[00:00:51] *** devinus has left #chromium
[00:01:01] <ananta> dmaclach: I just restarted the chrome frame builders. They should turn green in this run.
[00:02:21] <mmoss> ananta: did you mean to kill chrome frame (ie6) during an update?
[00:02:23] *** eggy has quit IRC
[00:02:37] *** mdm_chromium has quit IRC
[00:02:45] *** mrichards has joined #chromium
[00:02:57] *** mrichards is now known as eggy
[00:03:10] <rsesek> thakis: you up for a CL?
[00:03:18] <thakis> rsesek: sure
[00:03:56] <ananta> mmoss: yes.
[00:03:57] <thakis> man, i'm 95% done with something the whole day already
[00:04:02] <thakis> sucks
[00:04:18] <mmoss> ananta: looks like it's come back OK, but I think that's generally a bad idea and can confuse svn on the next run (lingering locks and such)
[00:04:26] <mmoss> ananta: in case you weren't aware
[00:04:41] <rsesek> thakis: sent
[00:04:43] <ananta> mmoss: ok
[00:04:58] <ojan> wow. the tree is super behind.
[00:05:35] <ojan> dmaclach, mmoss, tyoshino: I have a webkit roll ready to commit. Can you let me know when it's ok to commit it?
[00:05:40] <mmoss> ojan: you should have seen it before I closed :)
[00:05:48] <ojan> mmoss: i'm sure
[00:05:52] <mmoss> ojan: ok, soon
[00:06:05] <ojan> mmoss: i can wait. just wanted to get in your queueu
[00:07:06] <mmoss> ojan: wait like, 'til everyone else goes home and you can have the tree to yourself? or wait like, sometime this hour?
[00:07:52] *** johnny_g has quit IRC
[00:08:09] <ojan> mmoss: the latter
[00:08:22] <mmoss> ojan: ok, that's what I figured
[00:08:27] <ojan> mmoss: i'd like to get another roll in today as well if possible
[00:08:35] <mmoss> ojan: I think it's just going to stay closed for a while now
[00:09:03] *** fusion44 has quit IRC
[00:09:24] <mmoss> you all had a good run the last hour or so, sorry it had to end
[00:09:53] <thakis> rsesek: ui guys are looped in on that?
[00:10:38] *** Omen_20 has quit IRC
[00:16:28] <andybons> thakis: i'm talking to cole about it as soon as he's back as part of the top chrome stuff.
[00:18:19] *** pfeldman_ has joined #chromium
[00:19:25] *** Gina has quit IRC
[00:20:21] *** awolfson has quit IRC
[00:21:54] *** pfeldman has quit IRC
[00:22:01] <mmoss> mpcomplete: http://buildbot.jail.google.com/buildbot/chromium/builders/Chromium%20Linux/builds/13889/steps/check%20perms/logs/stdio
[00:22:13] <andybons> thakis: in the meantime i'll take responsibilty for making sure it's reverted otherwise
[00:22:14] <mmoss> mpcomplete: check perms failed
[00:22:25] *** pfeldman_ is now known as pfeldman
[00:22:32] <mpcomplete> mmoss: ok, on it
[00:22:35] <mmoss> thanks
[00:22:43] *** Omen_20 has joined #chromium
[00:22:57] <thakis> andybons: ok
[00:23:45] <trungl-bot> Tree closed by mmoss at chromium dot org: Tree is closed. Catching up still, watching webkit roll. ETA 3:40 PST.
[00:24:34] *** bratsche has quit IRC
[00:28:07] *** Gina has joined #chromium
[00:28:37] *** andybons has quit IRC
[00:29:45] *** coyo|pingout has quit IRC
[00:29:46] *** markmentovai has quit IRC
[00:33:16] *** fta_ has joined #chromium
[00:33:54] *** fta has quit IRC
[00:34:00] *** fta_ is now known as fta
[00:34:33] *** awidegreen has quit IRC
[00:34:47] <mmoss> ojan: any concern about Webkit Linux?
[00:35:19] <ojan> mmoss: doh. i messed up the expectaitons file. i'll fix now.
[00:38:13] *** bratsche has joined #chromium
[00:38:49] *** bent-mozilla has quit IRC
[00:41:41] <mmoss> ojan: Chromium Mac Builder (dbg)? link failures
[00:46:37] <ojan> mmoss: don't see anything in the webkit roll that would cause that.
[00:46:38] *** rdsmith has quit IRC
[00:46:44] <ojan> mmoss: hoping a clobber will fix it
[00:46:51] <ojan> mmoss: esp since the release bot compiled
[00:47:25] <mmoss> ojan: clobber it is then
[00:47:30] <ojan> mmoss: and webkit mac dbg compiled
[00:47:39] <ojan> mmoss: i allready kicked it off
[00:47:39] <mmoss> ojan: yeah
[00:48:43] <pkasting> FWIW, the ETA was ten minutes ago... might want to update it
[00:48:56] <mmoss> pkasting: or I might just open :)
[00:49:07] <pkasting> I am not opposed to that!
[00:50:08] <trungl-bot> Tree opened by mmoss at chromium dot org: Tree is open.
[00:50:45] *** rohitrao has quit IRC
[00:50:54] *** Venom_X has quit IRC
[00:51:19] *** novas0x2a has joined #chromium
[00:53:55] *** chocobo__ has joined #chromium
[00:53:55] *** ChanServ sets mode: +v chocobo__
[00:54:03] <ojan> mmoss: nope...it's a real failure. i'll rollback. :(
[00:54:16] <chocobo__> sheriffs: can I commit a change with a grd change?
[00:56:19] *** pfeldman has quit IRC
[00:56:26] *** General13372 has joined #chromium
[00:57:06] *** Erkan_Yilmaz_ has joined #chromium
[00:57:32] <ojan> mmoss: ok. rolled back. sorry. i thought it had run through all the bots. :(
[00:58:47] *** jrmuizel has quit IRC
[00:58:57] *** chocobo___ has joined #chromium
[00:58:57] *** ChanServ sets mode: +v chocobo___
[00:59:28] *** AaronMT has quit IRC
[00:59:45] *** General1337 has quit IRC
[01:00:25] *** Erkan_Yilmaz__ has quit IRC
[01:00:37] *** pdelgallego has quit IRC
[01:02:14] *** chocobo__ has quit IRC
[01:02:14] *** chocobo___ is now known as chocobo__
[01:02:50] <dmaclach> chocobo__: we're pretty gray right now, anybody else got grd changes?
[01:02:51] <ojan> argh. i don't knwo what happened. the webkit roll definitely compiled on the mac-dbg try bot
[01:03:01] <mmoss> chocobo__: hold off a bit, otherwise it's going to batch up with a dozen other commits
[01:03:21] <mmoss> ojan: did you say you rolled back?
[01:03:27] <ojan> mmoss: yeah
[01:03:53] <mmoss> ojan: why am I not seeing another commit?
[01:04:07] <chocobo__> sure, i will wait a bit. thanks
[01:04:24] <ojan> mmoss: argh. i guess i didn't. too much breaking at once. :(
[01:04:50] <aboodman> it is clear to me now that i need to setup the git tree that pulls from svn
[01:04:56] <mmoss> I'm going to close for a minute, ojan go ahead and commit, chocobo__ you'll be next
[01:05:02] <aboodman> having the delay from the mirror syncing doesn't work anymore
[01:05:18] <trungl-bot> Tree closed by mmoss at chromium dot org: Tree is closed. Catching up.
[01:05:36] *** novas0x2a has quit IRC
[01:05:42] <chocobo__> ok, let me know when
[01:06:19] *** yutak_home has joined #chromium
[01:06:20] *** ChanServ sets mode: +v yutak_home
[01:06:34] <ojan> mmoss: ok...committed
[01:06:57] <mmoss> aboodman: any idea if we've looked into triggers to get commits to replicate more real-time-ish?
[01:07:03] <ojan> mmoss: never mind...it failed.
[01:07:22] <aboodman> mmoss: dunno. evmar told me that he just sets up a tree that syncs to svn for these situations.
[01:07:27] <awong> ojan: want me to revert for you?  I can do it fast..
[01:07:43] *** viro101 has quit IRC
[01:08:00] <mmoss> aboodman: yeah, that was the original git setup (I still use it too)
[01:08:49] <ojan> ok. geez. finally committed for real.
[01:17:47] *** bratsche has quit IRC
[01:18:14] <dcheng> On Linux, is there a good way to compare a string16 against constants? L"foo" doesn't work since the chars in L"foo" are 4 bytes wide.
[01:18:21] <mmoss> jvoung? anyone?
[01:18:29] *** yutak_home has quit IRC
[01:18:41] <awong> dcheng: convert it to a string16 first?
[01:18:47] <jamesr> dcheng: UTF8ToUTF16("foo") ?
[01:18:52] <jamesr> or ASCIIto or whatever the call is
[01:19:23] <dcheng> So gcc has no prefix to use 16-bit chars for string constants?
[01:19:28] <awong> dcheng: no.
[01:19:56] <awong> wchar_t is 4-bytes in gcc to avoid needing UTF-16.
[01:23:00] *** chocobo__ has quit IRC
[01:23:29] <trungl-bot> Tree closed by mmoss at chromium dot org: Tree is closed. Catching up. Linux dbg-shlib -> jvoung
[01:23:52] *** chocobo__ has joined #chromium
[01:23:52] *** ChanServ sets mode: +v chocobo__
[01:25:00] <mmoss> chocobo__: grd change ready?
[01:25:22] *** xji has quit IRC
[01:26:20] *** xji has joined #chromium
[01:26:20] *** ChanServ sets mode: +v xji
[01:26:33] *** janm has quit IRC
[01:30:46] <chocobo__> yes, should i check in?
[01:31:44] <mmoss> chocobo__: yeah, now is good
[01:32:14] *** jrmuizel has joined #chromium
[01:32:16] <chocobo__> mmoss: done. thanks.
[01:32:36] *** bratsche has joined #chromium
[01:33:52] *** fta_ has joined #chromium
[01:34:23] *** fta has quit IRC
[01:34:32] *** fta_ is now known as fta
[01:36:50] *** abarth has quit IRC
[01:37:00] *** abarth has joined #chromium
[01:37:01] *** ChanServ sets mode: +v abarth
[01:39:06] *** wash has joined #chromium
[01:40:41] <trungl-bot> Tree closed by mmoss at chromium dot org: Tree is closed. Catching up.
[01:41:08] <awong> mmoss: should we revert jvoung's change?  Alternately, I could try to fix it.
[01:41:17] <awong> but the chagne actulaly has a number of style issues too :(
[01:41:18] <mmoss> awong: fix is in
[01:41:21] <awong> ok
[01:42:19] *** rsesek has quit IRC
[01:43:22] <ojan> i can't reproduce the mac compile failure locally :(
[01:46:01] <reiz> arg, I broke chrimium mac valgrind
[01:46:07] *** reiz is now known as leiz
[01:49:27] *** FullFlannelJacke has quit IRC
[01:51:23] *** chrisccoulson has quit IRC
[01:51:32] *** ebraminio has joined #chromium
[01:53:30] *** phajdan-jr|afk is now known as phajdan-jr
[01:53:47] <phajdan-jr> okay, who wants to LGTM a change that lowers ui_tests timeouts? :) http://codereview.chromium.org/3008002
[01:54:22] <awong> I do, but I'm not sure if I have the right. :P
[01:54:47] <mmoss> anybody have a clue what's going on with Chromium Builder (dbg)?
[01:55:15] <miranda_afkwdog> phajdan-jr: there you go.  I love shorter timeouts!
[01:55:20] <mmoss> fatal error LNK1318: Unexpected PDB error; UNKNOWN (24) ''
[01:55:21] *** miranda_afkwdog is now known as mirandac
[01:55:53] <awong> mmoss: sounds like a clobber?
[01:56:01] <phajdan-jr> mirandac: excellent! will wait a while for the trybots to finish, just to make sure it won't explode
[01:56:04] <mmoss> yeah, maybe
[01:56:21] <phajdan-jr> sounds like "punch Windows" to me
[01:57:04] <mirandac> phajdan-jr: I'm guessing you've run the ui_tests, and the shorter timeouts caused no major additional destruction?
[01:58:25] <mmoss> sheriffs are going away
[01:58:28] <phajdan-jr> mirandac: I'll let the trybots confirm this. I mean, the timeouts used to be at these values, so it should work.
[01:58:51] <mirandac> excellent!
[01:59:26] <mmoss> I'm going to open. There's a grd change (52573) that hasn't started on all builders yet, so you might want to hold off if you don't want to get caught up with that
[02:00:53] <trungl-bot> Tree opened by mmoss at chromium dot org: Tree is open. No Sheriffs.
