W3C home > Mailing lists > Public > public-css-testsuite@w3.org > September 2010

Review Report on sections 9.2.4 to 9.3.2 ~= 169 testcases

From: Gérard Talbot <css21testsuite@gtalbot.org>
Date: Sun, 5 Sep 2010 15:58:20 -0700
Message-ID: <2dd31102b3734bbfe4383b5b1ccd910b.squirrel@cp3.shieldhost.com>
To: "public-css-testsuite@w3.org" <public-css-testsuite@w3.org>
Cc: "Arron Eicholz" <arron.eicholz@microsoft.com>, "Boris Zbarsky" <bzbarsky@mit.edu>, "L. David Baron" <dbaron@dbaron.org>
Hello all,

This report is mostly for Fantasai, Arron Eicholz, Boris Zbarsky and
David Baron

Summary (mostly for Fantasai and Arron)
***************************************

- 2 testcases with fractional pixel problem found, this time affecting
Chrome 6 and Safari 5! See
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-091.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-092.htm
- if containing block of a positioned box is defined as padding box of
nearest positioned ancestor, then what is the content box of the
containing block referring to? I think the wording should be adjusted to
avoid causing confusion, to ease understanding
- offset and relative positioning: the offset (moving
up|left|bottom|right) is applied to the box regardless of its containing
block. So, the assert text should make that clear.
- 14 proposed replacements to testcases
- 1 proposed testcase to add:
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/position-absolute-percentage-inherit-001.htm
- I suggested that 2 testcases be rejected:
http://test.csswg.org/suites/css2.1/20100815/html4/display-applies-to-001.htm
http://test.csswg.org/suites/css2.1/20100815/html4/root-box-003.htm
- often in position-relative-*.htm testcases
<link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#positioning-scheme">
should be replaced with
<link rel="help" title="9.4.3 Relative positioning"
href="http://www.w3.org/TR/CSS21/visuren.html#position-props">
-
http://test.csswg.org/suites/css2.1/20100815/html4/position-relative-nested-001.htm
has many problems:
entire rewritten testcase:
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/position-relative-nested-001.htm
-
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-113.htm
assert text has to be corrected



*************************************
Section 9.2.4 The 'display' property
*************************************


----------------------

Author: Ian Hickson

http://test.csswg.org/suites/css2.1/20100815/html4/block-in-inline-001.htm

<link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#display-prop">
should be removed

and

<link rel="help" title="9.2.1.1 Anonymous block boxes"
href="http://www.w3.org/TR/CSS21/visuren.html#anonymous-block-level">
<meta name="flags" content="">
<meta name="assert" content="When an inline box contains a block box,
the inline box is broken around the block.">
should be added

----------------------

Author: Ian Hickson

http://test.csswg.org/suites/css2.1/20100815/html4/block-in-inline-002.htm

<link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#display-prop">
should be removed

and

<link rel="help" title="9.2.1.1 Anonymous block boxes"
href="http://www.w3.org/TR/CSS21/visuren.html#anonymous-block-level">
<meta name="flags" content="">
<meta name="assert" content="When an inline box contains a block box,
the inline box is broken around the block.">
should be added

----------------------

Author: Ian Hickson

http://test.csswg.org/suites/css2.1/20100815/html4/block-in-inline-003.htm

To add:
<meta name="flags" content="">
<meta name="assert" content="">

----------------------

Author: Ian Hickson

http://test.csswg.org/suites/css2.1/20100815/html4/block-in-inline-004.htm

To add: <meta name="assert" content="">

----------------------

Author: Ian Hickson

http://test.csswg.org/suites/css2.1/20100815/html4/block-in-inline-005.htm
and
http://test.csswg.org/suites/css2.1/20100815/html4/block-in-inline-006.htm

  <meta name="flags" content="dom">

should be replaced with

<meta http-equiv="Content-Script-Type" content="text/javascript">
<meta name="flags" content="dom interact">
<meta name="assert" content="When an inline box contains a block box,
the inline box is broken around the block.">

----------------------

Author: Ian Hickson

http://test.csswg.org/suites/css2.1/20100815/html4/block-in-inline-007.htm

  <meta name="flags" content="dom">

should be replaced with

<meta name="flags" content="">
<meta name="assert" content="When an inline box contains block boxes,
the inline box is broken around the blocks.">

----------------------

Author: Ian Hickson

http://test.csswg.org/suites/css2.1/20100815/html4/block-in-inline-008.htm

  <meta name="flags" content="dom">

should be replaced with

  <meta name="flags" content="">

----------------------

Author: Ian Hickson

http://test.csswg.org/suites/css2.1/20100815/html4/c561-list-displ-000.htm

- There is no class pc defined anywhere.
- div { color: navy; white-space: nowrap; }
If all the white-spaces are non-line-wrappable, then all of the &nbsp;
can be safely removed. As coded, the testcase "overdoes" this aspect.

