NamesList, Code Charts, ISO/IEC 10646

[Richard Wordingham]

On Mon, 04 May 2015 19:59:26 +0300
Eli Zaretskii <eliz at gnu.org> wrote:

> > Date: Mon, 4 May 2015 17:42:26 +0100
> > From: Richard Wordingham <richard.wordingham at ntlworld.com>

> > The clear issue raised was of knowing whether a character's
> > glyph would change with the bidi context.  One solution that
> > immediately comes to mind is to display the character in a pick
> > list according to the user's locale.

> User's locale has nothing to do with bidi context, so this would be
> simply wrong.

If paragraph embedding level is determined by an overriding profile but
there is nothing explicit, should not the locale determine the
directionality?  If so, the locale will often work for indicating
whether a character should be displayed in its left-to-right form or
its right-to-left form.

Are you, for example, suggesting that a code chart in Arabic should
display U+0028 LEFT PARENTHESIS and U+0029 RIGHT PARENTHESIS using the
same glyphs as for one in the English language?

Richard.