Barry Leiba | 2 Sep 14:10 2003
Picon

Re: Last Calling the LISTEXT draft?


> LIST () "" % should not return it, but for backwards compatibility I 
> think the old style LIST should return it when it can.

Right, that's why it was left there: compatibility with clients that expected
a plain LIST to return CHILDREN info.  I agree with the concern, and with the
suggested compromise.  Pete & Mark, does that work for you?

Barry
--
Barry Leiba, Internet Messaging Technology  (leiba <at> watson.ibm.com)
http://www.research.ibm.com/people/l/leiba

Mark Crispin | 2 Sep 19:55 2003

Re: Last Calling the LISTEXT draft?


On Tue, 2 Sep 2003, Barry Leiba wrote:
> > LIST () "" % should not return it, but for backwards compatibility I
> > think the old style LIST should return it when it can.
> Right, that's why it was left there: compatibility with clients that expected
> a plain LIST to return CHILDREN info.  I agree with the concern, and with the
> suggested compromise.  Pete & Mark, does that work for you?

I'd prefer some wording to the effect that servers may choose not to have
"old style" list return CHILDREN information, and thus the only reliable
way to get CHILDREN information is to use LISTEXT.

-- Mark --

http://staff.washington.edu/mrc
Science does not emerge from voting, party politics, or public debate.
Si vis pacem, para bellum.

Ken Murchison | 2 Sep 20:27 2003

Re: Last Calling the LISTEXT draft?


Quoting Mark Crispin <MRC <at> CAC.Washington.EDU>:

> 
> On Tue, 2 Sep 2003, Barry Leiba wrote:
> > > LIST () "" % should not return it, but for backwards compatibility I
> > > think the old style LIST should return it when it can.
> > Right, that's why it was left there: compatibility with clients that
> expected
> > a plain LIST to return CHILDREN info.  I agree with the concern, and with
> the
> > suggested compromise.  Pete & Mark, does that work for you?
> 
> I'd prefer some wording to the effect that servers may choose not to have
> "old style" list return CHILDREN information, and thus the only reliable
> way to get CHILDREN information is to use LISTEXT.

All the server has to do is no longer advertise CHILDREN, and then it no 
longer has to return CHILDREN info for "old style" LIST.

--

-- 
Kenneth Murchison     Oceana Matrix Ltd.
Software Engineer     21 Princeton Place
716-662-8973 x26      Orchard Park, NY 14127
--PGP Public Key--    http://www.oceana.com/~ken/ksm.pgp

Arnt Gulbrandsen | 2 Sep 20:31 2003
Picon

Re: Last Calling the LISTEXT draft?


Mark Crispin writes:
> I'd prefer some wording to the effect that servers may choose not to 
> have "old style" list return CHILDREN information, and thus the only 
> reliable way to get CHILDREN information is to use LISTEXT.

That sounds bad to me. If listext and CHILDREN are both present, then a 
client that supports CHILDREN but not listext would be hampered.

If that's desirable, a more honest way would be to provide a 
LIST=CHILDREN that says effectively "the functionality of 3348 is 
present, but not the syntax".

--Arnt

Mark Crispin | 2 Sep 20:39 2003

Re: Last Calling the LISTEXT draft?


On Tue, 2 Sep 2003, Ken Murchison wrote:
> > I'd prefer some wording to the effect that servers may choose not to have
> > "old style" list return CHILDREN information, and thus the only reliable
> > way to get CHILDREN information is to use LISTEXT.
> All the server has to do is no longer advertise CHILDREN, and then it no
> longer has to return CHILDREN info for "old style" LIST.

But doesn't CHILDREN have to be advertised if the server supports it in
LISTEXT?

-- Mark --

http://staff.washington.edu/mrc
Science does not emerge from voting, party politics, or public debate.
Si vis pacem, para bellum.

Arnt Gulbrandsen | 2 Sep 20:47 2003
Picon

Re: Last Calling the LISTEXT draft?


Ken Murchison writes:
> All the server has to do is no longer advertise CHILDREN, and then it 
> no longer has to return CHILDREN info for "old style" LIST.

How would the client know that it can send LIST (CHILDREN), unless by 
seeing that capability? The latest draft names two capabilities: 
LIST-SUBSCRIBED and LIST-REMOTE. It does not specify a capability for 
CHILDREN.

--Arnt

Mark Crispin | 2 Sep 20:51 2003

Re: Last Calling the LISTEXT draft?


On Tue, 2 Sep 2003, Arnt Gulbrandsen wrote:
> How would the client know that it can send LIST (CHILDREN), unless by
> seeing that capability? The latest draft names two capabilities:
> LIST-SUBSCRIBED and LIST-REMOTE. It does not specify a capability for
> CHILDREN.

I agree that LIST-CHILDREN should be there.

-- Mark --

http://staff.washington.edu/mrc
Science does not emerge from voting, party politics, or public debate.
Si vis pacem, para bellum.

Arnt Gulbrandsen | 2 Sep 20:56 2003
Picon

Re: Last Calling the LISTEXT draft?


Is it permitted to ask why it's LIST-* instead of LIST=*? IMAP seems to 
lean towards the latter in general.

--Arnt

Ken Murchison | 3 Sep 00:04 2003

Re: Last Calling the LISTEXT draft?


Mark Crispin wrote:
> On Tue, 2 Sep 2003, Ken Murchison wrote:
> 
>>>I'd prefer some wording to the effect that servers may choose not to have
>>>"old style" list return CHILDREN information, and thus the only reliable
>>>way to get CHILDREN information is to use LISTEXT.
>>
>>All the server has to do is no longer advertise CHILDREN, and then it no
>>longer has to return CHILDREN info for "old style" LIST.
> 
> 
> But doesn't CHILDREN have to be advertised if the server supports it in
> LISTEXT?

By my reading of the drafts (00 - 04), CHILDREN is part o fthe base 
LISTEXT spec.  The LISTEXT capability means that LIST (CHILDREN) is a 
valid option.

--

-- 
Kenneth Murchison     Oceana Matrix Ltd.
Software Engineer     21 Princeton Place
716-662-8973 x26      Orchard Park, NY 14127
--PGP Public Key--    http://www.oceana.com/~ken/ksm.pgp

Pete Maclean | 3 Sep 01:29 2003

Re: Last Calling the LISTEXT draft?


> > LIST () "" % should not return it, but for backwards compatibility I
> > think the old style LIST should return it when it can.
>
>Right, that's why it was left there: compatibility with clients that expected
>a plain LIST to return CHILDREN info.  I agree with the concern, and with the
>suggested compromise.  Pete & Mark, does that work for you?

If Mark agrees, then it's fine with me.

Pete


Gmane