- update to 4.2.0:
* An XML document provided as a string (i.e. decoded) now
correctly has its content encoding parsed.
* Boolean attributes are now automatically configured for
templates in non-XML mode, presuming that we're being used
to generate HTML.
* This means that the same loading mechanism can be used for
both XML-and HTML-based templates.
* Boolean attributes (those configured using the optional
`boolean_attributes` parameter) now work with $-expression
interpolation.
* Unlike content and regular attributes, a special check for a
falsy value is now done for boolean attributes, such that
boolean logic can be used for interpolation expressions
(only a truthy value will include the attribute).
This reverts a change in behavior introduced in 3.8.0.
* Fix format spec applying for f-strings.
* Drop support for Python 2.7, 3.5, 3.6.
* Add support for set- and dict comprehensions
* Remove the following functions resp. modules:
``.utils.text_()``
``.utils.unescape()``
``.compat``

* A RepeatDict no longer inherits from dict since it does not actually provide
* Added feature gate enable_comment_interpolation which controls whether
Add option restricted_namespace which controls whether to
restrict namespaces to those defined and used by the page
Fixed attribute HTML entity escaping issue where an entity such
Simplify exception tracking, reducing bytecode size