[XeTeX] Strange issue with tanwin, arabxetex

[Nathan Sidoli]

Dear François,

Thanks for looking into this.

I have been fiddling with it a bit more and I noticed that toggling the 
options between voc and novoc has an effect on this.

Here is a minimal example:

\documentclass{article}

\usepackage[novoc]{arabxetex}% toggle with voc
%\oldtanwin
\newfontfamily\arabicfont[Script=Arabic,Scale=1.4]{Traditional Arabic}% 
Traditional Arabic, Amiri
\newcommand{\A}{\textarabic}

\begin{document}

\A{_h.t"aN} $\quad$ \A{m_tl_t"aN}

\bigskip

\A{خطًا}  $\quad$ \A{مثلثًا}

\A{ًخطا}  $\quad$ \A{ًمثلثا}

\setRL
\begin{arab}
إذا أضيف إلى خَطّ مستقيم مَعْلُوم سطح متوازي الأضلاع معلوم ينقص عِن تمامه سَطْحًا 
متوازي الأضلاع معلوم الصورة، فإن أضلاع السطح الباقي معلومة
\end{arab}
\setLR

\end{document}

When we toggle to voc, short vowels that have been keyed into the 
unicode appear, but the short vowel that has been hard keyed ("aN) into 
the ascii disappears. (Also, strangely the tanwin that appears above the 
final alif, in the Persian style, is floating at the front of the word.)

The novoc option strips out any small vowels and any tanwin, but 
(strangely) does not strip out the sukun. Now, however the tanwin hard 
keyed in the ascii appears.

(If others are not seeing this behavior, I can send a .pdf)

Best,

Nathan



On 7/30/15 6:52 PM, FC wrote:
> Dear Nathan,
>
> When I had a look at ArabXeTeX last month (for the first time since 
> 2010 when I gave it over to a new maintainer) I could not find 
> anything that had changed since the last version I released (which 
> corresponds to https://github.com/fc7/arabxetex/releases/tag/1.1.4). 
> Therefore a build issue where the teckit mapping may have been 
> corrupted is more likely to be the reason, because the source files - 
> as far as I can judge - have not changed since 2010 and thus probably 
> not between TL 2014 and TL 2015.
>
> Regards,
> François
>
> 2015-07-29 13:28 GMT+02:00 Herbert Schulz <herbs at wideopenwest.com 
> <mailto:herbs at wideopenwest.com>>:
>
>     > On Jul 29, 2015, at 2:19 AM, Nathan Camillo Sidoli <sidoli at waseda.jp
>     <mailto:sidoli at waseda.jp>> wrote:
>     >
>     > Hello Everyone,
>     >
>     > Over the weekend I updated both my system (to Mac OS 10.10), and
>     TeXlive (to 2015).
>     >
>     > Now I have an issue with tanwin for indefinite accusatives in
>     ArabXeTeX. Here is a minimal example:
>     >
>     >
>     > \usepackage[novoc,fdf2noalif]{arabxetex}
>     > %\oldtanwin
>     > \newfontfamily\arabicfont[Script=Arabic,Scale=1.4]{Amiri}%
>     Traditional Arabic, Amiri
>     > \newcommand{\A}{\textarabic}
>     >
>     > \begin{document}
>     >
>     > \A{_h.t"aN} $\quad$ \A{m_tl_t"aN}
>     >
>     > \bigskip
>     >
>     > \A{خطًا}  $\quad$ \A{مثلثًا}
>     >
>     > \A{%
>     > إذا أضيف إلى خط مستقيم معلوم سطح متوازي الأضلاع معلوم
>     > ينقص عن تمامه سطحًا متوازي الأضلاع معلوم الصورة، فإّنأضلاع السطح
>     الباقي معلومة
>     > }
>     >
>     > \end{document}
>     >
>     > When I compile this document, a tanwin in the unicode input
>     (etc., خطًا) is stripped out in the output. (If I just use
>     polyglossia this does not happen, but for various reasons I am
>     committed to also using ArabXeTeX.)
>     >
>     >
>     > This behavior did not happen before the upgrade - that is, the
>     tanwin appeared as in the input. I have tried switching back to my
>     TeXlive 2014 distribution, but now this behavior is also happening
>     when I compile with that distribution as well. That is, it seems
>     to have been produced by updating my OS. I'm not sure what could
>     be causing this behavior and any suggestions would be welcome.
>     >
>     > Best,
>     >
>     > Nathan
>
>     Howdy,
>
>     On the offhand possibility that there has been an update of
>     arabxetex please also run TeX Live Utility and update all the
>     files. You can find TeX Live Utility in /Applications/TeX.
>
>     Good Luck,
>
>     Herb Schulz
>     (herbs at wideopenwest dot com)
>
>
>
>
>
>
>
>
>
>     --------------------------------------------------
>     Subscriptions, Archive, and List information, etc.:
>     http://tug.org/mailman/listinfo/xetex
>
>
>
>
>
> --------------------------------------------------
> Subscriptions, Archive, and List information, etc.:
>    http://tug.org/mailman/listinfo/xetex

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://tug.org/pipermail/xetex/attachments/20150730/d303a7fc/attachment.html>