Options:
Previous day, Next day
- # Session Start: Thu Feb 12 00:00:00 2015
- # Session Ident: #testing
- # [00:25] * Joins: gitterbot (~nodebot@public.cloak)
- # [00:34] * Quits: Ms2ger (~Ms2ger@public.cloak) ("nn")
- # [00:38] <MikeSmith> I have the dependencies for https://github.com/w3c/web-platform-tests/issues/1603
- # [00:39] <MikeSmith> 2dcontext/tools/gentest.py
- # [00:40] <MikeSmith> I'll re-run it now and open a PR
- # [00:42] <MikeSmith> 「<Error>: The function ‘CGFontGetGlyphPath’ is obsolete and will be removed in an upcoming update. Unfortunately, this application, or a library it uses, is using this obsolete function, and is thereby contributing to an overall degradation of system performance.」
- # [01:24] * Joins: scott_gonzalez (~scott_gonzalez@public.cloak)
- # [01:29] * Joins: darobin (rberjon@public.cloak)
- # [01:36] * Quits: darobin (rberjon@public.cloak) (Ping timeout: 180 seconds)
- # [01:36] * Quits: ptressel (~chatzilla@public.cloak) ("bbl")
- # [01:39] * Quits: AutomatedTester (~AutomatedTester@public.cloak) (Client closed connection)
- # [02:34] * Joins: gitbot (~gitbot@public.cloak)
- # [02:34] -gitbot:#testing- [web-platform-tests] haoxli opened pull request #1618: Add test for user denial of captured file (master...mediacapture/user-denial-test) https://github.com/w3c/web-platform-tests/pull/1618
- # [02:34] * Parts: gitbot (~gitbot@public.cloak) (gitbot)
- # [02:46] * terri is now known as terri_offline
- # [02:56] * Quits: bhill2 (~bhill2@public.cloak) (Client closed connection)
- # [03:55] * Quits: gsnedders (~gsnedders@public.cloak) (Client closed connection)
- # [04:11] * Quits: artb (~ArtB@public.cloak) ("Leaving.")
- # [05:03] * Joins: gsnedders (~gsnedders@public.cloak)
- # [05:35] * Quits: gsnedders (~gsnedders@public.cloak) (Client closed connection)
- # [06:36] * heycam is now known as heycam|away
- # [06:43] * Joins: gitbot (~gitbot@public.cloak)
- # [06:43] -gitbot:#testing- [web-platform-tests] sideshowbarker created sideshowbarker/2dcontext (+1 new commit): https://github.com/w3c/web-platform-tests/commit/fc08667b8907
- # [06:43] -gitbot:#testing- web-platform-tests/sideshowbarker/2dcontext fc08667 Michael[tm] Smith: Sync 2dcontext tests to #1599. Closes #1603.
- # [06:43] * Parts: gitbot (~gitbot@public.cloak) (gitbot)
- # [06:44] * Joins: gitbot (~gitbot@public.cloak)
- # [06:44] -gitbot:#testing- [web-platform-tests] sideshowbarker opened pull request #1619: Sync 2dcontext tests to #1599. Closes #1603. (master...sideshowbarker/2dcontext) https://github.com/w3c/web-platform-tests/pull/1619
- # [06:44] * Parts: gitbot (~gitbot@public.cloak) (gitbot)
- # [06:49] * Joins: gsnedders (~gsnedders@public.cloak)
- # [07:14] * Joins: gitbot (~gitbot@public.cloak)
- # [07:14] -gitbot:#testing- [web-platform-tests] zqzhang closed pull request #1599: Update test size.attributes.parse.whitespace.html (master...webplatform/update-cavnas) https://github.com/w3c/web-platform-tests/pull/1599
- # [07:14] * Parts: gitbot (~gitbot@public.cloak) (gitbot)
- # [07:14] * Joins: gitbot (~gitbot@public.cloak)
- # [07:14] -gitbot:#testing- [web-platform-tests] zqzhang pushed 3 new commits to master: https://github.com/w3c/web-platform-tests/compare/a696a72752e2...dfab8f2c1034
- # [07:14] -gitbot:#testing- web-platform-tests/master 547d042 chenxi: Update test size.attributes.parse.whitespace.html...