To add:

<meta name="flags" content="">
<meta name="assert" content="">

----------------------

Author: Ian Hickson

http://test.csswg.org/suites/css2.1/20100815/html4/clear-applies-to-000.htm

To add:

<meta name="flags" content="">
<meta name="assert" content="">

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/descendant-display-none-001.htm
http://test.csswg.org/suites/css2.1/20100815/html4/descendant-display-override-001.htm
http://test.csswg.org/suites/css2.1/20100815/html4/display-001.htm
http://test.csswg.org/suites/css2.1/20100815/html4/display-002.htm
http://test.csswg.org/suites/css2.1/20100815/html4/display-003.htm
http://test.csswg.org/suites/css2.1/20100815/html4/display-004.htm
http://test.csswg.org/suites/css2.1/20100815/html4/display-005.htm
http://test.csswg.org/suites/css2.1/20100815/html4/display-006.htm
http://test.csswg.org/suites/css2.1/20100815/html4/display-007.htm
http://test.csswg.org/suites/css2.1/20100815/html4/display-008.htm
http://test.csswg.org/suites/css2.1/20100815/html4/display-009.htm
http://test.csswg.org/suites/css2.1/20100815/html4/display-010.htm
http://test.csswg.org/suites/css2.1/20100815/html4/display-011.htm
http://test.csswg.org/suites/css2.1/20100815/html4/display-012.htm
http://test.csswg.org/suites/css2.1/20100815/html4/display-013.htm
http://test.csswg.org/suites/css2.1/20100815/html4/display-014.htm
http://test.csswg.org/suites/css2.1/20100815/html4/display-015.htm
http://test.csswg.org/suites/css2.1/20100815/html4/display-016.htm
http://test.csswg.org/suites/css2.1/20100815/html4/display-017.htm

Approved

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/display-applies-to-001.htm

By default, <a> elements are inline-level in user agent stylesheets. So,
how can that testcase actually test setting 'display: inline' to an <a>
element? In other words, how could the test possibly fail?

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/display-applies-to-002.htm

The code does not meet the intent of the testcase.

<meta name="assert" content="Display set to 'inline' applies the correct
rendering behavior for the 'abbr' element.">

        <style type="text/css">
            div div
            {
                display: inline;
            }
        </style>

        <p>Test passes if all the "Filler Text" below is on the same
line.</p>
        <div>
            Filler Text
            <abbr>Filler Text</abbr>
            Filler Text
        </div>

- By default, <abbr> elements are inline-level.
- There is no nested div.
- Display set to inline for abbr is not accomplished in the test anyway.

Proposed replacement (with blockquote instead of abbr):
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/display-applies-to-002.htm

----------------------

Author: Boris Zbarsky

http://test.csswg.org/suites/css2.1/20100815/html4/display-change-001.htm

<meta http-equiv="Content-Script-Type" content="text/javascript">
<meta name="assert" content="">

should be added

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/display-initial-001.htm

"The initial display value for all elements is 'inline'."

User agent stylesheets all set initial display value for all elements
listed in DTD.
"Note that although the initial value of 'display' is 'inline', rules in
the user agent's default style sheet may override this value."
http://www.w3.org/TR/CSS21/visuren.html#display-prop

So, I do not know how it would be possible to test this without
resorting to invalid markup code.

Possible replacement in that sense:
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/display-initial-001.htm

There are 2 generated content testcases which test this:
http://test.csswg.org/suites/css2.1/20100815/html4/after-content-display-018.htm
and
http://test.csswg.org/suites/css2.1/20100815/html4/before-content-display-018.htm

----------------------

Author: David Baron

http://test.csswg.org/suites/css2.1/20100815/html4/inline-block-000.htm

There are 2 meta assert and no explicitly defined expected results.
As is, the testcase is passed if CSS support is disabled.

