Adding DumpAccessibilityTree Tests

Adding DumpAccessibilityTree Tests

HTML Tags: del, dfn, img, menu, ol, option, q, source
Aria role: alert, img, timer
Aria state/property: aria-readonly="false", aria-hidden="false"

BUG=170582
Committed: https://crrev.com/baa385c24c57df62f64a5f1fecee739dc2dcc3ba
Cr-Commit-Position: refs/heads/master@{#293692}

[shreeramk, 2014-08-25 11:45:43]

Could you please suggest me the changes?
Thanks!

[dmazzoni, 2014-08-25 15:00:18]

https://codereview.chromium.org/496313003/diff/1/content/browser/accessibilit...
File content/browser/accessibility/aria-readonly.html (right):

https://codereview.chromium.org/496313003/diff/1/content/browser/accessibilit...
content/browser/accessibility/aria-readonly.html:3: <div role="textbox"
aria-readonly="false" tabindex="0">readonly=false</div>
I'd suggest one example with aria-readonly=false, and another with
aria-readonly=true.

https://codereview.chromium.org/496313003/diff/1/content/test/data/accessibil...
File content/test/data/accessibility/aria-hidden.html (right):

https://codereview.chromium.org/496313003/diff/1/content/test/data/accessibil...
content/test/data/accessibility/aria-hidden.html:3: <div
aria-hidden="false">hidden=false</div>
Same - add one with aria-hidden=true also

Also, aria-hidden is supposed to override display:none, so it'd be nice to test
that

[shreeramk, 2014-08-25 15:36:57]

from now onwards, I ll take care of all the values of a particular state in the
same test file.

https://codereview.chromium.org/496313003/diff/1/content/browser/accessibilit...
File content/browser/accessibility/aria-readonly.html (right):

https://codereview.chromium.org/496313003/diff/1/content/browser/accessibilit...
content/browser/accessibility/aria-readonly.html:3: <div role="textbox"
aria-readonly="false" tabindex="0">readonly=false</div>
On 2014/08/25 15:00:17, dmazzoni wrote:
> I'd suggest one example with aria-readonly=false, and another with
> aria-readonly=true.

Ok sure. I added only readonly=false as only that was mentioned in the bug.

https://codereview.chromium.org/496313003/diff/1/content/test/data/accessibil...
File content/test/data/accessibility/aria-hidden.html (right):

https://codereview.chromium.org/496313003/diff/1/content/test/data/accessibil...
content/test/data/accessibility/aria-hidden.html:3: <div
aria-hidden="false">hidden=false</div>
On 2014/08/25 15:00:18, dmazzoni wrote:
> Same - add one with aria-hidden=true also
> 
> Also, aria-hidden is supposed to override display:none, so it'd be nice to
test
> that
> 

Ok Sure :) (same reason).

[shreeramk, 2014-08-26 12:47:38]

PTAL :)

Thanks!

[dmazzoni, 2014-08-26 15:18:59]

Thanks for this contribution!

https://codereview.chromium.org/496313003/diff/20001/content/browser/accessib...
File content/browser/accessibility/aria-readonly.html (right):

https://codereview.chromium.org/496313003/diff/20001/content/browser/accessib...
content/browser/accessibility/aria-readonly.html:6: <div role="textbox"
aria-readonly="false" tabindex="0">This tests is for aria-readonly=false</div>
This tests -> This test

https://codereview.chromium.org/496313003/diff/20001/content/browser/accessib...
File content/browser/accessibility/dump_accessibility_tree_browsertest.cc
(right):

https://codereview.chromium.org/496313003/diff/20001/content/browser/accessib...
content/browser/accessibility/dump_accessibility_tree_browsertest.cc:308:
IN_PROC_BROWSER_TEST_F(DumpAccessibilityTreeTest, AccessibilityAriaalert) {
To be consistent with the rest of the file, capitalize the first letter of the
role after Aria, like:

