Bidi paragraph direction in terminal emulators
Richard Wordingham via Unicode
unicode at unicode.org
Sat Feb 9 17:46:38 CST 2019
On Sat, 9 Feb 2019 13:02:55 -0800
"Asmus Freytag \(c\) via Unicode" <unicode at unicode.org> wrote:
> To force Hindi crosswords mode you need to segment the string into
> each having a variable number of characters, and then assign a single
> position to them. Now some syllables are wider than others, so you
> could use the single/double width paradigm. The result may be
> somewhat legible for Devanagari, but even some of the closely related
> scripts may not fit that well.
It is also possible that whole syllables are used because there are
> To give you an idea, here is an Arabi crossword. It uses the isolated
> shape of
> all letters and writes all words unconnected. That's two things that
> may be acceptable for a puzzle, but not for text output.
> (try typing 3 vertical as a word to see the difference - it's 4x
Crosswords suffer from the need to be read vertically as well as
horizontally. Can Arabic naturally be written vertically?
In any case, Arabic typewriters exist and, so far as I understand,
work. The problem rather seems to be one of standardising the
Procrustean technique to be used. It seems from what Khaled Hosny
wrote that monospace for letters is the usual solution already.
The design difficulty for Arabic is rather that horizontally adjacency
may sometimes need to be treated as accidental rather than as an
invitation to cursively join..
More information about the Unicode