Options:
Previous day, Next day
- # Session Start: Fri Nov 06 00:00:00 2015
- # Session Ident: #testing
- # [00:05] * Joins: evansjeny (~evansjeny@public.cloak)
- # [00:31] * terri is now known as terri_offline
- # [00:38] * terri_offline is now known as terri
- # [00:40] * heycam is now known as heycam|away
- # [00:51] * terri is now known as terri_offline
- # [01:01] * heycam|away is now known as heycam
- # [01:27] * Quits: evansjeny (~evansjeny@public.cloak) ("Page closed")
- # [01:32] * Joins: jyasskin (~textual@public.cloak)
- # [02:09] * Quits: plinss (~plinss@public.cloak) (Ping timeout: 180 seconds)
- # [02:18] * Joins: plinss (~plinss@public.cloak)
- # [02:36] * heycam is now known as heycam|away
- # [02:49] * terri_offline is now known as terri
- # [02:52] * Quits: plh (plehegar@public.cloak) ("Leaving")
- # [02:59] * terri is now known as terri_offline
- # [03:07] * Quits: jyasskin (~textual@public.cloak) ("My computer has gone to sleep. ZZZzzz…")
- # [03:39] * heycam|away is now known as heycam
- # [03:52] * Joins: gitbot (~gitbot@public.cloak)
- # [03:52] -gitbot:#testing- [csswg-test] syncbot created robust-get_new_commits at d5069a3 (+0 new commits): https://github.com/w3c/csswg-test/commits/robust-get_new_commits
- # [03:52] * Parts: gitbot (~gitbot@public.cloak)
- # [03:52] * Joins: gitbot (~gitbot@public.cloak)
- # [03:52] -gitbot:#testing- [csswg-test] syncbot created take-2 (+1 new commit): https://github.com/w3c/csswg-test/commit/6d53125d80f8
- # [03:52] -gitbot:#testing- csswg-test/take-2 6d53125 Chris Rebert: Try checking for the existence of .hg subdirectory...
- # [03:52] * Parts: gitbot (~gitbot@public.cloak)
- # [03:53] * Quits: ArtB (~ArtB@public.cloak) ("Leaving.")
- # [03:59] * Joins: ArtB (~ArtB@public.cloak)
- # [04:04] * Quits: ArtB (~ArtB@public.cloak) ("Leaving.")
- # [04:08] * Joins: gitbot (~gitbot@public.cloak)
- # [04:08] -gitbot:#testing- [testharness.js] DLehenbauer opened pull request #160: Issue #157: DOM hierarchy error with XHTML and SVG (master...fix-xhtml-svg-dom-hierarchy-error) https://github.com/w3c/testharness.js/pull/160
- # [04:08] * Parts: gitbot (~gitbot@public.cloak)
- # [04:42] * Joins: gitbot (~gitbot@public.cloak)
- # [04:42] -gitbot:#testing- [csswg-test] cvrebert deleted patch-1 at 0538f60: https://github.com/w3c/csswg-test/commit/0538f60
- # [04:42] * Parts: gitbot (~gitbot@public.cloak)
- # [04:42] * Joins: gitbot (~gitbot@public.cloak)
- # [04:42] -gitbot:#testing- [csswg-test] cvrebert deleted fix-submodules at 6b37cc6: https://github.com/w3c/csswg-test/commit/6b37cc6
- # [04:42] * Parts: gitbot (~gitbot@public.cloak)
- # [04:42] * Joins: gitbot (~gitbot@public.cloak)
- # [04:42] -gitbot:#testing- [csswg-test] cvrebert deleted seed-directory at dc0ab3f: https://github.com/w3c/csswg-test/commit/dc0ab3f
- # [04:42] * Parts: gitbot (~gitbot@public.cloak)
- # [05:03] * Joins: mixbabii (~mixbabii@public.cloak)
- # [05:09] * Quits: kochi (~kochi@public.cloak) (Ping timeout: 180 seconds)
- # [05:10] * Quits: kochi1 (~kochi@public.cloak) (Ping timeout: 180 seconds)
- # [05:18] * Joins: kochi (~kochi@public.cloak)
- # [05:20] * Joins: kochi1 (~kochi@public.cloak)
- # [05:31] * Quits: mixbabii (~mixbabii@public.cloak) (Ping timeout: 180 seconds)
- # [07:37] * heycam is now known as heycam|away
- # [08:29] * Joins: jyasskin (~textual@public.cloak)
- # [08:57] * Joins: zcorpan (~zcorpan@public.cloak)
- # [09:01] * Joins: rego (~smuxi@public.cloak)
- # [09:04] * Quits: jyasskin (~textual@public.cloak) ("My computer has gone to sleep. ZZZzzz…")
- # [09:07] * Quits: rego (~smuxi@public.cloak) (Client closed connection)
- # [09:14] * Joins: dom (dom@public.cloak)
- # [09:35] * Joins: Ms2ger (~Ms2ger@public.cloak)
- # [09:59] <Ms2ger> Whoa, a Gecko dev writing a csswg test: https://hg.mozilla.org/integration/mozilla-inbound/rev/97e3492d6080a0a3d2eda07a4e36208b92758469
- # [10:01] <jgraham> Miracles, they happen?
