Search code examples
schememit-scheme

contradicting results for environment-parent and environment-has-parent? in MIT-scheme

From the official MIT-scheme texinfo manual:

Environments are first-class objects in MIT/GNU Scheme. An environment consists of some bindings and possibly a parent environment, from which other bindings are inherited.

... some other info omitted ...

-- procedure: environment-has-parent? environment
Returns #t if environment has a parent environment; otherwise returns #f.

-- procedure: environment-parent environment
Returns the parent environment of environment. It is an error if environment has no parent.

However, in MIT-scheme 10.1.10 on Debian buster 10, I do:

(environment-has-parent? (the-environment))

;Value: #t

(environment-parent (the-environment))

;Value: #f

These two results seem to contradict. (the-environment) has a parent, but its parent is false. What gives? I was expecting the second call to return some environment and print it, like ;Value: #[environment 12] or something, but instead it returns a boolean.

Edit: as one commenter pointed out below #f is an environment and we can get its bindings and everything. Of course, the bindings are not stored in the #f object, but still, that is an incredibly strange and undocumented convention they are using.

Solution

  • In MIT Scheme 10.1.10, #f is effectively treated as the top-level environment, and this is simply a convention that is undocumented in the manual.