Re: [Json] Consensus call: ABNF nits

[Norbert Lindenberg <[e-mail]>]

On Jun 21, 2013, at 9:40 , Paul Hoffman wrote:

> There are two proposals for dealing with nits in the ABNF:
>  0) Leave the ABNF in the current draft as-is
>  1) Change the ABNF in the draft to use the values in "Proposal 1" below
> Note that later proposals might make technical changes to the ABNF itself; those proposals are not what we are discussing in this consensus call.
> 
> Please respond to this message with a list of proposals you could accept, ordered from highest to lowest. Do not list proposals you cannot live with. If you cannot accept any of the proposals, please respond and say why.

Below are comments on three issues with proposal 1. The second issue, if not addressed, would make proposal 1 unacceptable to me. Otherwise I prefer proposal 1 over the old version.

> char = unescaped / (
>    escape (
>        %x22 /           ; "    quotation mark  U+0022
>        %x5C /           ; \    reverse solidus U+005C
>        %x2F /           ; /    solidus         U+002F
>        %x62 /           ; b    backspace       U+0008
>        %x66 /           ; f    form feed       U+000C
>        %x6E /           ; n    line feed       U+000A
>        %x72 /           ; r    carriage return U+000D
>        %x74 /           ; t    tab             U+0009
>        %x75 4HEXDIG ) ) ; uXXXX                U+XXXX

As noted earlier, the last line isn't quite correct - \uXXXX can not only represent U+XXXX, but also be part of U+YYYYYY, where YYYYYY depends on a preceding or following \uXXXX sequence.

Proposal: Replace "U+XXXX" with "U+XXXX or part of U+YYYYYY".

> escape = %x5C              ; \  U+005C
> quotation-mark = %x22      ; "  U+0022
> unescaped = %x20-21 / %x23-5B / %x5D-10FFFF  ; any Unicode scalar
>   ; value, except those that must be escaped (C0 control
>   ; characters, "QUOTATION MARK" [U+0022], and "REVERSE SOLIDUS"
>   ; [U+005C])

Describing "unescaped" as "any Unicode scalar value, except..." is wrong. The correct term is "Unicode code point". Unicode scalar values do not include the range U+D800 through U+DFFFF.
http://www.unicode.org/glossary/#unicode_scalar_value

> HEXDIG = DIGIT / %x41-46 / %x61-66   ; 0-9, A-F, or a-f
>       ; HEXDIG equivalent to HEXDIG rule in [RFC5234]

To complete the effort of disambiguating Unicode characters in the grammar, the comment should include "U+0030 through U+0039, U+0041 through U+0046, U+0061 through U+0066". We don't want them to be confused with ０-９, Ａ-Ｆ, or ａ-ｆ.

Norbert