Should HTMLClass::history and _HTMLItem::history have the same signature?

Hi all:

In my user.item.html template I call:

   tal:replace="structure python:context.history(limit=limit)"

I got a traceback because history doesn't support limit if this code
is rendered using the url ending in:

  user?@template=item

If the user ends in:

  user1

the code works fine. I think what is happening is in the user1 case I
am invoking _HTMLItem's history method with signature:

  history(self, direction='descending', dre=re.compile(r'^\d+$'),
                limit=None, showall=False ):

while in the user?@template=item case I am invoking HTMLClass's
history that has the signature:

  history(self)

I think these two signatures should be compatible. I suggest changing
the HTMLClass's definition to:

  history(self, **args)

so it accepts (but ignores) any arguments that would be valid for
_HTMLItem::history().

I could also add all the args for _HTMLItem::history() to
HTMLClass::history(). However this increases the maintenance burden as
any future arguments to _HTMLItem::history() would have to be added to
HTMLClass::history().

Thoughts? 

--
				-- rouilj
John Rouillard
===========================================================================
My employers don't acknowledge my existence much less my opinions.