<?xml version="1.0" encoding="UTF-8" standalone="yes" ?>
<!DOCTYPE bugzilla SYSTEM "https://www.w3.org/Bugs/Public/page.cgi?id=bugzilla.dtd">

<bugzilla version="5.0.4"
          urlbase="https://www.w3.org/Bugs/Public/"
          
          maintainer="sysbot+bugzilla@w3.org"
>

    <bug>
          <bug_id>19772</bug_id>
          
          <creation_ts>2012-10-30 10:16:02 +0000</creation_ts>
          <short_desc>if statement not needed for cobobox MSAA+UIA Express</short_desc>
          <delta_ts>2013-06-13 17:19:11 +0000</delta_ts>
          <reporter_accessible>1</reporter_accessible>
          <cclist_accessible>1</cclist_accessible>
          <classification_id>1</classification_id>
          <classification>Unclassified</classification>
          <product>ARIA</product>
          <component>Core AAM</component>
          <version>1.0</version>
          <rep_platform>PC</rep_platform>
          <op_sys>Windows NT</op_sys>
          <bug_status>CLOSED</bug_status>
          <resolution>FIXED</resolution>
          
          
          <bug_file_loc></bug_file_loc>
          <status_whiteboard></status_whiteboard>
          <keywords></keywords>
          <priority>P2</priority>
          <bug_severity>normal</bug_severity>
          <target_milestone>---</target_milestone>
          
          
          <everconfirmed>1</everconfirmed>
          <reporter name="Cynthia Shelly">cyns</reporter>
          <assigned_to name="Cynthia Shelly">cyns</assigned_to>
          <cc>andi.snowweaver</cc>
    
    <cc>clown</cc>
    
    <cc>dbolter</cc>
    
    <cc>surkov.alexander</cc>
          
          <qa_contact name="ARIA UA Implementors">aria-ua-impl</qa_contact>

      

      

      

          <comment_sort_order>oldest_to_newest</comment_sort_order>  
          <long_desc isprivate="0" >
    <commentid>77390</commentid>
    <comment_count>0</comment_count>
    <who name="Cynthia Shelly">cyns</who>
    <bug_when>2012-10-30 10:16:02 +0000</bug_when>
    <thetext>remove &quot;If aria-expanded (state) is not &quot;true&quot;, expose STATE_SYSTEM_COLLAPSED&apos; from UIA Express column for combox.  This is normal behavior for aria-expanded property and is covered in property table.

Check with David if this is needed for IA2.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>77959</commentid>
    <comment_count>1</comment_count>
    <who name="Andi Snow-Weaver">andi.snowweaver</who>
    <bug_when>2012-11-06 17:33:19 +0000</bug_when>
    <thetext>Done for MSAA+UIAExpress.

David, does this text need to be removed from the MSAA+IA2 column?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>78026</commentid>
    <comment_count>2</comment_count>
    <who name="alexander surkov">surkov.alexander</who>
    <bug_when>2012-11-08 01:27:26 +0000</bug_when>
    <thetext>It&apos;s not really covered by aria-expanded mapping:

&quot;aria-expanded is undefined (state) (default)&quot; has &quot;Not mapped&quot; for all platforms including both UIA and IA2.

Bug doesn&apos;t seems valid.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>78053</commentid>
    <comment_count>3</comment_count>
    <who name="Andi Snow-Weaver">andi.snowweaver</who>
    <bug_when>2012-11-08 14:55:59 +0000</bug_when>
    <thetext>Cynthia,

Can you please respond to Alex&apos;s comment in the bug? Do we need to change the aria-expanded undefined row now that we have removed this text?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>83509</commentid>
    <comment_count>4</comment_count>
    <who name="David Bolter">dbolter</who>
    <bug_when>2013-02-21 18:23:32 +0000</bug_when>
    <thetext>(In reply to comment #2)
&gt; It&apos;s not really covered by aria-expanded mapping:
&gt; 
&gt; &quot;aria-expanded is undefined (state) (default)&quot; has &quot;Not mapped&quot; for all
&gt; platforms including both UIA and IA2.

Alex we discussed this briefly during a W3C call but weren&apos;t 100% sure what you meant here.

&gt; 
&gt; Bug doesn&apos;t seems valid.


