Re: bug in forward-visible-line: Patch

[Luc Teirlinck <teirllm@dms.auburn.edu>]

Miles Bader wrote:

   BTW, one problem with the name `visiblep' is that it could be confused
   to mean the same thing as `pos-visible-in-window-p'.  OTOH, I don't
   think I don't think `invisiblep' has this problem, since the term of
   `invisible' in emacs is strongly associated with the invisible property.

As I already pointed out in my reply to David Kastrup, I failed to
notice at the time when I proposed this function that there already
was a function, `line-move-invisible', with essentially the same
functionality as invisiblep.  This function currently works fine in
the absence of invisibility properties that are lists with two or more
elements, but malfunctions in the presence of such invisibility
properties.  This is a bug that should be fixed and can easily be
fixed by either using the new version of `line-move-invisible' which I
proposed in a later posting or by using the C functionality and making
`line-move-invisible' a primitive, as Stefan suggested.  The advantage
of the latter solution would be that if `line-move-invisible' used the
same function as display to determine visibility (Stefan seems to
suggest that this would be easy to implement), then it would actually
be "theoretically impossible" for `line-move-invisible' to get it
wrong.

Sincerely,

Luc.