Re: [Json] Limitations on number size?

[Nico Williams <[e-mail]>]

On Mon, Jun 3, 2013 at 5:10 PM, Paul Hoffman <[e-mail]> wrote:
> Greetings. I'm forwarding this fragment with permission from the AppsAWG mailing list. It might warrant clarification in -bis document.
>
> Begin forwarded message:
>
>> From: Nico Williams <[e-mail]>
>> On Mon, Jun 3, 2013 at 10:21 AM, Manger, James H
>> <[e-mail]> wrote:
>>>> Imagine JSON had bignums (doesn't, but imagine).
>>>
>>> JavaScript might not, but JSON actually does have indefinite length bignums. A JSON number can have any number of decimal digits.
>>
>> Yeah, but it is constrained to IEEE 754 64-bit floating point values, no?
>
> I don't see any limitation in RFC 4627, but it would be interesting if other folks here believe there is a limitation as Nico does.

Nor does the draft ECMAScript 6 say that.  But a) ECMAScript 6 does
restrict ECMAScript numeric values to IEEE 754 64-bit values
(including NaN, positive infinity, and negative infinity).

I have been told by the maintainer of at least one JSON parser/encoder
that they will not add support for anything other than IEEE 754 64-bit
values.  I'm not sure if I should name the project, but I'll say this:
it's damn near the best C JSON library I've yet seen.  The point is
that the perception that JSON is a subset of JavaScript (ECMAScript)
is out there, and that because the latter is constrained as described,
so should the former.

I think that JSON should be taken as so constrained by ECMAScript, but
more importantly, any constraint, or even no constraint, should be
stated explicitly.

Nico
--