AccessibilityAriaAlert

and similarly for others in the file

https://codereview.chromium.org/496313003/diff/20001/content/test/data/access...
File content/test/data/accessibility/aria-hidden.html (right):

https://codereview.chromium.org/496313003/diff/20001/content/test/data/access...
content/test/data/accessibility/aria-hidden.html:11: <div id="overridedisplay"
aria-hidden="false">hidden=false</div>
This doesn't really test all of the permutations to make sure they all interact
properly. I'd test these six combinations:

display: block, no aria-hidden attribute
display: none, no aria-hidden attribute
display: block, aria-hidden=true
display: none, aria-hidden=true
display: block, aria-hidden=false
display: none, aria-hidden=false

https://codereview.chromium.org/496313003/diff/20001/content/test/data/access...
File content/test/data/accessibility/aria-img.html (right):

https://codereview.chromium.org/496313003/diff/20001/content/test/data/access...
content/test/data/accessibility/aria-img.html:6: <div role="img"> ... This tests
is for aria role="img" ...</div>
This tests -> This test (fix throughout this changelist)

https://codereview.chromium.org/496313003/diff/20001/content/test/data/access...
File content/test/data/accessibility/aria-readonly-expected-mac.txt (right):

https://codereview.chromium.org/496313003/diff/20001/content/test/data/access...
content/test/data/accessibility/aria-readonly-expected-mac.txt:1: AXWebArea
This expectation shows that the html wasn't a good example.

Aria-readonly applies mostly to text boxes. I'd use an input element:

<input aria-readonly="true" value="Uneditable text">
<input aria-readonly="false" value="Editable text">

https://codereview.chromium.org/496313003/diff/20001/content/test/data/access...
File content/test/data/accessibility/del-expected-win.txt (right):

https://codereview.chromium.org/496313003/diff/20001/content/test/data/access...
content/test/data/accessibility/del-expected-win.txt:3: ROLE_SYSTEM_STATICTEXT
name='I am '
This expectation basically says that DEL doesn't map to anything.

The definitive reference is here - make sure it agrees that the DEL element
doesn't map to a specific role or attribute in either Windows or Mac:

http://www.w3.org/TR/html-aapi/

https://codereview.chromium.org/496313003/diff/20001/content/test/data/access...
File content/test/data/accessibility/menu-type-context-expected-win.txt (right):

https://codereview.chromium.org/496313003/diff/20001/content/test/data/access...
content/test/data/accessibility/menu-type-context-expected-win.txt:3:
ROLE_SYSTEM_STATICTEXT name='Right-click inside this box to see the context
menu!'
This is not correct.

According to http://www.w3.org/TR/html-aapi/, this is supposed to be in a
ROLE_SYSTEM_MENUITEM.

What you should do is make your best guess as to what the correct output of this
test should be in the expectation file, then put a <#SKIP line at the top
(search for other expectation files with "skip" in them to see the syntax).

You don't have to fix the bug in this test, but don't check in a test that makes
lt look like we're doing it correctly, since we're not handling this one
correctly yet!

https://codereview.chromium.org/496313003/diff/20001/content/test/data/access...
File content/test/data/accessibility/option-in-datalist-expected-mac.txt
(right):

https://codereview.chromium.org/496313003/diff/20001/content/test/data/access...
content/test/data/accessibility/option-in-datalist-expected-mac.txt:1: AXWebArea
This clearly isn't right!

I don't think the datalist element does much when it's not connected to an input
element, right?

Maybe it doesn't make sense to test this one. I'd either get rid of it, or put
the "skip" tag in the expectation

[shreeramk, 2014-08-27 07:36:22]

All comments has been taken care in this patch set, except one for del.html

The spec says for del, that it doesn't have an accessible. Used styles are
mapped into text attributes on its text container.

So, I think it won't map to any attribute

[shreeramk, 2014-08-27 12:47:31]

PTAL!

Still I am getting the same expected result for aria-readonly on mac and win.

