mozilla :: #developers

16 May 2017
00:00pulsebotCheck-in: - Nicholas Nethercote - Bug 1362894 - Make profiler_call_{enter,exit} |inline|. r=mstange.
00:01pulsebotCheck-in: - Shane Caraveo - Bug 1360796 fix intermittent assert, r=mattw
00:05pulsebotCheck-in: - 50 changesets - Merge m-c to autoland a=merge
00:16pulsebotCheck-in: - Eric Rahm - Bug 1356025 - Add Capacity checks to nsTSubstring constructor. r=froydnj
00:19botondLooks like m-u has hit 400k revision :)
00:43pulsebotCheck-in: - cku - servo: Merge #16862 - stylo: Pass Cached ImageValue from stylo back to gecko (from CJKu:bug-1310885-part-6); r=heycam
00:51pulsebotCheck-in: - Andrew McCreight - Bug 1363544 - Call into nsXPConnect directly in the loader and XPCShellImpl. r=mrbkap
00:52pulsebotCheck-in: - Andrew McCreight - Bug 1363555 - Make IS_TEAROFF_CLASS into a function. r=mrbkap
01:47pulsebotCheck-in: - cku - servo: Merge #16864 - Stylo: Change mask-repeat initial value from no-repeat to repeat (from CJKu:bug-1364273); r=heycam
01:57pulsebotCheck-in: - Phil Ringnalda - Bug 1358559 - Sync the disabling of browser_all_files_referenced.js on linux32 to the redundant and duplicative run of the same test in the devtools suite
02:02pulsebotCheck-in: - Phil Ringnalda - Bug 1351197 - Disable test_TelemetryModules.js on Android on release or beta, where it crashes
02:03pulsebotCheck-in: - Daosheng Mu - Bug 1364856 - When VRLayerChild receives __delete__, stop sending IPC messages; r=kip
02:08pulsebotCheck-in: - Fischer.json - Bug 1330315 - Add a telemetry probe to track how the Preferences are opened, r=mconley
02:09birtlesThe & symbols in the new Bugzilla font are fancy
02:10birtlesMaybe not great for reading code, but fancy
02:10dylanbirtles: that's Fira Mono, because windows courier new is really awful on many screens.
02:11birtlesdylan: yeah, it's not awful, but &s are really hard to read -- and there are a lot of &s in rust code
02:12dylanlinux users get Liberation Mono[1] and mac users Menlo. I'm considering shipping Liberation Mono as a web font too.
02:12dylan[1] Technically we serve Courier New for those too, but due to magic font rules that gets translated to Liberation Mono.
02:14birtlesdylan: Liberation Mono looks better to me than Fira Mono for & at least -- but I suspect I'll just get used to Fira Mono before long
02:15birtlesdylan: also, do you know if the red circle with your number of outstanding review requests was removed on purpose?
02:15pulsebotCheck-in: - Jesse Ruderman - Bug 779426 - Crashtest. r=mats, a=test-only
02:15pulsebotCheck-in: - Markus Stange - Bug 1355340 - Don't attempt to accumulate line scrolls for line scroll events. r=masayuki, a=ritu
02:15birtlesoh wait, it's back
02:15birtlesnever mind
02:15birtlesit does some funny things sometimes, like going negative
02:26pulsebotCheck-in: - JW Wang - Bug 1364834 - Fix rounding error in WaveDataDecoder::ProcessDecode(). r=jya
02:32dylanbirtles: the review count math is lazy for the sake of performance.
02:37njnso... the back, forward and reload buttons appear to now be unmoveable in Nightly?
