On Sun, 24 Jun 2001, Francis Dupont wrote:
> In your previous mail you wrote:
>
> > => no, there are some codes which are written without switches.
>
> of course there are. RFC 2553 has been written in order to minimize the
> need for those switches.
>
> > Of course they work only on systems where IPv6 is fully integrated.
>
> i don't think so.
>
> => but we are sating the same thing. Perhaps we have different meanings
> for "integration"?
maybe i have not well understood what you've written, or, more probably,
i haven't explained myself very well.
i was trying to say this:
RFC 2553 has been written in order to minimize the need for compile-time
switches. however, i don't think that we should focus on that minimization,
but instead on the definition of an API extension that can lead to
easier porting procedures.
as itojun has explained us, IPv4 mapped addresses present us many security
problems that the developers will have to handle. unspecified behaviour
for bind(2) - read my answer to itojun - will eventually produce platform
specific code.
RFC2553 does not address these problems. in the long run, this policy will
cause many headaches to the developers. so, minimization of the compile
time switches is not an absolute (or good-in-itself) target to be reached.
the solution for this problems may be reached, IMHO, by making the two
socket families AF_INET and AF_INET6 indipendent (orthogonal).
> => I have the opposite policy so results are different. My target is
> integrated systems where IPv6 is *not* optional, no more than IP is!
we have different policies.
> i have seen that many programmers find difficult the process of porting
> their apps to ipv6. i think that we should write a small informational
> document that explains how to write good ipv6-compliant code.
> itojun's "Implementing AF-indipendent apps" is a very good document.
> however, it is obsolete and it doesn't give many informations.
>
> => I have done one but it is more than obsolete now. But my target
> is not AF-independence, it is IP version independence, even if a true
> AF independence (i.e. codes which can work with ISO CLNP) is a plus.
why don't you give us the url? it may be interesting.
> PS: codes full of switches are at least unreadable...
i don't think so. as soon as my patch for oftpd is integrated you'll
have one example. for now, refer to USAGI's inetd from netkit-base.
i find that the code is quite readable.
--
Aequam memento rebus in arduis servare mentem...
Mauro Tortonesi [EMAIL PROTECTED]
Ferrara Linux User Group http://www.ferrara.linux.it
Project6 - IPv6 for Linux http://project6.ferrara.linux.it
--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page: http://playground.sun.com/ipng
FTP archive: ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------
