- Issue created by @nod_
- First commit to issue fork.
I do not get why we have this test case:
I believe this might be a non-issue, as I included the test case directly from the HTML Standard page mainly to test that CKEditor or our HtmlBuilder doesn't mangle up the existing markup if that's what the original source code is.
I believe browser quirks like these should be left for the content editor to handle themselves rather than us converting it for them, especially if it might involve changing string values like:
const example = 'Consider this string: <!-- <script>';
to:
const example = 'Consider this string: \x3C!-- \x3Cscript>';
It can lead to unintentional bugs.
And with regards to:
And that is transformed into:
<script> const example = 'Consider this string: &lt;!-- &lt;script&gt;'; console.log(example); &lt;/script&gt; &lt;/body&gt;</script>
Is it actually being double encoded to HTML in your tests, or is the D.O formatter in the issue encoding it for you? As my local tests doesn't involve any encoded HTML entities as that's what 🐛 JavaScript operators in Needs work was meant to mitigate in the first place.
Nonetheless, I've added an updated test-case to showcase the behaviour that'll also work for test-only pipelines even though we are no longer using a
@dataProvider
.- Merge request !7291Draft: Issue #3437394: [DrupalHtmlEngine] Follow-up for script and style tags in CKEditor 5 → (Open) created by codebymikey