Proposed replacement:
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/inline-block-000.htm

----------------------

Author: David Baron

http://test.csswg.org/suites/css2.1/20100815/html4/inline-block-valign-001.htm

There are 2 meta assert, no explicitly defined expected results, 1
validation markup error. Testing inline-block, like in this testcase,
may be best done with a div, not a span.

Proposed replacement:
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/inline-block-valign-001.htm

----------------------

Author: Ian Hickson

http://test.csswg.org/suites/css2.1/20100815/html4/root-box-002.htm

- several validation markup errors
- meta assert missing

I spent quite a lot of time on this testcase (it's definitely an edge
case) and I propose this replacement:
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/root-box-002.htm

----------------------

Author: Ian Hickson

http://test.csswg.org/suites/css2.1/20100815/html4/root-box-003.htm

 <title>CSS Test: Styling the root element: display:none (page should be
a big green expanse)</title>

 <style type="text/css">
   html { display: none; background: green; color: red; }

If it is not supposed to be displayed, then how could it be painted? Why
should it be painted?

"display of 'none' does not create an invisible box; it creates no box
at all."
http://www.w3.org/TR/CSS21/visuren.html#display-prop

I do not think this testcase should be approved.

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/run-in-006.htm

<meta name="assert" content="Run-in boxes are converted to block boxes
when the run-in box runs into a table box.">

http://lists.w3.org/Archives/Public/www-style/2010Sep/0077.html

Approved

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/run-in-007.htm

Approved

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/run-in-008.htm
http://test.csswg.org/suites/css2.1/20100815/html4/run-in-009.htm
http://test.csswg.org/suites/css2.1/20100815/html4/run-in-010.htm
http://test.csswg.org/suites/css2.1/20100815/html4/run-in-011.htm
http://test.csswg.org/suites/css2.1/20100815/html4/run-in-012.htm
http://test.csswg.org/suites/css2.1/20100815/html4/run-in-013.htm

http://lists.w3.org/Archives/Public/www-style/2010Sep/0077.html

Approved

----------------------

*******************************
Section 9.3 Positioning schemes
*******************************

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-absolute-001.htm

Approved

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-relative-004.htm

The testcase has

  <link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#positioning-scheme">
        <meta name="flags" content="">
        <meta name="assert" content="Relative positioning with left or
right does not change size of a box.">

Ideally, the test should have left and right values and width: auto
(such that it would affect size when styled with position: absolute) and
then test displayed result.

"
For relatively positioned elements, 'left' and 'right' move the box(es)
horizontally, without changing their size. 'left' moves the boxes to the
right, and 'right' moves them to the left. Since boxes are not split or
stretched as a result of 'left' or 'right', the computed values are
always: left = -right.
"
Section 9.4.3 Relative positioning
http://www.w3.org/TR/CSS21/visuren.html#relative-positioning

Because of that I think
  <link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#positioning-scheme">

should be replaced with

  <link rel="help" title="9.4.3 Relative positioning"
href="http://www.w3.org/TR/CSS21/visuren.html#relative-positioning">

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-relative-005.htm

I propose to replace

<link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#positioning-scheme">

   <meta name="assert" content="Relatively positioned element with the
left specified offsets the element from the left of the containing
block.">

with

<link rel="help"  title="9.4.3 Relative positioning"
href="http://www.w3.org/TR/CSS21/visuren.html#position-props">

   <meta name="assert" content="Relatively positioned box with the left
specified offsets the left edge of the box itself from its position
in normal flow.">

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-relative-006.htm

I propose to replace

   <link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#positioning-scheme">
   <meta name="flags" content="">
   <meta name="assert" content="Relatively positioned box with the right
specified offsets the element from the right of the containing
block.">

with

   <link rel="help" title="9.4.3 Relative positioning"
href="http://www.w3.org/TR/CSS21/visuren.html#position-props">
   <meta name="flags" content="">
   <meta name="assert" content="Relatively positioned element with the
right specified offsets the edge of the box itself from its position
in normal flow.">

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-relative-007.htm

The assert is misleading or inappropriate.

The spec states

"
Since boxes are not split or stretched as a result of 'left' or 'right',
the computed values are always: left = -right.

If both 'left' and 'right' are 'auto' (their initial values), the
computed values are '0' (i.e., the boxes stay in their original
position).

