[XeTeX] Tonos v. oxia

[Philip Taylor (RHBNC)]

Consider the following code :

\font \greekfont = "Palatino Linotype"
\greekfont
U+1F7D : GREEK SMALL LETTER OMEGA WITH OXIA — ώ \par
U+03CE : GREEK SMALL LETTER OMEGA WITH TONOS — ώ \par
\end

The final (Greek) character on each of the two key lines is the character 
described by the preceding text. Yet after processing using XeTeX, the PDF 
output contains not (as expected) two different omegas bur instead two 
identical ones, both being U+1F7D : GREEK SMALL LETTER OMEGA WITH OXIA — ώ.  
Note that Unicode normalisation should not be taking place as I have 
deliberately left XeTeXinputnormalisation at its default value —

\XeTeXinputnormalization ⟨Integer⟩

Specify whether XƎTEX is to perform normalisation on the input text and, if so, 
what type
of normalisation to use. See http://unicode.org/reports/tr15/ for a description 
of Unicode
normalisation. The ⟨Integer⟩ value can be:

0 (default) do not perform normalisation.
1 normalise to NFC form, using precomposed characters where possible instead 
[of] base characters with combining marks.
2 normalise to NFD form, using base characters with combining marks instead of 
precomposed characters.
What is going on here, please ?

--
Philip Taylor

This email, its contents and any attachments are intended solely for the 
addressee and may contain confidential information. In certain circumstances, 
it may also be subject to legal privilege. Any unauthorised use, disclosure, or 
copying is not permitted. If you have received this email in error, please 
notify us and immediately and permanently delete it. Any views or opinions 
expressed in personal emails are solely those of the author and do not 
necessarily represent those of Royal Holloway, University of London. It is your 
responsibility to ensure that this email and any attachments are virus free.

Re: [XeTeX] Tonos v. oxia

[Ulrike Fischer]

Am Sun, 13 Nov 2022 09:58:28 + schrieb Philip Taylor (RHBNC):

> Consider the following code : 
> 
> \font \greekfont = "Palatino Linotype" \greekfont U+1F7D : GREEK
> SMALL LETTER OMEGA WITH OXIA — ώ \par U+03CE : GREEK SMALL LETTER
> OMEGA WITH TONOS — ώ \par \end 
> 
> The final (Greek) character on each of the two key lines is the
> character described by the preceding text. Yet after processing
> using XeTeX, the PDF output contains not (as expected) two
> different omegas bur instead two identical ones, both being
> U+1F7D : GREEK SMALL LETTER OMEGA WITH OXIA — ώ.  

> What is going on here, please ?

It is the font. With arial I get two different glyphs. And in the
lua-file of pala.ttf one can find the glyphs in the duplicates
table:

 ["duplicates"]={
   ...
  [974]={
[8061]=true,
   },
  ...
}

(but luaotfload normalizes by default and outputs always 03CE also
for arial anyway)
 



-- 
Ulrike Fischer 
http://www.troubleshooting-tex.de/

Re: [XeTeX] Tonos v. oxia

[Philip Taylor (Hellenic Institute)]

On 13/11/2022 10:34, Ulrike Fischer wrote:

It is the font. With arial I get two different glyphs.

Facinating — so do I (see below).  Thank you Ulrike — much appreciated.

  And in the lua-file of pala.ttf one can find the glyphs in the duplicates
table:

  ["duplicates"]={
...
   [974]={
 [8061]=true,
},
   ...
}

(but luaotfload normalizes by default and outputs always 03CE also
for arial anyway)


OK, not familiar with lua font files, so the above is not immediately 
meaningful to me, but the very fact that it is font-dependent is of 
itself very useful information.


\font \greekfont = "Palatino Linotype"
\font \greekfont = "Arial Unicode MS"
\greekfont
U+1F7D : GREEK SMALL LETTER OMEGA WITH OXIA — ώ \par
U+03CE : GREEK SMALL LETTER OMEGA WITH TONOS — ώ \par
\end

--
/Philip Taylor/