Sounds like we can close this bug as INVALID/WORKSFORME right? (I&apos;d agree)</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>83525</commentid>
    <comment_count>5</comment_count>
    <who name="alexander surkov">surkov.alexander</who>
    <bug_when>2013-02-22 02:05:59 +0000</bug_when>
    <thetext>(In reply to comment #4)
&gt; (In reply to comment #2)
&gt; &gt; It&apos;s not really covered by aria-expanded mapping:
&gt; &gt; 
&gt; &gt; &quot;aria-expanded is undefined (state) (default)&quot; has &quot;Not mapped&quot; for all
&gt; &gt; platforms including both UIA and IA2.
&gt; 
&gt; Alex we discussed this briefly during a W3C call but weren&apos;t 100% sure what
&gt; you meant here.

I don&apos;t recall details. But I think the problem is.

Cynthia asked to remove &quot;If aria-expanded (state) is not &quot;true&quot;, expose STATE_SYSTEM_COLLAPSED&apos; from UIA Express column for combox&quot; because &quot;This is normal behavior for aria-expanded property and is covered in property table&quot;. However aria-expanded property table says &quot;aria-expanded is undefined (state) (default)&quot; has &quot;Not mapped&quot; for all platforms&quot;. That means if those words are removed then combobox having not &quot;true&quot; aria-expanded shouldn&apos;t expose STATE_SYSTEM_COLLAPSED.

&gt; &gt; Bug doesn&apos;t seems valid.
&gt; 
&gt; Sounds like we can close this bug as INVALID/WORKSFORME right? (I&apos;d agree)

You can&apos;t call it invalid or worksforme until you back out the fix in MSAA+UIAExpress case (see comment #1).</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>84781</commentid>
    <comment_count>6</comment_count>
    <who name="David Bolter">dbolter</who>
    <bug_when>2013-03-21 18:22:04 +0000</bug_when>
    <thetext>(In reply to comment #5)
&gt; (In reply to comment #4)
&gt; &gt; (In reply to comment #2)
&gt; &gt; &gt; It&apos;s not really covered by aria-expanded mapping:
&gt; &gt; &gt; 
&gt; &gt; &gt; &quot;aria-expanded is undefined (state) (default)&quot; has &quot;Not mapped&quot; for all
&gt; &gt; &gt; platforms including both UIA and IA2.
&gt; &gt; 
&gt; &gt; Alex we discussed this briefly during a W3C call but weren&apos;t 100% sure what
&gt; &gt; you meant here.
&gt; 
&gt; I don&apos;t recall details. But I think the problem is.
&gt; 
&gt; Cynthia asked to remove &quot;If aria-expanded (state) is not &quot;true&quot;, expose
&gt; STATE_SYSTEM_COLLAPSED&apos; from UIA Express column for combox&quot; because &quot;This is
&gt; normal behavior for aria-expanded property and is covered in property
&gt; table&quot;. However aria-expanded property table says &quot;aria-expanded is
&gt; undefined (state) (default)&quot; has &quot;Not mapped&quot; for all platforms&quot;. That means
&gt; if those words are removed then combobox having not &quot;true&quot; aria-expanded
&gt; shouldn&apos;t expose STATE_SYSTEM_COLLAPSED.

This made my head hurt but I think I got it :)

Cynthia are you sure you don&apos;t want to state that UIA Express should expose STATE_SYSTEM_COLLAPSED for the case where role=combobox and aria-expanded is undefined?

&gt; 
&gt; &gt; &gt; Bug doesn&apos;t seems valid.
&gt; &gt; 
&gt; &gt; Sounds like we can close this bug as INVALID/WORKSFORME right? (I&apos;d agree)
&gt; 
&gt; You can&apos;t call it invalid or worksforme until you back out the fix in
&gt; MSAA+UIAExpress case (see comment #1).

It took me a while but I think here you are simply pointing out a technicality that if we close this as is then it would be as &quot;FIXED&quot; not INVALID/WORKSFORME right?

Alex are you OK leaving this one up to Cynthia/MS or are you concerned about this text WRT a possible FF UI Express implementation?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>84819</commentid>
    <comment_count>7</comment_count>
    <who name="alexander surkov">surkov.alexander</who>
    <bug_when>2013-03-22 04:25:58 +0000</bug_when>
    <thetext>If we leave out it as it is then it means that

&lt;div role=&quot;combobox&quot;&gt; is neither collapsed nor expanded in UIA Express and it is collapsed in IA2. In other words IA2 and UIA Express have different requirements about MSAA states. In case of Firefox both IA2 and UIA Express shares same MSAA tree so that will make us to break the spec this or that way.

