RE: [xml-dev] An XML document is not well-formed if encoding="..." does

[Liam R E Quin <liam@w3.org>]

On Sun, 2012-12-30 at 19:00 +0100, Hermann Stamm-Wilbrandt wrote:

> I think one of the main reasons for using UTF-8 in XML processing systems
> is that it is the default encoding in stylesheets (when  encoding="..."
> is missing in <xsl:output>.
> 
> I do not know the internals of other XSLT processors, but DataPower XSLT
> processor internal encoding is UTF-8 ([1], slide 11).

UTF-8 is massively easier to deal with than any of the others for
Unicode in C or C++, because it's compatible with the native string
handling. Anything else will often tend to be much slower, too, and to
use noticeably more memory - the C compiler can turn operations on the
built-in string (char *) type into inline assembly code, for example,
whereas a C++ object method call might involve an indirected function
call. so there tends to be a lot of pressure to use UTF-8.

My understanding is that the DataPower XSLT engine is used in embedded
routing systems in which case these issues may well be very important to
the developers. It'd be interesting to know to what extent (if any) that
was a factor.

Liam

-- 
Liam Quin - XML Activity Lead, W3C, http://www.w3.org/People/Quin/
Pictures from old books: http://fromoldbooks.org/
Ankh: irc.sorcery.net irc.gnome.org freenode/#xml