[02:01:11] *** Viclame1 has joined #chromium
[02:02:05] *** bweinstein has quit IRC
[02:02:45] *** Viclame has quit IRC
[02:02:53] *** FullFlannelJacke has joined #chromium
[02:05:11] *** mdm_chromium has joined #chromium
[02:06:42] *** GeekShadow has quit IRC
[02:10:01] *** Erkan_Yilmaz__ has joined #chromium
[02:12:21] *** awolfson has joined #chromium
[02:13:12] *** Erkan_Yilmaz has joined #chromium
[02:13:15] *** Erkan_Yilmaz_ has quit IRC
[02:14:40] *** yutak_home has joined #chromium
[02:14:41] *** ChanServ sets mode: +v yutak_home
[02:15:29] *** Erkan_Yilmaz_ has joined #chromium
[02:15:56] *** Erkan_Yilmaz__ has quit IRC
[02:18:13] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "unit_tests" on "Mac10.6 Tests (dbg)(1)" from 52574: apatrick at chromium dot org, chocobo at chromium dot org, derat at chromium dot org, erikkay at chromium dot org, hclam at chromium dot org, jiesun at google dot com, jvoung at google dot com, kathyw at chromium dot org, ojan at google dot com, tony at chromium dot org (:tony^work), vrk at google dot com)
[02:18:30] *** Erkan_Yilmaz__ has joined #chromium
[02:18:58] *** dmaclach has quit IRC
[02:19:00] *** Erkan_Yilmaz has quit IRC
[02:19:17] <aboodman> that has to be a new record
[02:19:32] <aboodman> 18 minutes
[02:20:05] <aboodman> i can eliminate kathyw if that helps :)
[02:20:10] <aboodman> docs changes only
[02:21:41] *** Erkan_Yilmaz_ has quit IRC
[02:22:25] *** Erkan_Yilmaz has joined #chromium
[02:23:55] <tony^work> i touched a .sh file that isn't run by the bots
[02:24:45] *** Erkan_Yilmaz__ has quit IRC
[02:29:30] *** joshia has quit IRC
[02:33:01] *** phajdan-jr has quit IRC
[02:33:26] <trungl-bot> Tree closed by tony at chromium dot org (:tony^work): Tree is closed (flaky test, but letting the bots catch up now)
[02:33:43] *** trungl has quit IRC
[02:37:05] *** ukai has joined #chromium
[02:39:40] *** eseidel has quit IRC
[02:39:47] *** fta_ has joined #chromium
[02:41:13] *** fta has quit IRC
[02:41:14] *** fta_ is now known as fta
[02:41:43] *** jrmuizel has quit IRC
[02:42:32] *** thakis_ has joined #chromium
[02:42:32] *** ChanServ sets mode: +v thakis_
[02:42:49] <tony^work> so, someone marked DispatchSimple as flakey
[02:42:52] <tony^work> but it should be disabled
[02:42:58] <tony^work> it doesn't fail, it crashes
[02:45:00] *** jcivelli has quit IRC
[02:45:20] <tony^work> anyway, I will reopen since things seem ok
[02:45:23] <tony^work> and disable the test
[02:46:34] <trungl-bot> Tree opened by tony at chromium dot org (:tony^work): tree is open, please land slowly
[02:50:30] * aboodman goes on a commit spree
[02:51:20] *** RT|Chatzilla has joined #chromium
[02:52:12] <thakis_> aboodman: the file list at the end of http://code.google.com/p/chromium/issues/detail?id=29995 is way way longer than the one at http://codereview.chromium.org/2932007
[02:52:16] <thakis_> not sure if that's intended?
[02:52:35] <aboodman> thakis_: they are generated changes.
[02:52:35] <thakis_> also, i love that this bug is fixed
[02:52:39] *** thakis_ has left #chromium
[02:52:44] <aboodman> wait...
[02:52:46] *** thakis_ has joined #chromium
[02:52:46] *** ChanServ sets mode: +v thakis_
[02:52:47] *** thakis_ has left #chromium
[02:53:06] *** thakis_ has joined #chromium
[02:53:06] *** ChanServ sets mode: +v thakis_
[02:53:08] <thakis_> oh, i see
[02:53:25] <aboodman> argh, wrong bug attribution
[02:53:35] <aboodman> darn
[02:53:38] <aboodman> i hate when that happens
[02:53:43] <aboodman> no way to fix retroactively
[02:56:09] <aboodman> thakis_: thanks for letting me know. something weird happened with the change descriptions. i will revert and fix.
[02:57:17] *** thakis_ has quit IRC
[03:02:07] *** yutak_home has quit IRC
[03:04:59] *** kurrik has quit IRC
[03:08:17] <atwilson> Anyone around who can answer a question for me about GTK widgets? I'm getting a crash on shutdown with my status tray code, and I'm trying to figure out the right fix.
[03:08:54] *** estes has quit IRC
[03:12:00] *** feldstein has quit IRC
[03:12:25] <estade> atwilson: yea
[03:12:35] <estade> atwilson: probably you should use an ownedwidgetgtk
[03:13:06] *** viro101 has joined #chromium
[03:13:14] <atwilson> estade: I'm wondering what that means :)
[03:13:14] <estade> alternatively you could NULL out your widget when it gets destroyed with g_signal_connect(widget, "destroy", G_CALLBACK(gtk_widget_destroyed), &widget);
[03:13:32] <atwilson> There's code in browser_list_gtk that destroys all visible widgets.
[03:13:48] <atwilson> But my status_tray code also closes all status icons when it is destroyed.
[03:14:04] <atwilson> The problem is that the "destroy all widgets" code in browser_list_gtk does it behind my back.
[03:14:17] *** sanjeevr has quit IRC
[03:14:20] <estade> atwilson: no, it only destroys windows
[03:14:25] <atwilson> Hmmm.
[03:14:40] *** mazda has joined #chromium
[03:15:09] *** hbono has joined #chromium
[03:15:10] <estade> gtk_widget_destroy only sends the "destroy" signal
[03:15:16] <atwilson> estade: perhaps. I'm not up on gtk APIs (tfarina wrote the gtk version of status icons).
[03:15:20] *** rafaelw has left #chromium
[03:15:38] <atwilson> status_icon_gtk unrefs my icon, but it yields:
[03:15:39] <atwilson> [13788:13788:2455032693:FATAL:chrome/app/chrome_dll_main.cc(250)] Gtk: gtk_widget_destroy: assertion `GTK_IS_WIDGET (widget)' failed
[03:15:55] <atwilson> If I comment out the code in browser_list_gtk so it doesn't call gtk_widget_destroy, everything is fine.
[03:16:00] *** ukai has quit IRC
[03:16:35] <estade> you can either connect to the "destroy" signal and act accordingly
[03:16:41] *** tonyg-cr has quit IRC
[03:17:04] *** aroben has quit IRC
[03:17:08] <estade> or you can use ownedwidgetgtk to hold onto a ref to the widget
[03:17:08] <atwilson> Ah, so it sounds like I want to hook the destroy signal and clean up my references?
[03:17:59] <estade> it's often easier to use ownedwidgetgtk. gtk_widget_destroy just sends the "destroy" signal, and if you hold a reference then you *might* drop it in response to that signal. When a widget's refcount drops to 0, it is actually destroyed
[03:18:13] <estade> ownedwidgetgtk will not drop its ref when the widget gets the destroy signal
[03:18:28] <estade> you probably don't actually have references to the widget in question
[03:18:39] <atwilson> I don't, you're right.
[03:18:43] <atwilson> I have a GtkStatusIcon.
[03:19:31] *** ukai has joined #chromium
[03:19:42] <atwilson> And according to the docs, GtkStatusIcon only has 3 signals (activate, popup-menu, size-changed), so I'm not at all sure that it gets a destroy.
[03:20:53] <trungl-bot> Tree opened by aa at chromium dot org: tree is open (CookiesNoPermission already reverted)
[03:20:53] *** hrna has joined #chromium
[03:20:57] *** bmizerany has quit IRC
[03:22:23] <atwilson> estade: So, I'm not quite clear about what I should be doing. We're destroying all top-level windows, which GtkStatusIcon clearly doesn't like, but it doesn't seem to give me any hooks to either maintain references to the underlying window, or anything.
[03:23:09] <estade> yea now that I look at that class, it seems probable that the browser list code is problematic
[03:23:13] <aboodman> fwiw, my patch passed tryservers: http://build.chromium.org/buildbot/try-server/builders/mac/builds/37986
[03:23:17] <aboodman> :(
[03:24:03] *** leeight has joined #chromium
[03:24:04] <estade> need a better filter than GTK_WIDGET_VISIBLE
[03:24:19] *** leeight has left #chromium
[03:24:44] <atwilson> estade: There are other slightly kludgey things I can do (like have browser_list clean up the status icons before calling AllBrowsersClosed()), but if there's a better filter criteria I can use, that'd be cleaner?
[03:26:56] *** bmizerany has joined #chromium
[03:27:16] *** eseidel has joined #chromium
[03:27:18] <estade> atwilson: figure out the type of the window that is being destroyed (perhaps GtkTrayIcon) and check for that (GTK_IS_TRAY_ICON)
[03:27:55] *** bmizerany has quit IRC
[03:28:35] <atwilson> estade: To be clear - you aren't saying that GTK_IS_TRAY_ICON already exists (because I don't think it does), right?
[03:28:56] <estade> atwilson: for every gtkwidget type, gtk defines GTK_IS_TYPE_NAME
[03:29:25] <atwilson> OK, I'll play around with that. Thanks for pointing me in the right direction.
[03:29:58] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "unit_tests" on "Mac10.6 Tests (dbg)(1)" from 52585: aa at chromium dot org)
[03:31:41] <aboodman> that unit test is unrelated to my change
[03:38:04] *** gavinp2 has quit IRC
[03:38:11] *** trungl has joined #chromium
[03:38:11] *** ChanServ sets mode: +v trungl
[03:39:05] <aboodman> i am going to mark it flaky
[03:40:31] *** hrna has quit IRC
[03:43:05] *** leeight has joined #chromium
[03:43:10] *** leeight has left #chromium
[03:46:07] <trungl-bot> Tree opened by aa at chromium dot org: Tree is open (browser_tests[CookiesNoPermission] already reverted)
[03:46:25] *** tedoc2000 has quit IRC
[03:47:22] *** pathorn has quit IRC
[03:51:02] *** eseidel has quit IRC
[03:51:09] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Chromium Mac Builder (dbg)" from 52587: ojan at google dot com)
[03:51:29] *** fta_ has joined #chromium
[03:52:53] *** fta has quit IRC
[03:52:58] *** fta_ is now known as fta
[03:53:37] *** abarth has quit IRC
[03:59:25] *** csilv has left #chromium
[04:01:01] *** WarChild has joined #chromium
[04:01:08] <dhollowa> the mac compile failure looks like a distcc problem
[04:01:26] <dhollowa> SVGNames.cpp failed, retried locally and got a different result
[04:01:33] *** WarChild has quit IRC
[04:03:18] <trungl-bot> Tree closed by dhollowa at chromium dot org: Tree is closed (distcc issue on mac, wait to see if it clears)
[04:03:31] *** ebraminio1 has joined #chromium
[04:03:58] *** ebraminio has quit IRC
[04:04:50] *** kermit has quit IRC
[04:06:34] *** kermit has joined #chromium
[04:08:20] <trungl-bot> Tree opened by dhollowa at chromium dot org: Tree is open
[04:11:33] *** jshin has quit IRC
[04:13:24] *** vandemar has joined #chromium
[04:15:04] <tyoshino> aboodman: before you reland r52581, please take a look at CookiesNoPermission failures on mac bots
[04:16:06] <vandemar> chromium 5.0.375.99 (and many previous beta and devel versions) doesn't render &lang; and &rang; in linux (instead they show as typical missing-char rectangles).  Firefox renders them fine on the same system, and both FF and Chrome render them fine on windows.  Is there anything I can do about it?
[04:16:48] <erikchen> are mac try-bots down?
[04:16:51] *** rniwa has quit IRC
[04:18:12] <dhollowa> vandemar: please log a bug at http://crbug.com
[04:20:28] *** stuartmorgan has quit IRC
[04:24:04] *** MikeSmithW3C has quit IRC
[04:24:42] *** MikeSmithW3C has joined #chromium
[04:25:51] *** jrmuizel has joined #chromium
[04:29:31] *** Sarten-X has joined #chromium
[04:30:21] *** mrossetti_ has joined #chromium
[04:33:14] *** mrossetti has quit IRC
[04:33:40] <tyoshino> erikchen: not working?
[04:35:40] *** boomerang has joined #chromium
[04:35:48] *** boomerang has left #chromium
[04:37:51] *** til_ has joined #chromium
[04:37:51] *** til has quit IRC
[04:37:51] *** til_ is now known as til
[04:43:44] *** Viclame1 has left #chromium
[04:44:14] <leiz> erikchen: they seem to be working fine
[04:48:53] *** cbentzel has quit IRC
[04:49:25] *** thakis_ has joined #chromium
[04:49:25] *** ChanServ sets mode: +v thakis_
[04:50:55] *** thakis_ has quit IRC
[04:52:21] *** huckphin has joined #chromium
[04:52:47] *** fta_ has joined #chromium
[04:54:14] *** fta has quit IRC
[04:54:22] *** fta_ is now known as fta
[04:55:17] *** ebraminio1 has quit IRC
[04:58:56] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Chromium Linux Builder x64" from 52600: tfarina at chromium dot org)
[05:00:12] *** tfarina has joined #chromium
[05:00:17] <tfarina> that was mine
[05:00:24] <tfarina> it needs to be reverted
[05:00:35] <tfarina> it was a bad merge :(
[05:00:45] <tyoshino> ok. could you?
[05:01:07] <tfarina> tyoshino: I don't know the drover syntax.
[05:01:07] *** til has quit IRC
[05:01:11] <tyoshino> ok
[05:01:15] <tyoshino> i'll do
[05:01:26] <tfarina> tyoshino: I'd like to do though
[05:01:40] <tfarina> tyoshino: so I can learn if it happens other time.
[05:01:40] <tyoshino> drover --revert 52600
[05:01:48] *** til has joined #chromium
[05:01:49] <tyoshino> in some tmp dir
[05:02:01] <tfarina> tyoshino: I need to be on the trunk branch?
[05:02:24] <tfarina> tyoshino: ?
[05:02:41] <tyoshino> nothing is needed other than depot_tools.
[05:03:04] <tfarina> tyoshino: if I do it in src/ it will works?
[05:03:34] <tfarina> tyoshino: I'm on src and in the trunk branch.
[05:03:35] <tyoshino> not sure. please just make empty directory and cd to it then run the command
[05:03:40] <tyoshino> http://dev.chromium.org/developers/how-tos/drover
[05:03:45] <tyoshino> as this instructs
[05:03:57] <tfarina> ok
[05:06:07] <tfarina> tyoshino: http://codereview.chromium.org/3033004
[05:07:06] <tyoshino> lg
[05:07:14] *** jamesr has quit IRC
[05:07:43] <tyoshino> thanks
[05:08:35] <tfarina> tyoshino: thanks for the help :)
[05:09:32] <tfarina> tyoshino: can I update the tree status?
