Re: [Json] [Technical Errata Reported] RFC8259 (7603)

[Rob Sayre <[e-mail]>]

On Tue, Aug 15, 2023 at 8:35 AM Tim Bray <[e-mail]> wrote:

> Um, no. The introduction says "A string is a sequence of zero or more
> Unicode characters [UNICODE].” And section 8, Strings, says "All Unicode
> characters may be placed within the quotation marks, except for the
> characters that MUST be escaped:  quotation mark, reverse solidus, and the
> control characters (U+0000 through U+001F).”  Granted, the language in 8.2
> suffers from the same problem."
>

Sorry, I pasted the correction instead of the original introduction text.

I think the correction is fine as it stands, because the document goes into
detail in the strings paragraphs. I showed 8.2 to illustrate that the WG
accepted the ugly unicode handling that's been inherited.

Even in your correction, you don't mention the quotation mark, for example.
So I don't think you need to mention the control characters either.

thanks,
Rob





>
> On Aug 15, 2023 at 8:30:14 AM, Rob Sayre <[e-mail]> wrote:
>
>> No, sorry, it's a bug.The issue here is that the introduction says:
>>
>> "A string is a sequence of zero or more Unicode code points [UNICODE]."
>>
>> but 8.2 says:
>>
>> "When all the strings represented in a JSON text are composed entirely of
>> Unicode characters [UNICODE] (however escaped)..."
>>
>> thanks,
>> Rob
>>
>>
>> On Tue, Aug 15, 2023 at 7:27 AM Carsten Bormann <[e-mail]> wrote:
>>
>>> I don’t want to prolong this exchange, but I need to point out:
>>>
>>> > On 15. Aug 2023, at 08:41, Rob Sayre <[e-mail]> wrote:
>>> >
>>> > I think you are in the rough here.
>>>
>>> Wrong script, we are not in a WG meeting here.
>>>
>>> We have an approved document.
>>> That says something (and has done so for 5.5 years since publication).
>>> There is a proposal to make a material change.
>>> This change is not needed (and hasn’t been for 5.5 years).
>>>
>>> Until we start a -bis process, this change should not be made.
>>>
>>> If the people who have to decide this don’t want to outright reject the
>>> change, they can always HFDU.
>>> (This would still give the proposed change some luster that it shouldn’t
>>> get, but it is a valid option.)
>>>
>>> Grüße, Carsten
>>>
>>>