00:38:05 -BitBot- (wpt) [issue] stephenmcgruer closed #22113: Origin policy tests not running on wpt.fyi - https://git.io/JvwvW 00:38:06 -BitBot- (wpt) [PR] stephenmcgruer merged #22284 by stephenmcgruer into master: Remove testharnessreport.js include in origin-policy tests subframe - https://git.io/JvPzN 02:09:17 -BitBot- (wpt) [PR] chromium-wpt-export-bot merged #22386 by chromium-wpt-export-bot into master: [WPT migration] Move align-baseline.html test from css3/flexbox to WPT - https://git.io/Jvyvi 04:00:33 -BitBot- (wpt) [PR] chromium-wpt-export-bot requested #22409 merge into master: [css-flexbox] Move anonymous-block.html test from css3/flexbox to WPT - https://git.io/Jvyad 05:44:40 -BitBot- (wpt) [PR] chromium-wpt-export-bot requested #22410 merge into master: Revert "Add tests for PointerCapture at Shadow DOM elements" - https://git.io/JvyoJ 06:04:16 -BitBot- (wpt) [PR] annevk requested #22411 merge into master: COOP: use BroadcastChannel as opener isn't always there - https://git.io/JvyoH 06:42:56 -BitBot- (wpt) [PR] annevk merged #22411 by annevk into master: COOP: use BroadcastChannel as opener isn't always there - https://git.io/JvyoH 08:23:38 -BitBot- (wpt) [PR] jgraham requested #22412 merge into master: Update wptserve version number for release - https://git.io/JvyXo 08:39:58 -BitBot- (wpt) [PR] Ms2ger requested #22413 merge into master: Revert "Unskip linting unit test by making it work with python3" - https://git.io/Jvy1J 09:11:05 -BitBot- (wpt) [issue] drott closed #22355: css/css-fonts/matching/ tests refresh to pass on FF, Safari? - https://git.io/JvM4b 09:11:05 -BitBot- (wpt) [PR] drott merged #22392 by drott into master: Add STAT table, fix name IDs, suppress font-synthesis, add fvar instance - https://git.io/JvyqK 09:28:03 -BitBot- (wpt) [issue] drott reopened #22355: css/css-fonts/matching/ tests refresh to pass on FF, Safari? - https://git.io/JvM4b 10:10:15 -BitBot- (wpt) [PR] chromium-wpt-export-bot requested #22414 merge into master: Fix handling of trailing whitespace on url('#foo ') - https://git.io/Jvyys 10:26:30 -BitBot- (wpt) [PR] drott requested #22415 merge into master: Add @jfkthame to suggested reviewers for css-fonts - https://git.io/Jvyyd 11:19:33 -BitBot- (wpt) [PR] chromium-wpt-export-bot merged #22404 by chromium-wpt-export-bot into master: [css-flexbox] Migrate auto-height-with-flex.html to WPT - https://git.io/Jvy4S 11:44:45 -BitBot- (wpt) [PR] chromium-wpt-export-bot requested #22416 merge into master: Allow specifying variable font range descriptors in either direction - https://git.io/Jvy9h 11:47:43 -BitBot- (wpt) [PR] stephenmcgruer merged #22413 by stephenmcgruer into master: Revert "Unskip linting unit test by making it work with python3" - https://git.io/Jvy1J 11:57:48 -BitBot- (wpt) [PR] clopez merged #22347 by clopez into master: css-grid: Use documents.font.ready() for tests with Ahem font. - https://git.io/JvMJw 13:13:25 -BitBot- (wpt) [PR] chromium-wpt-export-bot requested #22417 merge into master: [css-flexbox] Move auto-height-column-with-border-and-padding.html test from css3/flexbox to WPT - https://git.io/Jvy58 13:14:06 -BitBot- (wpt) [PR] annevk merged #22361 by annevk into master: Encoding: use Wasm to get a SharedArrayBuffer instance - https://git.io/JvMgj 13:17:27 -BitBot- (wpt) [PR] annevk merged #22363 by annevk into master: WebAudio/OrientationSensor/Portals: use Wasm to get a SharedArrayBuff… - https://git.io/JvMVS 13:18:22 -BitBot- (wpt) [PR] annevk merged #22359 by annevk into master: XHR/CSS/Compression: use Wasm to get at a SharedArrayBuffer instancd - https://git.io/JvMun 13:20:51 -BitBot- (wpt) [PR] chromium-wpt-export-bot merged #22414 by chromium-wpt-export-bot into master: Fix handling of trailing whitespace on url('#foo ') - https://git.io/Jvyys 13:27:10 jgraham: I'm happy to review https://github.com/web-platform-tests/wpt/pull/22412 