Re: [Json] Using a non-whitespace separator (Re: Working Group Last Call on draft-ietf-json-text-sequence)

[Nico Williams <[e-mail]>]

On Sun, Jun 1, 2014 at 6:10 PM, Paul Hoffman <[e-mail]> wrote:
> On Jun 1, 2014, at 3:35 PM, Nico Williams <[e-mail]> wrote:
>> On Sun, Jun 1, 2014 at 12:08 AM, Tim Bray <[e-mail]> wrote:
>>> No. There should be only one way to do things.
>>
>> I'm not terribly fond of this.
>>
>> It'd be easier if we picked a Unicode whitespace character that's not
>> used in the JSON whitespace rule and must be escaped in strings,
>> preferably one that terminals and such generally handle as a
>> whitespace.
>
> What would a whitespace character be "easier" than a character not allowed in Unicode? It seems to me that a character that could not exist in a string and therefore never needs to be escaped is "easier".

Because I want text-based tools to be able to at least display (and
grep, and...) JSON text sequence contents much as they can JSON text
contents.

>> Failing that a character that terminals will ignore.
>
> I suspect most terminals would ignore FEFF. But I don't think that's actually as important as the fact that it doesn't require escaping.

Right, it has to be a character that must be escaped, and its encoding
must require a single byte, otherwise we can't recover simply from
incomplete appends.

>> But since the problem we're talking about is only relevant to log-like
>> writing patterns, I'd prefer to make this optional for writers,
>> mandatory for readers.
>
> Make what optional for writers and mandatory for readers?

I'd like what jq does now to continue to be valid.  I'd like to add an
alternate separator for the log-file case.

Nico
--