Mailing List Archive

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Stuart D. Gathman wrote:
> On Wed, 17 Jan 2007, Norman Maurer wrote:
> > In fact we apply unknown-modifier BECAUSE "redirect=" is not a
> > redirect modifier by the gramma. So if you guys want to enforce the
> > interpretations you need to integrate the explanation in the spec.
>
> This is an excellent point. It would put Julian's argument to rest
> once and for all, if true.
>
> Let me restate it. Modifier is defined as:
>
> modifier = redirect | explanation | unknown-modifier
>
> Therefore, exp= does not match explanation, but it *does* match
> unknown modifier, and is not a syntax error.

OMG! You must be kidding! If we followed this interpretation, we could
throw away the grammar definitions for all the known modifiers, because
they'd be simply irrelevant.

Are you really saying that "unknown-modifier" should have been defined as:

unknown-modifier = unknown-modifier-name "=" macro-string
unknown-modifier-name = ( "a" / "b" / "c" / "d" ) name-tail /
"e" (
"" /
( DIGIT / "-" / "_" / "." ) name-tail /
( "a" / ... / "w" ) name-tail /
"x" (
"" /
( DIGIT / "-" / "_" / "." ) name-tail /
( "a" / ... / "o" ) name-tail /
( "q" / ... / "z" ) name-tail
) /
( "y" / "z" ) name-tail
) /
( "f" / ... / "q" ) name-tail /
"r" (
; you get the point!
) /
( "s" / ... / "z" ) name-tail
name-tail = *( ALPHA / DIGIT / "-" / "_" / "." )

And that still would have allowed "op=#34534l.5,34" despite the "op=" spec
according to your interpretation!

For me, it is completely obvious that the <name> in the <unknown-modifier>
definition excludes the names of known modifiers (for whatever a given
implementation considers "known"). *sigh*

> Similarly, "exp=-all" does not match explanation, but it does match
> unknown-modifier, so it is not a permerror.

...

> This means the grammar *is* consistent with 6.2/4, empty domain-spec
> is an unknown-modifier according to the grammar, and MUST be ignored.
>
> Sorry Julian, it looks like you'll just have to start a campaign for
> v=spf3.

Whatever.

This is starting to piss me off.

Why don't you look into what the spec is _really_ trying to say for a
change instead of what its letters are?

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFFrsgGwL7PKlBZWjsRAiC3AJ93Bs1MYEMU/YDn6UIRKGxfRnb0WACeMOTx
cXokPxRdElritKnE96CZ+ls=
=rvHy
-----END PGP SIGNATURE-----

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?list_id=1007

On Thu, 18 Jan 2007, Julian Mehnle wrote:

> OMG! You must be kidding! If we followed this interpretation, we could
> throw away the grammar definitions for all the known modifiers, because
> they'd be simply irrelevant.
>
> Are you really saying that "unknown-modifier" should have been defined as:
>
> unknown-modifier = unknown-modifier-name "=" macro-string
> unknown-modifier-name = ( "a" / "b" / "c" / "d" ) name-tail /

No. It just needs a note that says

unknown-modifier = name "=" macro-string
; where name is not any known modifier name

You don't have to spell out such things in detail, but you do have
to say something.

--
Stuart D. Gathman <stuart@bmsi.com>
Business Management Systems Inc. Phone: 703 591-0911 Fax: 703 591-6154
"Confutatis maledictis, flammis acribus addictis" - background song for
a Microsoft sponsored "Where do you want to go from here?" commercial.

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?list_id=1007

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Julian Mehnle wrote:
> Stuart D. Gathman wrote:
> > Sorry Julian, it looks like you'll just have to start a campaign for
> > v=spf3.
>
> Whatever.
>
> This is starting to piss me off.
>
> [...]

I apologize for the harsh tone. I was in a bad mood when I wrote this
because, well, this discussion had started to piss me off (for reasons
legitimate or not). ;-) However, I should not have expressed it in the
way I did, which perhaps offended some (foremost Stuart).

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFFruOEwL7PKlBZWjsRAgY8AJsFWLkhIflo8tW1PQeisE5HA++oCQCePE/a
TYvcs7ZPENzhAnBnODBhh6M=
=Xl/E
-----END PGP SIGNATURE-----

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?list_id=1007

On Thu, 18 Jan 2007, Julian Mehnle wrote:

> Julian Mehnle wrote:
> > Stuart D. Gathman wrote:
> > > Sorry Julian, it looks like you'll just have to start a campaign for
> > > v=spf3.
> >
> > Whatever.
> >
> > This is starting to piss me off.
> >
> > [...]
>
> I apologize for the harsh tone. I was in a bad mood when I wrote this
> because, well, this discussion had started to piss me off (for reasons
> legitimate or not). ;-) However, I should not have expressed it in the
> way I did, which perhaps offended some (foremost Stuart).

No problem. Nit-picking can become very frustrating - even when
entirely justified (such as when creating/interpreting an RFC).

--
Stuart D. Gathman <stuart@bmsi.com>
Business Management Systems Inc. Phone: 703 591-0911 Fax: 703 591-6154
"Confutatis maledictis, flammis acribus addictis" - background song for
a Microsoft sponsored "Where do you want to go from here?" commercial.

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?list_id=1007

Stuart D. Gathman wrote:

> Sorry Julian, it looks like you'll just have to start a campaign for v=spf3.

That's an oversimplification, a malformed 2822-From or Date header field
also is no "other header field".

Frank


-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?list_id=1007

On Fri, 19 Jan 2007, Frank Ellermann wrote:

> Stuart D. Gathman wrote:
>
> > Sorry Julian, it looks like you'll just have to start a campaign for v=spf3.
>
> That's an oversimplification, a malformed 2822-From or Date header field
> also is no "other header field".

That was a joke. Sorry. Technically, the grammar today *does* literally
say that a malformed err= is, in fact an unknown-modifier, and Norman
is technically correct. However, this is clearly not what was
intended, and the test suite assumes just what you've said. The errata I
added makes this explicit:

unknown-modifier clause is too greedy in ABNF

In section 4.6.1/2, unknown-modifier should be changed to:

unknown-modifier = name "=" macro-string
; where name is not any known modifier

Rationale:

The unknown-modifier clause is too greedy, causing any syntax errors in
redirect or explanation to simply become unknown-modifier. For instance,
"exp=-all" would match unknown-modifier. This was not the intent.

--
Stuart D. Gathman <stuart@bmsi.com>
Business Management Systems Inc. Phone: 703 591-0911 Fax: 703 591-6154
"Confutatis maledictis, flammis acribus addictis" - background song for
a Microsoft sponsored "Where do you want to go from here?" commercial.

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?list_id=1007