[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Scheme-reports] fresh empty strings

On 2012-01-23, at 15:04, John Cowan wrote:
> In order to allocate a string as an immediate, you also have to prove
> that it's immutable.
In the absence of a procedure for mutating the length of a string, empty strings are immutable. 

I guess I'd like to see a use case where it really paid off to define (eq? (string) (string)). 
In the absence of one, I personally don't care if the result is unspecified. 

Also, it's important to note that empty strings and empty lists aren't really analogous; empty lists are of course special values that mark the cdr of a proper list; they are only accidentally `lists with no elements'. To put it another way, the internal representation of strings and vectors (not to mention hashtables) is opaque, while lists are anything but opaque. 

Eagerly anticipating my next visit from the animal control people for flogging a dead horse...

-- vincent
Scheme-reports mailing list