Since there are hashtables and variables, it is a natural idea to think about using variables to be keys for
hastables, or values from hashtables to be keys from hastables (may be recursively).
In other word, it is natural to think about using these syntaxes
$(v)
$(h[k])
$(h[$(v)])
$(h1[$(h2[$(v)])])
$($v))
$($(v)[$(w)])
and so on .....
Kazimir should accept any of these syntaxes, but there is one restriction: there should be no space in the syntax. This means that $(h [k]) or $(h[ k ] ) are no valid syntax.