Defines what happens when relative JSON pointer starts with zero

[Relequestual]

Resolves #1121

[gregsdennis]

https://tools.ietf.org/id/draft-handrews-relative-json-pointer-00.html#rfc.section.3

The spec already says that relative JSON pointers must start with positive integers.

Maybe we just remove "non-negative".

[Relequestual]

The spec already says that relative JSON pointers must start with positive integers.

Maybe we just remove "non-negative".

It quantifies in the ABNF...

"0", or digits without a leading "0"

This is also shown in the proceeding examples...

Starting from the value "baz" (inside "foo"), the following JSON strings evaluate to the accompanying values:
"0" "baz"
...

It's essentially a no-op, but is valid.
The use is when apending # to get the key or index of the location.

(FYI latest is https://datatracker.ietf.org/doc/html/draft-bhutton-relative-json-pointer-00)

[gregsdennis]

Then we should either change the section I linked or change the ANBF. They conflict. Personally, I'm happy with either.

(It seems "non-negative" is the current term in the latest.)

[Relequestual]

You do not believe zero is a non-negative integer?
We changed the language from "positive" to "non-negative" so as to include zero.

[gregsdennis]

My concern was alignment. I didn't care which it was. Positive and non-negative are both fine as long as the spec is consistent.

[Relequestual]

It is consistent in the current version. You were viewing the previous 😅

[Relequestual]

As there has been little interest, I'm removing this PR and associated issue from the milestone.
It can be added back again should anyone show interest, but it shouldn't be a blocker.

[handrews]

While I find the phrasing slightly awkward, everything else I can think of is also slightly awkward. This is definitely correct on a technical level, should we just go ahead and merge this?