if you can give me some context :) 13:28:56 smcgruer_[EST]: Apparently mozdownload is using wptserve from pypi for its tests and the last release has some bug that's subsequently been fixed 13:30:03 So whimboo asked me to make a new release 13:49:00 -BitBot- (wpt) [PR] chromium-wpt-export-bot requested #22418 merge into master: wpt: remove weird pattern from webrtc simplecall-no-ssrcs - https://git.io/JvydC 15:13:30 foolip: I feel like maybe "I'm happy to support this change" made me sound too happy about you leaving the team ;) 15:21:24 -BitBot- (wpt) [PR] stephenmcgruer merged #22412 by stephenmcgruer into master: Update wptserve version number for release - https://git.io/JvyXo 15:21:37 Do'h, just realized I forgot to lgtm and merge that PR after asking you James, sorry -_- 15:23:28 No worries 15:24:09 jgraham: haha not at all :) 16:10:07 -BitBot- (wpt) [PR] Hexcles merged #22237 by Hexcles into master: MixedContent: Add static-import from dedicated/shared workers WPTs - https://git.io/Jv6F6 16:12:57 -BitBot- (wpt) [PR] chromium-wpt-export-bot requested #22419 merge into master: ReferrerPolicy: Add WPTs for static-import in dedicated/shared workers - https://git.io/Jvyjv 16:13:50 -BitBot- (wpt) [PR] chromium-wpt-export-bot requested #22420 merge into master: UpgradeInsecureRequest: Add WPTs for static-import in dedicated/shared workers - https://git.io/JvyjU 16:25:41 Are and <meta name="assert"...> required in CSS tests? https://web-platform-tests.org/writing-tests/css-metadata.html isn't very clear to me. 16:29:05 <smcgruer_[EST]> 'The test assertion is optional, but is is highly recommended to include one' 16:29:14 <smcgruer_[EST]> So the latter is definitely optional (but highly recommended) 16:30:00 <smcgruer_[EST]> https://web-platform-tests.org/writing-tests/test-templates.html?highlight=title is interesting, I've never seen that page before 16:32:31 <Hexcles> So css-metadata.html lists 3 "additional requirements". Spec links are definitely required (enforced by lint), assertions are explicitly "optional", and what about the "flags"? 16:33:26 <Hexcles> Based on my experience, it's not required, but the wording is basically the same as spec links and there is an example for no flags, which seems to imply it's required? 16:35:35 <jgraham> gsnedders will remember; I think the CSS build system uses flags for something but in practice I expect they're as accurate as all metadata that doesn't have an obvious effect i.e. very inaccurate 16:36:51 <jgraham> We've generally pushed back hard on having requirements for metadata on the basis that a) it's often wrong and b) it causes people to just write browser-specific tests instead (I've seen examples of people refusing to submit testsuites to the W3C due to metadata requirements) 16:37:20 <jgraham> (this is also why I'm not thrilled with the metadata requirement for QUIC, although I don't have a better suggestion there) 16:38:48 <Hexcles> What about <title>? That doesn't seem CSS-specific. Do we require/recommend <title> in general/for reftests in WPT? 16:40:11 <smcgruer_[EST]> I would suggest we recommend it at least 16:45:32 <jgraham> We at least use <title> for testharness tests (as the default test name), so it makes more sense to require that (plus it's really common practice to have a title in pages, so not anything like as much overhead as specifying whether the test works in a bunch of nonstandard configurations or linking to the spec or whatever) 16:58:39 <jgraham> Did we come to a definite conclusion about assert_precondition? 