379461 - [CSS21] Make XHTML <body> magic just like HTML <body>

Status: VERIFIED FIXED

(Core :: CSS Parsing and Computation, defect)

Description

[Simon Pieters [:zcorpan]]

User-Agent:       Opera/9.20 (Windows NT 5.1; U; en)
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9a5pre) Gecko/20070502 Minefield/3.0a5pre

<body> is magic in HTML but not in XHTML, and there are bugs in some edge cases. (Note that this bug is invalid per the current spec, but the plan is to change the spec: http://lists.w3.org/Archives/Public/www-style/2007Apr/0196.html )

Reproducible: Always

Steps to Reproduce:
1. Run the test cases.
2.
3.
Actual Results:  
'background-color', 'background-image' and 'overflow' is not handled equivalently between HTML and XHTML, and there are some bugs in the edge cases as well.

Expected Results:  
'background-color', 'background-image' and 'overflow' should work the same in both HTML and XHTML when applied to <body>. Edge cases should be handled as what the current spec says should happen for HTML.

Comment 1

[Simon Pieters [:zcorpan]]

Sorry, the Actual Results should say:

'background-color' and 'background-image' are not handled equivalently between HTML and XHTML, and there are some bugs in the edge cases as well.


(i.e., all tests on overflow pass already.)

Comment 2

[Simon Pieters [:zcorpan]]

Attachment: test case: background on body should fill the canvas

Comment 3

[Simon Pieters [:zcorpan]]

Attachment: test case: background on body should fill the canvas even when it has bogus siblings

Comment 4

[Simon Pieters [:zcorpan]]

Attachment: test case: background on body should not fill the canvas when root is not html:html

Comment 5

[Simon Pieters [:zcorpan]]

Attachment: test case: when background on body is propagated to the canvas, it shouldn't have a background itself

Comment 6

[Simon Pieters [:zcorpan]]

Attachment: test case: background on body when propagated to the canvas should be anchored at the root element

Comment 7

[Frank Wein [:mcsmurf]]

"but the plan is"...can we close that bug for now and open a new one when the spec has been changed for sure?

Comment 8

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

What do other browsers that support XHTML do here?

Comment 9

[Simon Pieters [:zcorpan]]

(In reply to comment #7)
> "but the plan is"...can we close that bug for now and open a new one when the
> spec has been changed for sure?

No, the spec won't change until implementors do this. The CSSWG said:

[[
This means that we reserve the possibility of making BODY in XHTML1 an 
exception, just like BODY in HTML, if we consider that the state of 
implementations makes it necessary. We'll make that decision no sooner 
than six months after publication of the document as a Candidate 
Recommendation.
]] -- http://lists.w3.org/Archives/Public/www-style/2007Apr/0196.html

Comment 10

[Simon Pieters [:zcorpan]]

(In reply to comment #8)
> What do other browsers that support XHTML do here?

Opera treats 'background-color' and 'background-image' the same between HTML and XHTML, but treats 'overflow' differently.

WebKit treats all of 'background-color', 'background-image' and 'overflow' differently.

Gecko treats 'background-color' and 'background-image' differently between HTML and XHTML, but treats 'overflow' equivalently.

I've added a table with the results at the bottom of http://simon.html5.org/test/css/magic-body/

HTH,

Comment 11

[Simon Pieters [:zcorpan]]

Attachment: overflow/004.htm: overflow on non-HTML body not should propagate to the viewport

Found out that this overflow test actually fails... (also the equivalent .xht).

Comment 12

[Simon Pieters [:zcorpan]]

Attachment: overflow/004.xht: overflow on non-HTML body not should propagate to the viewport

Comment 13

[Tomasz Kaźmierczak]

I can confirm the bug with body an background-color style.
In HTML mode firefox fills all the canvas with the body's bg-color, but in XHTML mode it fills only the body itself, and everything outside is white.
Opena and Konqueror do not have the problem.

Comment 14

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

attachment 264695 [details] and attachment 264696 [details] work for me on trunk.

Comment 15

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

Attachment: patch

Been meaning to get this in.

Comment 16

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

Comment on attachment 303605 [details] [diff] [review]
patch

An interdiff -w /dev/null this-patch shows:

@@ -3151,3 +3151,2 @@
         if (htmlDoc) {
-          if (!document->IsCaseSensitive()) { // HTML, not XHTML
             nsCOMPtr<nsIDOMHTMLElement> body;
@@ -3172,2 +3171 @@
         }
-      }

Comment 17

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 303605 [details] [diff] [review]
patch

r+sr=bzbarsky

Comment 18

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

Comment on attachment 303605 [details] [diff] [review]
patch

Very low risk patch to make it easier for authors to switch from HTML to XHTML.

Comment 19

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

Attachment: patch

Comment 20

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

Attachment: patch

Missed a case for backgrounds and another one relevant to scrollbars.  Added tests for both.

Comment 21

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

Comment on attachment 303676 [details] [diff] [review]
patch

comment 18 still applies for this approval request

Comment 22

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

Attachment: additional reftest adjustments

I also needed to make these adjustments to existing reftests, plus run 379461-3  (added in attachment 303676 [details] [diff] [review]; diffs to adjust not included in this patch) inside an iframe so that the scrollbars would show up in the drawWindow.

Comment 23

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

Landed 2008-02-19 08:46/47 -0800, backed out 2008-02-19 09:28 -0800 due to reftest failures (See comment 22), and relanded 2008-02-19 10:46 -0800 with reftest fixes.

Comment 24

[Marcia Knous [:marcia]]

verified fixed using  Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9b4pre) Gecko/2008022005 Minefield/3.0b4pre. Also verified on 10.5 - I verified using all of the test cases in the bug.

Comment 25

[Simon Pieters [:zcorpan]]

FYI: Using Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9b4pre) Gecko/2008022714 Minefield/3.0b4pre these tests fail:

http://simon.html5.org/test/css/magic-body/background-color/006-declarative.xht
http://simon.html5.org/test/css/magic-body/background-color/006.htm
http://simon.html5.org/test/css/magic-body/background-color/006.xht
(same with background-image)

http://simon.html5.org/test/css/magic-body/overflow/004.htm
http://simon.html5.org/test/css/magic-body/overflow/004.xht
(the declarative equivalent passes)

http://simon.html5.org/test/css/magic-body/overflow/006-declarative.xht
http://simon.html5.org/test/css/magic-body/overflow/006.htm
http://simon.html5.org/test/css/magic-body/overflow/006.xht

Comment 26

[Jeff Walden [:Waldo]]

(In reply to comment #25)

Can you file a new bug on these, please?

Comment 27

[Simon Pieters [:zcorpan]]

Filed bug 421469.