02:44pulsebotCheck-in: - Alastor Wu - Bug 1364296 - enable dom/media/test/test_unseekable.html on Android. r=jwwang
02:45pulsebotCheck-in: - Hiroyuki Ikezoe - Bug 1365091 - Enale transition-and-animation-with-different-durations.html on stylo. r=boris
02:45pulsebotCheck-in: - John Lin - Bug 1363276 - discard video output buffers according to presentation time rather than size. r=esawin
02:50pulsebotCheck-in: - Emilio Cobos lvarez - servo: Merge #16878 - stylo: Rework pseudo-elements to support pseudo selectors with state (from emilio:pseudos); r=bholley,xidorn,hiro
02:52pulsebotCheck-in: - cku - Bug 1364273 - (followup) update to reduce failure count of mask. r=cjku
02:58pulsebotCheck-in: - Chung-Sheng Fu - Bug 1342900 - Remove usage of nsTArray<nsAutoPtr<>> in Classifier. r=dimi
02:59pulsebotCheck-in: - Emilio Cobos lvarez - Bug 1364412: Simplify Servo_HasAuthorSpecifiedRules looking at the pseudo style. r=bholley
02:59pulsebotCheck-in: - Emilio Cobos lvarez - Bug 1364412: Don&#39;t ignore the pseudo-element argument in ResolvePseudoElementStyle. r=heycam
02:59pulsebotCheck-in: - Emilio Cobos lvarez - Bug 1364412: Reftest expectation updates. r=bholley
02:59pulsebotCheck-in: - Emilio Cobos lvarez - Bug 1364412: Convert pseudo-elements to an enum. r=hiro,xidorn
02:59pulsebotCheck-in: - Emilio Cobos lvarez - Bug 1364412: Expose pseudo-element flags, and properly reject pseudos in non-UA sheets. r=bholley
03:00pulsebotCheck-in: - Emilio Cobos lvarez - Bug 1364412: Properly reject to parse pseudo-element states that don&#39;t support state. r=bholley
03:00pulsebotCheck-in: - Stephen A Pohl - Bug 429824: Let OSX&#39;s native menu system handle keyboard shortcuts that we did not handle ourselves via XBL prototype handlers. This allows for handling of custom shortcuts, but does not allow for reassignment of existing shortcuts. r=mstange
03:00pulsebotCheck-in: - Stephen A Pohl - Bug 429824: Properly forward native OSX events to the native menu bar if they haven&#39;t been handled by the child process in e10s. r=mstange,masayuki
03:05birtlesnjn: they&#39;ve been broken for a couple of days I think -- also the &quot;...&quot; this space intentionally left blank&quot; mystery button
03:05njnbirtles: I noticed that too. Very odd
03:11m_katobirtles: bug 1355322 and bug 1355323
03:13pulsebotCheck-in: - Jim Chen - Bug 1362231 - Remove GeckoInterface.getProfile; r=snorp
03:13pulsebotCheck-in: - Jim Chen - Bug 1363167 - Remove GeckoInterface.getActivity; r=snorp
03:13pulsebotCheck-in: - Jim Chen - Bug 1325112 - 1. Remove GeckoInterface.isOfficial; r=nalexander
03:13pulsebotCheck-in: - Jim Chen - Bug 1325112 - 2. Purge caches in debug geckoview_example; r=rbarker
03:13pulsebotCheck-in: - Jim Chen - Bug 1362192 - Remove GeckoInterface.getDefaultUAString; r=snorp
03:13pulsebotCheck-in: - Jim Chen - Bug 1363567 - 1. Remove addPluginView/removePluginView methods; r=rbarker
03:13pulsebotCheck-in: - Jim Chen - Bug 1363567 - 2. Move onFullScreenPluginHidden to GeckoApp; r=rbarker
03:13pulsebotCheck-in: - Jim Chen - Bug 1363567 - 3. Remove GeckoInterface.getPluginContainer; r=me
03:21freesamaelkarlt: Possibly available for some questions about our gtk dpi system?
03:21karltfreesamael: sure
03:26freesamaelkarlt: One specific thing I&#39;m confused of is nsWindow::GetDPI() vs. gfxPlatformGtk::GetDPI(). It feels to me that we should use nsWindow::GetDPI in most case but nsWindow::GetDefaultScaleInternal() is using gfxPlatformGtk&#39;s one.
03:27freesamaelkarlt: Do you recall what the purpose was to have different GetDPI impl. ?
03:31freesamaelWhat I&#39;m trying to do is removing the GetDPI sync IPC call.
03:31freesamaelI&#39;m looking at the possibility of putting DPIs into nsIScreen, and make PuppetWidget::GetDPI use the dpi of the screen it belongs to. So I&#39;m trying to understand how the current implementation works.