17:00:32 <jgraham> I guess the last comment is "does this require an RFC" 17:01:05 <jgraham> OK, I think I should write an RFC for this 17:02:41 <smcgruer_[EST]> I was waiting for "should I write an RFC" XD 17:02:45 <smcgruer_[EST]> I'm happy to write it 17:18:37 <jgraham> Well if you want to do it I'm even happier with that solution ;) 17:21:58 <smcgruer_[EST]> +1, expect it later this week :) 17:50:59 -BitBot- (wpt) [PR] chromium-wpt-export-bot requested #22421 merge into master: [css-flexbox] Move box-sizing-min-max-sizes.html test to WPT web tests - https://git.io/JvSfC 17:51:40 -BitBot- (wpt) [PR] chromium-wpt-export-bot requested #22422 merge into master: Apply adjustments for 'text-anchor' after 'textLength' for <textPath> - https://git.io/JvSfW 18:17:09 <muan> Hello! I'm looking into writing a test for this bug https://html-is.glitch.me/firefox-space-focus.html. I've given it a go here: https://github.com/muan/wpt/blob/keydown/uievents/order-of-events/focus-events/focus-management-expectations.html but it is just passing everywhere (I expect it to fail in Firefox with an extra click event recorded). I'm assuming some event simulations might be missing. Has anyone dealt with these kind of issues before and have poi[CUT] 18:55:24 <annevk> can someone rubberstamp https://github.com/web-platform-tests/wpt/pull/22385? 18:57:51 <jgraham> My rubberstamp is in full working order 18:58:21 <annevk> \o/ 18:58:22 -BitBot- (wpt) [PR] annevk merged #22385 by annevk into master: HTML: use Wasm to get a SharedArrayBuffer instance - https://git.io/JvDAz 19:01:23 -BitBot- (wpt) [issue] annevk closed #22358: SharedArrayBuffer constructor is not always exposed - https://git.io/JvMEv 19:04:10 <smcgruer_[EST]> Was asked by the animations team for help debugging their failure in https://wpt.fyi/results/css/css-animations/CSSAnimation-ready.tentative.html?label=experimental&label=master&aligned, but it seems really hard to debug unhandled promises :/ 19:04:49 <smcgruer_[EST]> Essentially this boils down to 'some promise, somewhere, rejected with an AbortError and you weren't handling that promise for some reason' 19:07:26 <gsnedders> Hexcles, jgraham: the flags are what are used to make some tests -manual or -visual without those filename suffixes 19:08:23 <gsnedders> the CSS2 ones I have some confidence about given Opera's import mechanism relied on them and there haven't been that many new tests added since then to CSS2 :D 19:08:45 <Hexcles> If a test doesn't need any flag, then it can omit the whole meta tag, right? 19:08:53 <gsnedders> yes 19:09:01 <Hexcles> got it 19:09:08 <Hexcles> I'll probably clarify the doc a bit 19:10:29 <Hexcles> So I guess there's no practical need for <meta name="flags" content=""> (example 3 in https://web-platform-tests.org/writing-tests/css-metadata.html#requirement-flags) 19:11:29 <annevk> Hexcles: jgraham: you know who muan can ask? 19:12:04 * Hexcles looking at the code 19:12:20 * annevk has to go unfortunately; been a long day 19:12:30 <Hexcles> Bye annevk 19:12:39 <jgraham> annevk: I had a look at the demo, will try to figure out what's going on 19:12:40 <muan> Thanks annevk 19:12:41 <jgraham> muan: ^ 19:12:57 <muan> Thanks, jgraham! 19:21:17 -BitBot- (wpt) [PR] chromium-wpt-export-bot merged #22400 by chromium-wpt-export-bot into master: [css-flexbox] Move box-sizing.html test from css3/flexbox to WPT - https://git.io/Jvylj 19:27:09 <jgraham> muan: So if I remove the testdriver bits from your test and just run it manually it still passes in Firefox, so I think the problem is that the test isn't doing the same thing as the demo rather than that the testdriver stuff isn't working as expected 19:30:36 <muan> jgraham: by "remove the testdriver bits + run it manually" do you mean pressing the keys yourself or what do you replace the Actions() with? 19:31:22 -BitBot- (wpt) [PR] chromium-wpt-export-bot requested #22423 merge into master: Isolate activation-gated APIs from user-activation WPTs. - https://git.io/JvSTB 19:33:15 <muan> @jgraham I'm using testdriver actions as I am assuming that's the best way to simulate an actual space keypress from a human, which should come with keydown/keyup/keypress... This is in comparison to dispatching the keyboard events in JS. Is there a better way to do that? 19:34:10 <Hexcles> muan you're correct that's the recommended way; I think jgraham meant pressing the key manually 19:35:12 <jgraham> muan: I mean pressing the keys manually for the sake of debugging 19:35:23 <jgraham> Testdriver is the correct way to automate 19:35:47 <muan> Ah ok, let me double check! 