W3C home > Mailing lists > Public > wai-xtech@w3.org > August 2012

Re: Minutes from the ARIA Task Force Teleconference on 20 August

From: Benjamin Hawkes-Lewis <bhawkeslewis@googlemail.com>
Date: Wed, 22 Aug 2012 23:58:21 +0100
Message-ID: <CAEhSh3cY1YppN4R+2+zx87pgTFm72GSQzU8ZK5hw33oJMmBYKQ@mail.gmail.com>
To: WAI XTech <wai-xtech@w3.org>
Cc: Richard Schwerdtfeger <schwer@us.ibm.com>
On Wed, Aug 22, 2012 at 10:25 PM, Janina Sajka <janina@rednote.net> wrote:
>                                        Protocols and Formats Working Group Teleconference
>                                                           20 Aug 2012


>    RS:


>    browsers do not fix parsing errors in hidden content
>    SO, AT would have to both parseand repair the hidden content
>    This is way aoutside the scope of AT.

Wow. Just … wow. :(

Very disappointing to hear that PFWG is planning to make a Change
Proposal based on such misunderstandings.

I've explained twice that HTML WG is certainly not proposing that AT
should parse and repair HTML:



Richard's eccentric claim that "browsers do not fix parsing errors in
hidden content" is demonstrably wrong.

Try this simple test case in a modern browser (I used Firefox, Safari, Chrome):

  <!doctype html>
  <title>Parse and repair of hidden content</title>
  <base href=http://example.com>
  <div hidden aria-hidden=true style=display:none>
  <p>1<b>2<i>3</b>4</i>5</p><a href=foo></a>
  document.querySelector("pre").textContent =
     document.querySelector("[hidden]").innerHTML +
     "\n" +

Here's the test case in the DOM viewer:


It renders:

  <p>1<b>2<i>3</i></b><i>4</i>5</p><a href="foo"></a>


This test case proves that browsers fix incorrectly nested tags,
resolve href URLs, and parse markup inside an element with @hidden set
into the host DOM, as is required by HTML5.

Benjamin Hawkes-Lewis
Received on Wednesday, 22 August 2012 22:59:09 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:51:46 UTC