Re: Identifying the face between STRETCH and right fringe.

[Eli Zaretskii]

> Date: Tue, 20 Nov 2018 08:39:16 -0800
> From: Keith David Bershatsky <address@hidden>
> 
> Step 3:  Observe that the result is different on Emacs --with-ns, versus 
> --with-x and also on a Windows machine.  Specifically, the STRETCH on an NS 
> platform is seen spanning all the way to the right fringe.  On an X11 and NT 
> platform, the STRETCH is only visible for the width of the STRETCH (depicted 
> in the dump-glyph-row for each platform below).

It's not the stretch glyph that spans all the window width, it's the
face of the stretch that gets extended to the window end.  The stretch
glyph is still 7-character wide:

> ;;; NS
> 
> Row     Start       End Used oE><\CTZFesm     X    Y    W    H    V    A    P
> ==============================================================================
>  11       384       385    2 010010101000     0  176   56   16   16   12   12
>            -1        -1           0
>            -1        -1
>            -1        -1
>  Glyph#  Type       Pos   O   W     Code      C Face LR
>       0     C       384   B   7 0x0000bb      .   30 00
>       1     S       384   B  49 0x000000          29 00
                               ^^
49 = 7 * 7

> QUESTION #1:  In terms of identifying the face between the STRETCH and the 
> right fringe, how can I programmatically know the difference between the NS 
> situation, versus the NT and X11 situation?
> 
> QUESTION #2:  Is the difference in behavior between the different platforms 
> "a bug", and should the X11 and NT ports be "fixed" so that they behave like 
> the NS port in this situation?

I will look into this when I have time.  I don't yet know whether the
bug is on NS or on the other 2 platforms, though I tend to think the
former.