On 7/15/17, 1:29 PM, "Dave Crocker" <dhc(_at_)dcrocker(_dot_)net> wrote:
(sigh. re-re-sent to try for a valid tony address too...)
(resent, to get a working murray address. /d)
On 7/15/2017 9:10 AM, RFC Errata System wrote:
> Original Text
> -------------
> tag-spec = [FWS] tag-name [FWS] "=" [FWS] tag-value [FWS]
> tag-name = ALPHA *ALNUMPUNC
> tag-value = [ tval *( 1*(WSP / FWS) tval ) ]
> ; Prohibits WSP and FWS at beginning and end
>
> Corrected Text
> --------------
> tag-spec = [FWS] tag-name [FWS] "=" [FWS] [tag-value [FWS]]
> tag-name = ALPHA *ALNUMPUNC
> tag-value = tval *( 1*(WSP / FWS) tval )
> ; Prohibits WSP and FWS at beginning and end
>
> Notes
> -----
> The ABNF in the document permits two FWS rules to appear in the row. This
results in permitting a line with only whitespace in the header which falls
into obsolete syntax in RFC 5322 (Appendix B rule 12). The corrected text
disallows this by eliding the second FWS when the tag-value is empty/omitted.
Hitting 'obsolete' RFC5322 syntax doesn't bother me all that much, given
that there is now a long history of non-enforcement; the constructs were
never seriously deprecated.
But the proposed change does seem cleaner to me.
Right, “FWS optional-tval FWS” to “FWS optional-(tval and FWS)” is indeed
cleaner.
Tony
_______________________________________________
NOTE WELL: This list operates according to
http://mipassoc.org/dkim/ietf-list-rules.html