- # [11:02] <gsnedders> I thought that didn't happen?
- # [11:08] <AutomatedTester> would css-om tests be in wpt or csswg tests?
- # [11:08] <AutomatedTester> I am guessing the latter
- # [11:15] <gsnedders> the latter
- # [11:15] <gsnedders> if you're testing the a CSS WG spec, it should be in csswg-test
- # [11:20] <Ms2ger> AutomatedTester, though depending on how annoyed I am by the csswg, I could accept it into wpt too :)
- # [11:22] <AutomatedTester> Ms2ger: it's against zcorpan's spec and he is a nice chap so...
- # [11:26] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [11:28] * Joins: zcorpan (~zcorpan@public.cloak)
- # [11:34] <jgraham> You can ask zcorpan then :)
- # [11:35] <gsnedders> AutomatedTester: also I'm increasingly inclined to just take things into csswg-test and ask for forgiveness later
- # [11:35] <AutomatedTester> I can't find elementFromPoint specific tests
- # [11:36] <gsnedders> CSSOM ain't that well tested, so…
- # [11:36] <gsnedders> cssom-view-1/elementFromPosition.html
- # [11:45] <jgraham> https://github.com/w3c/web-platform-tests/issues/2304 fwiw
- # [11:48] <zcorpan> AutomatedTester: would be awesome with more cssom and cssom-view tests. i'll probably be writing tests myself as well over the next 12 months
- # [11:49] <zcorpan> AutomatedTester: i haven't really looked at the existing tests
- # [11:49] <AutomatedTester> zcorpan: where would you like them?
- # [11:49] <AutomatedTester> zcorpan: wpt or csswg tests?
- # [11:49] <AutomatedTester> zcorpan: I am looking to implement elementsFromPoint and want to know where to put tests
- # [11:50] <zcorpan> AutomatedTester: doesn't matter for me personally, but i suppose csswg makes more sense. i think tests in csswg also magically get linked in the spec itself with bikeshed (we should have that working for wpt too though)
- # [11:50] <AutomatedTester> ok
- # [11:50] <zcorpan> AutomatedTester: note that testharness.js can be used in csswg also AIUI
- # [11:50] <AutomatedTester> ok
- # [11:51] <zcorpan> feel free to ping me for review
- # [11:52] <jgraham> The main advantage of wpt is that it will be run in gecko
- # [11:52] <jgraham> But csswg test won't
- # [11:53] * Joins: ArtB (~ArtB@public.cloak)
- # [11:53] <jgraham> In order to fix that I probably need to make the wptrunner reftest harness faster in gecko; apparently the other reftest harness has some special magic
- # [11:53] <AutomatedTester> jgraham: if test_run: test_result = "pass" ?