[05:10:03] <trungl-bot> Tree opened by tyoshino at chromium dot org: Tree is open (52600 reverted -> cycling green)
[05:10:05] <tyoshino> tfarina: changed
[05:10:25] <tfarina> heh, cool thanks again.
[05:11:09] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Chromium XP" from 52600: ananta at chromium dot org (:iyengar), dhollowa at chromium dot org, gman at chromium dot org, jar at chromium dot org, tfarina at chromium dot org, thestig at chromium dot org (:leiz))
[05:11:16] *** yusukes_ has joined #chromium
[05:11:16] *** ChanServ sets mode: +v yusukes_
[05:11:21] <tyoshino> hmm
[05:11:53] <tfarina> tyoshino: what happened?
[05:11:54] <tyoshino> caused due to delayed build run
[05:11:56] <tyoshino> ?
[05:11:58] <tyoshino> looking
[05:12:43] *** MikeSmithW3C^ has joined #chromium
[05:12:51] *** estes has joined #chromium
[05:13:21] <tfarina> tyoshino: looks like it needs more time to catch the revert?
[05:14:17] <oshima> looks like the same issue.
[05:14:23] <oshima> C:\b\slave\chromium-rel-xp\build\src\chrome\browser\tab_contents\tab_contents_delegate.cc(26) : error C2039: 'CanBlur' : is not a member of 'TabContentsDelegate'
[05:14:50] <oshima> should be safe to re-open
[05:14:50] <tyoshino> tfarina: yes. i think. will be fine when 52601 is run on official bots
[05:15:22] <tyoshino> oshima: thanks. agreed
[05:16:07] *** MikeSmithW3C has quit IRC
[05:16:10] <tfarina> oshima: yup.
[05:16:12] <trungl-bot> Tree opened by tyoshino at chromium dot org: Tree is open (52600 reverted -> cycling green)
[05:18:54] *** huckphin has quit IRC
[05:21:39] *** eseidel has joined #chromium
[05:25:21] *** abarth has joined #chromium
[05:25:21] *** ChanServ sets mode: +v abarth
[05:25:50] <dhollowa> the AuthWatcherTest.AuthenticateBadAuthToken might be me
[05:25:53] <dhollowa> looking into it.
[05:26:27] *** mazda has quit IRC
[05:28:23] <oshima> tyoshino: can you review this? http://codereview.chromium.org/3011005
[05:30:34] <tyoshino> lg
[05:31:08] <oshima> thx
[05:32:28] *** mazda has joined #chromium
[05:32:28] *** ChanServ sets mode: +v mazda
[05:35:22] *** rsleevi has joined #chromium
[05:37:18] *** fta_ has joined #chromium
[05:37:53] *** fta has quit IRC
[05:38:07] *** fta_ is now known as fta
[05:40:03] *** estes has quit IRC
[05:41:11] *** estes has joined #chromium
[05:41:11] *** dmaclach has joined #chromium
[05:43:18] *** estes has quit IRC
[05:43:21] *** victorw has left #chromium
[05:43:51] *** tonyg-cr has joined #chromium
[05:43:51] *** ChanServ sets mode: +v tonyg-cr
[05:44:33] <tfarina> tyoshino: uploaded a CL that should fix the bad merge, sent to the trybots though.
[05:45:49] *** eseidel has quit IRC
[05:45:54] *** eseidel has joined #chromium
[05:46:32] *** tyoshino_m has joined #chromium
[05:46:32] *** ChanServ sets mode: +v tyoshino_m
[05:46:54] <tyoshino> tfarina: ok. i'll take a quick lok
[05:51:11] <tyoshino> hmm. looks good but patching failed
[05:52:25] <tyoshino> on try server
[05:53:16] <tfarina> tyoshino: I dunno, maybe it didn't catch yet. I will try it again later.
[05:54:02] *** Gina has quit IRC
[05:55:43] <tyoshino_m> ok
[05:58:20] *** trungl has quit IRC
[05:59:13] *** dmaclach_ has joined #chromium
[05:59:31] *** erikkay has quit IRC
[06:00:30] *** viro101 has quit IRC
[06:02:14] *** oshima has left #chromium
[06:02:14] <dhollowa> sync tests were in fact mine.  i've uploaded patch.  should cycle green on mac.
[06:03:03] *** dmaclach has quit IRC
[06:03:03] *** dmaclach_ is now known as dmaclach
[06:03:21] *** tyoshino_m has quit IRC
[06:03:57] *** trungl has joined #chromium
[06:03:57] *** ChanServ sets mode: +v trungl
[06:04:42] *** jrmuizel has quit IRC
[06:04:44] *** eseidel_ has joined #chromium
[06:08:05] *** eseidel has quit IRC
[06:08:05] *** eseidel_ is now known as eseidel
[06:14:54] <trungl> trungl-bot: fortune
[06:14:55] <trungl-bot> trungl: Deadwood, n.: Anyone in your company who is more senior than you are.
[06:17:11] *** tonyg-cr has quit IRC
[06:18:02] *** vithos has quit IRC
[06:20:58] *** thakis_ has joined #chromium
[06:20:58] *** ChanServ sets mode: +v thakis_
[06:21:32] *** thakis__ has joined #chromium
[06:21:32] *** ChanServ sets mode: +v thakis__
[06:22:07] <trungl> thakis_!
[06:22:17] <trungl> and thakis__!
[06:22:26] <trungl> and let's not forget thakis!
[06:22:29] <thakis__> trungl: !
[06:22:46] <thakis__> ?and no one else
[06:22:54] <thakis__> maybe trungl_afk, but he's not here anyway
[06:23:10] <trungl> what about trungl-bot?
[06:23:13] * thakis__ is happy he's not lonely like that
[06:23:14] <trungl> do bots not count?
[06:25:57] *** thakis_ has quit IRC
[06:26:11] <rsleevi> and then there were two
[06:29:04] <trungl> nooooooooooo!
[06:30:11] *** rohitkc has joined #chromium
[06:31:19] *** victorw has joined #chromium
[06:31:19] *** ChanServ sets mode: +v victorw
[06:32:42] *** fqian has quit IRC
[06:33:08] *** pfeldman_ has joined #chromium
[06:34:08] <trungl-bot> Tree opened by tyoshino at chromium dot org: Tree is open (WebKit valgrind -> vrk?)
[06:34:40] <pfeldman_> ojan: ping
[06:37:32] *** bweinstein has joined #chromium
[06:37:39] *** eseidel has quit IRC
[06:39:10] *** dmaclach has quit IRC
[06:44:16] <trungl-bot> Tree opened by tyoshino at chromium dot org: Tree is open (WebKit valgrind -> vrk?, remoting valgrind -> hclam?)
[06:45:00] *** yusukes_ has quit IRC
[06:45:17] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "base_unittests" on "Chromium Linux x64" from 52608: dhollowa at chromium dot org, viettrungluu at chromium dot org (:trungl))
[06:45:40] *** fta_ has joined #chromium
[06:46:22] <trungl> looking
[06:46:54] *** fta has quit IRC
[06:47:06] *** fta_ is now known as fta
[06:47:18] <tyoshino> looks finished, but the result sais killed
[06:50:58] *** satish_ has joined #chromium
[06:51:32] <satish_> anyone looking at the tree closure?
[06:51:45] <tyoshino> i'm
[06:52:45] <tyoshino> test itself finished cleanly, but post test stuff such as copying dashboard file, didn't finish
[06:53:38] *** eseidel has joined #chromium
[06:53:48] <trungl> hmmm, I think I'll
[06:53:50] <trungl> revert
[06:53:53] <trungl> I'm suspicious
[06:54:11] <tyoshino> which one?
[06:54:52] <trungl> 52608
[06:57:15] <trungl> trungl-bot: settreestatus Tree closed (viettrungluu reverting r52608)
[06:57:16] <trungl-bot> trungl: Set tree status to: "Tree closed (viettrungluu reverting r52608) [set by :trungl]".
[06:57:18] *** az has quit IRC
[06:57:24] <trungl-bot> Tree closed by viettrungluu at chromium dot org (:trungl): Tree closed (viettrungluu reverting r52608) [set by :trungl]
[06:57:39] * trungl must do something about that.
[06:57:57] <tyoshino> ok.
[06:58:14] <trungl> trungl-bot: settreestatus Tree closed (r52608 reverted, waiting for cycle)
[06:58:15] <trungl-bot> trungl: Set tree status to: "Tree closed (r52608 reverted, waiting for cycle) [set by :trungl]".
[06:58:24] <trungl-bot> Tree closed by viettrungluu at chromium dot org (:trungl): Tree closed (r52608 reverted, waiting for cycle) [set by :trungl]
[07:03:44] *** satish_ has quit IRC
[07:09:58] <pfeldman_> ojan: any good news on the roll?
[07:11:23] *** yusukes_ has joined #chromium
[07:11:24] *** ChanServ sets mode: +v yusukes_
[07:13:56] * trungl will get around to making the buildbot console faster someday.
[07:14:29] <abarth> trungl: faster, stronger, better
[07:14:48] <trungl> I'd settle for just one of them
[07:16:22] <abarth> trungl: i'm interested in your bot
[07:16:34] <abarth> trungl: is it in svn somewhere?
[07:17:20] <trungl> abarth: no (not yet?)
[07:18:06] <abarth> i wrote sherrifbot (the webkit bot) in python using irclib
[07:18:13] <abarth> but i'm not entirely happy with irclib
[07:18:29] <abarth> the main problem is i'm using two threads
[07:18:30] <trungl> abarth: trungl-bot is written as a supybot plugin
[07:18:45] <abarth> one for IRC and one for doing actual work
[07:18:55] <abarth> and there doesn't seem to be a way for the main thread to signal the IRC thread
[07:19:15] <abarth> oh, i looked at that
[07:19:29] <abarth> it seemed like it did a lot of the command parsing and stuff for you
[07:19:33] <trungl> abarth: (the trungl-bot code is surely some of the worst python code ever, but ... I didn't actually know any python when I started)
[07:19:33] <abarth> which seemed cool
[07:19:53] <abarth> is it single threaded?
[07:20:10] <trungl> abarth: it can operate in a multithreaded mode
[07:20:12] <abarth> some of sherrifbots commands take a long time
[07:20:18] <trungl> (which, in the case of trungl-bot, it does)
[07:20:34] <trungl> trungl-bot: halp
[07:20:36] <trungl-bot> trungl: Valid commands (use "halp <command>" for more): bug, commit, commits, cookie, fortune, halp, help, lkgr, offices, settreestatus, time, treestatus, uptime, webkitbug, webkitcommit, webkitcommits, whois. (If you whisper to me, I'll whisper back.)
[07:20:37] <abarth> ok, i'll look into it more
[07:20:45] <trungl> trungl-bot: commits
[07:20:47] <trungl> trungl-bot halp
[07:21:07] <trungl> this *should* work
[07:21:13] *** pfeldman has joined #chromium
[07:21:13] <trungl-bot> trungl: r52613 viettrungluu at chromium dot org: Revert 52608 - Add and alternative GetAppOutput() to process_util that takes a timeout. // r52612 tkent at chromium dot org: Remove inspector_resources target and use the upstreaming // r52611 tkent at chromium dot org: Implement spin buttons for <input type=number> // r52610 shess at chromium dot org: Land new resource for content-setting images. // r52609 shess at chromium dot org: (1 more message)
[07:21:14] <trungl-bot> trungl: Valid commands (use "halp <command>" for more): bug, commit, commits, cookie, fortune, halp, help, lkgr, offices, settreestatus, time, treestatus, uptime, webkitbug, webkitcommit, webkitcommits, whois. (If you whisper to me, I'll whisper back.)
[07:21:26] <trungl> argh
[07:21:26] *** loislo has joined #chromium
[07:21:32] <abarth> what does commits do?
[07:21:34] <trungl> did not work as expected
[07:21:46] <trungl> abarth: lists recent commits
[07:21:53] <trungl> (it takes a while)
[07:22:09] <trungl> it should have responded to "halp" before commits
[07:22:23] <abarth> oh, the commands each get their own thread
[07:22:25] <abarth> ?
[07:22:32] <abarth> sheriffbot queues them up
[07:22:36] <abarth> and services them sequentially
[07:22:50] <trungl> trungl-bot: commits
[07:22:53] <trungl> trungl-bot: ping
[07:22:54] <trungl-bot> trungl: r52613 viettrungluu at chromium dot org: Revert 52608 - Add and alternative GetAppOutput() to process_util that takes a timeout. // r52612 tkent at chromium dot org: Remove inspector_resources target and use the upstreaming // r52611 tkent at chromium dot org: Implement spin buttons for <input type=number> // r52610 shess at chromium dot org: Land new resource for content-setting images. // r52609 shess at chromium dot org: (1 more message)
[07:22:55] <trungl-bot> pong
[07:23:02] <pfeldman> tony^work: ping
[07:23:09] <trungl> trungl-bot: webkitcommits
[07:23:10] <trungl> trungl-bot: ping
[07:23:10] <trungl-bot> pong
[07:23:11] <trungl-bot> trungl: r63518 tkent at chromium dot org / r63517 abarth at webkit dot org / r63516 victorw at chromium dot org / r63515 ggaren at apple dot com / r63514 tkent at chromium dot org / r63513 abarth at webkit dot org / r63512 abarth at webkit dot org / r63511 morrita at google dot com / r63510 yuzo at google dot com / r63509 arv at chromium dot org
[07:23:16] <trungl> there we go
[07:23:16] *** tonyg-cr has joined #chromium
[07:23:16] *** ChanServ sets mode: +v tonyg-cr
[07:23:17] <abarth> trungl: i see
[07:23:33] <abarth> trungl: i should say that some commands sheriffbot can service from the irc thread
[07:23:39] <abarth> without round tripping ot the main thread
[07:23:45] <abarth> like the eliza functionality
[07:24:12] *** pfeldman_ has quit IRC
[07:24:49] <trungl> I wish I had more fine-grained control over threading
[07:24:53] <trungl> (in supybot)
[07:25:02] <trungl> maybe it's possible
[07:25:20] <trungl> hmmm
[07:25:30] * trungl should re-write trungl-bot.
