Friki 2.1.1 release candidate 1
Release Notes

Following the two "beta" releases of Friki 2.1.1, here is a third
one. I think this one is solid enough to be counted as a "release
candidate". It has been tested on Windows, Linux and MacOSX, with
all the major browsers on each platform, and using a selection of
different character sets. Thanks to the great work from a crack team
of beta testers we seem to have nailed all the remaining oddities.
If there are no "show stoppers" in the next seven days, I will
release this as 2.1.1 final. After that, any further changes and
fixes will be part of Friki 2.1.2.

Note that this release does not include an optimised "OndeDotFour"
binary release. I still haven't got to the bottom of some annoying
exceptions which only occur when using the java.util.regex regular
expression APIs, so I recommend that all users install the
OneDotThree release (which uses the more robust "jregex" regular
expression code). If you wish to experiment, you can still build
your own OndDotFour release if you download the source release.

Note also that some minor changes have been made to the way page
files are named on disk, to the default page templates, and to the
default settings in "friki.prp". So if you have installed any of the
previous alpha or beta releases of Friki 2.1.1, I suggest you either
move/delete the contents of your Friki data directory, or give this
new release a fresh, empty one. You should be able to copy back in
any pages which use only basic ASCII characters in their names, but
I'd advise creating a new page for any which use non-ASCII letters
in their names, and pasting-in or recreating the content.  Sorry.

See the change notes for details of what has changed since the
previous release.