On 12/2/24 12:12, Alessandro Vesely wrote:
> On Sun 01/Dec/2024 22:24:08 +0100 Barry Leiba wrote:
>> It does look like the ABNF should be changed to allow "0" or "1" to be
>> mixed with "d" and "s", though I do think that "0" and "1" are
>> mutually exclusive with each other.
>
> Now that the meaning shifted to enirely different reports for d and s, any
> combination must be allowed, including 0:s:d. E.g.:
>
> dmarc-fo = "0" / "1" *( ":" dmarc-afrf)
> dmarc-afrf = "d" / "s"
The text for this tag changed in draft 37:
[...] the values may appear in the list in any order.
The ABNF for dmarc-fo was updated and require 0 or 1 to be first in the
value list, if present. However the wording above necessitates that the
ABNF for dmarc-fo be changed to:
dmarc-fo = "0" / "0:d" / "0:d:s"
/ "0:s" / "0:s:d"
/ "1" / "1:d" / "1:d:s"
/ "1:s" / "1:s:d"
/ "d" / "d:0" / "d:0:s"
/ "d:1" / "d:1:s"
/ "d:s" / "d:s:0"
/ "d:s:1"
/ "s" / "s:0" / "s:0:d"
/ "s:1" / "s:1:d"
/ "s:d" / "s:d:0"
/ "s:d:1"
which seems a tad excessive.
Should we restrict the text to require 0 and 1 to be first in the list,
or should we go for a variation of Alessandros suggestion:
dmarc-fo = ("0" / "1") *(":" dmarc-afrf)
/ dmarc-afrf [":" ("0" / "1")] [":" dmarc-afrf]
/ dmarc-afrf [":" dmarc-afrf] [":" ("0" / "1")]
dmarc-afrf = "d" / "s"
Daniel K.
_______________________________________________
dmarc mailing list -- [email protected]
To unsubscribe send an email to [email protected]
