This is more or less a continuation of the thread at
http://tug.org/pipermail/xetex/2010-October/018741.html. Unfortunately, I
don't know how to include this message in that thread, since I don't
actually receive email from this list. Please excuse me. Anyway, Khaled's
suggestion from that thread worked; using the following source, Greek and
Latin characters appear in the desired font within math mode.
\documentclass{article}
\usepackage{amsmath}
\usepackage{unicode-math}
\setmainfont{Linux Libertine O}
\setmathfont{xits-math.otf}
\setmathfont[range=\mathit/{latin,Latin,greek,Greek}]{Linux Libertine O
Italic}
\begin{document}
Since $X$ is club in $\kappa$, so is the derivative $X\prime$. Therefore
each superset of $X'$ is stationary. Blah blah. It follows that $\{\xi <
\kappa : \xi \text{ is Mahlo} \} \supseteq \{\xi < \kappa : \xi \text{ is
the } \xi^{\text{th}} \text{ Mahlo} \}$ is stationary in $\kappa$, and that
$\kappa$ is the $\kappa^{\hspace{0.2mm}\text{th}} \text{ Mahlo}$.
\end{document}
Yet, as you'll see if you compile it, the primes are tiny and
misplaced. (The related thread from a few days ago, at
http://tug.org/pipermail/xetex/2010-October/018636.html, isn't helping me
here.) Furthermore, if in the final line of the preamble one uses, say,
Minion Pro instead of Libertine, the primes vanish. If the issue is that
unicode-math is looking for the prime glyph in Libertine (or Minion, as the
case may be) how can I tell it to look in XITS instead? Also, as I feared
would happen, the kerning on exponents, commas, and the like isn't ideal. Is
there a way to get around this that's more efficient than tweaking hspace as
needed?
Thanks again,
Cole
--------------------------------------------------
Subscriptions, Archive, and List information, etc.:
http://tug.org/mailman/listinfo/xetex
