Re: Last Call: draft-chakrabarti-ipv6-addrselect-api (IPv6 Socket API for Address Selection) to Informational RFC

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 2007-05-21 12:47, Jari Arkko wrote:
Brian,

Section 8 states

   Applications MUST not set contradictory flags at the same time.
I think the document should specify what the API must do if
contradictory flags are set. That would be the normative MUST, not
the above statement of the obvious.

Thanks for your review, Brian!

I agree with the principle that you state above. However,
the document does have a definition of contradictory
flags (Section 2) and explicitly describes what the API
must do when getting them (Section 6) i.e., causes
error EINVAL. Or EAI_BADEXTFLAGS for getaddrinfo
(Section 7).

Ah, sorry, I'd missed that.


I do agree though that there's something wrong in the
text in Section 10 about application requirements.
Perhaps s/MUST not/should not/. This may apply to
some other text in Section 10, too.

Yes, I think it is strange to use the normative words
in this way.

    Brian

--
NEW: Preferred email for non-IBM matters: brian.e.carpenter@xxxxxxxxx

_______________________________________________

Ietf@xxxxxxxx
https://www1.ietf.org/mailman/listinfo/ietf

[Index of Archives]     [IETF Annoucements]     [IETF]     [IP Storage]     [Yosemite News]     [Linux SCTP]     [Linux Newbies]     [Fedora Users]