Re: Seeking clarification of ABNF "name formation" reference

[Philip Guenther <[e-mail]>]

On Mon, 21 May 2007, Dave Crocker wrote:
> In the ABNF document, there is a precedence chart:
>
>> <section title="Operator Precedence">
>>
>>             <t> The various mechanisms described above have the following
>>                precedence, from highest (binding tightest) at the top, to
>>                lowest (loosest) at the bottom: <list>
>>
>>                   <t>Strings, Names formation</t>
>>                   <t>Comment</t>
...
>>                </list>
>>             </t>
>
>
> The reference to "Name formation" has been in this form since RFC 2234.
>
> Oddly, we can't figure out what it is supposed to mean, exactly.
>
> We could make some guesses, but I'd rather check with the group.

My guess was that it's a reference to the 'prose' value form, ala:
 	tag      = 1*<any ASTRING-CHAR except "+">

Inside the angle-brackets, only close-angle-bracket is special, just as 
inside a string, only double-quote is special, so those forms have a 
higher precedence than comment (semicolon isn't special) and repetition 
(asterisk isn't special), and...

So then I looked at the RFC and discovered that the prose value form is 
*only* mentioned in the ABNF definition of ABNF!  Was its prose 
description misplaced and accidentally used to shim a wobbly table?  If 
so, that would explain the dangling reference.


Philip Guenther