03:31pulsebotCheck-in: - Nick Alexander - Bug 1365060 - Move MOZ_{NATIVE_DEVICES, INSTALL_TRACKING} to moz.configure. r=chmanchester
03:33karltfreesamael: which api has a sync ipc call?
03:35freesamaelkarlt: In some cases (not really often, in fact) PuppetWidget::GetDPI causes a sync IPC call tho PBrowser.
03:36ihsiao|sheriffdutyemilio: ping, stylo build bustage,
03:36ihsiao|sheriffdutyheycam: hi ^
03:37emilioihsiao|sheriffduty: is that after
03:37karltfreesamael: looks like the nsIWidget one is a physical dpi. i.e. using actual screen measurements:
03:37emilioihsiao|sheriffduty: I did a few pushes after that that should&#39;ve helped
03:37ihsiao|sheriffdutyit&#39;s from servo: Merge #16878
03:38karltfreesamael: the gfxPlatform one is a &quot;logical&quot; (i.e. fake) dpi for font scaling
03:38ihsiao|sheriffdutyemilio: this one
03:38emilioihsiao|sheriffduty: yeah, the later pushes should&#39;ve fixed the bustage
03:38ihsiao|sheriffdutyemilio: ok, so it should be fixed
03:39emilioihsiao|sheriffduty: let me know if they don&#39;t though! And thanks for pinging :)
03:39ihsiao|sheriffdutygot it, thanks
03:39karltfreesamael: the gfxPlatform number would be just relative to a base of 96
03:47karltfreesamael: moving nsIWidget::GetDPI() to the screen sounds sensible
03:49pulsebotCheck-in: - Brian Birtles - Bug 1355348 - Drop extra eRestyle_StyleAttribute hint from Element::SetSMILOverrideStyleDeclaration; r=dbaron
03:49freesamaelkarlt: One other problem I haven&#39;t figured out is multi-screen support. ScreenHelperGtk is using Xinerama to enumerate monitors:
03:49freesamaelI don&#39;t know how I can apply gdk_screen_get_height_mm with the XDisplay screenInfo
03:51freesamaelkarlt: I&#39;m curious why wouldn&#39;t us using something like gdk_display_get_monitor to enumerate monitors
03:52* freesamael has absolutely no experience with gtk and X...
03:54karltfreesamael: just that gdk didn&#39;t know about monitors when that code was written; using monitors is likely the right thing to do now
03:55freesamaelkarlt: Cool! Is there a minimum gtk version we need to support? I&#39;m hoping not to break everything when changing the code
03:55karltfreesamael: gtk 3.4 is our minimum
03:56karltfreesamael: i expect that has the monitor functions
03:56freesamaelkarlt: Cool! Thanks a lot! I&#39;ll make it a try and may ask for mentoring when I need :)
03:57karltfreesamael: sure, thanks for sorting this out
04:04pulsebotCheck-in: - Glenn Watson - servo: Merge #16860 - Update WR (preserve-3d, AA improvements) (from glennw:preserve3d); r=mbrubeck
04:14njnare 32-bit Mac builds still a thing? or can I remove code specific to that platform?
04:14njnglandium: do you know? ^^
04:31geraldnjn: Mac OS X 10.7 Lion and later only run in 64-bit. ESR45 supports 10.6 (hence 32 bits too), but starting with ff49 we only support 10.9, which is 64-bit only.
04:31njngerald: thanks!
04:31globnjn: also
04:54pulsebotCheck-in: - Mike Hommey - Bug 1363992 - Remove support for system jemalloc. r=njn
04:54pulsebotCheck-in: - Mike Hommey - Bug 1363992 - Remove support for making jemalloc4 the default. r=njn
04:54pulsebotCheck-in: - Mike Hommey - Bug 1363992 - Remove jemalloc 4. r=njn
05:09pulsebotCheck-in: - Hiroyuki Ikezoe - Bug 1364412 - Update reftest and mochitest expectations for pseudo-elements selectors. r=me
05:31drnoihsiao|sheriffduty: have you seen on Windows build this error: Failed to download
05:32glandiumdrno: there&#39;s a problem to get the required authentication tokens
05:33drnoglandium: okay. just wanted to make sure someone is aware of a/the problem :)
05:50ihsiao|sheriffdutyheycam: could you help to check the reftest Rs16 failure like
05:51heycamyeah, those patches should have made that test pass
05:51heycamso I&#39;ll adjust it
05:55* heycam messed up his autoland checkout, so must wait for it to clone again
05:57hiroheycam: I can handle it, I missed the one.
