Took stab at expression documentation.

- Changed user exceptions to include HTML markup.
- Got rid of regular expression input types.

Updated __call__ so queries can be used in documents.

Changed name of acquire method to aq_acquire.  Explicit.acquire is
an alias.

Added latest validation rules.

Added template dict to override arguments.

  - Available for Impleicit and Explicit,

  - Does not filter names with leading underscore,

  - Accepts optional 'filter' and 'extra' arguments.  If 'filter'
    is provided, then it must be a callable object and it is
    called with four arguments:

      inst -- The object in which an attribute is found

      name -- The attribute name

      v -- The attribute value

      extra -- The 'extra' value passed to 'acquire' or None.

    The filter function should return 1 if the attribute should
    be returned by acquire and 0 otherwise.  It may also raise an
    error, in which case the error is propigated.

This is, still a work in progress.

Added level attribute used in preventing excessive recursion.

Added get method.

Made some changes to synchonize with cDocumentTemplate.

Added a get method on TemplateDicts to do lookup without
(non-DocumentTemplate) method calls.

This is, still a work in progress.

This is, still a work in progress.

Fixed some error generation bugs.

Made parse error generation a little more HTML friendly.

Added a regex-like parser that doesn't have regex's tendency to hang
the process.  Maybe I'll be able to use re in the future. ;-)

Added errQuote to aid in parse error message generation.

rewritten on every start up.