| Summary: | [JSC] Workaround test262.report bug by making $ properties enumerable | ||||||
|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Yusuke Suzuki <ysuzuki> | ||||
| Component: | New Bugs | Assignee: | Yusuke Suzuki <ysuzuki> | ||||
| Status: | RESOLVED FIXED | ||||||
| Severity: | Normal | CC: | ashvayka, ews-watchlist, keith_miller, mark.lam, msaboff, saam, tzagallo, webkit-bug-importer | ||||
| Priority: | P2 | Keywords: | InRadar | ||||
| Version: | WebKit Nightly Build | ||||||
| Hardware: | Unspecified | ||||||
| OS: | Unspecified | ||||||
| Attachments: |
|
||||||
|
Description
Yusuke Suzuki
2021-07-10 15:05:15 PDT
Created attachment 433277 [details]
Patch
Comment on attachment 433277 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=433277&action=review > Source/JavaScriptCore/ChangeLog:8 > + test262.report's harness has a bug and it reports incorrect results for JSC on tests using $262 properties. Is it a bug though? https://github.com/tc39/test262/blob/main/INTERPRETING.md#host-defined-functions requires $262 and print to be non-enumerable (which they are), yet it doesn't clarify attributes of $262 properties. "ordinary object with the following properties" kinda implies PropertyAttribute::None to me. Should we author a PR to clarify it? eshost has a workaround for non-enumerability that this patch fixes: https://github.com/bterlson/eshost/blob/76630bd6bf2d8401f92579f0d706273a495ce19e/runtimes/jsc.js#L6-L13. It would be nice to remove it as well. Committed r279810 (239573@main): <https://commits.webkit.org/239573@main> All reviewed patches have been landed. Closing bug and clearing flags on attachment 433277 [details]. |