W3C home > Mailing lists > Public > public-css-testsuite@w3.org > November 2012

background-334 and background-size-008: reviewed and approved

From: Gérard Talbot <css21testsuite@gtalbot.org>
Date: Fri, 9 Nov 2012 19:38:44 -0500
Message-ID: <44eed3f656dbada43d7784e354038056.squirrel@ed-sh-cp3.entirelydigital.com>
To: "Zhang, Zhiqiang" <zhiqiang.zhang@intel.com>
Cc: "Public CSS testsuite mailing list@w3.org" <public-css-testsuite@w3.org>, "Santos, Thiago" <thiago.santos@intel.com>, "Zhang, Haili" <haili.zhang@intel.com>, "Yu, Ling L" <ling.l.yu@intel.com>, "Yang, Lei A" <lei.a.yang@intel.com>
Le Ven 9 novembre 2012 3:28, Zhang, Zhiqiang a écrit :

> Hi Gerard,
> background-334 and background-size-008 need to be approved, with your
> *-review.html added into the test code:



Both tests have been reviewed and approved:


Both tests are not linked to a reftest right now.

I have created


For background-size-008 test, you can link it to
ref-filled-green-100px-square.xht in



1- Null end tags (" /") have been removed since these are not needed in



1- Null end tags (" /") have been removed since these are not needed in

Line 25:  background-image: url(support/60x60-green.png);

has been replaced with

         background-image: url(support/50x50-green.png);

The reason is that scaling an image from 60px to 100px involves an
66.6666% increase: such percentage could be rounded up to 67% or down to
66% or up to 66.7% or down to 66.6% and then the increase could be 39px
or 41px instead of 40px. This phenomenon [1] actually happens in Opera
12.10 under Linux KDE 4.9.3 (i686 32bits); it does not happen in Windows
XP SP3 with Opera 12.10. Because such rounding down of fraction
phenomenon in Opera 12.10 does not in any way demonstrate a failure in
implementing background-size, then best is to use an image dimensions
that will scale without creating fractional pixel and rounding up or
down. So, the safest way to elegantly deal with such phenomenon or to
elegantly work around inaccurate platform APIs would be to use the
image so that whatever the background-size declaration is, then there
should be no fraction of a pixel created. Or to use an image which, when
scaled up or down, will not involve fractions: this explains why I
replaced 60x60 with 50x50 in background-size-008 test.

I have asked Řyvind Stenhaug from Opera to confirm my understanding of
such issue.

[1]: This phenomenon can be observed with Opera 12.10 under Linux KDE with

Contributions to the CSS 2.1 test suite:

CSS 2.1 Test suite RC6, March 23rd 2011:

CSS 2.1 test suite harness:

Contributing to to CSS 2.1 test suite:
Received on Saturday, 10 November 2012 00:39:18 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:13:25 UTC