Re: standard quotes for PDF

[Han-Wen Nienhuys]

In article <address@hidden>,
Karl Berry <address@hidden> wrote:
>    kept getting bugreports by people that cut & paste coding and were
>
>It seems odd to me that I don't recall ever seeing such a report about
>any other Texinfo document.  Is there something different about the
>Lilypond manual that makes the pasting behave differently?  E.g., that
>UTF-8 stuff which Werner just posted?
>
>    Cut & pasting from evince PDF into gedit UTF-8 
>
>Sorry, I'm still confused.  This isn't a function of the pdf viewer
>and/or editor being pasted into, rather than the pdf file?

Thanks for the attachments.

It's for sure that TeX and texinfo have a special mechanism to
substitute U+2019 for ASCII 27; I'm not aware of generic substitution
mechanisms translating Unicode back to ASCII, but I'm no expert. 

>I tried this Texinfo code:
>
>    @code{g'code}
>    @example
>      g'example
>    @end example
>    @verbatim
>      g'verbatim
>    @end verbatim
>
>Ran it through pdftex, viewed it with xpdf, cut and pasted into emacs,
>and got the expected 0x27 ' characters (in all cases).

Yes, it does that with Xpdf over here as well; for some reason I can't
get it to paste into emacs, but it does work when pasting into Gedit.

However, with Adobe Reader, I get curly quotes pasted into Gedit.  The
same happens with KPdf (the KDE PDF viewer) and Evince (the Gnome
one.)  Given that Adobe defined the PDF standard, I consider their
behavior standard, and I suspect some kind of bug or misfeature in
Xpdf.

Maybe Werner can shed some light on this matter?