- # [11:54] <AutomatedTester> that would make it super fast
- # [11:54] <AutomatedTester> :P
- # [12:19] * Joins: gitbot (~gitbot@public.cloak)
- # [12:19] -gitbot:#testing- [web-platform-tests] martiansideofthemoon opened pull request #2308: adding condition to run LastModified test (master...my-code-fix) https://github.com/w3c/web-platform-tests/pull/2308
- # [12:19] * Parts: gitbot (~gitbot@public.cloak)
- # [12:22] <gsnedders> jgraham: it's not entirely clear how to do so, given the differences. certainly some need complexity that the normal reftest runner can't provide.
- # [12:23] <gsnedders> (ofc that's treating the current reftest runner as an immutable, but I don't think there's any desire to add the complexity needed for cross-origin testing)
- # [12:26] <jgraham> gsnedders: I mainly mean in the way it gets the screenshots
- # [12:29] <gsnedders> does wptrunner not just use WebDriver?
- # [12:30] <jgraham> Yes, which paints to an internal canvas and converts to base64 and sends it back over the wire
- # [12:30] <jgraham> Apparently the reftest harness doesn't do that anymore, but does something faster
- # [12:30] <gsnedders> whereas you need to move the comparison of the two canvases into the browser
- # [12:30] <gsnedders> to match it for speed
- # [12:32] <AutomatedTester> webdriver only puts that on an internal canvas for gecko
- # [12:33] <AutomatedTester> webdriver does other horrifying things for other browsers
- # [12:33] <gsnedders> yeah, but the cost is sending it over the wire, primarily
- # [12:33] <jgraham> Well it looks like the reftest harness is also smart about reusing canvases rather than always building new ones
- # [12:34] <jgraham> I'm sure this is a possible optimisation but marionette will need to be taught about it
- # [12:40] * Joins: gitbot (~gitbot@public.cloak)
- # [12:40] -gitbot:#testing- [wptrunner] gsnedders opened pull request #155: Use glob on include/exclude paths (master...glob_paths) https://github.com/w3c/wptrunner/pull/155
- # [12:40] * Parts: gitbot (~gitbot@public.cloak)
- # [12:45] * Joins: gitbot (~gitbot@public.cloak)
- # [12:45] -gitbot:#testing- [wptrunner] jgraham closed pull request #155: Use glob on include/exclude paths (master...glob_paths) https://github.com/w3c/wptrunner/pull/155
- # [12:45] * Parts: gitbot (~gitbot@public.cloak)
- # [12:45] * Joins: gitbot (~gitbot@public.cloak)
- # [12:45] -gitbot:#testing- [wptrunner] jgraham pushed 2 new commits to master: https://github.com/w3c/wptrunner/compare/d1d81bd43763...de53aee43147
- # [12:45] -gitbot:#testing- wptrunner/master 1b2f9e6 Geoffrey Sneddon: Use glob on include/exclude paths.
- # [12:45] -gitbot:#testing- wptrunner/master de53aee jgraham: Merge pull request #155 from gsnedders/glob_paths...
- # [12:45] * Parts: gitbot (~gitbot@public.cloak)
- # [12:47] <gsnedders> wait is my patch not broken?
- # [12:47] <gsnedders> why does url only get set once in the loop?
- # [12:48] <gsnedders> I'm getting more and more confused by this code.
- # [12:49] <jgraham> Oh yes, your patch is broken
- # [12:49] <jgraham> I sohuld have paid attention to the part of my brain that said "this looks too easy"
- # [12:50] <gsnedders> I thought it looked too easy. But I couldn't work out why.
- # [12:52] <jgraham> So you can make it work
- # [12:53] <jgraham> Also I rather assumed you had tested it before making a PR
- # [12:53] <jgraham> Anyway try hoisting the glob out to where add_include is called
- # [12:53] <jgraham> If there any any matches, call add_include once for each match, otherwise call add_include with the raw argument
- # [12:54] <gsnedders> I had tested it and it appeared to work.
- # [12:54] <gsnedders> Which is the really weird part.
- # [12:54] <gsnedders> I can't justify how it worked.