- # [07:14] -gitbot:#testing- web-platform-tests/master 092f472 chenxi: Update tests.yaml for size.attributes.parse.whitespace
- # [07:14] -gitbot:#testing- web-platform-tests/master dfab8f2 Zhiqiang Zhang: Merge pull request #1599 from chenxix/webplatform/update-cavnas...
- # [07:14] * Parts: gitbot (~gitbot@public.cloak) (gitbot)
- # [07:14] * Joins: gitbot (~gitbot@public.cloak)
- # [07:14] -gitbot:#testing- [web-platform-tests] zqzhang closed pull request #1619: Sync 2dcontext tests to #1599. Closes #1603. (master...sideshowbarker/2dcontext) https://github.com/w3c/web-platform-tests/pull/1619
- # [07:14] * Parts: gitbot (~gitbot@public.cloak) (gitbot)
- # [07:14] * Joins: gitbot (~gitbot@public.cloak)
- # [07:14] -gitbot:#testing- [web-platform-tests] zqzhang deleted sideshowbarker/2dcontext at fc08667: https://github.com/w3c/web-platform-tests/commit/fc08667
- # [07:14] * Parts: gitbot (~gitbot@public.cloak) (gitbot)
- # [07:23] * Joins: gitbot (~gitbot@public.cloak)
- # [07:23] -gitbot:#testing- [web-platform-tests] kangxu opened pull request #1620: Update idlharness tests for Battery Status API to latest CR (master...battery-status/update3) https://github.com/w3c/web-platform-tests/pull/1620
- # [07:23] * Parts: gitbot (~gitbot@public.cloak) (gitbot)
- # [07:30] * Joins: gitbot (~gitbot@public.cloak)
- # [07:30] -gitbot:#testing- [web-platform-tests] zqzhang closed pull request #1617: Rewrite the tests for synthetic PopStateEvents. (master...popstateevent) https://github.com/w3c/web-platform-tests/pull/1617
- # [07:30] * Parts: gitbot (~gitbot@public.cloak) (gitbot)
- # [07:30] * Joins: gitbot (~gitbot@public.cloak)
- # [07:30] -gitbot:#testing- [web-platform-tests] zqzhang pushed 2 new commits to master: https://github.com/w3c/web-platform-tests/compare/b77f29ada78c...51d507531d26
- # [07:30] -gitbot:#testing- web-platform-tests/master 77abd41 Ms2ger: Rewrite the tests for synthetic PopStateEvents.
- # [07:30] -gitbot:#testing- web-platform-tests/master 51d5075 Zhiqiang Zhang: Merge pull request #1617 from Ms2ger/popstateevent...
- # [07:30] * Parts: gitbot (~gitbot@public.cloak) (gitbot)
- # [08:22] * Joins: zcorpan (~zcorpan@public.cloak)
- # [09:36] * Joins: anssik (~uid10742@public.cloak)
- # [10:11] * Joins: darobin (rberjon@public.cloak)
- # [11:46] <zcorpan> MikeSmith: hmm, looks like the _assertSame is still expecting \r. but i think now we should revert the change and edit escapeHTML in gentest.py to emit 
 for \r
- # [11:47] <MikeSmith> zcorpan: Oh
- # [11:50] <MikeSmith> I didn't bother to actually (re)test anything when I ran the script
- # [11:52] <MikeSmith> anyway agreed we should revert it
- # [11:54] <MikeSmith> I can deal with regenerating the files after the fix at least
- # [11:55] <MikeSmith> On my mobile now
- # [12:47] <zcorpan> MikeSmith: ok, thx. i'll see if i can provide a fix sans regen
- # [13:02] * Quits: anssik (~uid10742@public.cloak) ("Connection closed for inactivity")
- # [13:08] * Quits: terri_offline (~terri@public.cloak) (Ping timeout: 180 seconds)
- # [13:47] * Joins: Lachy (~Lachy@public.cloak)
- # [13:53] <jgraham> !summon Ms2ger
- # [14:07] * Joins: plh (plehegar@public.cloak)
- # [14:09] * Joins: terri_offline (~terri@public.cloak)
- # [14:09] * terri_offline is now known as terri
- # [14:25] <ato> dev.w3.org down?
