263305 2023-10-17 23:26:31 -0700 [meta] Implement close requests and close watchers 2026-01-12 09:03:31 -0800 1 1 1 Unclassified WebKit DOM WebKit Nightly Build Unspecified Unspecified REOPENED BrowserCompat, InRadar, WPTImpact P2 Normal --- 274717 274719 274832 287873 266379 266554 266989 1 d webkit-unassigned afrehner.work eoconnor karlcow lwarlow mike ryan webkit-bug-importer oldest_to_newest 1985799 0 d 2023-10-17 23:26:31 -0700 This includes both the new CloseWatcher API, and hooking it up to <dialog> and popover="" so that they have the same behavior. (Including in edge cases like many of them being opened simultaneously.) See: * https://html.spec.whatwg.org/#close-requests-and-close-watchers * https://github.com/WICG/close-watcher * https://github.com/whatwg/html/pull/9462 * https://github.com/web-platform-tests/wpt/tree/master/close-watcher * https://source.chromium.org/chromium/chromium/src/+/main:third_party/blink/renderer/core/html/closewatcher/ * https://source.chromium.org/chromium/chromium/src/+/main:content/browser/closewatcher/ 1985958 1 webkit-bug-importer 2023-10-18 14:00:43 -0700 <rdar://problem/117163294> 2000284 2 lwarlow 2023-12-17 11:07:57 -0800 Pull request: https://github.com/WebKit/WebKit/pull/21939 2030131 3 d 2024-04-21 23:29:07 -0700 Small spec change coming here to improve web compat: https://github.com/whatwg/html/pull/10291 2076617 4 karlcow 2024-11-20 23:18:25 -0800 Safari is failing that Firefox and Chrome are passing. https://wpt.fyi/results/close-watcher/user-activation/y-dialog-disconnected.html 2076618 5 d 2024-11-20 23:21:34 -0800 I mean, Safari is failing all the close watcher tests, right? Not just that one. https://wpt.fyi/results/close-watcher?label=experimental&label=master&aligned 2090579 6 ews-feeder 2025-01-29 15:53:01 -0800 Committed 289529@main (74e0b308c997): <https://commits.webkit.org/289529@main> Reviewed commits have been landed. Closing PR #21939 and removing active labels. 2090590 7 lwarlow 2025-01-29 16:22:35 -0800 Reopening so this can be used to track the full implementation that PR should have used a sub issue rather than the meta one.