什么是“空”，“程序”？ 和“符号？” 在下面的代码中意味着什么？

Question

I am going through the following documentation for implementing a Lisp Interpreter in Python: http://norvig.com/lispy.html

In the standard_env function, a dictionary has been defined for mapping certain symbols or variables to their corresponding functions or values. However, I am not being able to understand the following entries in the env dictionary:

env = {}
env.update({
    'equal?':  op.eq, 
    'list?':   lambda x: isinstance(x,list),
    'null?':   lambda x: x == [], 
    'number?': lambda x: isinstance(x, (int, float)),   
    'procedure?': callable,
    'symbol?': lambda x: isinstance(x, str),
})

What does the '?' at the end of the key mean? Could you please provide some examples where these entries will be used?

Answer 1

Syntactically ? is just one of the many characters allowed in identifiers in Lisp. So it's simply part of the function name.

By convention function names ending in ? are used for functions that return boolean values.

Specifically equal? checks whether its two arguments are equal, null? checks whether its argument is an empty list and all the other ones are type checks, ie they check whether their argument is of the given type.

Could you please provide some examples where these entries will be used?

Functions returning booleans are most often used in if or cond conditions, so you might see something like this as an example of a function using null? :

(define (sum lst)
  (if (null? lst)
    0
    (+ (first lst) (sum (rest lst)))))

Answer 2

I'm not familiar with Lisp, but looking at the definitions of these terms, this looks like these are being used to check what the parameters are, ie 'list?' maps to a function that tests if the parameter is a list, 'procedure?' tests if the parameter is a procedure (or at least is callable), 'null?' tests for an empty list (perhaps this is how nulls are represented in the interpreter), etc.

Answer 3

In most Lisp dialects, identifiers (actually, symbol names) can contain punctuation like ? (or ! or - )

Conventionally, in Scheme, predicate names (naming functions returning a testable boolean) have their name ending with ? (for example list? ), and side effecting functions have their name ending with ! (for example vector-set! ); see also R5RS and SICP .

In Common Lisp, predicate names often end with P (eg listp )

These naming conventions are just useful conventions. You could name your function foo!bar?x even if it is not a predicate.