- # [14:26] <Yves> slow, it seems
- # [14:29] <ato> Yes, the document loaded finally.
- # [14:31] <jgraham> darobin: So how does one refer to a term defined in another spec with ReSpec?
- # [14:31] <jgraham> I can see how to refer to the whole spec, but not a specific definition
- # [14:35] <darobin> jgraham: that's not supported at this point (though there have been discussions about making it work in all tools)
- # [14:37] <darobin> if there's a pressing need I can probably add support pretty quickly, it's just a question of pushing it up the priority list
- # [14:39] <jgraham> Not really a pressing need, I was just looking at WebDriver and wondering what it would take to reuse the request and response objects defined in fetch
- # [14:39] <jgraham> And, I guess, lots of algorithms defined in HTML
- # [14:44] <jgraham> It's going to be pretty annoying if every time you have to do something in HTML it looks like "queue a task [HTML] to create a script [HTML] for /source/ with environment settings object [HTML] (whatever it should be)"
- # [14:44] <jgraham> s/in HTML/in WebDriver/
- # [14:51] * Joins: playjax (~chatzilla@public.cloak)
- # [14:55] <darobin> jgraham: oh, yeah, I definitely wouldn't do it that way
- # [14:56] <darobin> there are several options here
- # [14:57] <darobin> based on what people do, one is simply to have "then you <a href="http://...#queue" class="externalRef">queue a task</a>
- # [14:57] <darobin> the other is to have a section at the top with "The following concepts are defined in HTML [HTML]"
- # [14:57] <darobin> then "<li><a href="to html"><dfn>queue a task..."
- # [14:58] <jgraham> Yeah, I guess explicit links work
- # [15:13] * Joins: AutomatedTester (~AutomatedTester@public.cloak)
- # [15:20] * Quits: playjax (~chatzilla@public.cloak) (Client closed connection)
- # [15:26] * Joins: artb (~ArtB@public.cloak)
- # [15:43] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [16:51] * Joins: bhill2 (~bhill2@public.cloak)
- # [16:54] * Joins: Ms2ger (~Ms2ger@public.cloak)
- # [16:54] * Joins: zcorpan (~zcorpan@public.cloak)
- # [17:01] * Quits: zcorpan (~zcorpan@public.cloak) (Ping timeout: 180 seconds)
- # [17:10] * Quits: AutomatedTester (~AutomatedTester@public.cloak) (Client closed connection)
- # [17:13] * Joins: Wilto (~Wilto@public.cloak)
- # [17:14] * Joins: yoav (~Adium@public.cloak)
- # [17:14] <yoav> darobin: Critic? Seriousely???
- # [17:15] <Ms2ger> yoav, ?
- # [17:15] <darobin> yoav: you don't like Critic?
- # [17:15] <Wilto> The old cartoon? It was okay, I guess.
- # [17:15] <darobin> yoav: you get used to it :)
- # [17:15] <yoav> I like busting darobin's chops better
- # [17:15] <darobin> lol
- # [17:15] * darobin had forgtten about the /topic, it's been there a looooooooong time
- # [17:16] <plh> lol
- # [17:16] <darobin> yoav and Wilto show up and immediately my playlist switches to "Sexbomb" — coincidence?
- # [17:16] <Wilto> absolutely not
- # [17:16] <darobin> right, I did think it was what you two sing whenever you join a channel I'm on
- # [17:16] <jgraham> Oooh, it's a Ms2ger
- # [17:17] * Ms2ger bows
- # [17:17] <Ms2ger> What do you need now?
- # [17:17] <darobin> jgraham, Ms2ger: do you know if anyone has experience converting Blink layout tests to testharness?
- # [17:17] <darobin> the Responsive Images people would like to do that
- # [17:17] <jgraham> Ms2ger, MikeSmith: Any chance that one of you will have the chance to look at my two reftest-related reviews any time soon?
- # [17:18] <jgraham> darobin: Would they like to convert some Service Worker tests whilst they're at it?