- # [12:54] <jgraham> It's rather messy, but add_include really does only add one include
- # [12:55] <gsnedders> I don't get quite how this fits in with the URL parsing, tbh.
- # [12:55] <jgraham> What URL parsing?
- # [12:56] <gsnedders> _get_components, etc.
- # [12:56] <gsnedders> also the query/fragment special-casing in there
- # [12:57] <jgraham> You can have a test that requires a query string to set something
- # [12:57] <gsnedders> but does it only take relative URLs?
- # [12:58] <jgraham> Relative urls and paths
- # [12:58] <jgraham> Well "paths"
- # [12:58] <jgraham> path/to/tests/html/sometest.html?foo=bar is valid
- # [12:59] <jgraham> and it will strip the query
- # [12:59] <jgraham> find that path/to/tests/html/sometest.html is a file
- # [12:59] <gsnedders> also what's the expected behaviour of including path/? does *Everything* under path get run, or just the subset in the manifest?
- # [12:59] <jgraham> And that html/sometest.html?foo=bar is the test
- # [13:00] <jgraham> You can only ever run tests in the manifest
- # [13:00] <jgraham> That's rather the point
- # [13:00] <gsnedders> That's what I thought.
- # [13:00] <gsnedders> But then I included css-fonts-3 and it started trying to run all HTML, XHTMl, and XHTML Print variants of tests.
- # [13:01] <jgraham> Well they're all in the manifest
- # [13:01] <jgraham> try css-fonts-3/html
- # [13:01] <gsnedders> how do they get skipped?
- # [13:01] <gsnedders> or ignored?
- # [13:01] <jgraham> Servo has an include.ini file to specify the precise directories to run
- # [13:02] <jgraham> https://github.com/servo/servo/blob/master/tests/wpt/include_css.ini
- # [13:02] <gsnedders> k
- # [13:03] <jgraham> The whole code is more defensible if you realise it's basically trying to build that manifest file from the command line arguments
- # [13:03] <jgraham> Anyway, lunch
- # [13:10] <gsnedders> so what I want is someway to subset the normal include, I guess?
- # [13:22] * Joins: gitbot (~gitbot@public.cloak)
- # [13:22] -gitbot:#testing- [wptrunner] gsnedders opened pull request #156: Fix the previous commit to actually handle multiple URLs (master...glob_paths_fix) https://github.com/w3c/wptrunner/pull/156
- # [13:22] * Parts: gitbot (~gitbot@public.cloak)
- # [13:42] * Joins: gitbot (~gitbot@public.cloak)
- # [13:42] -gitbot:#testing- [wptrunner] jgraham closed pull request #156: Fix the previous commit to actually handle multiple URLs (master...glob_paths_fix) https://github.com/w3c/wptrunner/pull/156
- # [13:42] * Parts: gitbot (~gitbot@public.cloak)
- # [13:49] * Joins: fwtnb (~fwtnb@public.cloak)
- # [14:03] <gsnedders> "Font names containing any white space need to be quoted. It is also recognized if it is not quoted." — because explicit assertions are useful!
- # [15:27] <AutomatedTester> my fork of wpt is only 3.5k commits behind...
- # [15:29] <gsnedders> hah
- # [15:29] <gsnedders> I know that feel.