If 'left' is 'auto', its computed value is minus the value of 'right'
(i.e., the boxes move to the left by the value of 'right').

If 'right' is specified as 'auto', its computed value is minus the value
of 'left'.
"
http://www.w3.org/TR/CSS21/visuren.html#relative-positioning


I propose to replace

   <link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#positioning-scheme">
   <meta name="flags" content="">
   <meta name="assert" content="Relatively positioned element with
'left' set to 'auto' and 'right' set to a value appears at expected
offset.">

with

   <link rel="help" title="9.4.3 Relative positioning"
href="http://www.w3.org/TR/CSS21/visuren.html#position-props">
   <meta name="flags" content="">
   <meta name="assert" content="A relatively positioned box with 'right'
set to a value moves the box to the left by the value of 'right'.">

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-relative-008.htm

The assert is misleading or inappropriate.

The spec states

"
Since boxes are not split or stretched as a result of 'left' or 'right',
the computed values are always: left = -right.

If both 'left' and 'right' are 'auto' (their initial values), the
computed values are '0' (i.e., the boxes stay in their original
position).

If 'left' is 'auto', its computed value is minus the value of 'right'
(i.e., the boxes move to the left by the value of 'right').

If 'right' is specified as 'auto', its computed value is minus the value
of 'left'.
"
http://www.w3.org/TR/CSS21/visuren.html#relative-positioning

I propose to replace

 <link rel="help" title="9.4.3 Relative positioning"
href="http://www.w3.org/TR/CSS21/visuren.html#positioning-scheme">
    <meta name="flags" content="">
    <meta name="assert" content="Relatively positioned element with
'right' set to 'auto' and 'left' set to a value appears at expected
offset.">

with

 <link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#position-props">
    <meta name="flags" content="">
    <meta name="assert" content="A relatively positioned box with 'left'
set to a value moves the box to the left by the value of 'right'.">

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-relative-009.htm

I propose to replace

<link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#positioning-scheme">

with

<link rel="help" title="9.4.3 Relative positioning"
href="http://www.w3.org/TR/CSS21/visuren.html#position-props">

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-relative-010.htm

I propose to replace

<link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#positioning-scheme">

with

<link rel="help" title="9.4.3 Relative positioning"
href="http://www.w3.org/TR/CSS21/visuren.html#position-props">

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-relative-013.htm

This test is not testing what its title and assert are stating. Both top
and bottom should be defined on a DHTML layer with height: auto or
height: some_value and then verify that the height has not changed due
to setting top and bottom.

"
If neither is 'auto', 'bottom' is ignored (i.e., the computed value of
'bottom' will be minus the value of 'top').
"

In the testcase, #div2 and #div3 are just moved from bottom toward top:
none of them are tested with both bottom and top set.

Proposed replacement:
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/position-relative-013.htm

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-relative-014.htm

I propose to replace

<link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#positioning-scheme">

with

<link rel="help" title="9.4.3 Relative positioning"
href="http://www.w3.org/TR/CSS21/visuren.html#position-props">

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-relative-015.htm

I propose to replace

<link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#positioning-scheme">

with

<link rel="help" title="9.4.3 Relative positioning"
href="http://www.w3.org/TR/CSS21/visuren.html#position-props">

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-relative-016.htm

I propose to replace

<link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#positioning-scheme">

with

<link rel="help" title="9.4.3 Relative positioning"
href="http://www.w3.org/TR/CSS21/visuren.html#position-props">


----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-relative-017.htm

I propose to replace

<link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#positioning-scheme">

with

<link rel="help" title="9.4.3 Relative positioning"
href="http://www.w3.org/TR/CSS21/visuren.html#position-props">

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-relative-018.htm

I propose to replace

<link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#positioning-scheme">

with

<link rel="help" title="9.4.3 Relative positioning"
href="http://www.w3.org/TR/CSS21/visuren.html#position-props">

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-relative-019.htm

I propose to replace

<link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#positioning-scheme">

with

<link rel="help" title="9.4.3 Relative positioning"
href="http://www.w3.org/TR/CSS21/visuren.html#position-props">

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-relative-020.htm
and
http://test.csswg.org/suites/css2.1/20100815/html4/position-relative-021.htm

In those 2 testcases, the wrapping #div1 does not require to have
position: relative; for the purpose of those testcases, only #div1's
child needs to be relatively positioned. The #div1 position could be
static.