[07:26:00] <abarth> the key thing for sheriffbot is he caches a ton of data in memory
[07:26:13] <abarth> because talking to build.webkit.org is slow
[07:26:21] <abarth> he also has an webkit working copy
[07:26:24] *** hamaji has joined #chromium
[07:26:27] <abarth> so he can make rollouts patches
[07:26:31] <abarth> which is also really slow
[07:27:00] <trungl> wow
[07:27:09] <abarth> trungl: the rollout feature is really popular
[07:27:13] <trungl> right
[07:27:53] <abarth> the code is all in http://trac.webkit.org/browser/trunk/WebKitTools/Scripts/webkitpy
[07:27:58] <abarth> but there's a lot of code in there
[07:28:16] <abarth> a bunch of it is in http://trac.webkit.org/browser/trunk/WebKitTools/Scripts/webkitpy/tool/bot
[07:28:30] <trungl> abarth: is it possible to play a drover-like trick to avoid doing all the patches on a single checkout?
[07:28:58] <abarth> i guess, but it might as well have a full checkout
[07:29:01] <abarth> its running on EC2
[07:29:06] <abarth> there's plenty of disk
[07:29:09] <abarth> that machine has like six checkouts
[07:29:13] <abarth> for a bunch of different bots
[07:29:36] <abarth> memory is tighter than disk
[07:29:43] <abarth> but the stuff in cache just gets paged out
[07:29:53] <abarth> and then he's basically read it from disk instead of from the network
[07:29:57] <abarth> so that's a big win
[07:30:03] <trungl> abarth: it's just that if the full checkout can be avoided, then one wouldn't have to worry about making the rollout patches serially
[07:30:23] <abarth> that hasn't been a problem
[07:30:38] <abarth> they need to be landed serially anyway
[07:30:44] <trungl> right
[07:30:52] <abarth> you can land them by hand, but usually, they go though the commit-queue bot
[07:31:05] <abarth> which takes much longer than preparing the rollouts
[07:31:11] <abarth> since it builds them
[07:32:32] <abarth> i'll look more into supybot
[07:32:44] <abarth> it sounds like it does a lot more than irclib does
[07:33:29] <trungl> it does, probably including a lot of stuff you don't want it to do
[07:33:51] <trungl> (though much of that can be removed/disabled)
[07:37:28] <trungl> tyoshino: I think we should be good to re-open, except for maybe the purple
[07:39:08] <tyoshino> trungl: i see. ok. base_unittests is fine.
[07:40:07] <tyoshino> webkit is flaky...
[07:40:26] <trungl> yeah, it seems to be sporadically purple
[07:40:32] <trungl> tyoshino: shall I re-open?
[07:40:41] <tyoshino> please
[07:40:42] <tyoshino> thanks
[07:40:54] <trungl> trungl-bot: settreestatus Tree is open
[07:40:55] <trungl-bot> trungl: Set tree status to: "Tree is open [set by :trungl]".
[07:41:48] <trungl-bot> Tree opened by viettrungluu at chromium dot org (:trungl): Tree is open [set by :trungl]
[07:42:08] *** michaeln has quit IRC
[07:48:26] *** tonyg-cr has quit IRC
[07:51:47] *** eseidel has quit IRC
[07:53:55] *** satish_ has joined #chromium
[07:55:32] *** fta_ has joined #chromium
[07:56:47] *** fta has quit IRC
[07:56:52] *** fta_ is now known as fta
[08:03:38] *** stevenjb has joined #chromium
[08:04:32] *** MichealH has joined #chromium
[08:04:45] *** roc has quit IRC
[08:12:50] *** Omen_20 has quit IRC
[08:15:28] *** yusukes_ has quit IRC
[08:15:41] <satish_> fishd: Hi Darin, yt?
[08:26:52] *** Omen_20 has joined #chromium
[08:28:23] *** satish_ has quit IRC
[08:34:52] *** abarth is now known as abarth|zZz
[08:40:41] *** tyoshino_m has joined #chromium
[08:40:41] *** ChanServ sets mode: +v tyoshino_m
[08:45:45] *** Erkan_Yilmaz has quit IRC
[08:46:24] *** eseidel has joined #chromium
[08:47:47] *** pdelgallego has joined #chromium
[08:48:26] *** MichealH has quit IRC
[08:49:04] *** pfeldman_ has joined #chromium
[08:49:07] *** viro101 has joined #chromium
[08:49:12] *** MichealH has joined #chromium
[08:52:32] <jochen__> moin
[08:52:52] *** pfeldman has quit IRC
[08:52:52] *** pfeldman_ is now known as pfeldman
[08:53:35] *** stevenjb has left #chromium
[08:53:47] *** janm has joined #chromium
[08:54:35] *** markusheintz has joined #chromium
[08:55:12] *** MichealH has quit IRC
[08:55:45] *** eseidel has quit IRC
[08:56:15] *** MichealH has joined #chromium
[08:56:20] *** Zaba has quit IRC
[08:56:34] *** MichealH has quit IRC
[08:58:02] *** eseidel has joined #chromium
[08:58:52] *** Zaba has joined #chromium
[09:07:17] *** jwalden has quit IRC
[09:08:48] *** fta_ has joined #chromium
[09:08:58] *** eseidel has quit IRC
[09:09:37] *** fta has quit IRC
[09:09:51] *** fta_ is now known as fta
[09:09:53] *** saurik has quit IRC
[09:12:09] *** rsleevi has left #chromium
[09:13:29] *** tyoshino_m has quit IRC
[09:13:51] *** dave_levin has quit IRC
[09:14:58] *** pfeldman has quit IRC
[09:16:47] *** satish_ has joined #chromium
[09:18:25] *** janm has quit IRC
[09:18:56] *** dave_levin has joined #chromium
[09:25:17] *** dave_levin has quit IRC
[09:26:01] *** trungl has quit IRC
[09:26:46] *** wash has quit IRC
[09:28:42] *** monreal has joined #chromium
[09:31:02] *** victorw has left #chromium
[09:31:13] *** bweinstein has quit IRC
[09:32:25] *** star-affinity has joined #chromium
[09:33:19] *** MikeSmithW3C^ is now known as MikeSmithW3C
[09:34:27] *** GeekShadow has joined #chromium
[09:42:24] *** saurik has joined #chromium
[09:52:17] *** az has joined #chromium
[09:52:38] *** viro101_ has joined #chromium
[09:52:39] *** viro101 has quit IRC
[09:52:40] *** viro101_ is now known as viro101
[09:52:58] *** mnissler_ has joined #chromium
[09:52:59] *** ChanServ sets mode: +v mnissler_
[09:53:03] *** mnissler_ has quit IRC
[09:53:06] *** mnissler has quit IRC
[09:53:27] *** mnissler has joined #chromium
[09:53:27] *** ChanServ sets mode: +v mnissler
[09:55:29] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Linux Perf (2)" from 52631: mnissler at chromium dot org, nirnimesh at chromium dot org, satorux at chromium dot org, tyoshino at chromium dot org)
[09:55:44] <mnissler> sorry, committing fix.
[09:55:55] *** bauerb has joined #chromium
[09:55:55] *** ChanServ sets mode: +v bauerb
[09:56:38] <tyoshino> ok
[09:58:29] *** [Nef] has joined #chromium
[10:05:20] <tyoshino> thanks for the fix
[10:06:41] <trungl-bot> Tree opened by mnissler at chromium dot org: Tree is open
[10:08:24] *** monreal_ has joined #chromium
[10:11:32] *** fta_ has joined #chromium
[10:11:56] *** monreal has quit IRC
[10:12:52] *** fta has quit IRC
[10:13:05] *** fta_ is now known as fta
[10:13:35] *** pfeldman has joined #chromium
[10:17:37] *** maikmerten has joined #chromium
[10:25:40] *** hrna has joined #chromium
[10:30:40] *** chrisccoulson has joined #chromium
[10:30:51] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Chromium Builder (dbg)" from 52633: mnissler at chromium dot org, tyoshino at chromium dot org)
[10:32:02] *** chrisccoulson has quit IRC
[10:32:50] *** chrisccoulson has joined #chromium
[10:32:56] <mnissler> PDB error again, reopening and watching the tree...
[10:33:43] *** Nola has joined #chromium
[10:33:53] <trungl-bot> Tree opened by mnissler at chromium dot org: Tree is open (mnissler watching Chromium Builder (dbg))
[10:34:06] <Nola> hi -- why does chrome look so "weird" ?
[10:34:31] <Nola> it does not look like Windows
[10:34:36] <Nola> the theme doesn't match
[10:34:45] <Nola> and there's no themes for Chrome that are anywhere near my windows theme
[10:34:52] <Nola> Why cant it just do a regular old native rendering?
[10:34:56] <Nola> for the window
[10:39:33] *** vithos has joined #chromium
[10:40:26] *** Peter` has quit IRC
[10:43:16] *** ROBOd has joined #chromium
[10:44:41] <tyoshino> clobber necessary?
[10:44:48] <mnissler> Chromium Builder (dbg) failed again with PDB error, but on a different binary
[10:45:04] <mnissler> tyoshino: You have my vote :)
[10:45:37] *** abarth|zZz has quit IRC
[10:46:14] *** mnissler_ has joined #chromium
[10:47:25] <mnissler_> tyoshino: Did you clobber?
[10:48:21] <tyoshino> not yet
[10:48:29] *** fta has quit IRC
[10:48:55] <tyoshino> will do -
[10:49:01] <mnissler_> thx
[10:49:41] *** fta has joined #chromium
[10:51:03] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "update" on "Chromium Builder (dbg)")
[10:51:22] <akalin2> that's not me, is it
[10:52:08] <mnissler_> tyoshino was going to clobber
[10:52:45] <akalin2> okay
[10:53:11] <tyoshino> yes. sorry.
[10:53:13] <tyoshino> please ignore
[10:53:19] <pfeldman> tyoshino: do you want to revert r52624?
[10:55:04] <tyoshino> pfeldman: sorry. for what?
[10:55:05] <trungl-bot> Tree opened by tyoshino at chromium dot org: Tree is open (mnissler tyoshino watching Chromium Builder (dbg))
[10:55:40] <tyoshino> maybe i missed something
[10:55:41] <pfeldman> tyoshino: XP browser tests
[10:56:30] <pfeldman> (it enabled failing test claiming it is working again?)
[10:57:27] <tyoshino> ok. thanks for the info
[10:57:41] <tyoshino> that should be reverted. agreed
[10:58:59] *** SkyLined has joined #chromium
[10:58:59] <jibot> SkyLined is a hacker working on Google Chrome security. & a hacker working on Google Chrome security (http://skypher.com) & messing up jibot & getting confused & a hacker working for Google on browser security and Chromium in particular (http://skypher.com)
[11:00:06] *** mnissler_ has quit IRC
[11:00:23] <pfeldman> tyoshino: i wonder if startup tests is a real failure. mnissler, should r52630 be under suspect?
[11:01:57] <mnissler> pfeldman, tyoshino: 52630 only changes logic in Linux options dialog.
[11:02:20] <SkyLined> ?forgetme
[11:02:21] <jibot> I've forgotten all about SkyLined
[11:02:43] *** SkyLined has left #chromium
[11:02:51] *** SkyLined has joined #chromium
[11:04:21] <mnissler> pfeldman, tyoshino : startup tests log indicate automation test failures... might by 52628, no?
[11:04:41] <pfeldman> mnissler: oh, i did not look at the logs
[11:08:29] <tyoshino> pfeldman, mnissler: thanks, it looks like caused by that change.
[11:09:27] <pfeldman> mnissler: sorry, you are my usual suspect based on the commit rate.
[11:11:22] <tyoshino> nirnimesh: yt?
[11:12:50] *** MikeSmithW3C^ has joined #chromium
[11:16:06] *** MikeSmithW3C has quit IRC
[11:17:37] <mnissler> pfeldman: no worries :)
[11:17:58] *** apavlov has joined #chromium
[11:18:29] <tyoshino> mnissler: Chromium Builder (dbg) compiled successfully
[11:18:51] <mnissler> tyoshino: \o/
[11:20:26] <trungl-bot> Tree opened by tyoshino at chromium dot org: Tree is open
[11:24:32] *** yusukes_ has joined #chromium
[11:24:32] *** ChanServ sets mode: +v yusukes_
[11:28:03] *** fta_ has joined #chromium
[11:28:47] *** fta has quit IRC
[11:28:49] *** fta_ is now known as fta
[11:33:58] *** fta_ has joined #chromium
[11:34:44] *** fta has quit IRC
[11:34:49] *** fta_ is now known as fta
[11:39:49] *** satish_ has quit IRC
[12:01:11] *** hbono has quit IRC
[12:11:06] *** davirtavares has joined #chromium
[12:13:38] *** fta_ has joined #chromium
[12:15:13] *** fta has quit IRC
[12:18:14] *** fta__ has joined #chromium
[12:18:32] *** fta__ is now known as fta
[12:19:22] *** fta_ has quit IRC
[12:23:43] *** tkent has quit IRC
[12:34:06] *** brot has left #chromium
[12:36:43] *** Peter` has joined #chromium
[12:37:50] *** rohitkc has quit IRC
[12:38:27] *** apavlov has quit IRC
[12:46:03] *** fta_ has joined #chromium
[12:46:13] *** apavlov has joined #chromium
[12:47:41] *** fta has quit IRC
[12:47:48] *** fta_ is now known as fta
[12:49:48] *** podivilov has joined #chromium
[12:52:29] *** podivilov has quit IRC
[12:56:24] *** General1337 has joined #chromium
[12:58:57] *** fta_ has joined #chromium
[12:59:40] *** General13372 has quit IRC
[12:59:52] *** fta has quit IRC
[13:00:03] *** fta_ is now known as fta
[13:05:53] *** masterov has quit IRC
[13:11:30] *** FullFlannelJacke has quit IRC
[13:13:25] *** fta has quit IRC
[13:14:56] *** fta has joined #chromium
[13:18:21] *** ager has joined #chromium
[13:20:13] *** ager has left #chromium
[13:20:17] *** ager has joined #chromium
[13:24:48] *** FullFlannelJacke has joined #chromium
[13:26:35] *** fta_ has joined #chromium
[13:27:24] <nirnimesh> tyoshino: thanks for reverting my CL which broke startup_tests dbg
[13:27:40] *** fta has quit IRC
[13:27:47] <nirnimesh> I've fixed it now, and will reland.
