Mailing List Archive: SPF: Devel

Questions on the test suite and draft

Index | Next | Previous | View Threaded

rao at gnu

Feb 2, 2004, 10:12 PM

Post #1 of 6 (1368 views)
Permalink

Questions on the test suite and draft

I'm currently working on a small SPF patch/client for Exim. I have a
few questions regarding the test suite (1.991) and RFC Draft 02.9.5
that have been bugging me.

1. Tests #89 - #96 / RFC Draft Appendix A

The TXT record for 70.spf1-test.mailzone.com is "v=spf1
exists:%{lr+=}.lp._spf.spf1-test.mailzone.com -all". Appendix A lists
valid macro delimiters -- the equal sign is not included. Was this an
accidental omission in the RFC Draft, or should tests #89 through #96
return "unknown" (parse error)?

2. Semantics of "include" mechanism are no longer clear

Since the extra result values were added in 02.9.5, it's not clear how
"include" should behave. For example, Test #82 seems to indicate that
if a recursive query returns "none", then "include" should return
"unknown". Could text be added to Section 4.2 that covers all
possible recursive return values, much like the "include" section from
the November 02.7 RFC Draft (when softfail was last seen)?

3. Section 7.2 typo

The "%{p2}.trusted-domains.example.net" example is listed twice in the
RFC Draft.

The test suite has been very useful. Many thanks for that. However,
I'll throw in my vote that it's time to split out test.txt into
test-{must,should,mailspfqueryonly}.txt.

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?listname@Ë`Ì{5¤¨wâÇSÓ°)h

christophe at saout

Feb 3, 2004, 3:44 AM

Post #2 of 6 (1314 views)
Permalink

Re: Questions on the test suite and draft [In reply to]

Am Di, den 03.02.2004 schrieb Paul Fisher um 06:12:

> 2. Semantics of "include" mechanism are no longer clear
>
> Since the extra result values were added in 02.9.5, it's not clear how
> "include" should behave. For example, Test #82 seems to indicate that
> if a recursive query returns "none", then "include" should return
> "unknown". Could text be added to Section 4.2 that covers all
> possible recursive return values, much like the "include" section from
> the November 02.7 RFC Draft (when softfail was last seen)?

When the recursive lookup returns either ("none", "fail", "softfail" or
"neutral") the include mechanism does not match. "pass" makes it return
the prefix (+ -> "pass", - -> "fail", ~ -> "softfail", ? -> "neutral").
Errors are directly returned.

You are right, the draft is somewhat unclear about this, but this is how
it makes most sense in the end (and how Mail::SPF::Query does it).

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?listname@Ë`Ì{5¤¨wâÇSÓ°)h

rich at copsewood

Feb 3, 2004, 4:36 AM

Post #3 of 6 (1342 views)
Permalink

Re: Questions on the test suite and draft [In reply to]

I am currently working on a Python SPF implementation,
partly to learn more about DNS and mail forwarding. This
passes my initial tests, but I would also like
to test this against the test suite referred to here. From
where can I download this, and is it
documented somewhere e.g. in other SPF sources ?

Thanks in advance,
Richard Kay

On Tue, Feb 03, 2004 at 12:12:31AM -0500, Paul Fisher wrote:
> I'm currently working on a small SPF patch/client for Exim. I have a
> few questions regarding the test suite (1.991) and RFC Draft 02.9.5
> that have been bugging me.
>
> 1. Tests #89 - #96 / RFC Draft Appendix A
>
> The TXT record for 70.spf1-test.mailzone.com is "v=spf1
> exists:%{lr+=}.lp._spf.spf1-test.mailzone.com -all". Appendix A lists
> valid macro delimiters -- the equal sign is not included. Was this an
> accidental omission in the RFC Draft, or should tests #89 through #96
> return "unknown" (parse error)?

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?listname@Ë`Ì{5¤¨wâÇSÓ°)h

terry at wayforward

Feb 3, 2004, 7:26 AM

Post #4 of 6 (1329 views)
Permalink

Re: Questions on the test suite and draft [In reply to]

Download Meng's Mail:SPF:Query package from
http://spf.pobox.com/downloads.html

There's a file test.txt, which consists of comments # and tuples of the
form
test-num sender ipaddr expected-result

I think tests 132 and higher are specific to the Perl implementation,
the rest should work immediately.

Several people have issues with test #75. The test file says an
implementation should FAIL, while several implementors have wondered
why it shouldn't result in UNKNOWN.

Cheers!