05:57heycamhiro: oh, thank you!
05:58pulsebotCheck-in: - Hiroyuki Ikezoe - Bug 1364412 - Update reftest expectations for pseudo-elements selectors. r=me
05:58hiroheycam, ihsiao|sheriffduty: Done. ^
05:59ihsiao|sheriffdutythanks hiro
05:59hiroihsiao|sheriffduty: np
06:11pboneIf I see code like &quot;class <SomeIdentifier> <AnotherIdentifier> { ...&quot; what does it mean? Is this inheritance?
06:11glandiumpbone: what is SomeIdentifier, and what is AnotherIdentifier?
06:11glandiumpbone: (inheritance is class Foo: Bar)
06:11pbonejust some identifiers, but I&#39;ll get a searchfox link with an example.
06:12pboneYeah, I thought inheritance was class NewClass : ParentClass so this is confusing me.
06:12glandiumI bet one of the identifiers is an annotation or something like that
06:13pboneOh one was a macro but ctags+vim took me to the wrong place..
06:14glandiumpbone: if the macro was not ALLCAPS, maybe there&#39;s a bug to file. What was it?
06:15pboneNah, it was all caps. I trusted ctags too much.
06:15pboneit was JS_PUBLIC_API there,
06:15pboneFollowing it in searchfox gives the macro definition (correct)
06:16pbonebut following with vim + ctags (at least the way I built the ctags file) gave a class.
06:16pboneclass MOZ_STACK_CLASS JS_PUBLIC_API(AutoEntryMonitor) {
06:16glandiumpbone: maybe try something smarter than ctags
06:16pboneBut that&#39;s in one of my build directory&#39;s include files, so I don&#39;t know where to find it for real.
06:17glandiumpbone: we have prefs for ycm in the tree, fwiw
06:17pboneglandium: Do you have a good suggestion? it works for me on C code, but this is my first C++ project in a long long time.
06:17pboneokay cool. I&#39;ll look at that.
06:17pulsebotCheck-in: - Bobby Holley - servo: Merge #16874 - Use a SmallVec when gathering applicable declarations (from bholley:applicable_declarations_smallvec); r=emilio
06:17pboneI may be asking more C++ questions in the next few weeks while I try to remember it.
06:18glandiumpbone: hand in there :)
06:18pboneglandium: thansk for your git-cinnabar tool BTW. DId I tell you this already?
06:19glandiumpbone: maybe, maybe not. Either way, np
06:22pboneycm looks really good. I never knew about this before.
06:25pulsebotCheck-in: - Matteo Ferretti - Bug 1348919 - forced grid cell infobar and area infobar position; r=gl
06:32pulsebotCheck-in: - Christoph Kerschbaumer - Bug 1362993 - Followup: Use gBrowser.addTab with explicit triggeringPricnipal instead of BrowserTestUtils. r=kmaglione
06:32pulsebotCheck-in: - Mark Hammond - Bug 1365150 - profile.get returns null instead of throwing, .add returns the new GUID. r=lchang
06:43njnhmm, what&#39;s the fix for this problem?
06:43njn 0:04.70 /home/njn/moz/mi4/tools/profiler/gecko/nsProfiler.cpp:573:27: error: ambiguous conversion from derived class &#39;nsProfiler&#39; to base class &#39;nsISupports&#39;:
06:43njn 0:04.70 class nsProfiler -> class nsIProfiler -> class nsISupports
06:43njn 0:04.70 class nsProfiler -> class nsIObserver -> class nsISupports
06:43njn 0:04.70 os->NotifyObservers(this, &quot;profiler-subprocess-gather&quot;, nullptr);
06:44heycamnjn: some macro with the word AMBIGUOUS in it, I think
06:44njnheycam: NS_ISUPPORTS_CAST, maybe?
06:45heycamnjn: oh, just for that call. yeah I think that&#39;s the one.
06:47njnheycam: yep, seems to work; thanks
07:05John-GaltLooks like Windows builds are busted on infra.
