"Headers attribute specified on a cell refers to cells in the same table element" [a25f45]: require headers to be th elements

[tombrunet]

Updates based on #1910

• Added to the expectation that the referenced element is expected to be a th
• Updated Pass 3 and Fail 4.

Closes issue(s):

• closes Rule a25f45 table headers - mismatch between rule applicability, examples and HTML spec #1910

Need for Call for Review:

This will require a 2 weeks Call for Review

---

Pull Request Etiquette

When creating PR:

• Make sure you're requesting to pull a branch (right side) to the develop branch (left side).
• Make sure you do not remove the "How to Review and Approve" section in your pull request description

After creating PR:

• Add yourself (and co-authors) as "Assignees" for PR.
• Add label to indicate if it's a Rule, Definition or Chore.
• Link the PR to any issue it solves. This will be done automatically by referencing the issue at the top of this comment in the indicated place.
• Optionally request feedback from anyone in particular by assigning them as "Reviewers".

When merging a PR:

• Close any issue that the PR resolves. This will happen automatically upon merging if the PR was correctly linked to the issue, e.g. by referencing the issue at the top of this comment.

How to Review And Approve

• Go to the “Files changed” tab
• Here you will have the option to leave comments on different lines.
• Once the review is completed, find the “Review changes” button in the top right, select “Approve” (if you are really confident in the rule) or "Request changes" and click “Submit review”.
• Make sure to also review the proposed Call for Review period. In case of disagreement, the longer period wins.

[tombrunet]

@WilcoFiers I put in a change to the expectation based on the issue description, but I'm a bit hesitant about that. I think ATs handle this regardless of whether or not it points to a td or a th. We certainly don't have any fail examples here to demonstrate that. I certainly agree that the examples should be in line with the spec. But, should we:

1. Keep this change to the expectation
2. Revert the expectation and add a failure case where a headers attribute points to a td element
3. Revert the change and do not add new test cases

[WilcoFiers]

I would feel much better about this if we had some test data showing that referencing tds, with and without role=row/columnheader doesn't work. Would it be possible to get someone to do some testing before we say this definitely fails WCAG?

[Jym77]

It is at least invalid HTML: headers attribute, and shouldn't create a relationship: Forming relationship, step 7 only keeps the "headers".

Due to non-interference with the host language, I'd be surprised if that worked. But I agree we should test it nonetheless.

[tombrunet]

I used this as a test with just regular HTML:
https://codesandbox.io/s/table-test-68d78y?file=/index.html

JAWS doesn't care if the referenced element is a td or a th, it reads the referenced headings in both cases. Surprisingly, NVDA and desktop Voiceover seem to ignore the headers attribute and use the automatic calculation - they won't read td's, but they read the th at the top of the column rather than the th that I referenced via headers.

So, 2 of the 3 desktop screen readers don't handle headers properly even when they're referencing a th. The 3rd doesn't care if it's a td or a th. Given that, I'm having a hard time seeing how this rule has any real world value - the results don't seem to be a factor of what the headers attribute is referencing.

[tombrunet]

To be clear, I was using Chrome on Mac, and Edge on Windows.

[WilcoFiers]

Probably don't need the role anymore if we do this. If we're going to fail headers referencing td elements we should have a new failed example for it. What we have now references an element inside the cell, which I do think is a problem.