Ecma-48 proposed styling controls update updated & math expression representation proposal update

[Kent Karlsson]

I've done an update to the ECMA-48 styling: proposed update.
https://github.com/kent-karlsson/control/blob/main/ecma-48-style-modernisation-2024.pdf
 
The major updates are:
 
1) Additional bidi modes: KISS (eliminating the odd tweeks),
   KISS2 (also eliminate the nestable bidi controls), and
   a mode where only "pure RTL runs in same script" are bidi
   reversed. The last mode (though not necessarily the control
   sequence) is suitable for editors of program *source code*
   (C++, Perl, PHP, bash, Java, ...) and data *source code*
   (HTML, XML, CSV, ..., custom). The "source code mode" is
   similar to how bidi must be handled in math expressions.
   
2) Variants for having arrow directions follow the text layout
   order (mirroring, rotation) which is given as default, and a
   temporary variant (end at bidi B or bidi S char) for not to
   mirror nor rotate arrows (arrows refer to external directions).
   (Note that mirroring and rotation data in Unicode take different
   approaches here; default (as yet) in Unicode for mirroring is
   that arrows refer to external (to the text) directions, whereas
   rotation (for CJK vertical) is that arrows refer to the text
   itself.
   
3) In Annex B, the wretched and totally unhelpful general category
   Cc is completely overridden by more useful general categories;
   similarly for bidi and line break properties.
   (Yes, "default-ignorable" should also be, well, ignored.)
 
As hinted in point 2 above, arrow mirroring data
(https://www.unicode.org/L2/L2022/22026r-non-bidi-mirroring.pdf)
would be helpful... It was called "non-bidi" because at the time
it was intended only for math expressions, as an edit support, there
must be no automatic (non-explicit) mirroring in math expressions
or source code. But as you see in point 2, I now think it should
be used also for bidi automatic mirroring; note again that automatic
(implicit) bidi mirroring does not apply at all to math expressions,
nor to source code.
 
 
I've also done an update to the (independent!) math expression
representation proposal. 
https://github.com/kent-karlsson/control/blob/main/math-layout-controls-2024.pdf
Minor updates, but I included the data in 
https://www.unicode.org/L2/L2022/22026r-non-bidi-mirroring.pdf
in an annex; I don't know what is happening, if anything, to that
poposal.
 
/Kent Karlsson

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://corp.unicode.org/pipermail/unicode/attachments/20240105/41fa21e1/attachment.htm>