[XeTeX] Hebrew with XeTeX

[Erwin Ochsenmeier]

Dear Jonathan,

You are right about the hiriq-yod. I checked the font mapping and  
FB1D is mapped as displayed in XeTeX. Question is then why it is  
rendered differently in Mellel, unless they use the combination and  
not the FB1D, which would be interesting. Still the combination as  
rendered by Mellel is visually similar to Biblia Hebraica. Maybe I  
should sent a link of this thread to tiro.

E. Ochsenmeier


Message: 4
Date: Tue, 19 Feb 2008 14:57:05 +0000
From: Jonathan Kew <jonathan_kew at sil.org>
Subject: Re: [XeTeX] Hebrew with Xetex
To: Unicode-based TeX for Mac OS X and other platforms <xetex at tug.org>
Message-ID: <F92F56D5-76DA-4F4A-B84C-14137E6EC100 at sil.org>
Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed

On 19 Feb 2008, at 11:50 am, Erwin Ochsenmeier wrote:

 > Here is a sample file with the code used and the problems highlighted

With regard to the position of the HIRIQ dot under YOD, I believe
XeTeX is using the precomposed glyph provided in SBL Hebrew (and
mapped to U+FB1D). The dot here is raised close to the YOD, whereas
the attachment point for a separate dot glyph is kept low.

XeTeX is justified in doing this because the Unicode character U+FB1D
is defined as being canonically equivalent to the sequence <U+05D9, U
+05B4>. It is an inconsistency in the font if the rendering of the
sequence and the precomposed character is not the same; the two
representations of the text are by definition synonymous and should
be indistinguishable.

Regarding the positioning of METEG, I'm not sure what is going on
here; it would require some analysis of the OpenType tables in the
font to see why it's not behaving as expected.

JK