an "empty vtable" (whatever that would mean; maybe just one 
> line of text saying that it's an empty collection) would still be 
> inserted, and that if the collection became non-nil, it could be 
> reverted and displayed properly.

Thanks.  Would you or Eric like to submit a patch along these lines?