07:06John-GaltNon-tc build, at least.
07:08Tomcat|sheriffdutyon m-c, inbound ?
07:16ihsiao|sheriffdutyTomcat|sheriffduty: on autoland, m-i, m-c
07:24John-GaltTomcat|sheriffduty: Everywhere
07:24John-Galttry, too
07:30Tomcat|sheriffdutyihsiao|sheriffduty: thats something for buildduty
07:30Tomcat|sheriffdutyihsiao|sheriffduty: it tries to download something and run into a 403
07:31pulsebotCheck-in: - Xidorn Quan - servo: Merge #16888 - Implement @counter-style for stylo (from upsuper:bug1328319); r=heycam,Manishearth,SimonSapin
07:34pulsebotCheck-in: - Xidorn Quan - Bug 1328319 part 1 - Move nsCSSCounterStyleRule into a separate header file. r=heycam
07:34pulsebotCheck-in: - Xidorn Quan - Bug 1328319 part 2 - Add basic integration of @counter-style. r=heycam
07:34pulsebotCheck-in: - Xidorn Quan - Bug 1328319 part 3 - Add support for auto value in nsCSSValue sugar. r=manishearth
07:34pulsebotCheck-in: - Xidorn Quan - Bug 1328319 part 4 - Add binding for setting pair value of nsCSSValue. r=manishearth
07:34pulsebotCheck-in: - Xidorn Quan - Bug 1328319 part 5 - Add bindings for list and pair list value of nsCSSValue. r=manishearth
07:34pulsebotCheck-in: - Xidorn Quan - Bug 1328319 part 6 - Make it possible to delay resolving counter style. r=heycam
07:34pulsebotCheck-in: - Xidorn Quan - Bug 1328319 part 7 - Enable querying counter-style rule on Servo backend. r=heycam
07:34pulsebotCheck-in: - Xidorn Quan - Bug 1328319 part 8 - Make stylo use counter-style for list-style-type and counter functions. r=heycam,SimonSapin
07:34pulsebotCheck-in: - Xidorn Quan - Bug 1328319 part 9 - Don&#39;t clean up retired counter styles for stylo. r=heycam
07:34pulsebotCheck-in: - Xidorn Quan - Bug 1328319 part 10 - Update test expectations. r=heycam
08:31Tomcat|sheriffdutyjhlin: ping
08:31jhlinTomcat|sheriffduty: pong
08:32Tomcat|sheriffdutyjhlin: hi ! seems your push to autoland cause a leak
08:32Tomcat|sheriffdutycould you take a look ?
08:36jhlinTomcat|sheriffduty: My push for bug 1363276? It is only run on Android.
08:36pulsebotCheck-in: - Xidorn Quan - Bug 1355352 - Reject unicode-range descriptor when its value contains invalid part. r=jfkthame
08:40Tomcat|sheriffdutyjhlin: hmm ok :)
08:40Tomcat|sheriffdutylets do more retrigger
08:58pulsebotCheck-in: - Bobby Holley - Bug 1364863 - Make stylo property of reftests dependent on the pref, not just the define. r=heycam
09:02pulsebotCheck-in: - Yoshi Huang - Bug 1363597: dont use data: URI for form action. r=smaug
09:02pulsebotCheck-in: - Yoshi Huang - Bug 1363612 - Part 1: move to seperate files. r=smaug
09:02pulsebotCheck-in: - Yoshi Huang - Bug 1363612 - Part 2: move files for bug 677495. r=smaug
09:02pulsebotCheck-in: - Yoshi Huang - Bug 1363612 - Part 3: fix for moz-binding: url. r=smaug
09:02pulsebotCheck-in: - Yoshi Huang - Bug 1363632 - rewrite test_497898.html. r=smaug
09:02pulsebotCheck-in: - Yoshi Huang - Bug 1364369 - update more files to use srcdoc attr. r=smaug
09:02pulsebotCheck-in: - Yoshi Huang - Bug 1363673 - fix tests that use data:image/png. r=smaug
09:44pulsebotCheck-in: - Iris Hsiao - Backed out 4 changesets (bug 1343912) for test_timer_flood.html perma-failing
09:45pulsebotCheck-in: - Julian Descottes - Bug 1365014 - support pref getters with default value in Services shim;r=tromey
09:45pulsebotCheck-in: - Julian Descottes - Bug 1365014 - inspector.html add shims for sdk, uuid, lazy requires;r=gl
