Tests that fail in UAs that implement the most recent spec.

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

Tests that fail in UAs that implement the most recent spec.

Ms2ger
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi all,

Servo is implementing the four/eight-digit hex colour syntax. This
causes us to fail some tests in the CSS 2.1 test suite. It seems to be
we should have a policy that a correct implementation of the most
recent specification should never cause test failures in any
conformance test suite, as that sends a message that actively works
against interoperability.

Thoughts?
Ms2ger
-----BEGIN PGP SIGNATURE-----

iQEcBAEBAgAGBQJVv6QNAAoJEOXgvIL+s8n23gcH/00kDjc0P7+Tw7CK3bKX7ynp
C646CrUQNfmWFaEWuSJJnttE2ZP+thUSJbho8hGso/plha5l0gB9hY3bX/CtCzoh
DU2b7irYRj2fp50pibj6uwJXRDYIAOtCOG4YZxh1Whmm90k55bCxK+p7SD80Y7Ah
m8SFci3gfmY6Wc5VeM7DQWMEmFeg+ErkjOqOt8/gg0WFXkzq605kJ76W2LBkmTWY
Em7oUGrvcYNKR6RQZsAtfYuPhV2IwldGLNtm+b2pr8kVHPhH0L63dlsRPJJJpHa1
hM2t9RBRAcucLtjVUPeU8b8SuE+0j7x7oHGZQBogi2LouXIyLzxW1uLVykzbCfM=
=CDab
-----END PGP SIGNATURE-----

Reply | Threaded
Open this post in threaded view
|

Re: Tests that fail in UAs that implement the most recent spec.

Tab Atkins Jr.
On Mon, Aug 3, 2015 at 10:25 AM, Ms2ger <[hidden email]> wrote:
> Servo is implementing the four/eight-digit hex colour syntax. This
> causes us to fail some tests in the CSS 2.1 test suite. It seems to be
> we should have a policy that a correct implementation of the most
> recent specification should never cause test failures in any
> conformance test suite, as that sends a message that actively works
> against interoperability.

Is this covered by the policy surrounding replacement specs?  The
Color module replaces part of 2.1, so 2.1's tests that disagree with
it are incorrect.

~TJ

Reply | Threaded
Open this post in threaded view
|

Re: Tests that fail in UAs that implement the most recent spec.

Simon Sapin
In reply to this post by Ms2ger
(Responding to
https://lists.w3.org/Archives/Public/public-css-testsuite/2015Aug/0000.html 
, in case email headers donโ€™t get set properly.)

CSSWG resolved today that that tests invalidated by later (levels of)
specs should allow either behavior. For refetests, that means having
more than one possible reference.

http://log.csswg.org/irc.w3.org/css/2015-08-25/#e581689

--
Simon Sapin

Reply | Threaded
Open this post in threaded view
|

Re: Tests that fail in UAs that implement the most recent spec.

L. David Baron
On Tuesday 2015-08-25 13:41 +0200, Simon Sapin wrote:
> CSSWG resolved today that that tests invalidated by later (levels of) specs
> should allow either behavior. For refetests, that means having more than one
> possible reference.

Or at least *may* mean that.  There are often other, simpler,
solutions, which I think are generally preferable.

-David

--
๐„ž   L. David Baron                         http://dbaron.org/   ๐„‚
๐„ข   Mozilla                          https://www.mozilla.org/   ๐„‚
             Before I built a wall I'd ask to know
             What I was walling in or walling out,
             And to whom I was like to give offense.
               - Robert Frost, Mending Wall (1914)

signature.asc (836 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Tests that fail in UAs that implement the most recent spec.

Simon Sapin
On 25/08/15 14:30, L. David Baron wrote:
> On Tuesday 2015-08-25 13:41 +0200, Simon Sapin wrote:
>> CSSWG resolved today that that tests invalidated by later (levels of) specs
>> should allow either behavior. For refetests, that means having more than one
>> possible reference.
>
> Or at least *may* mean that.  There are often other, simpler,
> solutions, which I think are generally preferable.

Yes, thatโ€™s a good point.

--
Simon Sapin