- # [17:18] * Joins: anselm (~anselm@public.cloak)
- # [17:18] <Ms2ger> darobin, I hear someone's converted some SW tests like that
- # [17:18] <jgraham> Ms2ger, MikeSmith: if not I can call in the cavalry
- # [17:18] <Ms2ger> jgraham, sound the horn
- # [17:18] <yoav> darobin: The problem is not converting (blink's LayoutTest system supports both)
- # [17:18] <darobin> jgraham: well, I was thinking: the RICG has a pretty sexy and active gang of people, if there were docs or even just general advice about conversions, they could take that on for RI stuff but some might get interested in other parts as well
- # [17:19] <Ms2ger> I'll be out for the weekend too
- # [17:20] <yoav> darobin: My main question is how/where does one create a w3c test suite that can later be passed around to other engines
- # [17:20] <Ms2ger> yoav, here :)
- # [17:20] <darobin> yoav: we might not be talking about the same thing re "converting"
- # [17:20] <Ms2ger> https://github.com/w3c/web-platform-tests
- # [17:21] <darobin> yoav: well, https://github.com/w3c/web-platform-tests/ is just that
- # [17:21] <yoav> darobin: OK
- # [17:21] <Ms2ger> http://testthewebforward.org/
- # [17:21] <darobin> for <picture>, I reckon it'd just go under /html/
- # [17:21] <darobin> and as Ms2ger points out, there's documentation on the TestTWF site
- # [17:21] * darobin thought we'd talked about this before but it might have been just with Marcos
- # [17:22] <yoav> One more hurdle I can think of, which is a major one, is the lack of cross-browser ways to set DPR
- # [17:22] <Ms2ger> DPR?
- # [17:22] <yoav> devicePixelRatio
- # [17:23] <darobin> Dead Pirate Roberts obviously
- # [17:23] <darobin> or, well, that too
- # [17:23] <Ms2ger> Ah
- # [17:23] <darobin> Dread*
- # [17:23] <yoav> in Blink/WebKit there's a test-only API that lets you do that
- # [17:23] <Ms2ger> Yeah, not even sure how we do that internally
- # [17:23] <yoav> maybe we can create a JS abstraction around that, if Gecko has something similar
- # [17:27] <jgraham> In gecko you need specialPowers to set a pref
- # [17:27] <Wilto> Little later today I’m gonna chat with Bocoup about maybe getting some hands on this.
- # [17:27] <jgraham> So basically if you have to do that for the tests you *can't* make them cross-browser
- # [17:28] <jgraham> (web-platform-tests aren't run in a way that allows you to get specialPowers in gecko)
- # [17:28] <jgraham> There were also some Mozilla people talking about maybe writing <picture> tests, but I'm not sure what happened to that idea
- # [17:29] <yoav> :/
- # [17:30] <Wilto> Hm.
- # [17:30] <yoav> That means that everything that requires a specific DPR to impact the result can't be cross browser
- # [17:30] <yoav> which is everything srcset
- # [17:31] <jgraham> Yeah, but it's sort of an intrinsic limitation of cross-browser testing. Until we standardise a test-only API for setting that kind of thing, but no one has really wanted to do that yet
- # [17:33] <yoav> what's the limitation on the moz side? testharness tests can't have specialPowers?
- # [17:33] <Ms2ger> Servo doesn't have special powers :)
- # [17:33] <Wilto> samesies
- # [17:36] <jgraham> The fundamental limitation here is that there isn't an API you can call in a cross-browser way that allows you to set what you want to set. Different vendors might have their own internal APIs, but there's no guarantee they are compatible. For single-vendor tests this doesn't matter; they just write tests that work with whatever API they happen to have. For cross-browser tests this does matter
- # [17:36] <darobin> jgraham: isn't that something that wptrunner could act on?
- # [17:36] <jgraham> darobin: How?
- # [17:36] <darobin> jgraham: well, it is running a browser instance, can it not manipulate privileged stuff?
- # [17:37] <yoav> jgraham: You could also add a JS abstraction on top that manages these different APIs
- # [17:37] <jgraham> The situation is that gecko has a thing, blink has a similar (but possibly different?) thing. Presumably webkit has something different to blink. IE might or might not have a thing. Servo has nothing.
