-
Notifications
You must be signed in to change notification settings - Fork 28
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
Add a prototype automated test file (not generated) #350
base: master
Are you sure you want to change the base?
Conversation
tests/checkbox/automated/test-01-navigate-to-unchecked-checkbox-reading.nvda.json
Outdated
Show resolved
Hide resolved
}, | ||
{ | ||
"assert_role": [ | ||
"checkbox" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
If there is going to be a test file for each AT, do is maybe make more sense for the normalization of role from checkbox
to check box
to live here? (I'm not sure...)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think it's better to have the mapping in the AT's repo (if open source, at least), so that the test can be applicable across versions of the AT even if they changed how a role is conveyed. If it's in aria-at
then it makes things annoying both for the AT vendor and the aria-at CG when it changes (though may be rare, but still).
…'nav' base the repo root
000cf6d
to
144ea0f
Compare
|
As indicated on a previous call, I don't agree with this. It creates fragmentation because NVDA is the only truly open source AT that we're likely to support in the short to medium term future. Even TalkBack on Android can't be considered open source in the way NVDA is, because the source is only made available along with public releases. You wrote:
Why does it make things annoying for the AT vendor? If they are using some part of ARIA-AT as a dependency in their build toolchain, they should be expected to update it along with other similar dependencies. If they feel that they are blocking on someone on the ARIA-AT side making a role mapping or other update, they can submit a PR or file an issue. Same goes for the ARIA-AT side. With this proposal, the ARIA-AT project would be responsible for maintaining and updating the large majority of role mappings, but not all of them. It may be rare for existing mappings to change, but new roles may be invented which are implemented in multiple ATs at around the same time. At that point, somebody from the ARIA-AT CG (or whoever) could get stuck in updating different AT mappings, only to find that, say, NVDA's mappings were maintained elsewhere and hadn't been updated yet. At which point I'd fully expect them to at least consider submitting a PR against NVDA, which would mean they'd done the work anyway. At the most basic level, having NVDA's role mappings be maintained by NVAccess and having all others be maintained by the ARIA-AT CG is confusing. Not to mention the fact that, AFAIK, NVAccess haven't agreed to take on that responsibility, so it may come down to somebody else to take it on anyway. At which point we'd just be spreading things across repos for no gain. |
See #349
This isn't ready to be merged, but creating a PR to facilitate review and discussion.