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

Re: The ". $" notation



On 3/19/13, David A. Wheeler <dwheeler@xxxxxxxxxxxx> wrote:
> Alan Manuel Gloria:
>> There's a subtle problem with ". $" though...
>>
>> First:
>>
>> a $ b
>> ===>
>> (a b) ; as described in the rationale for SUBLIST.
>>
>> Then:
>>
>> a . $ b
>> ===>
>> (a . b)
>>
>> ???
>
> Bug, and I think fixed.  Current development version of
> "unsweeten" and the ANTLR implementation seem to work.
> The Scheme implementation now does this:
>
> $ printf 'a . $ b\n\n' | ./unsweeten
> (a b)
>
>
> Supporting ". $" does have a slight annoyance; it creates a minor
> ambiguity in the grammar (basically, like a "dangling else" clause in many
> langauges).
> I've been able to avoid those so far.  It's not a crisis, because we can
> easily make it go first, but it is an annoyance.

Err mostly I was pointing out that this is actually an inconsistency -
SUBLIST normally does not wrap a single datum after it in an extra
layer of parens, while "." essentially removes an extra layer of
parens.  So maybe ". $" notation isn't as good as I thought.

Sincerely,
AmkG