Re: [Json] About JSON equality

[Francis Galiegue <[e-mail]>]

On Wed, Feb 20, 2013 at 12:10 AM, Joe Hildebrand (jhildebr)
<[e-mail]> wrote:
> Are you proposing that this be added to 4627bis, or that it be in a
> separate document?
>

That is what I am asking. I believe the more JSON is used, the more
people will want to "compare" JSON values, whatever they may be (and a
hint again: please make RFC 4627bis allow to transfer _any_ JSON
value, not just arrays and objects).

There are two problems here: strings and numbers.

For strings, Unicode code points and their representation in whatever
encoding should be considered equal, and parsers should be able to
handle whatever they are fed with.

For numbers, there is the much more annoying limitation that all
humanly representable numbers just cannot be represented in
programming languages, and that JSON (at least to my eyes) should not
feel compelled to represent NaN (since it is, essentially, language
dependent) nor infinity (since it is an abstract concept and not even
a number).

But both of these problems can be at least formulated as being present
-- the JSON Schema validation spec does. The real question becomes,
will future uses of JSON rely on value equality?

-- 
Francis Galiegue, [e-mail]
Try out your JSON Schemas: http://json-schema-validator.herokuapp.com