09:47pulsebotCheck-in: - Andreas Pehrson - Bug 1363253 - Ignore events specific to preview window also when decrementing. r=florian
10:10pulsebotCheck-in: - Nihanth Subramanya - Bug 1363842 - Implement new toolbar button hover and active background styles. r=dao
10:19pulsebotCheck-in: - Carsten &quot;Tomcat&quot; Book - Backed out changeset aed8ff339114 (bug 1364911) for memory leaks and making Bug/test failure bug 1340425 perma fail
10:19pulsebotCheck-in: - Jan Beich - Bug 1365026 - Skip places/arrow-down.svg on other Gtk platforms as well. r=dao
10:21pulsebotCheck-in: - ffxbld - No bug, Automated blocklist update from host bld-linux64-spot-389 - a=blocklist-update
10:34pulsebotCheck-in: - 55 changesets - merge autoland to mozilla-central a=merge
10:36smaughrm, why is location bar gray
10:36smaugin up-to-date m-i builds
10:49Gijssmaug: sounds like a regression from bug 1352366 (cc daleharvey )
10:50daleharveysmaug: using any theming (from os or firefox?)
10:50daleharveyand what platform
10:50smaugdaleharvey: this is default theme for Fedora 25 I think
10:51smaugno firefox themes
10:51smaugdaleharvey: both locationbar and searchbar have gray background
10:54daleharveycool cheers, definitely shouldnt be grey, they are both set to -moz-field which is as they were before but will find out whats up
10:57pulsebotCheck-in: - Nihanth Subramanya - Bug 1363869 - Increase vertical padding of toolbar buttons. r=dao
10:58smaugdaleharvey: uploaded a screenshot to the bug
10:58smaugHmm, do we have any image sharing tool
10:58smaugsomething similar to
11:00daleharveysmaug: cheers
11:01pulsebotCheck-in: - Henrik Skupin - Bug 1364245 - Extend delay for slow loading resource testcase to prevent intermittent. r=ato
11:10pulsebotCheck-in: - Masayuki Nakano - Bug 1303260 Get rid of unnecessary <input> and the code setting focus to it in parent window r=smaug
11:15pulsebotCheck-in: - Masayuki Nakano - Bug 1354443 part.1-1 Make nsPrintData refcountable r=dholbert
11:15pulsebotCheck-in: - Masayuki Nakano - Bug 1354443 part.1-2 Methods of nsPrintEngine should guarantee that objects owned by nsPrintData won&#39;t be released when they&#39;re referred by calling methods via their arguments r=dholbert
11:15pulsebotCheck-in: - Masayuki Nakano - Bug 1354443 part.2 nsPrintEngine::Disconnect() should clear mPrintObj r=dholbert
11:15pulsebotCheck-in: - Masayuki Nakano - Bug 1354443 part.3 While nsDocumentViewer is destroying mPrintEngine, null out its pointer & transfer ownership to local var r=dholbert
11:15pulsebotCheck-in: - Masayuki Nakano - Bug 1354443 part.4 nsPrintData should guarantee that object which has nsIWebProgressListener interface won&#39;t be released during calling a method of it r=dholbert
11:17pulsebotCheck-in: - Simon Sapin - servo: Merge #16870 - Add size_of tests for geckolib selectors (from servo:size_of); r=emilio
11:34daleharveysmaug: found the problem, cheers
11:53pulsebotCheck-in: - Mark Banner - Bug 1361976 - ESLint hooks should ensure that the ESLint setup is up to date. r=mossop
11:58GijsI basically can&#39;t use my locally built osx nightly (off current m-c) because of continuous crashing
11:59Gijstop of stack is apparently (according to apple&#39;s crash reporter) 0 XUL 0x00000001098fd325 JSAutoCompartment::JSAutoCompartment(JSContext*, JSScript*) + 21 (jsscript.h:1191)
11:59Gijsanyone seen this and/or know what bug to look at / what to pref off to make it not crash every minute or so?