On Tuesday, February 3, 2004, at 06:36 AM, Richard Kay wrote:

> I am currently working on a Python SPF implementation,
> partly to learn more about DNS and mail forwarding. This
> passes my initial tests, but I would also like
> to test this against the test suite referred to here. From
> where can I download this, and is it
> documented somewhere e.g. in other SPF sources ?
>
> Thanks in advance,
> Richard Kay
>
> On Tue, Feb 03, 2004 at 12:12:31AM -0500, Paul Fisher wrote:
>> I'm currently working on a small SPF patch/client for Exim. I have a
>> few questions regarding the test suite (1.991) and RFC Draft 02.9.5
>> that have been bugging me.
>>
>> 1. Tests #89 - #96 / RFC Draft Appendix A
>>
>> The TXT record for 70.spf1-test.mailzone.com is "v=spf1
>> exists:%{lr+=}.lp._spf.spf1-test.mailzone.com -all". Appendix A lists
>> valid macro delimiters -- the equal sign is not included. Was this an
>> accidental omission in the RFC Draft, or should tests #89 through #96
>> return "unknown" (parse error)?
>
> -------
> To unsubscribe, change your address, or temporarily deactivate your
> subscription,
> please go to
> http://v2.listbox.com/member/?listname@Ë`Ì{5¤¨wâÇSÓ°)h
>

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?listname@Ë`Ì{5¤¨wâÇSÓ°)h

wayne at midwestcs

Feb 3, 2004, 7:39 AM

Post #5 of 6 (1316 views)
Permalink

Re: Questions on the test suite and draft [In reply to]

In <20040203113655.GA491 [at] home> Richard Kay <rich [at] copsewood> writes:

> I am currently working on a Python SPF implementation,
> partly to learn more about DNS and mail forwarding. This
> passes my initial tests, but I would also like
> to test this against the test suite referred to here. From
> where can I download this, and is it
> documented somewhere e.g. in other SPF sources ?

Meng has asked me to try and coordinate the test data and programs.
Currently, the test data is kind of spread out all over the place.

I have *NOT* done all the cleanup on the test data that should be
done, but in the "release early, release often" philosophy, I have
just put up a directory on my website with what I have right now.

See: http://www.midwestcs.com/spf/tests/

Part of what I have done is change both the Perl and C systems to have
an spfquery program that implements all the features of SPF on the
command line. I then changed test.pl to call spfquery instead of
doing direct calls to the SPF libraries. So, in theory, all you need
to do is create a similar spfquery.

Please post comments, suggestions and corrections to the test suite
here. That way, everyone can be kept up to date.

-wayne

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?listname@Ë`Ì{5¤¨wâÇSÓ°)h

philip-spf at gladstonefamily

Feb 3, 2004, 6:53 PM

Post #6 of 6 (1330 views)
Permalink

Re: Questions on the test suite and draft [In reply to]

In 0.991, it does return unknown (according to the file) (this is the
test that does 'include' with no argument).

Philip

Terence Way wrote:

> Download Meng's Mail:SPF:Query package from
> http://spf.pobox.com/downloads.html
>
> There's a file test.txt, which consists of comments # and tuples of the
> form
> test-num sender ipaddr expected-result
>
> I think tests 132 and higher are specific to the Perl implementation,
> the rest should work immediately.
>
> Several people have issues with test #75. The test file says an
> implementation should FAIL, while several implementors have wondered why
> it shouldn't result in UNKNOWN.
>
> Cheers!
>
>
> On Tuesday, February 3, 2004, at 06:36 AM, Richard Kay wrote:
>
>> I am currently working on a Python SPF implementation,
>> partly to learn more about DNS and mail forwarding. This
>> passes my initial tests, but I would also like
>> to test this against the test suite referred to here. From
>> where can I download this, and is it
>> documented somewhere e.g. in other SPF sources ?
>>
>> Thanks in advance,
>> Richard Kay
>>
>> On Tue, Feb 03, 2004 at 12:12:31AM -0500, Paul Fisher wrote:
>>
>>> I'm currently working on a small SPF patch/client for Exim. I have a
>>> few questions regarding the test suite (1.991) and RFC Draft 02.9.5
>>> that have been bugging me.
>>>
>>> 1. Tests #89 - #96 / RFC Draft Appendix A
>>>
>>> The TXT record for 70.spf1-test.mailzone.com is "v=spf1
>>> exists:%{lr+=}.lp._spf.spf1-test.mailzone.com -all". Appendix A lists
>>> valid macro delimiters -- the equal sign is not included. Was this an
>>> accidental omission in the RFC Draft, or should tests #89 through #96
>>> return "unknown" (parse error)?
>>
>>
>> -------
>> To unsubscribe, change your address, or temporarily deactivate your
>> subscription,
>> please go to
>> http://v2.listbox.com/member/?listname@Ë`Ì{5¤¨wâÇSÓ°)h
>>
>
> -------
> To unsubscribe, change your address, or temporarily deactivate your
> subscription, please go to
> http://v2.listbox.com/member/?listname@Ë`Ì{5¤¨wâÇSÓ°)h
>
>

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?listname@Ë`Ì{5¤¨wâÇSÓ°)h

Index | Next | Previous | View Threaded

Interested in having your list archived? Contact Gossamer Threads