I propose to replace

<link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#positioning-scheme">

with

<link rel="help" title="9.4.3 Relative positioning"
href="http://www.w3.org/TR/CSS21/visuren.html#position-props">

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-relative-nested-001.htm

The test has various problems:
- doctype declaration,
- correct link to swatch-red.png as background-image
- rel. pos. top percentage of a auto-height content
- y background-position: center will center the center of the image at
the center of the padding box of the element. This is not what top: 50%
is doing or is supposed to do or will do. So, even if everything above
is fixed, the calculation wrt overlapping will be off by half the height
of the innermost box (<p>).
- <link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#positioning-scheme">
should be instead/rather
<link rel="help" title="9.3.2 Box offsets: 'top', 'right', 'bottom',
'left'" href="http://www.w3.org/TR/CSS21/visuren.html#position-props">

More info:
http://lists.w3.org/Archives/Public/public-css-testsuite/2010Sep/0006.html

Proposed replacement (although the assert and title are radically
different):
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/position-relative-nested-001.htm

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/positioning-float-001.htm

The 3 "Filler text" are not distinguishible from each other. So,
strictly speaking, we can not be sure that the test is correctly passed,
actually successful. If CSS support is disabled, then the test is passed
by all browsers.

Proposed replacement:
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/positioning-float-001.htm

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/positioning-float-002.htm

Proposed replacement:
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/positioning-float-002.htm

Another idea would be to create the example from the spec itself at the
end of section 9.5.1, like this:
http://www.gtalbot.org/BrowserBugsSection/MSIE7Bugs/CSS21Section95FloatExample.html

----------------------

****************************************************************
Section 9.3.1 Choosing a positioning scheme: 'position' property
****************************************************************

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-001.htm
http://test.csswg.org/suites/css2.1/20100815/html4/position-002.htm
http://test.csswg.org/suites/css2.1/20100815/html4/position-003.htm
http://test.csswg.org/suites/css2.1/20100815/html4/position-004.htm
http://test.csswg.org/suites/css2.1/20100815/html4/position-005.htm
http://test.csswg.org/suites/css2.1/20100815/html4/position-absolute-002.htm

Approved

[Missing is a testcase where an absolutely positioned box is dimensioned
and positioned using solely 'top', 'right', 'bottom' and 'left' offset
values with percentage units and when its positioned child inherit all
of the properties.
Testcase:
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/position-absolute-percentage-inherit-001.htm
]
----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-applies-to-001.htm
http://test.csswg.org/suites/css2.1/20100815/html4/position-applies-to-002.htm
http://test.csswg.org/suites/css2.1/20100815/html4/position-applies-to-003.htm
http://test.csswg.org/suites/css2.1/20100815/html4/position-applies-to-004.htm
http://test.csswg.org/suites/css2.1/20100815/html4/position-applies-to-005.htm
http://test.csswg.org/suites/css2.1/20100815/html4/position-applies-to-006.htm
http://test.csswg.org/suites/css2.1/20100815/html4/position-applies-to-007.htm
http://test.csswg.org/suites/css2.1/20100815/html4/position-applies-to-013.htm
http://test.csswg.org/suites/css2.1/20100815/html4/position-applies-to-014.htm
http://test.csswg.org/suites/css2.1/20100815/html4/position-applies-to-015.htm

In all those 10 testcases,

<link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#choose-position">

should be replaced with

<link rel="help" title="9.7 Relationships between 'display', 'position',
and 'float'"
href="http://www.w3.org/TR/CSS21/visuren.html#position-props">

because this is the only spot in the spec where elements with display
set to  table or sub-table elements are described that they can be
absolutely positioned and how the position of the box can be determined.

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-applies-to-008.htm
http://test.csswg.org/suites/css2.1/20100815/html4/position-applies-to-009.htm

Approved

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-applies-to-010.htm

The testcase should mention that a bullet list-marker must be produced
and the code should provide sufficient space for it.

Proposed replacement:
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/position-applies-to-010.htm

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-applies-to-011.htm
http://test.csswg.org/suites/css2.1/20100815/html4/position-applies-to-012.htm

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-fixed-001.htm

Approved

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-fixed-002.htm

Approved

[It could be possible to create another testcase based on a more
realistic website scenario. Also, that testcase tests with absolute
units (in); there is no test with percentage and em units and
with/without inherit.]