11:59Gijsthis happened with yesterday&#39;s m-c tip as well but it&#39;s still there today and now I&#39;m fed up.
12:07pulsebotCheck-in: - Sebastian Hengst - Backed out changeset ac59d773e2c2 (bug 1361976) for busting ESlint (module setup_helper missing). r=backout
12:09pulsebotCheck-in: - Mike de Boer - Bug 1354128 - Add Web Developer button to the photon app menu. r=Gijs
12:30pulsebotCheck-in: - Marco Castelluccio - Bug 1353882 - Disable TestArenaAllocator.AllocationsPerChunk and TestArenaAllocator.MemoryIsValid tests in code coverage builds. r=jmaher
12:51katschutten: why is telemetry.m.o not showing data more recent than a week ago?
12:52katschutten: (well, the nightly buildIDs stop at may 09)
13:05pulsebotCheck-in: - 43 changesets - merge mozilla-inbound to mozilla-central a=merge
13:09pulsebotCheck-in: - 58 changesets - Merge mozilla-central to mozilla-inbound on a CLOSED TREE
13:13pulsebotCheck-in: - 46 changesets - Merge mozilla-central to autoland on a CLOSED TREE
13:21chuttenkats: Backfilling after a short outage
13:22katschutten: ok, thanks
14:39pulsebotCheck-in: - ffxbld - Automatic version bump. CLOSED TREE NO BUG a=release
14:39pulsebotCheck-in: - ffxbld - No bug - Tagging 9bef2dfbd584c0f3ae52953bfc3aed64e308960d with FIREFOX_54_0b8_BUILD1, FIREFOX_54_0b8_RELEASE a=release CLOSED TREE
14:48dylanso if anyone notices that bzexport is faster -- you&#39;re welcome. An API it was using is now an order of magnitude faster.
14:52dylanwell, 99% of the time. If you&#39;re unlucky it will take the same amount of time. #cachingismagic
15:01stormimyk: hi
15:01mykhi stormi
15:02mykstormi: shall we chat about embedding?
15:02stormimyk: yes
15:02stormimyk: I made some slides to help explain our usage
15:03stormijesup: ready?
15:04stormiI will not easily have both access to IRC and to a microphone so I suggest we first start with IRC and switch if needed
15:04mykstormi: ok, sounds good
15:04mykstormi, jesup: let&#39;s chat in the #embedding channel
15:08pulsebotCheck-in: - ffxbld - No bug, Automated HSTS preload list update from host bld-linux64-spot-388 - a=hsts-update
15:08pulsebotCheck-in: - ffxbld - No bug, Automated HPKP preload list update from host bld-linux64-spot-388 - a=hpkp-update
15:08pulsebotCheck-in: - ffxbld - No bug, Automated blocklist update from host bld-linux64-spot-388 - a=blocklist-update
15:12pulsebotCheck-in: - ffxbld - No bug, Automated HSTS preload list update from host bld-linux64-spot-304 - a=hsts-update
15:12pulsebotCheck-in: - ffxbld - No bug, Automated HPKP preload list update from host bld-linux64-spot-304 - a=hpkp-update
15:12pulsebotCheck-in: - ffxbld - No bug, Automated blocklist update from host bld-linux64-spot-304 - a=blocklist-update
15:14pulsebotCheck-in: - ffxbld - No bug, Automated HSTS preload list update from host bld-linux64-spot-380 - a=hsts-update
15:14pulsebotCheck-in: - ffxbld - No bug, Automated HPKP preload list update from host bld-linux64-spot-380 - a=hpkp-update
15:14pulsebotCheck-in: - ffxbld - No bug, Automated blocklist update from host bld-linux64-spot-380 - a=blocklist-update
15:28bkellyare nightly updates still turned off?
15:28bsmedberggsvelto, at why did you change/have to change the return type from void to nsISupports?
15:29gsveltobsmedberg, so that I can return a promise, I need to wait for the operation to finish in order to test it
15:29RyanVMbkelly: yes, and today&#39;s are delayed by infra issues
15:29bsmedberggsvelto, are promises auto-coerced to nsISupports?