On the another hand if UIA bridge doesn&apos;t guess that neither collapsed nor expanded combobox is actually collapsed then screen readers should have a different logic for combobox processing depending on API. I don&apos;t think more diversity in this case is sensible.

Anyway we could do a guessing more but it&apos;d be great to hear Cynthia. It seems there is a bunch of options here:
1) It was done on purpose to comply with some UIA requirements
2) I missed some part of the spec and my guesses are wrong
3) It was a mistake</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>84849</commentid>
    <comment_count>8</comment_count>
    <who name="Joseph Scheuhammer">clown</who>
    <bug_when>2013-03-22 16:52:12 +0000</bug_when>
    <thetext>Here&apos;s another consideration.  The spec for combobox asserts that aria-expanded is a *required* state [1] -- see the &quot;Characteristics of combobox table&quot;.  The implication is that comboboxes are inherently expandable/collapsible.

That, in turn, means a combobox is either expanded or collapsed.  I think it&apos;s a safe bet that when first instantiated, the combobox is collapsed.  Rather, I can&apos;t think of a use case where a combobox is initially expanded without the user interacting with it. (It&apos;s possible that I&apos;m just not creative enough...).

This logic leads back to &quot;if aria-expanded is not &apos;true&apos;, expose STATE_SYSTEM_COLLAPSED&quot;.

[1] http://www.w3.org/WAI/PF/aria/roles#combobox</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>84880</commentid>
    <comment_count>9</comment_count>
    <who name="alexander surkov">surkov.alexander</who>
    <bug_when>2013-03-23 01:47:21 +0000</bug_when>
    <thetext>Shouldn&apos;t UAIG explicitly say that collapsed combobox means SYSTEM_STATE_COLLAPSED must be exposed in case of MSAA+UIA/MSAA+IA2 ?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>84925</commentid>
    <comment_count>10</comment_count>
    <who name="Joseph Scheuhammer">clown</who>
    <bug_when>2013-03-25 15:19:08 +0000</bug_when>
    <thetext>(In reply to comment #9)
&gt; Shouldn&apos;t UAIG explicitly say that collapsed combobox means
&gt; SYSTEM_STATE_COLLAPSED must be exposed in case of MSAA+UIA/MSAA+IA2 ?

It does for MSAA+IA2.
It doesn&apos;t for MSAA+UIA because Cynthia requested it be removed (the reason for this bugzilla).

IMHO, that edit should be undone; but, waiting for Cynthia&apos;s opinion.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>85111</commentid>
    <comment_count>11</comment_count>
    <who name="Cynthia Shelly">cyns</who>
    <bug_when>2013-03-28 17:32:50 +0000</bug_when>
    <thetext>Go ahead and put it back, pending david&apos;s approval of Issue 566.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>87181</commentid>
    <comment_count>12</comment_count>
    <who name="David Bolter">dbolter</who>
    <bug_when>2013-05-02 17:31:58 +0000</bug_when>
    <thetext>(In reply to comment #11)
&gt; Go ahead and put it back, pending david&apos;s approval of Issue 566.

Sure I&apos;m fine with reintroducing text like: &apos;If aria-expanded (state) is not &quot;true&quot;, expose STATE_SYSTEM_COLLAPSED&apos; to be put in the UIA Express Column for role combobox. (Issue 566)</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>87182</commentid>
    <comment_count>13</comment_count>
    <who name="Joseph Scheuhammer">clown</who>
    <bug_when>2013-05-02 17:32:54 +0000</bug_when>
    <thetext>(In reply to comment #12)
&gt; (In reply to comment #11)
&gt; &gt; Go ahead and put it back, pending david&apos;s approval of Issue 566.
&gt; 
&gt; Sure I&apos;m fine with reintroducing text like: &apos;If aria-expanded (state) is not
&gt; &quot;true&quot;, expose STATE_SYSTEM_COLLAPSED&apos; to be put in the UIA Express Column
&gt; for role combobox. (Issue 566)

Cynthia, feel free to close this given David&apos;s comment.</thetext>
  </long_desc>
      
      

    </bug>

</bugzilla>