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

fix(target-size): support translucent elements within pseudo-stacking-context ancestors with implicit z-order #4351

Draft
wants to merge 1 commit into
base: develop
Choose a base branch
from

Conversation

dbjorge
Copy link
Contributor

@dbjorge dbjorge commented Mar 1, 2024

Draft until researching/testing more on:

  • do elements with transform properties need similar treatment (and are there other cases where we should be treating "fake" stacking contexts as true?)
  • how to write more complete tests for this (particularly, distinguishing the cases that would be solved by the POSITION_LEVEL part of the change and the treat-as-true-stacking-context part)
  • double check browser support (validated current cases on stable chrome, firefox, safari)

Closes: #4350

Copy link
Contributor

@straker straker left a comment

Choose a reason for hiding this comment

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

I don't think this is the correct solution. The basic setup to reproduce the bug is that the parent needs to both be positioned without using z-index and use another CSS technique that creates a stacking context (transform works as well). Then a child with opacity will cause the stack to think the parent is on top (without this change).

Tested with other styles besides using opacity and things like transform or contain on the child also show the same problem. So it's not just opacity this should account for but any CSS style that creates a stacking context.

<div id="one" style="width: 50xp; height: 50px; position: relative; contain: layout;">
  <div id="two" style="width: 50xp; height: 50px; transform: scale(1)"></div>
</div>
<script>
axe.setup();
axe.commons.dom.getElementStack( document.getElementById('two') ) // [#one, #two] 
</script>

Further testing of different setups shows that the issue seems to be how the stacking order is determined for positioned auto elements. Take the following example:

<style>
  div {
    width: 100px;
    height: 100px;
  }
</style>
<div id="1" style="background: red;">
  <div id="2" style="contain: paint; background: rgba(0,0,255,0.5);"></div>
</div>
<div id="3" style="margin-top: -100px; background: green;">
  <div id="4" style="background: yellow;"></div>
</div>

In the browser this creates a yellow and 50% blue background. Axe's element stack and stacking order is correct. The #2 element creates a stacking context so replaces the parent stacking context (which is what everything else uses), making it on top element.

// axe.commons.dom.getElementStack('#4')
[
    "2",
    "4",
    "3",
    "1",
    "fixture"
]

// stacking order of every node in the element stack
[
    [
        {
            "stackLevel": 0.1,
            "treeOrder": 5,
            "vNode": "#2"
        }
    ],
    [
        {
            "stackLevel": 0,
            "treeOrder": 0
        }
    ],
   // ...
]

However if we change it to use positioned parents then everything is incorrect.

<style>
  div {
    width: 100px;
    height: 100px;
  }
</style>
<div id="1" style="position: absolute; top: 0; background: red;">
  <div id="2" style="contain: paint; background: rgba(0,0,255,0.5);"></div>
</div>
<div id="3" style="position: absolute; top: 0; background: green;">
  <div id="4" style="background: yellow;"></div>
</div>

In the browser this shows a solid yellow background. The parent elements both use fake stacking contexts so their stack orders are now 2 deep and positioned using POSITION_LEVEL, but the #2 element is a stacking context so overrides the parent stacking context, thus creating the bug.

// axe.commons.dom.getElementStack('#4')
[
    "2",
    "4",
    "3",
    "1",
    "fixture"
]

// stacking order
[
    [
        {
            "stackLevel": 0.1,
            "treeOrder": 5,
            "vNode": "#2"
        }
    ],
    [
        {
            "stackLevel": 0,
            "treeOrder": 0
        },
        {
            "stackLevel": 0.3,
            "treeOrder": 6,
            "vNode": "#3"
        }
    ],
    [
        {
            "stackLevel": 0,
            "treeOrder": 0
        },
        {
            "stackLevel": 0.3,
            "treeOrder": 6,
            "vNode": "#3"
        }
    ],
    [
        {
            "stackLevel": 0,
            "treeOrder": 0
        },
        {
            "stackLevel": 0.3,
            "treeOrder": 4,
            "vNode": "#1"
        }
    ],
    [
        {
            "stackLevel": 0,
            "treeOrder": 0
        }
    ],
    [
        {
            "stackLevel": 0,
            "treeOrder": 0
        }
    ],
    [
        {
            "stackLevel": 0,
            "treeOrder": 0
        }
    ]
]

This makes it seem that the #2 element should be draw in the same stack level as the other nodes so it's drawn in DOM order (making it the bottom of the stack)

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.

target-size: false positive when translucent target is inside a positioned "fake" stacking context
2 participants