W3C home > Mailing lists > Public > www-style@w3.org > October 2011

IE9 fails critical SVG-HTML integration & what is the status on this issue?

From: Kevin Ar18 <kevinar18@hotmail.com>
Date: Thu, 20 Oct 2011 22:08:42 -0400
Message-ID: <SNT110-W5307B3F9A13667A2B04A19AAE80@phx.gbl>
To: <public-fx@w3.org>, <www-style@w3.org>

Approximately 1 year ago (before the release of IE9) I brought up the subject of how browsers should handle pointer events/pointer interaction when you use SVG in an HTML page.

References:
http://wiki.csswg.org/spec/css3-ui#issue-5 (incomplete)
http://lists.w3.org/Archives/Public/www-style/2010Nov/0304.html (more details)

To quickly sum up, browsers need a way to support the following 6 states:

[1] Floating SVG elements that do not trigger events (critical)
[2] Inline SVG elements that do not trigger events
[3] Floating SVG elements that trigger events/block interaction.
[4] Inline SVG elements that trigger events/block interaction.

[5] Default behavior (regarding pointer events) for a floating SVG element
[6] Default behavior (regarding pointer events) for an inline SVG element

[1] is considered critical (there is no workaround); the rest are important, but optional, since there is always a workaround.

Last time I was involved in this discussion, many of these questions were still unanswered and/or needed to be added to the specs.


I am writing this for several reasons:

1. What is the current status on this issue?  I looked through a few logs, but is there any I should look at in particular?

2. Which of the 6 states has been written down in specs yet or decided on?

3. Perhaps most importantly, I would like to let the IE team know about the problem with ( State [1]: Floating SVG elements that do not trigger events)  IE 9 completely fails to support this and there is no workaround... Without IE support, SVG-HTML integration is impossible to do.  Private email didn't work, and I know the IE team watches these mailing lists, so I thought I'd post here.


4. I wrote testcases for states [1] and [2].  The attached file includes the tests I used.
48 Automated TestsFirefox 4.0B12pre (2011-02-16) - passes all, except "mousewheel"Opera 11.11, Build 2109 - passes critical 1 & 3, but fails semi-important 4 & 6 (pass: 1,2,3,5; fail: 4,6)Safari 5.0.5 (7533.21.1) r87071 - passes allIE 9.0.8112.16421 - fails critical tests 1 & 3 (pass 2,5; fail: 1,3,4,6)Manual TestsFirefox 10.0a1 (2011-10-20) - passes all testsOpera 11.11, Build 2109 - passes all critical tests (7.1 is "iffy": Opera lets you select text in 7.1)Safari 5.0.5 (7533.21.1) r87071 - passes all critical tests (does not work for: 5.1,6.1,7.2,8.4,8.5 because Safari does not have a context menu for SVG)IE 9.0.8112.16421 - interestingly enough, IE passes most of the critical stuff, but it is problematic in several ways:
				1-4 - shows extra options for svg -- this could be annoying/problematic5.2 fails: the cursor changes and it also shows the url in the status indicator at the bottom7.3 shows the I bar when it shouldn't8.1 shows teh I bar, but it shouldn't8.3 - the button highlights when you hover over; it shouldn't

	
5. I might be willing to write more testcases for all 6 states, based on what has been written into the specs and/or decided on in recent months. ... so would like to get caught up.

6. P.S. which mailing list(s) should I continue discussion on?  Last time I was told to CC several mailings lists with this topic.

Thanks,
Kevin

 		 	   		  


Received on Friday, 21 October 2011 02:09:12 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:45 GMT