- # [15:44] * Joins: ShaneM (~ShaneM@public.cloak)
- # [15:44] * Quits: ShaneM (~ShaneM@public.cloak) (Client closed connection)
- # [15:46] * Joins: ShaneM (~ShaneM@public.cloak)
- # [15:54] * Quits: fwtnb (~fwtnb@public.cloak) (Client closed connection)
- # [15:58] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [16:04] * Joins: plh (plehegar@public.cloak)
- # [16:30] * Quits: ShaneM (~ShaneM@public.cloak) (Ping timeout: 180 seconds)
- # [16:58] * Joins: zcorpan (~zcorpan@public.cloak)
- # [17:06] * Quits: zcorpan (~zcorpan@public.cloak) (Ping timeout: 180 seconds)
- # [17:08] * Joins: ShaneM (~ShaneM@public.cloak)
- # [17:17] * Joins: zcorpan (~zcorpan@public.cloak)
- # [17:18] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [17:58] * Joins: gitbot (~gitbot@public.cloak)
- # [17:58] -gitbot:#testing- [web-platform-tests] Ms2ger closed pull request #2306: Validate capture target authenticity (master...pointerevent-capture-target-authenticity) https://github.com/w3c/web-platform-tests/pull/2306
- # [17:58] * Parts: gitbot (~gitbot@public.cloak)
- # [17:58] * Joins: gitbot (~gitbot@public.cloak)
- # [17:58] -gitbot:#testing- [web-platform-tests] Ms2ger pushed 2 new commits to master: https://github.com/w3c/web-platform-tests/compare/2b817486a1b1...f581ba021596
- # [17:58] -gitbot:#testing- web-platform-tests/master 597e1ba Marius Stefan Bethge: pointerevents: validate capture target authenticity
- # [17:58] -gitbot:#testing- web-platform-tests/master f581ba0 Ms2ger: Merge pull request #2306 from bethge/pointerevent-capture-target-authenticity...
- # [17:58] * Parts: gitbot (~gitbot@public.cloak)
- # [17:58] * Joins: gitbot (~gitbot@public.cloak)
- # [17:58] -gitbot:#testing- [testharness.js] dontcallmedom created promise-throw (+1 new commit): https://github.com/w3c/testharness.js/commit/4676cbf3a77f
- # [17:58] -gitbot:#testing- testharness.js/promise-throw 4676cbf Dominique Hazael-Massieux: Adapt throw test for Promise-returning operations
- # [17:58] * Parts: gitbot (~gitbot@public.cloak)
- # [17:59] * Joins: gitbot (~gitbot@public.cloak)
- # [17:59] -gitbot:#testing- [testharness.js] dontcallmedom opened pull request #161: Adapt throw test for Promise-returning operations (master...promise-throw) https://github.com/w3c/testharness.js/pull/161
- # [17:59] * Parts: gitbot (~gitbot@public.cloak)
- # [18:12] * Joins: gitbot (~gitbot@public.cloak)
- # [18:12] -gitbot:#testing- [csswg-test] kojiishi pushed 1 new commit to master: https://github.com/w3c/csswg-test/commit/abaf54a8e7ec8969c9bd8e7b63fe5c1c759eedfd
- # [18:12] -gitbot:#testing- csswg-test/master abaf54a Koji Ishii: Fix line-box-height-vlr tests to use vlr as references...
- # [18:12] * Parts: gitbot (~gitbot@public.cloak)
- # [18:13] * Quits: Ms2ger (~Ms2ger@public.cloak) (Ping timeout: 180 seconds)
- # [18:23] * Joins: Ms2ger (~Ms2ger@public.cloak)
- # [18:39] * Quits: Ms2ger (~Ms2ger@public.cloak) ("nn")
- # [19:38] * Joins: jyasskin (~textual@public.cloak)
- # [19:44] * Quits: jyasskin (~textual@public.cloak) ("Textual IRC Client: www.textualapp.com")
- # [20:51] * Quits: ShaneM (~ShaneM@public.cloak) (Ping timeout: 180 seconds)
- # [21:58] * Joins: jyasskin (~textual@public.cloak)
- # [22:08] <AutomatedTester> I thought wpt got rid of the submodules...
- # [22:08] <AutomatedTester> I must have remembered incorrectly
- # [23:10] * Quits: dom (dom@public.cloak) (Ping timeout: 180 seconds)
- # [23:15] * Joins: gitbot (~gitbot@public.cloak)
- # [23:15] -gitbot:#testing- [web-platform-tests] AutomatedTester opened pull request #2309: [cssom-view] Add Element From Point tests (master...elementFromPoint) https://github.com/w3c/web-platform-tests/pull/2309
- # [23:15] * Parts: gitbot (~gitbot@public.cloak)
- # [23:27] * Quits: jyasskin (~textual@public.cloak) ("My computer has gone to sleep. ZZZzzz…")
- # Session Close: Sat Nov 07 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