[13:27:48] *** fta_ is now known as fta
[13:29:52] <tyoshino> ok. thanks
[13:31:26] *** Sarten-X has quit IRC
[13:36:30] *** andrix has joined #chromium
[13:42:42] *** fta_ has joined #chromium
[13:43:52] *** fta has quit IRC
[13:44:02] *** fta_ is now known as fta
[13:47:30] *** masterov has joined #chromium
[13:50:08] *** cbentzel has joined #chromium
[13:54:56] *** yutak has quit IRC
[13:55:05] <trungl-bot> Tree opened by tyoshino at chromium dot org: Tree is open (WebSocket test -> yutak)
[13:55:15] *** yutak has joined #chromium
[13:55:15] *** ChanServ sets mode: +v yutak
[14:03:17] <tyoshino> anyone who has priviledge to reboot Chromium Win bots here?
[14:07:14] <trungl-bot> Tree opened by yutak at chromium dot org: Tree is open (Vista Tests -> Seems like a bot issue, needs reboot?)
[14:12:14] *** Godhimself has joined #chromium
[14:12:50] *** ager has quit IRC
[14:14:16] *** cbentzel has quit IRC
[14:16:15] *** mazda has quit IRC
[14:17:53] *** zaspire has joined #chromium
[14:19:50] <yutak> Seems like we need to kill an old process in Vista Tests bot
[14:27:47] *** fta_ has joined #chromium
[14:28:36] *** fta has quit IRC
[14:28:47] *** fta_ is now known as fta
[14:32:05] *** Viclame has joined #chromium
[14:32:41] *** apavlov has quit IRC
[14:35:05] *** yutak has quit IRC
[14:35:34] *** yutak has joined #chromium
[14:35:34] *** ChanServ sets mode: +v yutak
[14:40:56] *** GeekShadow has quit IRC
[14:43:21] *** GeekShadow has joined #chromium
[14:43:38] *** steinex has left #chromium
[14:48:04] *** jrmuizel has joined #chromium
[14:53:31] *** fta_ has joined #chromium
[14:54:00] *** ager has joined #chromium
[14:54:26] *** AaronMT has joined #chromium
[14:54:56] *** fta has quit IRC
[14:55:02] *** fta_ is now known as fta
[14:56:53] *** yutak has quit IRC
[14:58:53] *** yann2 has joined #chromium
[14:59:51] <yann2> hello guys, I need some help to confirm a bug that I think is supposed to be fixed - it is regarding the "ended" signal for videos
[15:00:11] *** cbentzel has joined #chromium
[15:00:44] <yann2> it seems to rarely be sent, or not at all.... I am writing a small music player: http://waste.mandragor.org/audioplayer/audio.html  , when a song is finished it should go to the next, never does in chromium or rarely, works great in firefox
[15:01:21] *** yutak has joined #chromium
[15:01:21] *** ChanServ sets mode: +v yutak
[15:01:41] <yann2> I fund this : http://code.google.com/p/chromium/issues/detail?id=16768  , but it says it is fixed
[15:01:51] <trungl-bot> Tree closed by jochen at chromium dot org (:jochen__): Tree is closed (jochen landing grd change, Vista Tests -> Seems like a bot issue, needs reboot?)
[15:01:59] <jochen__> landing a grd change
[15:02:00] <yann2> using chrome 5.0.375.99 on Ubuntu 10.4
[15:02:19] <yann2> anyone able to reproduce?
[15:02:39] *** meE_ has joined #chromium
[15:03:40] *** apavlov has joined #chromium
[15:04:24] *** cbentzel has quit IRC
[15:05:11] <tyoshino> jochen_: ok.
[15:05:23] *** romainhuet has joined #chromium
[15:05:31] <shoe> yann2: which OS?
[15:05:44] <yann2> ubuntu 10.4
[15:06:08] <shoe> yann2: http://code.google.com/p/chromium/issues/detail?id=30452
[15:06:16] <awong_> does vista tests still need a reboot?
[15:07:34] <yann2> thanks shoe
[15:07:58] <shoe> yann2: actually, http://code.google.com/p/chromium/issues/detail?id=45074
[15:08:48] <awong_> toyoshino: ping...I can reboot vista tests if it still needs it...
[15:08:53] <tyoshino> oh
[15:09:01] <tyoshino> please.
[15:09:05] <awong_> okay
[15:09:18] <tyoshino> http://build.chromium.org/buildbot/waterfall/builders/Vista%20Tests
[15:09:20] <tyoshino> this bot.
[15:09:39] <yann2> shoe, it seems to say it has been fixed, any idea where I could get the nightly build?
[15:11:39] <shoe> I don't know.  are there nightlies for linux?
[15:12:28] *** Sarten-X has joined #chromium
[15:12:29] <yann2> I think I found http://build.chromium.org/buildbot/snapshots/chromium-rel-linux/ , thanks, I ll give it a try
[15:12:52] <awong_> tyoshino: rebooted.  BTW, if you have the buildbot password, there's a page with instructions internally on how to reboot bots, etc.
[15:14:09] *** Guest83504 has joined #chromium
[15:14:20] <yutak> Actually there are a few people who know the buildbot password here in Tokyo, but all of them have gone home already
[15:14:23] <tyoshino> awong_: thanks. i had done by myself in past times, but now i don't have pass
[15:15:01] <tyoshino> yep
[15:15:54] <awong_> ah
[15:16:05] <yann2> shoe, thanks a lot, it works great on the latest build :)
[15:16:31] *** Guest83504 is now known as BrendenB
[15:18:02] <trungl-bot> Tree closed by yutak at chromium dot org: Tree is closed (jochen landing grd change, Vista Tests -> Rebooted, should cycle green)
[15:18:21] <awong_> Am I right to assume that we'll probably keep the tree closed for another 30 mins or so for the grd change (trying to decide if I should go back to sleep).
[15:18:48] *** yann2 has left #chromium
[15:22:26] <jochen__> if you want to check in, check in
[15:22:28] *** asdsad has joined #chromium
[15:22:51] *** jrmuizel has quit IRC
[15:23:32] *** knighthawk has quit IRC
[15:24:05] *** hebz0rl has joined #chromium
[15:25:17] <awong_> tyoshino: the two main win builders have picked up the grd change. Is it okay if I checking my two changes?
[15:26:42] *** yutak has quit IRC
[15:26:51] *** yutak has joined #chromium
[15:26:51] *** ChanServ sets mode: +v yutak
[15:28:08] <tyoshino> awong_: ok
[15:28:19] <awong_> cool. thanks
[15:32:45] *** awolfson has quit IRC
[15:36:11] <trungl-bot> Tree opened by jochen at chromium dot org (:jochen__): Tree is open (Vista Tests -> Seems like a bot issue, needs reboot?)
[15:38:14] *** rohitrao has joined #chromium
[15:38:14] *** ChanServ sets mode: +v rohitrao
[15:38:17] *** fta_ has joined #chromium
[15:39:25] *** shoe has quit IRC
[15:40:28] *** wjmaclean_ has joined #chromium
[15:40:28] *** ChanServ sets mode: +v wjmaclean_
[15:40:31] *** fta has quit IRC
[15:40:41] *** js2 has quit IRC
[15:40:46] *** fta_ is now known as fta
[15:41:46] *** cbentzel has joined #chromium
[15:43:37] *** tbassetto has joined #chromium
[15:43:41] <awong_> a couple of the buildbots seem wedged.  http://build.chromium.org/buildbot/waterfall/builders/Linux%20Tests%20(dbg)(1)/builds/14030 has been running for 2 hours
[15:43:44] <awong_> I'm going to kick it.
[15:44:29] <pamg> Go awong_  go.
[15:45:45] <awong_> hah
[15:45:45] <tyoshino> k
[15:46:22] <tyoshino> thanks for rebooting the vista bot. yutak's issue has been resolved
[15:46:48] <awong_> awesome :)
[15:47:18] <trungl-bot> Tree opened by yutak at chromium dot org: Tree is open
[15:47:18] *** asdsad has quit IRC
[15:47:20] <yutak> :)
[15:48:32] <jochen__> the Chromium Builder (dbg) is the pdb link error
[15:53:05] *** asdsad has joined #chromium
[15:53:58] <tyoshino> it happened today and resolved by rebooting then.
[15:54:22] *** pinkerton has joined #chromium
[15:54:22] *** ChanServ sets mode: +v pinkerton
[15:55:20] *** lgombos has quit IRC
[15:55:58] <rohitrao> shess: someone in the DC office has been complaining about http://code.google.com/p/chromium/issues/detail?id=6626 .  Do you know anything about sqlite and fragmentation?
[15:56:05] *** lgombos has joined #chromium
[15:56:12] <rohitrao> I don't really understand the complaints on mac, cause I thought HFS took care of this somehow
[15:56:45] <awong_> tyoshino: does it need a reboot or a clobber?
[15:56:46] *** cbentzel_ has joined #chromium
[15:57:18] <awong_> tyoshino: I'll just reboot...
[15:57:34] <tyoshino> ah sorry
[15:57:42] <tyoshino> i searched log and remembered
[15:57:50] <tyoshino> just clobbering worked.
[15:58:26] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "base_unittests" on "Modules Mac10.6 (dbg)" from 52661: rsesek at chromium dot org, tyoshino at chromium dot org)
[15:59:27] *** ager has left #chromium
[15:59:50] <tyoshino> looking
[16:00:11] *** rsesek has joined #chromium
[16:00:11] *** ChanServ sets mode: +v rsesek
[16:00:33] <tyoshino> hmm
[16:00:37] <rsesek> I think we can just reopen
[16:00:51] <rsesek> both tests are FAILS_
[16:01:19] *** cbentzel_ has quit IRC
[16:01:27] <trungl-bot> Tree opened by rsesek at chromium dot org: Tree is open
[16:02:21] *** cbentzel has quit IRC
[16:03:38] <tyoshino> there're some backtrace of base_unittests itself
[16:03:48] <tyoshino> but, ok. open and wait and see
[16:03:49] *** jrmuizel has joined #chromium
[16:04:21] <rsesek> neither of our changes could have caused that, so next cycle will be the decider
[16:04:43] <tyoshino> all right
[16:05:48] *** dmaclach has joined #chromium
[16:06:38] *** cbentzel has joined #chromium
[16:07:03] *** dmaclach_ has joined #chromium
[16:07:03] *** dmaclach has quit IRC
[16:07:04] *** dmaclach_ is now known as dmaclach
[16:07:15] <awong_> tyoshino: FYI, Chromium Builder (dbg) passed on second build w/o the clobber.  Also, Webkit Linux (dbg) failures I think are flake.
[16:07:25] <awong_> my change doesn't touch anything near those areas...
[16:07:38] <tyoshino> awong_: ok. thanks
[16:07:42] *** asdsad has quit IRC
[16:10:05] <rohitrao> trungl-bot: hello
[16:10:05] <trungl-bot> rohitrao: Error: "hello" is not a valid command.
[16:10:08] <rohitrao> sad
[16:10:30] <rohitrao> trungl_afk: your bot won't exchange pleasantries with me
[16:14:44] *** awidegreen has joined #chromium
[16:14:54] *** yuiop has joined #chromium
[16:16:19] *** thakis__ has quit IRC
[16:17:08] <rsesek> tyoshino: base_unittests seems fine this cycle
[16:17:27] *** SkyLined2 has joined #chromium
[16:17:42] <tyoshino> rsesek: ok
[16:18:08] *** dmaclach has quit IRC
[16:20:03] <rsesek> are we allowed to use dot notation in Chromium ObjC?
[16:20:40] *** SkyLined has quit IRC
[16:21:06] *** romainhuet has quit IRC
[16:21:17] *** yuiop has quit IRC
[16:21:55] *** Godhimself has quit IRC
[16:22:28] <pinkerton> rsesek: no
[16:22:36] <rsesek> sadface okay
[16:22:49] <pinkerton> mark and i had that discussion a few weeks ago
[16:23:18] <rsesek> what came of it?
[16:23:31] <pinkerton> "no"
[16:23:34] <pinkerton> :D
[16:23:40] <rsesek> ha
[16:24:02] <pinkerton> the style guide allowing it still hasn't been written/finalized/pushed yet, so the rules still say "no"
[16:24:03] *** phajdan-jr has joined #chromium
[16:24:03] *** ChanServ sets mode: +v phajdan-jr
[16:24:24] <rsesek> yes I noticed the discrepancy between the ext/int version of the guide
[16:24:40] <pinkerton> ah, so we just haven't pushed it. hrm.
[16:24:51] <pinkerton> we really should do that then, tell mark to get off his lazy ass.
[16:25:14] <rsesek> I'll bug him after I finish my change to the guide :p
[16:25:19] <pinkerton> heh
[16:25:33] *** pathorn has joined #chromium
[16:25:33] *** ChanServ sets mode: +v pathorn
[16:26:24] *** js2 has joined #chromium
[16:27:39] *** romainhuet has joined #chromium
[16:28:27] *** til_ has joined #chromium
[16:28:27] *** til has quit IRC
[16:28:28] *** til_ is now known as til
[16:29:38] *** erikkay has joined #chromium
[16:29:38] *** ChanServ sets mode: +v erikkay
[16:32:01] *** SkyLined2 has quit IRC
[16:32:12] *** tedoc2000 has joined #chromium
[16:32:34] <phajdan-jr> yay! landed a change that lowers UI tests timeouts dramatically. Hopefully it'll be good for the tree. Watching.
