[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
bug#47235: 28.0.50; Suboptimal display of decomposed character sequences
From: |
Eli Zaretskii |
Subject: |
bug#47235: 28.0.50; Suboptimal display of decomposed character sequences |
Date: |
Thu, 18 Mar 2021 17:09:15 +0200 |
> From: Philipp Stephani <p.stephani2@gmail.com>
> Date: Thu, 18 Mar 2021 15:35:40 +0100
>
> On some systems (e.g. my current GNU/Linux system), "A" followed by
> U+0308 COMBINING DIAERESIS isn't displayed as a single glyph. See the
> attached screenshot, which compares this character sequence with the
> canonically-equivalent precomposed form "Ä". The precomposed glyph
> form looks better, and it would be great if Emacs could use it for the
> composite variant as well.
We do something like that for Hebrew, see hebrew-font-get-precomposed
and its call in hebrew-shape-gstring. That is based on data structure
we maintain manually. We could do that for other scripts as well,
but:
. if the glyph for the precomposed is not available in the font, you
will see "tofu", which is arguably worse than what happens today
. the HarfBuzz folks were really unhappy about such "manual
compositions", and said we should remove that code and rely on the
shaping engine to do this