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-fixed-003.htm

Approved

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-fixed-006.htm

id on html is not valid.

  <script type="text/javascript">
            var el = document.getElementById('test');
            document.write(window.getComputedStyle(el, null).position);
        </script>

can be replaced with

<script type="text/javascript">
document.write(window.getComputedStyle(document.documentElement,
null).position);
</script>

Referring to root element can be done:
http://www.w3.org/TR/DOM-Level-2-Core/core.html#ID-87CD092

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/position-relative-001.htm
http://test.csswg.org/suites/css2.1/20100815/html4/position-relative-002.htm
http://test.csswg.org/suites/css2.1/20100815/html4/position-static-001.htm

Approved

----------------------

***********************************************************
Section 9.3.2 Box offsets: 'top', 'right', 'bottom', 'left'
***********************************************************

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/bottom-004.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-005.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-006.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-007.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-008.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-016.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-017.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-018.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-019.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-020.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-028.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-029.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-030.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-031.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-032.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-040.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-041.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-042.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-043.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-044.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-052.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-053.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-054.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-055.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-056.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-064.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-065.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-066.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-067.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-068.htm


margin-top: 0.97in;
is supposed to be equal to 93px but it is equal to 93.12px.
This turns out to be okay (rounded down or truncated) with current
browsers but it may not be the case for all possible browsers out there
or future browsers.

Approved nevertheless

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/bottom-076.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-077.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-078.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-079.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-080.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-088.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-089.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-090.htm

            div
            {
                font: 12pt ahem;
                position: relative;
            }


There is no specifically required need for ahem font being defined.

Approved nevertheless

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/bottom-091.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-092.htm

Fractional pixel problem! This time with Chrome 6.0.472.53

  border-top: 1ex solid black; /* == 12px and not to 13px which would be
the rounding up of 12.8px */
  margin-top: 6.5ex; /* == 83px, rounded down from 83.2px */
  bottom: 7.5ex; /* == 96px */

Proposed replacements:
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/bottom-091.htm
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/bottom-092.htm

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/bottom-100.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-101.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-102.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-103.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-104.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-109.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-110.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-111.htm

Approved

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/bottom-112.htm

Wrt relative positioning,
"If both [top offset value and bottom offset value] are 'auto', their
computed values are both '0'."
http://www.w3.org/TR/CSS21/visuren.html#relative-positioning

That should be mentioned in the assert.

Proposed modification:

  <link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#position-props">
  <meta name="flags" content="">
  <meta name="assert" content="The 'bottom' property sets a value of
'auto'.">

changed into

  <link rel="help" title="9.4.3 Relative positioning"
href="http://www.w3.org/TR/CSS21/visuren.html#relative-positioning">
  <meta name="flags" content="">
  <meta name="assert" content="If both top offset value and bottom
offset value of a relatively positioned box are 'auto', then their
computed values are both '0'.">

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/bottom-113.htm

A lot has to be explained about such testcase in the testcase itself...

            #div1
            {
                border-top: 3px solid red;
                height: 1in;
                bottom: 100%;
            }
            #div2
            {
                border-top: 3px solid black;
                margin-top: 0.97in;
                bottom: inherit;
            }

#div1's bottom: 100% refers to 100% of its containing block's height but
since its containing block's height "is not specified explicitly (i.e.,
it depends on content height)" (see section 10.5:
http://www.w3.org/TR/CSS21/visudet.html#the-height-property ), then such
height is resolved as 'auto' which becomes 0; so, #div1's bottom edge is
not moving up, is not pushed upward.

On the other hand, #div2's 'bottom: inherit' refers to its parent's
computed value which is 100% (and not 0!) which refers to 100% of its
containing block's height which is 1in in this testcase. So, #div2's
bottom edge will be moved upward of 1in toward the top of the document
and then the #div2's black border should perfectly overlap #div#'s red
border.

The assert text should be updated accordingly to reflect all this.

