Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Dialogs in web pages in Firefox and Chrome now get browseMode if the parent document also has browseMode #7433

Merged
merged 6 commits into from
Aug 29, 2017

Conversation

michaelDCurran
Copy link
Member

Link to issue number:

Fixes #4493

Summary of the issue:

The web accessibility community has requested that dialogs be navigable the same way any normal web document is (I.e. has NVDA browseMode by default), unless they are embedded in an application. Currently NVDA treats dialogs similar to applications in that browseMode is not available by default.

Description of how this pull request fixes the issue:

Dialogs in Firefox and Chrome will now get browseMode by default if the parent document has a treeInterceptor.
This involved creating a new WebDialog NVDAObject class in ia2web and using this for dialogs in place of Application.
Also, a change was made to FocusableUnfocusableContainer behavior to allow a sub or super class to choose to allow setFocus to fallback to setting focus to the object itself rather than the first focusable descendant, if the object itself definitely is focusable. Many dialogs on the web do ahve a tabindex of -1 and the UX is better if we can set focus to the dialog not the descendant when providing browseMode on first activation so it reads from the top.

Testing performed:

A div with role=dialog in a standard web document, and also role=dialog inside an div with role=application in a web document.
Tested in both Chrome and Firefox.

Known issues with pull request:

This does not change Internet Explorer as we are no longer supporting new features, and Edge currently has limitations in its text patterns which make this impossible (namely there is no specific text pattern for each dialog to bound browseMode).

Change log entry:

Section: Changes

…a sub or super class to turn off automatic focusing of the first focusable descendant, by setting a special attribute.
@derekriemer
Copy link
Collaborator

Will this break dialogs which (kindly) focus some item in them, which are most important? also, what if there's a document within the dialog?

class WebDialog(Document):
"""
A dialog that will use a treeInterceptor if its parent currently does.
"""
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It might be good to explain the practical impact in real world terms; i.e. that web dialogs should use browse mode by default unless they're inside an ARIA application.

# For dialogs that will get browseMode and are focusable themselves (have a tabindex):
# Allow setting focus to the dialog directly, rather than its first focusable descendant
def _get__alwaysFocusFirstfocusableDescendant(self):
return not self.shouldCreateTreeInterceptor
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

As discussed, unless I'm missing something, I don't think the changes related to FocusableUnfocusableContainer are necessary. That behavior should only get used if the container is not focusable itself.

@@ -46,6 +46,21 @@ def _get_shouldCreateTreeInterceptor(self):
class Application(Document):
shouldCreateTreeInterceptor = False

class WebDialog(Document):
"""
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Consider moving this into behaviors, as it might be useful for Edge and/or MSHTML.

…heir label when embedded in a document, we should not do this when the dialog or application is the root of the buffer as we lose all the content if this particular application or dialog gets browse mode itself.
…should not be reported as a marker if they themselves are the root of the content.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

NVDA Screen Reader Does Not Read Text Within Modal Dialogs
4 participants