<div dir="ltr">Thanks for this but after reading this for several times, I still have difficulty understanding the actual problems. Maybe it is because I never faced any of these issues. So I would be grateful if you could explain these issues more clearly so that we can discuss them. On the other side (Knowing that luatex is different to XeTeX), I guess that unibidi is done by the ICU and I am not sure if any of these problems exist in XeTeX too, in XeTeX you have some kind of unicode bidi if (1) you use any opentype font which uses ICU and (2) the direction of each, say English word in RTL is LTR but not two words, direction of numbers is LTR and obviously specific characters are mirrored.<br>
<br><div class="gmail_quote">On Tue, Jul 12, 2011 at 11:51 PM, Taco Hoekwater <span dir="ltr"><<a href="mailto:taco@elvenkind.com">taco@elvenkind.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<br>
Hi,<br>
<br>
Now that the Unicode BiDi algorithm implementation discussion<br>
seems to restart, it is a good time to ask the people involved<br>
to solve the non-trivial problem of adjusting the algorithm<br>
to cater for the inclusion of<br>
<br>
* out-of-line node objects (\setbox, \box, \copy),<br>
* out-of-line tokens (like alignment templates and macro expansions),<br>
* vertical typesetting,<br>
* TeX grouping levels,<br>
* Interfering nodes (math, \special, \rule etc)<br>
<br>
Thanks in advance,<br><font color="#888888">
<br>
Taco<br>
</font></blockquote></div><br></div>