[16:33:20] <awong_> phajdan-jr: woohoo!
[16:33:30] <awong_> FYI, Webkit windows builder is borked.
[16:33:42] <awong_> not sure how to fix.  considering a reboot...anyone got an idea?
[16:33:51] <phajdan-jr> kick the bot
[16:33:54] <awong_> http://buildbot.jail.google.com/buildbot/chromium/builders/Webkit/
[16:33:59] <awong_> kick == reboot?
[16:34:27] <jochen__> phajdan-jr: \o/
[16:34:32] <phajdan-jr> rather a joke, but if it's windows, reboot won't hurt I think
[16:34:41] *** trungl has joined #chromium
[16:34:41] *** ChanServ sets mode: +v trungl
[16:34:46] <jochen__> pathorn: could you help me with a plugin test issue here
[16:34:49] <awong_> reboot it is.
[16:34:50] <jochen__> ehrm
[16:34:52] <jochen__> phajdan-jr:
[16:34:58] <phajdan-jr> jochen__: hi!
[16:35:06] <trungl> 'morning, Chromium
[16:35:07] <trungl-bot> trungl: Good morning!
[16:35:17] <phajdan-jr> jochen__: by the way, how is your work on the grd brokenness? it sounds great that there is a fix for that.
[16:35:38] <jochen__> postponed till next week
[16:35:42] <jochen__> http://codereview.chromium.org/3020005
[16:35:56] <jochen__> it appears that the selfdeleteplugin tests are executed in the wrong fixture
[16:36:05] <jochen__> where the plugin that they actual test isn't loaded
[16:36:08] *** bent-mozilla has joined #chromium
[16:36:37] <jochen__> i think the reason they work most of the time on windows is that once a complete ui_test run is done, all required plugins are loaded anyway
[16:37:18] <jochen__> but somehow i can't execute the tests locally on linux
[16:37:45] <awong_> jochen__: what happens when you exectue locally?
[16:38:00] <jochen__> the linux version of that plugin is a nacl plugin
[16:38:08] <jochen__> and the html files just claim not to find the plugin
[16:38:18] <jochen__> although the nacl plugin itself is listed in about:plugins
[16:38:27] <awong_> fun.
[16:38:37] *** awolfson has joined #chromium
[16:38:46] *** tyoshino has quit IRC
[16:38:52] <jochen__> even more fun, there are two libnpapi_layout_test_plugins in the source tree
[16:39:10] *** eseidel has joined #chromium
[16:39:28] <phajdan-jr> jochen__: are they at least the same, or forked?
[16:39:40] <jochen__> forked
[16:39:48] <awong_> :(
[16:39:48] <jochen__> the one is in webkit... it's not a nacl plugin obviously
[16:39:56] <jochen__> the other is in webkit/tools/npapi_layout..
[16:40:10] <awong_> if you try to load both, one is unlikely to load....
[16:40:52] <jochen__> the webkit one is not compiled on linux
[16:41:01] <jochen__> make gives a warning about only compiling one
[16:41:12] <jochen__> like, on a fresh checkout
[16:41:32] <awong_> ah. that's makes sense, and is probably actually a good thing.
[16:41:45] *** mrossetti has joined #chromium
[16:42:09] <jochen__> so anyway, for the tests failing after my build
[16:42:31] <jochen__> the TestDoubleDisplay should go away after soon it doesn't occur on the trybots anyway
[16:42:50] *** mrossetti has quit IRC
[16:42:58] <jochen__> and the SelfDeletePlugin one should actually always fail, it's an error that it every works
[16:43:13] <mmoss> anybody doing anything with all the red memory bots?
[16:43:29] <jochen__> i'm watching them
[16:43:30] <mmoss> been like that a while
[16:43:41] <mmoss> are they changing?
[16:43:46] <jochen__> yes
[16:43:54] <mmoss> fair enough
[16:44:37] <jochen__> i'd suggest to fix the selfdeleteplugin tests by actually moving them to the right fixture
[16:44:49] <jochen__> and disable them on linux until i figure out how to load the plugin
[16:45:03] <jochen__> because right now, they just work by pure chance
[16:45:09] <jochen__> and not by testing the plugin
[16:45:15] *** mrossetti has joined #chromium
[16:45:30] <mmoss> moving the to the right fixture?
[16:45:50] <jochen__> it's a layout plugin test, which is executed in the npapi plugin test fixture
[16:45:57] *** tyoshino has joined #chromium
[16:46:20] *** peper has quit IRC
[16:46:26] *** thakis_ has joined #chromium
[16:46:26] *** ChanServ sets mode: +v thakis_
[16:47:26] <jochen__> anyway, if we wait long enough, it'll just start to work again :)
[16:48:24] <mmoss> but still need to be disabled on linux? or you working on the loading issue?
[16:48:42] <jochen__> i'm working on it
[16:49:11] <mmoss> ok, I'll leave it alone unless things get messier
[16:49:58] *** stuartmorgan has joined #chromium
[16:49:58] *** ChanServ sets mode: +v stuartmorgan
[16:50:44] *** MichealH has joined #chromium
[16:52:17] *** Erkan_Yilmaz has joined #chromium
[16:53:28] *** Viclame has quit IRC
[16:53:34] *** monreal_ has quit IRC
[16:54:36] *** hamaji has quit IRC
[16:55:42] <phajdan-jr> sheriffs, I'm watching the tree slightly less carefully than before (to do some work). Please alert me immediately if any test fails that might be even remotely related to the timeout issues.
[16:56:46] <jochen__> i committed the fix for this selfdeleteplugin
[16:56:49] <jochen__> let's see what happens
[16:57:09] *** malavv_ has joined #chromium
[16:58:03] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Chromium Linux Builder (dbg)" from 52665: msneck at google dot com, tfarina at chromium dot org, tyoshino at chromium dot org)
[16:59:06] *** peper has joined #chromium
[17:00:27] *** dmaclach has joined #chromium
[17:00:46] *** dmaclach has quit IRC
[17:01:00] <erikchen> anyone looking into this?
[17:01:48] *** hamaji has joined #chromium
[17:04:48] *** fsamuel has joined #chromium
[17:05:56] <phajdan-jr> if not, then... "revert now, ask questions later" please
[17:06:26] <erikchen> looks like build-bots need to be kicked
[17:06:42] <tfarina> erikchen: I think it's flaky
[17:06:42] <erikchen> distcc[7180] (dcc_writex) ERROR: failed to write: Connection timed out
[17:06:43] <erikchen> distcc[7180] (dcc_writex) ERROR: failed to write: Broken pipe
[17:06:45] <phajdan-jr> oh yeah, compile timeout
[17:06:53] <tfarina> erikchen: yeah
[17:07:06] <tfarina> just wait for the next compilation I guess.
[17:07:09] <phajdan-jr> erikchen: no, these seem to be harmless, something weird happens later though
[17:07:10] <erikchen> am I allowed to mark tree as open?
[17:07:17] <phajdan-jr> erikchen: I think yes
[17:08:08] <phajdan-jr> okay, who owns DromaeoTest? The timeouts there are mine. :-/
[17:08:09] <trungl-bot> Tree opened by erikchen at google dot com: Tree is open
[17:10:11] <trungl-bot> Tree opened by phajdan.jr at chromium dot org (:phajdan-jr): Tree is open (phajdan.jr on dromeao_jslib_test and possibly others, timeouts related)
[17:13:52] <jochen__> lots of the windows redness should go away after 52669
[17:14:01] *** MikeSmithW3C has joined #chromium
[17:16:18] *** tonyg-cr has joined #chromium
[17:16:18] *** ChanServ sets mode: +v tonyg-cr
[17:16:30] *** MikeSmithW3C^ has quit IRC
[17:17:47] *** malavv__ has joined #chromium
[17:19:27] *** romainhuet has quit IRC
[17:19:33] <phajdan-jr> it's interesting that only dromaeo_jslib_test seems to be failing and other long tests are fine
[17:19:33] *** romainhuet has joined #chromium
[17:19:45] *** trungl has quit IRC
[17:19:47] *** malavv_ has quit IRC
[17:19:47] <phajdan-jr> sheriffs, I'm going to investigate locally for a while if you don't mind
[17:19:58] <phajdan-jr> ETA no more than 30 minutes
[17:20:07] *** yutak_home has joined #chromium
[17:20:08] *** ChanServ sets mode: +v yutak_home
[17:20:12] <phajdan-jr> but I can fix immediately at any time if you want
[17:20:23] <phajdan-jr> would just prefer to keep the timeout low
[17:20:31] *** malavv has quit IRC
[17:21:23] *** eseidel has quit IRC
[17:21:27] *** hamaji has quit IRC
[17:22:20] *** star-affinity has quit IRC
[17:22:31] <phajdan-jr> erikchen: compile net failed on your change
[17:22:36] <erikchen> ok
[17:24:50] <erikchen> fix going in soon
[17:26:39] *** agayev has joined #chromium
[17:26:59] *** mazda has joined #chromium
[17:27:58] *** agayev has left #chromium
[17:34:01] <erikchen> fix in
[17:38:09] *** mdwright_laptop has joined #chromium
[17:38:47] *** xiyuan has joined #chromium
[17:38:47] *** ChanServ sets mode: +v xiyuan
[17:38:49] *** shreyas_ has joined #chromium
[17:39:51] *** mazda has quit IRC
[17:40:07] *** shreyas_ has quit IRC
[17:40:54] *** mazda has joined #chromium
[17:40:54] *** ChanServ sets mode: +v mazda
[17:41:21] *** [Nef] has quit IRC
[17:41:24] <phajdan-jr> Dromaeo should be fixed after my latest commit. Watching the tree.
[17:42:43] <malavv__> Can someone help me, I'm a little confused, is the maximum priority in crbug, 1 or 3?
[17:43:09] <phajdan-jr> malavv__: 1 is highest priority
[17:43:15] <phajdan-jr> malavv__: and there might be 0
[17:43:50] <malavv__> k k thx
[17:45:56] *** Venom_X has joined #chromium
[17:46:10] *** phajdan-jr is now known as phajdan-jr|afk
[17:46:43] *** shreyas_ has joined #chromium
[17:47:03] *** SkyLined has joined #chromium
[17:47:22] *** shreyas_ has quit IRC
[17:47:56] *** pdelgallego has quit IRC
[17:48:33] *** SkyLined has joined #chromium
[17:48:39] *** thakis_ has quit IRC
[17:48:39] *** pdelgallego has joined #chromium
[17:49:12] *** timsteele has quit IRC
[17:49:37] *** rniwa has joined #chromium
[17:49:37] *** ChanServ sets mode: +v rniwa
[17:52:13] <mmoss> we have a grd change coming up, anyone else want to get in on it?
[17:52:19] *** SkyLined has quit IRC
[17:52:28] <xiyuan> mmoss: I am about to land CL http://codereview.chromium.org/2935011 which has GRD changes.
[17:52:39] <mmoss> ^^^^ that would be the one :)
[17:52:47] <mmoss> anyone else?
[17:53:45] <xiyuan> mmoss: let me know when I can land it... :)
[17:53:59] *** fta_ has joined #chromium
[17:55:03] *** fta has quit IRC
[17:55:15] *** fta_ is now known as fta
[17:56:25] <mmoss> xiyuan: go ahead
[17:56:38] <xiyuan> cool.. thanks
[17:57:39] <xiyuan> mmoss: change is in... :)
[18:01:59] *** phajdan-jr|afk is now known as phajdan-jr
[18:02:36] *** thakis_ has joined #chromium
[18:02:36] *** ChanServ sets mode: +v thakis_
[18:09:04] *** star-affinity has joined #chromium
[18:12:18] *** zaspire has quit IRC
[18:14:23] *** dmaclach has joined #chromium
[18:15:55] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "test_shell_tests" on "Webkit" from 52665: msneck at google dot com, tyoshino at chromium dot org)
[18:17:10] <mmoss> closure looks flakey: [2740:1428:5899625:ERROR:ssl_test_util.cc(219)] Failed to connect to server
[18:17:30] <mmoss> I'm going to open
[18:17:46] *** Venom_X is now known as Venom_bbiab
[18:18:27] *** jwalden has joined #chromium
[18:19:57] <trungl-bot> Tree opened by mmoss at chromium dot org: Tree is open (r52674 should make Windows green)
[18:20:58] *** star-affinity has quit IRC
[18:21:42] *** markmentovai has joined #chromium
[18:21:42] *** ChanServ sets mode: +v markmentovai
[18:23:25] *** eseidel has joined #chromium
[18:24:08] *** fqian has joined #chromium
[18:25:33] *** star-affinity has joined #chromium
[18:30:44] *** jrmuizel has quit IRC
[18:31:12] *** eseidel has quit IRC
[18:32:34] *** star-affinity has quit IRC
[18:33:10] *** andrix1 has joined #chromium
[18:33:51] *** pfeldman has quit IRC
[18:34:30] *** bweinstein has joined #chromium
[18:36:07] *** andrix has quit IRC
[18:38:09] *** jshin has joined #chromium
[18:38:09] *** ChanServ sets mode: +v jshin
[18:39:26] *** RT|Chatzilla has quit IRC
[18:40:26] *** jg has joined #chromium
[18:40:51] <jg> Hi, where can I get version 6 of google chrome?
[18:42:35] *** magn3ts has joined #chromium
[18:42:46] <magn3ts> Why would Chrome show the default 404 page rather than *my* 404 page?
[18:43:00] *** jg has quit IRC
[18:43:17] <magn3ts> firefox is the only browser that shows my own custom page. IE also displays its default page.
[18:44:19] <Namegduf> magn3ts: Size.
[18:44:32] <magn3ts> its tiny
[18:44:37] <Namegduf> If the page is really small, Chrome assumes it's a default 404 page and shows its internal one.
[18:44:47] <Namegduf> IE does similar.