- # [17:37] <darobin> then we need a convention to state how DPR-dependent tests run
- # [17:37] <jgraham> s/different to/similar to/
- # [17:38] <yoav> if(window.internals)// do Blink/WebKit stuff
- # [17:38] <Wilto> Could we create a method for use throughout the tests, that points to a function with something to the tune of “/* DO YOUR BROWSER-Y DPR STUFF HERE */”?
- # [17:38] <yoav> etc,etc
- # [17:38] <Wilto> He said JavaScriptially.
- # [17:39] <jgraham> In gecko at least you don't get the priviledged context just by being a test, I don't think, you have to do something extra (ms2ger might know what)
- # [17:39] <jgraham> But writing "cross browser" tests that are full of browser-specific code is not a winning strategy
- # [17:39] <Ms2ger> We install an addon in mochitests that exposes an API to the web page
- # [17:42] <jgraham> Right, so if you want to propose that browsers expose a common test-centric API in testing situations, that seems like a reasonable idea but it actually needs a specification
- # [17:43] <yoav> jgraham: I'm still not clear on why can't we start with a JS based API abstraction
- # [17:44] <yoav> if (super powers or whatever) setDprForGecko()
- # [17:44] <jgraham> Because putting browser-specific code in tests isn't allowed
- # [17:45] <yoav> we could have that abstraction outside of the tests themselves
- # [17:45] <Wilto> I mean, I can see where that would spiral out of control over time.
- # [17:45] <jgraham> and specialPowers isn't avaiable to web-platform-tests in Gecko
- # [17:46] <jgraham> If you want to abstract it into a library then that seems like it's basically standardising an API
- # [17:46] <jgraham> So I'm saying "let's have a conversation about whether standardising such an API is a good idea and if so what it should look like"
- # [17:47] <yoav> I'm all for that
- # [17:47] <jgraham> public-test-infra@lists.w3.org
- # [17:47] <yoav> But it'd be a shame if we'd have to stall all efforts of have a w3c test suit until such a spec gets hammered down
- # [17:48] <yoav> when we can prollyfill it
- # [17:48] <jgraham> Well you can do eerything that doesn't require setting prefs already
- # [17:48] <yoav> which is close to nothing
- # [17:49] <jgraham> Starting with that and having the bigger conversation in parallel seems like a good place to figure out how to proceed
- # [17:49] <yoav> almost all srcset tests rely on the DPR being either 1 or 2 or something well defined
- # [17:50] <jgraham> Well, if you create a polyfill right now it still won't work in gecko because the infrastructure doesn't support it, or in IE because we don't know how to polyfill this in IE.
- # [17:50] <yoav> although when I think of it, some of the logic they test is UA specific
- # [17:50] <jgraham> So I think having the conversation upfront is really quite necessary
- # [17:50] <yoav> ok
- # [17:53] * Quits: bhill2 (~bhill2@public.cloak) (Client closed connection)
- # [17:53] * Joins: bhill2 (~bhill2@public.cloak)
- # [18:00] * Quits: bhill2 (~bhill2@public.cloak) (Client closed connection)
- # [18:08] <Wilto> So, in terms of The Plan, then: where do we start? Fire off an email on the subject to public-test-infra@lists.w3.org ?
