[js] test order of host callout & typecheck during eval #16301
Conversation
This adds tests for tc39/ecma262#1504 which cannot be tested purely in JavaScript because it involves HostEnsureCanCompileStrings. As documented in a table on tc39/ecma262#1495 this passes on recent Chrome and Safari and fails on Firefox so this test probably shouldn't land right away.
|
Mind moving this to the |
|
@jugglinmike I don't mind at all, but it's testing language in ECMA262, not language in CSP3. |
|
It's kind of defined in both places. Storing the test in Alternatively, you might consider testing this in Test262 via some new host-defined behavior along the same lines as |
|
@jugglinmike Will move it over. |
I reformatted it to follow the guidance at https://github.com/web-platform-tests/wpt/blob/38f02ed8bb05f69ab4b061d17220f8f46709a0a3/content-security-policy/README.html#L37-L60 and moved the `<meta http-equiv=Content-Security-Policy>` content into a separate file per https://github.com/web-platform-tests/wpt/blob/38f02ed8bb05f69ab4b061d17220f8f46709a0a3/content-security-policy/README.html#L22-L24
|
@jugglinmike Moved and reworked to follow conventions local to that directory. |
content-security-policy/generic/eval-typecheck-callout-order.html
Outdated
Show resolved
Hide resolved
Thanks!
We can add "do not merge yet" to the pull request, or we could merge it with the ".tentative" file name marker. The latter is less likely to be forgotten, so if this isn't contentious, then that might be best. @mikewest do you have a preference? |
|
The test seems accurate to me, assuming tc39/ecma262#1504 lands. I'd be fine with y'all landing these tests now as |
... until tc39/ecma262#1495 lands.
|
The test is now tentative. $ git mv content-security-policy/generic/eval-typecheck-callout-order{,.tentative}.html
$ git mv content-security-policy/generic/eval-typecheck-callout-order{,.tentative}.html.headers |
* [js] test order of host callout & typecheck during eval This adds tests for tc39/ecma262#1504 which cannot be tested purely in JavaScript because it involves HostEnsureCanCompileStrings. As documented in a table on tc39/ecma262#1495 this passes on recent Chrome and Safari and fails on Firefox so this test probably shouldn't land right away. * Moved test to content-security-policy/generic I reformatted it to follow the guidance at https://github.com/web-platform-tests/wpt/blob/38f02ed8bb05f69ab4b061d17220f8f46709a0a3/content-security-policy/README.html#L37-L60 and moved the `<meta http-equiv=Content-Security-Policy>` content into a separate file per https://github.com/web-platform-tests/wpt/blob/38f02ed8bb05f69ab4b061d17220f8f46709a0a3/content-security-policy/README.html#L22-L24 * move `<title>` into `<head>` * Mark new test tentative ... until tc39/ecma262#1495 lands.
This adds tests for tc39/ecma262#1504 which cannot be tested purely in
JavaScript because it involves HostEnsureCanCompileStrings.
As documented in a table on tc39/ecma262#1495 this passes on recent
Chrome and Safari and fails on Firefox so this test probably shouldn't
land right away.