[18:44:59] <magn3ts> haha
[18:45:30] <magn3ts> well thats one way of doing things
[18:46:15] *** bauerb has quit IRC
[18:47:04] *** Erkan_Yilmaz_ has joined #chromium
[18:47:14] <magn3ts> any recommended ways to add junk data to it?
[18:49:35] <ojan> markmentovai, jrg: ping
[18:49:45] <phajdan-jr> magn3ts: HTML comment maybe
[18:50:11] <markmentovai> markmentovai:
[18:50:11] <ojan> markmentovai, jrg: trying to do a webkit roll. it fails to link on leopard debug due to OOM (can't map file).
[18:50:12] <markmentovai> i mean
[18:50:13] <markmentovai> ojan
[18:50:15] *** Erkan_Yilmaz has quit IRC
[18:50:20] <markmentovai> yes, that's correct.
[18:50:21] *** csilv has joined #chromium
[18:50:21] *** ChanServ sets mode: +v csilv
[18:50:31] <ojan> markmentovai, jrg: trying to figure out what to do so we can roll webkit sometime today.
[18:50:39] <markmentovai> we're moving as fast as we can, but getting these bots over to 10.6 is taking some time
[18:50:54] <markmentovai> we can disable svg in the mac build temporarily
[18:51:09] <ojan> markmentovai: yeah, that's what tony^work and i were thinking
[18:51:10] <markmentovai> that'll buy a tiny bit of time and will hopefully be enough to tide us over
[18:51:29] <ojan> markmentovai: we only need to do so for the chrome builders though, test_shell still links
[18:51:43] <ojan> markmentovai: so we shouldn't lose significant test coverage in the meantime
[18:52:15] <markmentovai> not sure if we have that sort of granularity without doing a master restart
[18:53:16] <tony^work> ok, I can disable svn on the mac dbg builder
[18:53:25] <tony^work> just change the .gyp/include.gypi file?
[18:53:35] <ojan> markmentovai: hm...so, would you suggest just marking all the SVG layout tests as failing on mac-dbg in the meantime?
[18:53:35] *** bweinstein has quit IRC
[18:53:38] <markmentovai> oh, yeah, that'll do it
[18:53:52] <tony^work> the layout tests should still pass
[18:53:55] *** shreyas_ has joined #chromium
[18:53:57] <markmentovai> just need to remember to undo that change in the future
[18:54:01] <tony^work> since I won't touch the mac webkit dbg builder
[18:54:06] *** thakis_ has quit IRC
[18:54:10] <markmentovai> yup
[18:54:16] <markmentovai> however...
[18:54:24] <markmentovai> trybots are still a concern, if they still have svg enabled
[18:54:25] <tony^work> once we replace them with SL, won't that clobber any changes on the current builder?
[18:54:40] <markmentovai> and any other bots that build all the way through to the app framework
[18:54:50] <tony^work> oh, hmm
[18:55:12] <markmentovai> we can edit the .gyp files to disable svg on mac by default and explicitly enable it on the mac layout test bots?
[18:55:21] *** shreyas_ has quit IRC
[18:55:23] <tony^work> what about the official build?
[18:55:30] <markmentovai> doh.
[18:55:33] <tony^work> oh, it's just dbg
[18:55:35] <tony^work> maybe that's ok?
[18:55:45] <tony^work> official builds will be release
[18:55:59] <markmentovai> i don't think we can disable svg on a per-configuration basis like that
[18:56:09] <markmentovai> it turns files on and off, right? it's not just a set of compiler macros.
[18:56:17] <tony^work> yes, that's right
[18:56:19] <tony^work> yeah, you're right
[18:56:20] <markmentovai> yeah, we can't do that.
[18:56:59] <tony^work> breaking all the mac trybots sounds bad
[18:57:03] <markmentovai> yup
[18:57:18] <markmentovai> ok, how 'bout this
[18:57:32] <markmentovai> we can disable svg if os==mac and branding!=chrome
[18:57:53] <markmentovai> and explicitly enable it on the mac layout bots
[18:57:58] *** viro101 has quit IRC
[18:59:00] <tony^work> ok
[18:59:13] <markmentovai> ojan: how's that sound?
[18:59:13] <tony^work> so just edit the ~/.gyp/include.gypi file on the mac webkit dbg builder?
[18:59:14] *** rniwa has quit IRC
[18:59:24] <markmentovai> tony^work: yeah. on both of them.
[18:59:33] <tony^work> ah, right both of them
[18:59:37] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Chromium Linux Builder x64" from 52688: darin at chromium dot org (:fishd))
[18:59:53] <markmentovai> tony^work: and on the canary/ies, too
[19:00:02] *** hamaji has joined #chromium
[19:00:04] <ojan> sgtm
[19:00:07] <tony^work> ah, yup
[19:00:23] <markmentovai> and also, file a bug to track this so we can undo all of these hackies once we're off 10.5 as a build platform.
[19:00:32] <fishd> i done broke 64 bit linux
[19:01:04] <fishd> reverting
[19:01:25] <tony^work> I'll file a bug now for tracking
[19:02:05] <tony^work> it would be nice if the release builders still had svg, so I may try to enable it on some of those machines as well
[19:02:13] *** chaitanyag has joined #chromium
[19:02:26] *** apavlov has quit IRC
[19:02:32] <tony^work> markmentovai: are you able to make the gyp changes to disable svg?
[19:02:52] <markmentovai> tony^work: not rapidly
[19:02:59] <tony^work> heh, ok, I'll look into it
[19:03:09] <markmentovai> it'll be at least 90 mins before i'll be able to do that
[19:03:11] *** saittam has joined #chromium
[19:03:44] *** MichealH_ has joined #chromium
[19:04:39] <trungl-bot> Tree opened by darin at chromium dot org (:fishd): Tree is open (reverted r52688, should cycle green)
[19:06:48] *** MichealH has quit IRC
[19:09:39] *** mnissler_ has joined #chromium
[19:10:38] *** GeekShad0w has joined #chromium
[19:11:04] *** GeekShadow has quit IRC
[19:11:06] *** andrix1 has quit IRC
[19:12:24] *** ananta has left #chromium
[19:13:15] *** hamaji has quit IRC
[19:13:44] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Chromium Mac Builder" from 52690: albertb at chromium dot org, darin at chromium dot org (:fishd), oshima at chromium dot org)
[19:15:44] *** loislo has quit IRC
[19:15:50] *** sbyer has left #chromium
[19:17:20] <eglaysher> So is this a real compile failure, or just a resurgence of the previous one?
[19:17:21] *** Erkan_Yilmaz__ has joined #chromium
[19:17:23] <mmoss> fishd: mac is same error as linux?
[19:17:39] <mmoss> I think it's ok to open
[19:17:47] *** fta_ has joined #chromium
[19:18:06] *** oshima has joined #chromium
[19:18:06] *** ChanServ sets mode: +v oshima
[19:18:47] <trungl-bot> Tree opened by mmoss at chromium dot org: Tree is open (reverted r52688, should cycle green)
[19:19:21] *** fta has quit IRC
[19:19:29] *** fta_ is now known as fta
[19:20:32] *** Erkan_Yilmaz_ has quit IRC
[19:20:56] *** mnissler_ has quit IRC
[19:21:12] *** saittam has quit IRC
[19:25:33] *** rniwa has joined #chromium
[19:25:33] *** ChanServ sets mode: +v rniwa
[19:26:29] *** jamesr has joined #chromium
[19:26:29] *** ChanServ sets mode: +v jamesr
[19:27:58] *** tbassetto has quit IRC
[19:28:43] *** Viclame has joined #chromium
[19:31:04] *** bweinstein has joined #chromium
[19:31:31] *** Riotta has joined #chromium
[19:31:39] <Riotta> hello
[19:32:10] <Riotta> does somebody can tell me where chromium stores saved passwords?
[19:32:11] <pinkerton> hello
[19:32:16] <pinkerton> depends on the OS
[19:32:29] <Riotta> linux
[19:32:48] <Riotta> is it one files or many?
[19:33:11] *** kees has quit IRC
[19:33:25] <chaitanyag> Hi... gcl upload cl is failing for me with: svn status returned unexpected output: C      chrome/chrome_browser.gypi. Any idea what's wrong?
[19:33:30] *** kees has joined #chromium
[19:33:36] <phajdan-jr> chaitanyag: resolve the conflict
[19:33:46] *** hsuh has joined #chromium
[19:34:28] *** satish_ has joined #chromium
[19:35:05] *** eseidel has joined #chromium
[19:35:34] <Riotta> I would like to backup only passwords stored by chromium and delete the rest of the profile is it possible ?
[19:35:44] <Riotta> on linux
[19:36:05] <satish_> fishd: Hi, yt
[19:36:57] <fishd> satish_: hi
[19:37:16] <chaitanyag> phajdan-jr: thanks
[19:37:22] <satish_> fishd: jorlow@ referred me to you. Any chance you'd have time to review a webkit patch for me? :) https://bugs.webkit.org/show_bug.cgi?id=42367
[19:38:19] *** rafaelw has joined #chromium
[19:38:33] <fishd> satish_: ok.. i put it in my queue.  working on 3 other webkit reviews at the moment
[19:38:46] *** mdwright_laptop has left #chromium
[19:39:05] <satish_> fishd: thanks a ton. :)
[19:39:19] <mirandac> sheriffs: I have a grd change.  It can wait, but I'd like to get it in today.  What do you suggest?
[19:39:22] *** rniwa has quit IRC
[19:40:36] <dmaclach> mirandac: assuming things are green once 52674 completes can we try it then? That'll also be about lunch time PST so probably a good time.
[19:40:50] <mirandac> dmaclach: sure, sounds good.
[19:42:59] *** [Nef] has joined #chromium
[19:45:58] *** Riotta has quit IRC
[19:46:57] *** lnostdal has quit IRC
[19:47:57] *** sbyer has joined #chromium
[19:47:57] *** ChanServ sets mode: +v sbyer
[19:48:45] <tonyg-cr> wow, builders are way behind
[19:49:05] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Linux Builder (ChromiumOS)" from 52699: ahendrickson at google dot com, darin at chromium dot org (:fishd), davemoore at chromium dot org, erg at chromium dot org (:eglaysher), jhawkins at chromium dot org, tonyg at chromium dot org (:tonyg-cr), vandebo at chromium dot org)
[19:50:07] <eglaysher> /b/slave/chromium-rel-linux-chromiumos/build/src/third_party/WebKit/WebCore/dom/OverflowEvent.idl:26:7:error: invalid preprocessing directive #FIXME
[19:50:10] <eglaysher> Hmmmmm
[19:51:07] <fishd> next error looks autofill related
[19:51:15] <fishd> cannot allocate an object of abstract type 'chromeos::AccountCreationTabContents'
[19:51:40] <mmoss> fishd: jhawkins I think
[19:51:44] <fishd> RenderViewHostDelegate::AutoFill::FillAutoFillFormData not implemented
[19:51:54] <mmoss> 52699
[19:52:06] <fishd> revert!
[19:52:20] *** maikmerten_ has joined #chromium
[19:52:33] <mmoss> fishd: no rush, tree needs a break anyhow
[19:52:38] <mmoss> jhawkins: you on that?
[19:52:55] <fishd> if we keep the tree closed, we'll just get a pile up of additional commits once it opens
[19:52:55] *** skanduk has joined #chromium
[19:52:58] <fishd> slinky time
[19:55:07] *** jamesr has quit IRC
[19:55:58] *** maikmerten has quit IRC
[19:56:33] <jam2> sheriffs: i have a grd change too, should i try to land it at the same time as mirandac?
[19:56:42] *** asdsad has joined #chromium
[19:56:49] *** johnny_g has joined #chromium
[19:57:33] <dmaclach> jam2: please
[19:57:56] <mirandac> dmaclach: jam2: just say the word when it's time...
[19:58:23] <jam2> mirandac: ok, i'll make the patch right now :)  i just want to add a string, I'll use it later
[19:58:30] <jam2> is that ok?
[19:58:39] *** Buglouse has joined #chromium
[19:58:42] *** Malmis has quit IRC
[19:58:44] <mirandac> jam2: sure thing, that's what I did, too :-)
[20:01:36] *** romainhuet has quit IRC
[20:01:56] <jam2> ok making the change right now
[20:02:08] <jam2> i only have to touch the grd file right?  (havent added a string in a _long_ time :) )
[20:02:42] *** satish_ has quit IRC
[20:02:54] *** Malmis has joined #chromium
[20:02:54] *** lnostdal has joined #chromium
[20:02:56] <mirandac> jam2: if you're not adding a string, it won't "count" as a change, because the ordering won't be disturbed...
[20:03:15] <dhollowa> mirandac, jam2: i'd like to get on this .grd train too please.
[20:03:17] <jam2> mirandac: i'm adding a string
[20:03:18] <mirandac> dhollowa just told me he has a grd change as well, he's getting it ready right now...
[20:03:22] <mirandac> heh
[20:03:46] <mirandac> jam2: ok, yeah -- you only have to change the grd file, right ( sorry, misunderstood)
[20:04:36] <jam2> np, thank you
[20:05:02] <jam2> i should add it to the bottom right?
[20:05:14] <trungl-bot> Tree closed by mmoss at chromium dot org: Tree is closed - chromeos compile -> jhawkins fixing; grd change coming
[20:05:32] <thakis> jam2: ask pkasting, he has rejiggered all the grd files recently
[20:05:38] <thakis> (to be alphabetical?)
[20:05:54] <jam2> got it, thanks.  will make alphabetical
[20:05:58] <mirandac> jam2: you can add it wherever it makes sense in the ordering, I think; if you add it to the bottom, it won't cause any problems, again because the ordering of all previous is maintained.
[20:06:14] <jam2> ah, so it seems makign them alphabetical is screwing us!
[20:06:18] <jam2> given the gyp issues
[20:06:29] <mirandac> it depends on the .grd file -- generated_resources.grd seems to be organized by grouping
[20:06:34] <mmoss> grd changes ready to go?
