[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
CAP 10: ABNF 'ordering' thought
While looking at the latest draft I
noticed that the ABNF ordering of mandatory and optional bits is somewhat
odd (or at least different than Im used to from RFC 2445). For example,
from Section 9.6. VQUERY Component::
queryprop = 1*(
; 'queryid' is OPTIONAL but MUST NOT
occur
; more than once. If the "TARGET"
property
; is supplied then the "QUERYID"
property
; MUST BE supplied.
;
queryid / target
; the following are OPTIONAL, and MAY
occur
; more than once
;
/ name / other-props
; the following MUST occur at least
once.
;
/ query
)
The mandatory property
is at the bottom after the optional ones. This seems a bit odd to
me given that in 2445 we ordered the mandatory first and then the optional
bits so that any "1*", "2*", etc. grammars had
the mandatory bits up front and then the optional bits. For example,
from Section 4.6 Calendar Components:
calprops = 2*(
; 'prodid' and
'version' are both REQUIRED,
; but MUST NOT
occur more than once
prodid /version
/
; 'calscale' and
'method' are optional,
; but MUST NOT
occur more than once
calscale
/
method
/
x-prop
)
I would ask that we try
to keep the same kind of ordering in our CAP ABNF. This is more a
preference and request than an actual "we must do this" kind
of note. I find it easier if we are consistant w/the base RFCs when
doing a quick analysis of the ABNF... (Back to injesting caffine
and waking up fully...)
Bruce
===========================================================================
Bruce Kahn
INet: Bruce_Kahn@xxxxxxxxxxxxxxxx
Messaging & Collaboration
Phone: 978.399.6496
IBM Software Group
FAX: and nothing but the FAX...
Standard disclaimers apply, even where prohibited by law...