Formal comment #194 (enhancement)

hash-table-mutable? vs hash-table-immutable?
Reported by: 	Daniel Villeneuve
Version: 	5.92

SUMMARY

The predicate about mutability of hash-tables would be better defined
as hash-table-immutable?.

DESCRIPTION

I suggest that the predicate to test about mutability/immutability of
hash-tables be named hash-table-immutable? for the following reasons:

a) This is the property that is given to hash-table-copy.

b) Among all Scheme collections, immutability is the exception, and
   this is what we want to single out.

c) I think that the most frequent idiom for this predicate is

(define (some-function-modifying-a-hash-table! ht)
  (if (not (hash-table-mutable? ht))
     (error ...)))

which can be written a little more succinctly as

(define (some-function-modifying-a-hash-table! ht)
  (if (hash-table-immutable? ht)
     (error ...)))

PROPOSAL

Replace hash-table-mutable? by hash-table-immutable?.


RESPONSE:

The editors believe that this choice is almost purely subjective, 
and do not plan to adopt the comment's suggestion.  However, the 
consistency issue raised in the comment's point (a) will be addressed 
by changing the specification of hash-table-copy.  In the next draft
of the report, hash-table-copy will return an immutable hash table
unless a second 'mutable' argument is provided and is true.