[XeTeX] Unicode roman numerals or not?

[Sven Siegmund]

Hello,

I wanted to test, whether XeTeX typesets unicode roman numerals
(U+2160–U+217F) when I let enumerate in roman numerals.

To test this, I wrote:

\documentclass[a4paper,twoside]{article}
\usepackage[english]{babel} % english typesetting
\usepackage{xunicode} % handle Unicode
\usepackage{xltxtra} % XeTeX extras
\usepackage{fontspec} % access TTF/OTF fonts
\setmainfont[Mapping=tex-text]{DejaVu Serif} % use this font and
handle `` "" -- ---
\setmonofont[Mapping=tex-text,Scale=MatchLowercase]{DejaVu Sans Mono}
\setsansfont[Mapping=tex-text,Scale=MatchLowercase]{DejaVu Sans}
\setmathrm[Mapping=tex-text]{DejaVu Serif}

\begin{document}

\renewcommand{\labelenumi}{\Roman{enumi}}
\begin{enumerate}
    \item one
    \item two
    \item three
\end{enumerate}

\end{document}

After I xelatexed this and opened the PDF in Acrobat Reader 8, it has
found the enumeration numbers when I searched for characters U+2160,
U+2161 or U+2162.

But it has found them equally well also when I have searched for a
single, double or triple capital I (U+0049). Now I don't know why this
happens. Is Adobe Acrobat 8 so smart and looks also for unicode roman
numerals when searching for corresponding letters or is XeTeX so smart
and internally offers both in the source of the PDF?

S.