Re: JSYNC

Oh, heh. I was not even considering using libyaml for any of this, but I
guess that's a good idea. I was just considering using the existing plethora
of JSON parser/emitters, and adding a consistent API layer over them.

The main point of JSYNC is simple, cross language, plain text serialization.
Human friendly, human editable is not a concern. Therefore I expect that
only JSYNC emitters will write JSYNC (not people).

I don't see this as a big deal, even using libyaml. But I will consider it,
and see how it plays out in the early adoption cycle.

Thanks, for reviewing what I've written so far. Hopefully I will have a spec
and JavaScript and Python initial implementation done by the end of May.

Cheers, Ingy

On Sat, May 1, 2010 at 8:45 AM, William Spitzak <[email protected]> wrote:

> Ingy dot Net wrote:
>
>     I'm not sure this is the best choice, as numbers can start with a
>>    dot. This is not in official JSON syntax, however, because they must
>>    start with a zero and are never quoted while strings always are, but
>>    there are plenty of semi-JSON syntaxes that do not follow these
>>    rules and this would be a serious incompatibility with them.
>>
>>
>> Can you elucidate with an example? I don't follow the logic.
>>
>
> libyaml does not provide any obvious api to determine if a scalar was
> quoted. Using it to read these files and decide if leading periods should be
> stripped would require you to peek into the "events" to determine if they
> were quoted, making it impossible to use the higher level structured api.
> Also I suspect it was not intended that you be able to look at the quoting
> of events and this is a side effect of using the same structure for output
> as input.
>

------------------------------------------------------------------------------

_______________________________________________
Yaml-core mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/yaml-core