[shreeramk, 2014-09-03 17:26:59]

On 2014/08/27 12:47:31, shreeram.k wrote:
> PTAL!
> 
> Still I am getting the same expected result for aria-readonly on mac and win.

PTAL!

[dmazzoni, 2014-09-03 21:30:02]

I think this is done once the aria-hidden expectations are in!

https://codereview.chromium.org/496313003/diff/60001/content/test/data/access...
File content/test/data/accessibility/aria-hidden.html (right):

https://codereview.chromium.org/496313003/diff/60001/content/test/data/access...
content/test/data/accessibility/aria-hidden.html:13: #overridedisplay { display:
none; }
unused?

https://codereview.chromium.org/496313003/diff/60001/content/test/data/access...
content/test/data/accessibility/aria-hidden.html:16: <body>
I don't see the expectations for aria-hidden

[dmazzoni, 2014-09-03 21:31:53]

https://codereview.chromium.org/496313003/diff/60001/content/browser/accessib...
File content/browser/accessibility/aria-readonly.html (right):

https://codereview.chromium.org/496313003/diff/60001/content/browser/accessib...
content/browser/accessibility/aria-readonly.html:6: <input aria-readonly="false"
tabindex="0" value="Editable text">This test is for aria-readonly=false</div>
The closing </div> may be confusing things?

[shreeramk, 2014-09-04 13:38:58]

PTAL!

For aria-hidden.html, I have uploaded the expected file on win and mac.

For aria-readonly, I generated the expected file on my win system and uploaded,
but it seems the try bot is generating something else.(Same as previous output)

[dmazzoni, 2014-09-04 15:08:38]

lgtm

OK, please just put the "skip" line in the expectation files for aria-readonly,
and I'll take a look and try to figure out why they're broken.

Thanks!

[dmazzoni, 2014-09-04 15:09:16]

To clarify, please feel free to commit once try jobs pass. Just skip any test
expectations that you can't get to work.

[shreeramk, 2014-09-05 13:20:48]

The CQ bit was checked by shreeram.k@samsung.com

[commit-bot: I haz the power, 2014-09-05 13:21:45]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/shreeram.k@samsung.com/496313003/200001

[commit-bot: I haz the power, 2014-09-05 15:28:26]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-09-05 15:28:28]

Try jobs failed on following builders:
  android_dbg_tests_recipe on tryserver.chromium.linux
(http://build.chromium.org/p/tryserver.chromium.linux/builders/android_dbg_tes...)

[dmazzoni, 2014-09-05 15:45:44]

You need -expected-android.txt files. It's fine with me if you leave those all
as "skip", but you can't leave them out of the changelist.

[shreeramk, 2014-09-05 16:36:42]

On 2014/09/05 15:45:44, dmazzoni wrote:
> You need -expected-android.txt files. It's fine with me if you leave those all
> as "skip", but you can't leave them out of the changelist.

Ok. Tomorrow I ll add the expectations for android.

[shreeramk, 2014-09-08 04:55:15]

The CQ bit was checked by shreeram.k@samsung.com

[commit-bot: I haz the power, 2014-09-08 04:55:46]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/shreeram.k@samsung.com/496313003/220001

[shreeramk, 2014-09-08 04:55:50]

The CQ bit was unchecked by shreeram.k@samsung.com

[shreeramk, 2014-09-08 07:22:22]

The CQ bit was checked by shreeram.k@samsung.com

[commit-bot: I haz the power, 2014-09-08 07:24:35]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/shreeram.k@samsung.com/496313003/220001

[commit-bot: I haz the power, 2014-09-08 09:19:27]

Committed patchset #12 (id:220001) as 3b2c2a21c4d6b6c840c287e6231e084fffbb77f4

[commit-bot: I haz the power, 2014-09-10 03:45:17]

Patchset 12 (id:??) landed as
https://crrev.com/baa385c24c57df62f64a5f1fecee739dc2dcc3ba
Cr-Commit-Position: refs/heads/master@{#293692}