Re: "Final" version of tty child frames

[Eli Zaretskii]

> Date: Fri, 17 Jan 2025 16:08:04 +0100
> Cc: gerd.moellmann@gmail.com, jared@finder.org, emacs-devel@gnu.org
> From: martin rudalics <rudalics@gmx.at>
> 
>  > Anyway, this is wrong for bidirectional display, where a glyph to the
>  > right of another one could have its charpos smaller.
> 
> This must be handled in row->reversed_p.

No, these are two different aspects of the bidirectional display.
row->reversed_p means that the whole glyph row was reversed for
display, so it starts at the right window edge and proceeds to the
left.  What I'm talking about is reordering of characters for display
even if row->reversed_p is zero.  This reordering could cause the
glyphs corresponding to 10 characters in buffer positions 1 to 10 to
be in the following order on display:

  1 2 8 7 5 6 4 3 9 10

Which means you cannot assume any relation between the right/left
direction on display and increasing/decreasing order of buffer
positions.