XHTML, whitespace fixes, beautifications, etc.)

hook. Stupid, stupid, stupid.

hacks that were in there before.  Tried to reduce the diff to the Zope 2.9 code
a little and moved some content type guessing code to PageTemplateFile where it
really belongs (that also avoids a circular import problem).

macro access happens.

as a base class)

old Zope 2 code.

traversal works.

zope.pagetemplate.pagetemplate.PageTemplate, we need to provide some Zope 2
specifics still.  This isn't a big deal, though, the wrapper code is small.

as an instance by Expressions.py). Minor style cleanup in ZRPythonExpr.py.

Some docstring fixes.

the original thing

aren't evaluated by RestrictedPython. Five needs this.

in Zope 2.10 now, so we could just as well register it here. Gets rid of some
stupid code duplication in Five as well...

* Made ZRPythonExpr subclass the one from zope.tales.pythonexpr.
* Made Expressions use ZRPythonExpr instead of the standard one
  from zope.tales.pythonexpr.

the old-style expressions. We now use the ones from Zope 3. If we can. Not sure yet :)

this still needs further clean up, is to say consolidation.
also, Five needs to be fixed regarding its (very funky) usage of Expressions.py

Basically, zope.app's path expressions use ITraversable adapters to
traverse through the object graph. This is the only sane thing to do,
really. However, we still want to take care of Zope2's OFS.Traversable
which most promimently allowes the __bobo_traverse__ hook.  (The irony
is that OFS.Traversable uses Zope 3-style *URL* traversal these days;
yes, *URL* traversal, not *object graph* traversal).

Still confused?

Basically, if we encounter something that is OFS.Traversable'ish, use
restrictedTraverse. Use Zope3-style object graph traversal in all other
cases....

correct argument order in useful implementation