19:37:21 <jgraham> muan: So I think if I add a setTimeout to spin the event loop before checking the events I see a failure 19:45:12 <jgraham> Hmm, although that is not now failing 19:49:58 <muan> jgraham: hmm the script inside of then() in a setTimeout? It still passes for me. What do you mean by "that is not now failing"? 19:50:14 -BitBot- (wpt) [PR] chromium-wpt-export-bot merged #22421 by chromium-wpt-export-bot into master: [css-flexbox] Move box-sizing-min-max-sizes.html test to WPT web tests - https://git.io/JvSfC 19:51:52 <jgraham> muan: I got it to fail when running manually but can't using testdriver 19:54:38 <muan> jgraham Ah I see. yeah, manually pressing the keys I also see that one extra event was recorded in firefox which causes it to fail. 20:00:30 <jgraham> So I notice the keyCode is different when the actual key is pressed 20:02:58 <jgraham> idk if that is related to the bug or not 20:04:41 <jgraham> muan: I'm away now but https://gist.github.com/jgraham/24f4dd737060e054d204905ff50e1cea is a slimmed down version of the test that I've been using 20:04:58 <jgraham> It still doesn't work, but I think you need to talk to smaug 20:05:36 <muan> jgraham Got it. Thanks for taking a look! 20:10:30 <Hexcles> muan jgraham I think this is https://bugzilla.mozilla.org/show_bug.cgi?id=1417955 20:11:03 <Hexcles> https://w3c.github.io/webdriver/#keyboard-actions 20:12:44 <muan> Hexcles: Oh sweet! That is it. That falied the test! 20:13:01 <muan> (And still pass in Chrome/Safari) 20:16:04 <muan> jgraham Hexcles Thank you both! I'll see what I can clean up and send a pull request. 20:16:19 <Hexcles> I'm no input/webdriver expert; this is rather confusing to me... What is \ue00d? It's not a valid Unicode codepoint? 20:18:28 -BitBot- (wpt) [PR] chromium-wpt-export-bot merged #22335 by chromium-wpt-export-bot into master: Automate activation propagation tests for same/cross-origin frames. - https://git.io/Jv17b 20:53:08 -BitBot- (wpt) [PR] muan requested #22424 merge into master: Add a test for unexpected activation triggered during space keydown driven focus management - https://git.io/JvSI0 21:15:18 <jgraham> WebDriver uses the PUA to represent unprintable keys like shift. I don't really know why space came along for the ride 21:15:26 <jgraham> It's Selenium legacy really 21:18:09 -BitBot- (wpt) [PR] chromium-wpt-export-bot merged #22410 by chromium-wpt-export-bot into master: Revert "Add tests for PointerCapture at Shadow DOM elements" - https://git.io/JvyoJ 21:27:38 -BitBot- (wpt) [PR] chromium-wpt-export-bot requested #22425 merge into master: [css-flexbox] Migrate column-flex-child-with-overflow-scroll.html to WPT - https://git.io/JvSLm 21:44:28 -BitBot- (wpt) [PR] jfkthame merged #22415 by jfkthame into master: Add @jfkthame to suggested reviewers for css-fonts - https://git.io/Jvyyd 22:12:20 -BitBot- (wpt) [PR] AutomatedTester requested #22426 merge into master: WebDriver - Add test for Drag and Drop API - https://git.io/JvSth 22:23:18 -BitBot- (wpt) [PR] Hexcles requested #22427 merge into master: Clarify CSS metadata requirements - https://git.io/JvSqG 22:40:16 -BitBot- (wpt) [PR] chromium-wpt-export-bot requested #22428 merge into master: Land the 2 passing tests for pointer capture in shadow dom - https://git.io/JvSqS 23:13:16 -BitBot- (wpt) [PR] chromium-wpt-export-bot merged #22425 by chromium-wpt-export-bot into master: [css-flexbox] Migrate column-flex-child-with-overflow-scroll.html to WPT - https://git.io/JvSLm