- # [18:09] <Ms2ger> Sounds good
- # [18:20] <jgraham> Yep
- # [18:20] <jgraham> You could, perhaps, CC browser-tools-testing, but set reply-to public-test-infra
- # [18:22] * Quits: Wilto (~Wilto@public.cloak) (Ping timeout: 180 seconds)
- # [18:24] * Quits: darobin (rberjon@public.cloak) (Client closed connection)
- # [18:29] * Joins: Wilto (~Wilto@public.cloak)
- # [18:31] * Quits: Lachy (~Lachy@public.cloak) ("My MacBook Pro has gone to sleep. ZZZzzz…")
- # [18:35] * Quits: anselm (~anselm@public.cloak) (Client closed connection)
- # [18:35] * Joins: Lachy (~Lachy@public.cloak)
- # [18:41] * Joins: bhill2 (~bhill2@public.cloak)
- # [18:42] * Joins: bhill2_ (~bhill2@public.cloak)
- # [18:45] * Quits: Lachy (~Lachy@public.cloak) ("My MacBook Pro has gone to sleep. ZZZzzz…")
- # [18:48] * Joins: fff (~fff@public.cloak)
- # [18:48] * Quits: bhill2 (~bhill2@public.cloak) (Ping timeout: 180 seconds)
- # [18:54] * Quits: fff (~fff@public.cloak) (Ping timeout: 180 seconds)
- # [19:05] * Joins: Lachy (~Lachy@public.cloak)
- # [19:36] * Joins: bhill2 (~bhill2@public.cloak)
- # [19:36] * Quits: bhill2_ (~bhill2@public.cloak) (Client closed connection)
- # [19:37] * Quits: bhill2 (~bhill2@public.cloak) (Client closed connection)
- # [19:43] * Quits: Lachy (~Lachy@public.cloak) ("My MacBook Pro has gone to sleep. ZZZzzz…")
- # [19:47] * Joins: bhill2 (~bhill2@public.cloak)
- # [19:49] * Quits: bhill2 (~bhill2@public.cloak) (Client closed connection)
- # [19:49] * Joins: bhill2 (~bhill2@public.cloak)
- # [20:07] * Quits: yoav (~Adium@public.cloak) ("Leaving.")
- # [20:15] * Joins: zcorpan (~zcorpan@public.cloak)
- # [20:15] * Quits: bhill2 (~bhill2@public.cloak) (Client closed connection)
- # [20:22] * Quits: Wilto (~Wilto@public.cloak) ("Lost terminal")
- # [20:24] * Joins: gitbot (~gitbot@public.cloak)
- # [20:24] -gitbot:#testing- [web-platform-tests] Ms2ger opened pull request #1621: Define a strigifier on the URLUtils interfaces. (master...url-html-idl) https://github.com/w3c/web-platform-tests/pull/1621
- # [20:24] * Parts: gitbot (~gitbot@public.cloak) (gitbot)
- # [20:25] * Joins: bhill2 (~bhill2@public.cloak)
- # [20:27] * Quits: bhill2 (~bhill2@public.cloak) (Client closed connection)
- # [20:27] * Joins: bhill2 (~bhill2@public.cloak)
- # [20:38] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [20:38] * Joins: AutomatedTester (~AutomatedTester@public.cloak)
- # [21:01] * Joins: guest (~guest@public.cloak)
- # [21:04] * Quits: guest (~guest@public.cloak) ("Page closed")
- # [21:29] * Joins: darobin (rberjon@public.cloak)
- # [21:36] * Joins: anselm (~anselm@public.cloak)
- # [21:43] * Quits: anselm (~anselm@public.cloak) (Ping timeout: 180 seconds)
- # [21:48] * Joins: zcorpan (~zcorpan@public.cloak)
- # [21:51] * terri is now known as terri_offline
- # [21:55] * Quits: zcorpan (~zcorpan@public.cloak) (Ping timeout: 180 seconds)
- # [22:03] * terri_offline is now known as terri
- # [22:12] * Quits: plh (plehegar@public.cloak) ("Leaving")
- # [22:37] * Quits: darobin (rberjon@public.cloak) (Client closed connection)
- # [22:38] * Joins: darobin (rberjon@public.cloak)
- # [22:39] * Joins: anselm (~anselm@public.cloak)
- # [22:45] * Quits: darobin (rberjon@public.cloak) (Ping timeout: 180 seconds)
- # [23:03] * heycam|away is now known as heycam
- # [23:05] * Joins: Wilto (~Wilto@public.cloak)
- # [23:12] * Quits: Wilto (~Wilto@public.cloak) (Ping timeout: 180 seconds)
- # [23:13] * Joins: Wilto (~Wilto@public.cloak)
- # [23:25] * Quits: anselm (~anselm@public.cloak) (Client closed connection)
- # [23:26] * Joins: anselm_ (~anselm@public.cloak)
- # [23:28] * Quits: Wilto (~Wilto@public.cloak) (Ping timeout: 180 seconds)
- # Session Close: Fri Feb 13 00:00:00 2015
Previous day, Next day
Think these logs are useful? Then please donate to show your gratitude (and keep them up, of course). Thanks! — Krijn