See also: IRC log
trackbot, start meeting
<scribe> Meeting: ARIA Testing
<scribe> scribe: jongund
JG: I have not heard back from Deque people on creating tests
RS: ARIA modal is an issue
... On the OS X plateform prunes, how do we represent that in
the results
<richardschwerdtfeger> https://www.w3.org/wiki/ARIA_1.1_Testable_Statements#aria-modal.3D.22true.22_on_a_dialog
JG: Trying what are the different
types values/ results we need to represent
... That one sounds like it might be hard and to do a manul
test this round
JD: What I think would be to do,
start a top level container with some children, and then make
the dialog appear and then get a child level count
... When the modal appears the other children will be gone and
there will only be one
RS: You need to check that ARIA modal, how do you want me to do this
JD: Cynthia of proposed JSON
object, she had an array
... I would have an array of element assertions
... first part of the test has three children...
RS: We need to put some additional text, the element X has so many children, but we have to make sure stuff is gone
JD: SO the child count one
... The reason I say child count because it should be available
on all platforms
RS: So you only want to do a child count
JD: SO when you get the modal dialog, it has a single child, the parent ....
JG: We have the capability to say something is not defined
JD: We need to know that some elements are not there
JG: We need to define what we want to test
JD: We need to find an object and then test its property, determining if some element does not exist is harder and less reliable
<richardschwerdtfeger> https://www.w3.org/wiki/ARIA_1.1_Testable_Statements#aria-modal.3D.22true.22_on_a_dialog
JD: I like the sibling idea
... Can you do a test where you don't have aria modal and then
have modal
RS: We have modal=false the next test
JD: If we have the sibling test,
in one test there will be no siblings and in the next test
there will be siblings
... I would phrase it this way, the parent element with
role=test has N child count, so instead of saying sibling we
say how many children does the parent have
... DOes that make sense RS
... I think you did it right, but the wrong place in the
wiki
... Your statement is only true for the AXPI platform
RS: I will put that in there
JG: Other issues?
RS: No
JD: Shane has a sample ATTA, so I am hoping to a demo at TPAC for my plateform
JG: Good news!
JD: I working on it and think I
am close
... My platform accessibility listener, we need to block the
main thread
... The other thing I am hoping, ... , IA2
JG: You have code, we can translate it to C#
JD: NVDA does IA2, if it doesn't have to be in C# and then it could be done Python
JG: Send me a link to your code and we can take a look at it
JD: I have a question, can Jon participate
RS: I need to leave, can you take a look what I did for ATK
JD: Technically no, but I know
what you mean
... I am looking for state model is set
... If it is not in the state set....
RS: You want it to say....
JD: In column 2 ...
... discussion of ATK mapping details for state modal ...
... What you have is parsable, but not technically correct
RS: What do we do with shown, what do you want me to do?
JD: Present versus absent
...
RS: I am going to put true
... I am leaving
DS: The thing I am interested in
is creating tests for SVG, JD has been working with Fred
... Fred is planning on becoming an invited expert
... I want to work with JD and Fred on SVG testing
JD: All the little kinks are
worked out for core ARIA, they can be used for SVG
testing
... We have all these tables and work out the issues, we should
be able to use Freds existing tables
DS: I will joining these calls on an as needed basis
JD: Do we have anyone working on OS X?
JG: I am working on it
JD: If you are planning on automated testing who will do a platform tools
DS: We know there are some limitations with automated testing and accessibility
This is scribe.perl Revision: 1.144 of Date: 2015/11/17 08:39:34 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Found Scribe: jongund Inferring ScribeNick: jongund WARNING: No "Topic:" lines found. Present: Joanmarie_Diggs shepazu WARNING: Fewer than 3 people found for Present list! Regrets: cythnia Found Date: 14 Sep 2016 Guessing minutes URL: http://www.w3.org/2016/09/14-aria-test-minutes.html People with action items: WARNING: No "Topic: ..." lines found! Resulting HTML may have an empty (invalid) <ol>...</ol>. Explanation: "Topic: ..." lines are used to indicate the start of new discussion topics or agenda items, such as: <dbooth> Topic: Review of Amy's report[End of scribe.perl diagnostic output]