Re: [Yaml-core] portland spec hacking

[Brian I. <in...@tt...>]

On 29/01/04 20:57 -0500, Clark C. Evans wrote:
> Hello all.  Oren and I flew into Portland for the last four days to
> hack on the YAML spec with Brian.  The result of our work, not quite
> ready for prime time, is located at http://yaml.org/spec/29jan2004.html
> if you'd like to take a peak.
> 
> Syntax changes:
>   - allowing for stuctured keys (?:) to omit the value part (:)

ie:

    ---
    ? this
    ? is a
    ? set
    --- #equivalent to
    ? this
    :
    ? is a
    :
    ? set
    --- #which is equivalent to
    this:
    is a:
    set:
    --- {this, is a, set} #see below
    ...

>   - adding {set, syntax, shorthand} 
>     equivalent to {set: '', syntax: '', shorthand: ''}

Actually it's:

      equivalent to {set: , syntax: , shorthand: }

to be pedantic.
      
>   - adding [ordered: mapping, syntax : short hand]
>     equivalent to [{ordered: mapping}, {syntax: short hand}]

This gives PHP folks another way to serialize their ordered maps.

    ---
    - Jan: January
    - Feb: February
    - Mar: March
    --- [
      Jan: January,
      Feb: February,
      Mar: March,
    ]
    ...

>   - empty tag (!) no longer permitted
> 
> Conceptual changes:
>   - we no longer have default tagging, at least the spec does
>     not specify such a (broken) mechanism, instead the YAML 
>     processor is required to resolve all "implicit" tags

And a parser must report whether a scalar was plain or not. (As opposed
to quoted, literal, folded). So that the following are distinguishable:

    ---
    string: ''
    null:
    boolean: yes
    string: 'yes'

> Document changes:
>   - we now use DocBook for our source data
>   - we have pretty pictures
>   - the intro is completely rewirtten
>   - the entire model section was rewritten to 
>     reflect several concerns:
> 
>       * it was too formal
>       * it lacked an overview (this was actually in the
>         mis-named design section, which was moved)
>       * idea of a native model was flawed (for many reasons)
>       * it lacked many items which were instead detailed
>         in the syntax productions
>       * it was too long (we made this one worse...;)
>       * it was expressed interms of 'state' rather than
>         process, when defining the behavior of a YAML processor 
>         is the whole point of the specification
> 
> Anyway, it is still quite a ways off, and we made so many
> changes that it is most likely not consistent.  If you'd 
> like to read it over though, and provide comments on the
> direction, please do so.
> 
> Cheers!
> 
> Clark
> 
> 
> -------------------------------------------------------
> The SF.Net email is sponsored by EclipseCon 2004
> Premiere Conference on Open Tools Development and Integration
> See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
> http://www.eclipsecon.org/osdn
> _______________________________________________
> Yaml-core mailing list
> Yam...@li...
> https://lists.sourceforge.net/lists/listinfo/yaml-core