15:30gsveltobsmedberg, apparently... I&#39;ve looked around into other code returning promises and they do it either by using nsISupports or jsval as the return type
15:30bsmedbergI am surprised that the `await cs.addCrash()` code works properly
15:31bsmedbergand a little worried
15:32* bsmedberg goes to read the promise impl, maybe it&#39;s using classinfo magic
15:32gsveltoI&#39;m not 100% sure that&#39;s the proper way to do it, but that&#39;s what other bits of code do
15:32gsveltobsmedberg, BTW you&#39;ll notice I added an option to the CrashManager constructor in the CrashManagerTest object, that&#39;s for enabling the CrashStore to work properly
15:34bsmedberggsvelto, aha
15:35bsmedbergit is magic
15:35gsveltobsmedberg, is it OK though? Or should I do it differently?
15:35bsmedbergyeah it&#39;s fine then
15:36gsveltogreat, thanks
15:42Gijsbaku: re: reader mode fix, sorry, I wrote a fairly similar patch but with more comments and an explicit nulling out (to make it more obvious in code that trying to use &#39;doc&#39; is a bad idea) in bug 1364056 - can you stamp that, given that it does more or less the same thing? :)
15:43bakuGijs: r+
15:43Gijsbaku: thanks!
15:56tfeon firefox / desktop, how can i run a single js test file?
15:57bsmedbergtfe, what kind of test?
15:57tfea browser based test
15:57daleharvey./mach test path/to/test.js usually works
15:57tferelated to
16:22katsis anybody else getting an error building on windows, unable to find mt.exe during the configure checks?
16:28pulsebotCheck-in: - Kris Maglione - Bug 1365256: Don&#39;t register chrome for disabled, restartful add-ons. r=aswan a=RyanVM CLOSED TREE
16:30RyanVMkats: sounds like it&#39;s not finding the Windows SDK, but no, I haven&#39;t seen that
16:31bzDo we have any sort of ETA on inbound reopening?
16:32Aryxkats: bug 1356493
16:32gregglind(anyone have a sense for good ways to turn use browserify/webpack/some other tool to amke a single ouptut .jsm library. use case: shield-study-utils)
16:33gregglind(which wants to be included in other addons without a ton of hassle.)
16:33katsRyanVM: Aryx: thanks
16:33Aryxbz: sorry, no idea
16:33RyanVMbz: I don&#39;t think so, sorry
16:34bzAryx, RyanVM: No need to apologize. Just checking so I can plan time.
16:37ccorcoranI wrote an xpcshell test, ran it on treeherder. osx & windows are as expected, but the test didn&#39;t run on linux or android. Why would this be?
16:37ccorcoranthe test is &quot;launchWithoutHang_bug1360493.js&quot;
16:39RyanVMccorcoran: [task 2017-05-16T14:57:23.929505Z] 14:57:23 INFO - TEST-START | toolkit/xre/test/launchWithoutHang_bug1360493.js
16:39RyanVM[task 2017-05-16T14:57:40.047027Z] 14:57:40 INFO - TEST-PASS | toolkit/xre/test/launchWithoutHang_bug1360493.js | took 16117ms
16:39Standard8ccorcoran: I doubt this is it, but the convention with xpcshell tests is to format the filename as test_*.js and generally to drop the bug number
16:40RyanVMccorcoran: ^ is from the linux32 opt X8 run -
16:40RyanVMccorcoran: shows it running and failing on Android too
16:41ccorcoranohh ok i thought all the Xn runs are identical
16:41ccorcoranit&#39;s distributing them i see
16:41ccorcoranStandard8: thanks; i&#39;ll take it into consideration
17:09ehsandoes anyone know of a replacement for the mass password reset extension that works on Nightly?
17:10Gijswho does top crash work these days?
17:12rstrongehsan: I don&#39;t but I got around it by launching a release build with my profile and then going back to nightly
17:13rstrongmight have used beta
17:13ehsanrstrong, oh... going back to older builds on the same profile is dangerous, it can destroy data...
17:13* ehsan prefers editing passwords manually :/
17:14rstrongI know :) I made a backup just in case. No problems so far and I did it days ago
17:14ehsanone day hopefully we won&#39;t force LDAP password resets any more :(
17:14ehsanrstrong, glad you thought of making a backup :)
17:14* rstrong remembers when we used to try to be backwards compatible
17:17freddybehsan: you can search by pa