Proposed modification:
  <meta name="assert" content="The 'bottom' property sets a value of
'inherit', inheriting the computed value from the parent element.">

changed to

  <meta name="assert" content="The 'bottom' property sets a value of
'inherit', inheriting the computed value from the parent element. When
the bottom offset is specified as a percentage, then the computed
value is the percentage. Percentage value of bottom refers to the
containing block's height as long as such containing block's height is
specified explicitly (i.e., it does not depend on its content
height).">

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/bottom-applies-to-001.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-applies-to-002.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-applies-to-003.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-applies-to-004.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-applies-to-005.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-applies-to-006.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-applies-to-007.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-applies-to-013.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-applies-to-014.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-applies-to-015.htm

In all those 10 testcases,

<link rel="help"
href="http://www.w3.org/TR/CSS21/visuren.html#position-props">

should be replaced with

<link rel="help" title="9.7 Relationships between 'display', 'position',
and 'float'"
href="http://www.w3.org/TR/CSS21/visuren.html#position-props">

because this is the only spot in the spec where elements with display
set to  table or sub-table elements are described that they can be
absolutely positioned and how the position of the box can be determined.

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/bottom-applies-to-008.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-applies-to-009.htm

Approved

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/bottom-applies-to-010.htm

The testcase should allocate sufficient horizontal space to display a
bullet and the expected results should mention that a bullet list-marker
must be produced.

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/bottom-applies-to-011.htm
http://test.csswg.org/suites/css2.1/20100815/html4/bottom-applies-to-012.htm

Approved

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/bottom-offset-001.htm

  <meta name="assert" content="The 'bottom' property, for absolute
positioning, specifies the offset of the element in relation to the
containing block's bottom content edge.">

but that is rather difficult to understand and awkward, I'd say.
Containing block for absolute positioned box is formed by padding box of
nearest positioned ancestor.

"
If the element has 'position: absolute', the containing block is
established by the nearest ancestor with a 'position' of 'absolute',
'relative' or 'fixed', in the following way: (...) the containing block
is formed by the padding edge of the [nearest positioned] ancestor.
"
Section 10.1 Containing block
http://www.w3.org/TR/CSS21/visudet.html#containing-block-details

To talk of content edge of containing block is for sure going to confuse
CSS beginners and CSS intermediates as well.

I see the same wrong explanation and assert in this testcase:
http://test.csswg.org/suites/css2.1/20100815/html4/containing-block-001.htm

Those testcases bottom-offset-001.htm and containing-block-001.htm
should be redone to include a padding belt around the content box and a
border belt and then be tested accordingly.

Proposed replacement:
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/bottom-offset-001.htm

[Note:
http://test.csswg.org/suites/css2.1/20100815/html4/containing-block-017.htm
http://test.csswg.org/suites/css2.1/20100815/html4/containing-block-018.htm
http://test.csswg.org/suites/css2.1/20100815/html4/containing-block-019.htm
and
http://test.csswg.org/suites/css2.1/20100815/html4/containing-block-020.htm
have the correct explanations but the nested abspos block element (in
those tests) has top:auto and left:auto
]

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/bottom-offset-002.htm

I suggest to replace

   <title>CSS Test: Position 'bottom' property defines the offset from
the bottom content edge of the parent box for relative
positioning</title>

and

   <meta name="assert" content="The 'bottom' property, for relative
positioning, specifies the offset of the element in relation to the
parent's bottom content edge.">

with

   <title>CSS Test: 'bottom' offset - relative positioning</title>

and

   <meta name="assert" content="The 'bottom' property, for relative
positioning, specifies the offset to be applied to the bottom edge of
the box itself.">

The testcase has, uses, resorts to 2 relatively positioned boxes but
that is not necessary since a relatively positioned box is positioned
with regards to its normal flow position, not with regards to its parent
(positioned or not) box. In fact, I believe it would be better to remove
anything and everything not required by the testcase goals.

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/bottom-offset-003.htm

Approved

----------------------

Author: Microsoft

http://test.csswg.org/suites/css2.1/20100815/html4/bottom-offset-percentage-001.htm

Proposed replacement:
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/bottom-offset-percentage-001.htm

----------------------


regards, Gérard
-- 
Contributions to the CSS 2.1 test suite:
http://www.gtalbot.org/BrowserBugsSection/css21testsuite/

CSS 2.1 test suite (beta 3; August 15th 2010):
http://test.csswg.org/suites/css2.1/20100815/html4/toc.html

CSS 2.1 test suite contributors:
http://test.csswg.org/source/contributors/
Received on Sunday, 5 September 2010 22:59:00 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 20 September 2010 17:52:03 GMT