Fonts & Character Sets ⇒ Kerning and alternate glyphs with Minion Pro

[AleCes]

Hem... Sorry I surely did, but I thought I would subscribe to a forum while I actually ended up having my mailbox flooded with messages from other people, didn't even know why.
Anyhow, I grasped how it works: I just sent a message through the mailing list and I'll let you all know if they can solve it.
Thank you.

[frabjous]

I did mention, twice, that it was an email list, not a forum. All the mail on the mailing list goes out to everyone on the list ,so that's why you're getting it. But it is probably the most direct way to contact the XeTeX developers, as far as I know.

[AleCes]

Yeah, sorry, I didn't pay heed. I'll let you know if something comes out if it.

[AleCes]

No bail-out in sight from TUG. Anyone here can help??? I really need that frigging font to work well!

[AleCes]

It should be something concerning the GPOS: Glyph Positioning Table.
Any hints? Do you know of a package that allows me to activate it?

[AleCes]

OK, guys from tug.org they confirmed my suspicions: it's all about XeTeX not being able to read the GPOS table. Waiting for a workaround, I'll have to adjust manually kerning of this combination, but, since it occurs frequently I'd like to know whether it's possible to define it as valid for all occurrences in the prologue. On tug.org they told me it can't be done, or rather that I should first modify XeTeX code.
Any bail-out coming from this community?

[frabjous]

I imagine the people on the XeTeX list know what they're talking about.

Now that you've brought it to their attention, there's a good chance someone will be motivated to work on it. Do keep in mind however that this not commercial software, but open source software developed by volunteers in their free time, and sometimes problems cannot be fixed right away. No doubt they'd appreciate help if you did know programming (or could learn quickly) and could take a look at the source code.

I am curious, however, whether you ever tried compiling your document with LuaLaTeX, since it supports the fontspec package too, and uses a different engine.

[AleCes]

I am curious, however, whether you ever tried compiling your document with LuaLaTeX, since it supports the fontspec package too, and uses a different engine.

Yeah, someone mentioned it on tug.org too. Does it support polyglossia too?

[frabjous]

I'm afraid I don't know. I have very little experience with it. Maybe someone else will be able to shed some light.

[AleCes]

OK, I forgot to tell you it's been finally ascertained that the problem originates from a bug in XeTeX which doesn't kern it the right way. It's definitively sure there's nothing wrong with the font, as I'd supposed from the very beginning.
Workaround: manually \kern-2.ex the relevant combination. Try this, it should work now:

Code: Select all

Code, edit and compile here:

1
2
3
4
5
6
7
8
9
10
11
12
13
\documentclass{article}
\usepackage{fontspec}
\defaultfontfeatures{Mapping=tex-text}
\setmainfont{Minion Pro}
\usepackage{polyglossia}
\setmainlanguage{serbian}
\usepackage{xunicode}
\usepackage{xltxtra}
\begin{document}
\fontsize{24}{24}\selectfont
У\kern-.2ex бити
\end{document}
 
 
הההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההההה
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

Run LaTeX here