Re: Question on YAML 1.2 spec Example 9.5

Oops.

You are absolutely right; this is a bad example. Thanks for catching it
- I really appreciate you delving into the details like that!

The thing is, in theory we _could_ get away with not making % an
indicator. The 1.2 spec contains a new explanation of the role of '---',
and this new way of thinking allows for that. I had that in mind when I
wrote the example... However - % has always been an indicator and we
don't think changing it right now is a good idea.

This throws into suspicion the need for '---' to terminate directives
(and here I was thinking I found some use for '---'. Sigh.) but we are
_really_ not going to get rid of '---'. Chalk the combination up to
"historical reasons" ;-)

Besides, who knows, perhaps we'll want to do something nifty with % on
YAML 9.7 or something ;-)

I'll fix the example and some wording in the part discussing '---' to
that effect.

Thanks again,

    Oren Ben-Kiki

On Thu, 2009-04-23 at 12:32 -0700, Joshua Choi wrote:
> To the YAML 1.2 spec writers,
> 
> In the current draft's Example 9.5:
> 
> %YAML 1.2
> ---
> % of matches: 20
> ...
> %YAML1.2
> ---
> # Empty
> ...
> 
> "% of matches: 20" seems to be a valid node, but I don't see how
> that's possible. "% of matches" is acting as a plain-style mapping
> key. Plain scalars cannot start with an indicator character such as %.
> How, then, is this allowed? (For the record, according to Instant
> YAML, PyYaml says that this is indeed invalid in YAML 1.1.)
> 
> Sincerely,
> Joshua Choi


------------------------------------------------------------------------------
Crystal Reports - New Free Runtime and 30 Day Trial
Check out the new simplified licensign option that enables unlimited
royalty-free distribution of the report engine for externally facing 
server and web deployment.
http://p.sf.net/sfu/businessobjects