[XeTeX] Polytonic greek and XeLaTeX

[Jonathan Kew]

On 31 Dec 2007, at 12:13 am, Johannes Engel wrote:

> Jonathan Kew schrieb:
>> Maybe it's not relevant here, but I don't know what you mean by a
>> "composed" glyph compared to a "prepared" one.
> OK, see the difference:
> This one is a composed one: Ὀ (i.e. a capital Omicron and a psili),
> this one is a prepared one: Ὀ (i.e. one letter Omicron with psili).

OK, I figured you must mean something like that, but wasn't clear  
about the terms; I'd usually refer to these as "decomposed" versus  
"precomposed" representations. They look identical when I read your  
email, though. (That would depend on the capabilities of the email  
client and the fonts it's using. In principle, these two  
representations *should* be indistinguishable, but many systems and  
fonts don't quite achieve that.)

Interestingly, when I used copy-and-paste to get the text from your  
email into an editor, both of these lines (and your earlier sample)  
ended up with the precomposed character -- so something (either my  
mail client or editor) seems to be applying Unicode normalization  
rules (NFC). That's why the original test worked for me with Minion Pro.

So the explanation is that your text contained decomposed sequences,  
but Minion Pro doesn't support these. Indeed, looking into the font,  
I see that it lacks the combining marks from the Unicode 03xx block.  
So it will only work with accented Greek (or Latin, for that matter)  
if the data uses precomposed characters.

JK