bug#37641: major/minor tick faces bleed into empty lines at the end of buffer

[Eli Zaretskii <eliz@gnu.org>]

> From: Juanma Barranquero <lekktu@gmail.com>
> Date: Tue, 8 Oct 2019 04:38:18 +0200
> Cc: 37641@debbugs.gnu.org
> 
> > Looks good, but please simplify by having 2-level if, the outer one
> > checking when to display a number, the inner which face to use for
> > that.  There's no need to test beyond_zv more than once, and AFAIR
> > beyond_zv and it->what == IT_EOB are equivalent.
> 
> beyond_zv and it->what == IT_EOB are similar, but not equivalent. I see the difference when deleting empty
> lines at the end of the buffer (specifically, when deleting from the last line, and not past the last line).
> 
> With the `what' check, the face still bleed info the first post-EOB line. Checking beyond_zv it works (see
> attached images).

Sorry, I don't think I understand what the images show me.  They seem
identical.  Which face bleeds and where?  Please point out what should
I be looking at to understand the difference.

Did you try to arrange for the last line to be a multiple of one of
the ticks as well?  Also, what happens if you use the beyond_zv test
in all the conditions, or use the it->what test in all the conditions?

IOW, I don't understand why we need two different conditions regarding
EOB for displaying a number with different faces.  What am I missing?

> As for simplifying the check, it is possible to check beyond_zv only once, with the minor downside of having
> two paths to set lnum_face (which works as the default face, and the face post-EOB).

You could simply start with

  tem_it.face_id = lnum_face_id;

and then have a series of tests for replacing that with another face
ID; it would at least save you the 'else' clause.

Thanks.