[20:07:32] <dhollowa> mmoss: i'm ready
[20:07:32] <jam2> yep one sec, sending for reveiwe to mirandac right now
[20:07:49] <mirandac> if we only organize by time, always adding to the end, it will be much more difficult to maintain strings by, for example, which dialog they belong to... (same if we organize alphabetically only).  I am speaking of the big one, generated_resources.grd, mainly
[20:08:44] <mirandac> mmoss: ready to submit
[20:08:50] <mmoss> ok
[20:08:56] <jam2> one sec please!
[20:09:14] <jam2> mirandac: http://codereview.chromium.org/2834050
[20:09:59] <mirandac> jam2: lgtm
[20:10:16] <jam2> thanks, ok ready tocommit
[20:10:25] *** johnny_g_ has joined #chromium
[20:10:25] <jam2> i guess one of us will get out of date and will quickly have ot update their file :)
[20:10:41] <mirandac> heh, good point
[20:10:47] <jam2> you can go first
[20:10:51] <dhollowa> game on
[20:10:53] <mirandac> ok!
[20:10:54] <dhollowa> :-)
[20:11:00] <mirandac> sent
[20:11:05] <jam2> u mean committed?
[20:11:06] <jam2> just checking
[20:11:06] <mirandac> yes
[20:11:10] <mirandac> sorry :-)
[20:11:19] <mirandac> 52705
[20:11:43] <dhollowa> cheater!
[20:11:50] <jam2> committed 52706
[20:12:50] *** johnny_g has quit IRC
[20:12:50] *** johnny_g_ is now known as johnny_g
[20:12:54] *** eseidel_ has joined #chromium
[20:14:19] <trungl-bot> Tree closed by dmaclach at chromium dot org: Tree is closed - chromeos compile -> jhawkins fixing; grd change landed
[20:15:14] *** eseidel has quit IRC
[20:15:14] *** eseidel_ is now known as eseidel
[20:16:35] *** sytse has quit IRC
[20:17:02] <dhollowa> working on it... merging.
[20:18:48] <mmoss> anyone know what's going on with Builder Chromium XP mini_installer_test?
[20:19:55] *** mrossetti has quit IRC
[20:20:43] <mirandac> mmoss: oh-oh, looks like it's related to FRUI issues.  looking.
[20:20:58] <mmoss> mirandac: thanks
[20:21:16] *** MichealH_ is now known as MichealH
[20:21:34] *** MichealH has joined #chromium
[20:22:05] <shess> rohitrao: are you comfortable with the while (i--) idiom?  Main reason I used for() instead was so that i was captured in the for() scope not the function scope.
[20:22:05] <dhollowa> problems with merge somehow...  "Command "git svn dcommit --no-rebase" failed."
[20:22:19] *** fqian has quit IRC
[20:22:21] <dhollowa> Transaction is out of date: File '/trunk/src/chrome/app/generated_resources.grd' is out of date
[20:22:37] *** fqian has joined #chromium
[20:22:38] *** fta_ has joined #chromium
[20:22:41] <mirandac> mmoss: yes, this is me.  the tester is trying to close the FirstRunUI window, which no longer exists.  Will revert my change if I can't fix quickly.
[20:22:54] <dhollowa> anyone have any idea how to clear this?
[20:23:01] *** rafaelw has quit IRC
[20:23:03] <rohitrao> shess: I understand the for loop now, I had just never seen it before and it took me a few minutes to figure out why it worked
[20:23:12] <phajdan-jr> dhollowa: git rebase your favorite way
[20:23:18] <phajdan-jr> dhollowa: I usually do git cl rebase
[20:23:20] <rohitrao> shess: I don't know if other people will find it confusing or not :)
[20:23:29] *** Precea[BNC] is now known as Precea
[20:23:44] <dhollowa> i did a git pull of trunk
[20:23:46] *** fta has quit IRC
[20:23:55] <dhollowa> then a git merge -v trunk on my branch
[20:23:58] *** fta_ is now known as fta
[20:24:00] <shess> rohitrao: probably putting the loop variable on a separate line would make it more readable.  Right now the weight of the line is in the initialization.
[20:24:02] <tfarina> git pull --rebase?
[20:24:02] <dhollowa> thought that'd be the same, no?
[20:24:25] <mmoss> tonyg-cr, ojan: do we need to flag some tests after the WEB_TIMING change? http://buildbot.jail.google.com/buildbot/chromium/builders/Webkit%20Linux/builds/26325
[20:24:29] <phajdan-jr> dhollowa: not sure, maybe svn git mirror is slightly behind
[20:24:45] <phajdan-jr> dhollowa: that's why I'm using git with regular svn, and not the git:// tree btw
[20:25:03] <rohitrao> shess: do you need the index? can you just use a reverse iterator?
[20:25:14] <tonyg-cr> mmoss: the updated test expectations were checked in with the flag flip in webkit
[20:25:21] <dhollowa> git cl rebase worked
[20:25:23] <dhollowa> thanks all.
[20:25:24] <tonyg-cr> so they may fail until the next roll
[20:25:51] <shess> rohitrao: originally was.  Made things wonderfully verbose.
[20:25:58] <mmoss> tonyg-cr: ah, ok. when is the roll?
[20:26:24] <rohitrao> shess: perhaps, but easier for me to read ;)
[20:26:27] <markmentovai> tony^work: 90 minutes are up. were you able to get the svg thing done as discussed?
[20:26:37] <tony^work> markmentovai: turns out someone already rolled webkit
[20:26:40] <shess> rohitrao: AFAICT, views has both in the same order.  I'll just do that and let someone log a bug, maybe.
[20:26:41] <tony^work> 10 hours ago
[20:26:43] <tony^work> and it didn't break
[20:26:46] <tonyg-cr> mmoss: that is a question for ojan
[20:26:47] <markmentovai> oh, that's cute
[20:26:52] <tony^work> but I uploaded my change for future reference
[20:26:53] <rohitrao> shess: I dunno, I might be nitpicking too much, but I hate having to think about code
[20:26:57] <markmentovai> k
[20:27:11] <rohitrao> shess: also, if those two for loops stay one line long, no braces
[20:27:13] <shess> rohitrao: [this is all probably like how when you move house, the movers pack EVERYTHING in a room, including trash.]
[20:27:29] <shess> rohitrao: really?  I thought for loops got braces all the time.
[20:27:39] <tonyg-cr> mmoss: we could skip them until the roll if it is going to be a while
[20:27:45] <phajdan-jr> oshima: the autocomplete test you've marked FLAKY should be DISABLED, because it's crashy. Sent you details in the review e-mail.
[20:27:47] <rohitrao> shess: huh, in that case, fix the one a few lines up ;)
[20:27:49] <rohitrao> let me check
[20:28:17] <mmoss> tonyg-cr: let's see what ojan says
[20:28:40] <oshima> phajdan-jr: flaky does not work with crash?
[20:28:51] <oshima> s/flaky/FLAKY_/
[20:29:00] <dhollowa> landed 52707 .grd
[20:29:15] <phajdan-jr> oshima: no, it'll still turn things red
[20:29:26] *** skanduk has quit IRC
[20:29:32] <phajdan-jr> oshima: I'm trying to clean up the confusion about prefixes, but it'll take time
[20:29:41] <oshima> phajdan-jr: ok. will fix it
[20:29:47] <phajdan-jr> oshima: ideally, we should be back to just DISABLED in... a year? ;-)
[20:29:49] <ojan> mmoss, tonyg-cr: no ETA on the next roll. don't block on it. set the tests as failing in the downstream expectations file instead.
[20:29:59] <rohitrao> shess: style guide didn't seem to say, i'm pro-braces
[20:30:11] <ojan> mmoss, tonyg-cr: we can remove it with the webkit roll.
[20:30:27] <markmentovai> rohitrao: anything i can disambiguate for you?
[20:30:33] *** Cheery has joined #chromium
[20:30:34] <oshima> phajdan-jr: yeah, i don't like FLAKY_. it's just hiding issue.
[20:30:43] <Cheery> hm.. I just noticed something worrysome
[20:30:49] <shess> rohitrao: I'm braces-aren't-optional myself, but have come to accept that single-line if() clauses with no else can get away with out them in the interests of team harmony.
[20:30:51] <Cheery> Websocket request starts with GET
[20:30:57] <Cheery> but..
[20:31:00] <oshima> but understand it's hard to remove them completely in ui tests.
[20:31:00] <rohitrao> markmentovai: do we omit braces for a single-line for clause?
[20:31:14] <markmentovai> you may but you're not required to
[20:31:14] <Cheery> RFC 2616 says that GET requests may be cached
[20:31:23] <rohitrao> markmentovai: ok, thanks
[20:31:42] <Cheery> therefore WebSockets are violating HTTP specs.
[20:31:58] <markmentovai> except in extreme cases, nobody's gonna call you out for using braces when you're in doubt.
[20:32:21] <Cheery> Or am I entirely incorrect there?
[20:32:22] <shess> while (1) {{{ doit(); }}}
[20:32:40] <phajdan-jr> oshima: I hate FLAKY. In fact, the only thing blocking proper resolution is the problem with machines. Once we can get a few new bots... :)
[20:32:49] <Cheery> This is quite important, because if I'm in correct. Websocket is going to be an entirely useless in production environments.
[20:32:54] <markmentovai> shess: use true instead of 1, then we can address the braces.
[20:33:17] *** loislo has joined #chromium
[20:33:42] <shess> markmentovai: for(;;) {{{ doit(); }}}?
[20:34:05] <thakis> webkit folks: is layout_test_results the windows layout test results bot ( http://build.chromium.org/buildbot/layout_test_results/webkit-rel-webkit-org/52703/ )?
[20:34:11] <markmentovai> ok, but i think while(true) is more readable
[20:34:14] <shess> though while (((true))) {{{ doit(); }}} is getting there.  maybe I don't want to be there.
[20:34:34] <markmentovai> well, there's while ((true == true))
[20:34:46] <mmoss> jochen__: Builder Vista Tests (dbg)(3) and Builder Vista Tests (dbg)(4) seem to be failing since your r52655 this morning
[20:35:00] <jochen__> mmoss: yes, just uploading a fix for that
[20:35:05] <mmoss> jochen__: thanks
[20:35:16] <jochen__> the problem is, i can't reproduce it on trybots nor on my own vista box
[20:35:26] <jochen__> so i just guess it fixes it
[20:36:11] *** jamesr has joined #chromium
[20:36:11] *** ChanServ sets mode: +v jamesr
[20:36:26] *** markusheintz has quit IRC
[20:36:52] <dhollowa> markmentovai: or more concisely while (!!1)
[20:36:55] <oshima> phajdan-jr: http://codereview.chromium.org/3039007
[20:37:25] *** rafaelw_wfh has joined #chromium
[20:37:43] <shess> dhollowa: need to make it symmetrical.  while ((!!11)) {{ doit(); }}
[20:37:58] <Cheery> having POST in websocket would ensure it reaches the server
[20:38:00] <markmentovai> dmaclach, mmoss, tyoshino: the tree status is "Tree is closed - chromeos compile -> jhawkins fixing; grd change landed" - looks like jhawkins fixed and chromeos compiled. grd change landed? ok? is that why the tree's closed now?
[20:38:00] <shess> while ((11&&11)) {{ ;; }}
[20:38:20] <rohitrao> bool kFalse = true; while ((((kFalse == true)))
[20:38:25] <mmoss> markmentovai: refresh :)
[20:38:41] <markmentovai> that's service
[20:38:44] <trungl-bot> Tree opened by mmoss at chromium dot org: Tree is open
[20:38:51] *** fqian has quit IRC
[20:39:03] *** satish_ has joined #chromium
[20:40:36] <pinkerton> sigh, i thought the deadline was sunday night. i can't wait to sheriff on monday.
[20:40:58] *** asdsad has quit IRC
[20:41:18] <shess> pinkerton: omg
[20:41:33] <pinkerton> youbetcha
[20:42:04] <markmentovai> pinkerton: you can keep the tree closed all day monday
[20:42:14] <pinkerton> i plan on it
[20:42:21] <markmentovai> that would be the status quo anyway
[20:42:49] <pinkerton> so it seems
[20:42:59] *** fqian has joined #chromium
[20:43:34] <phajdan-jr> oshima: LGTM
[20:44:05] <ojan> tonyg-cr: looks like enabling webtiming downstream caused another test to fail on the canaries: http://test-results.appspot.com/dashboards/flakiness_dashboard.html#showExpectations=true&showLargeExpectations=true&tests=fast%2Fdom%2Fprototype-inheritance.html&useWebKitCanary=true
[20:44:20] <oshima> thx
[20:44:32] <ojan> tonyg-cr: it's failing on the non-canaries as well of course
[20:44:45] <tonyg-cr> ojan: looking
[20:45:10] <rohitrao> wait, when is the deadline
[20:45:33] <pinkerton> quitting a long-running chrome build, it's pegging the cpu and still "quitting"
[20:45:49] <markmentovai> rohitrao: monday
[20:46:12] <rohitrao> but not monday @3am?
[20:46:15] <pinkerton> ouch
[20:46:20] <pinkerton> i had to force-quit 3 renderers
[20:46:23] <pinkerton> then the app quit
[20:46:25] <rohitrao> haha
[20:46:27] <pinkerton> that seems really bad
[20:46:29] <markmentovai> in the past, monday has meant ?you have the whole work day?
[20:46:45] <markmentovai> ?up until the beginning of tuesday?s work day?
[20:47:16] *** timsteele has joined #chromium
[20:47:17] <jibot> timsteele is a Chromium Googler.
[20:47:26] <markmentovai> duh.
[20:47:46] <jochen__> who put that jibot in here?
[20:47:50] <jochen__> it's kinda annoying
[20:48:01] <phajdan-jr> jochen__: ping
[20:48:14] <phajdan-jr> jochen__: I'm surprised that GetActiveTab made tests more reliable
[20:48:20] <phajdan-jr> jochen__: I'd be interested in the details
[20:48:22] *** stuartmorgan has quit IRC
[20:48:38] <pinkerton> rsesek!
[20:48:43] <phajdan-jr> jochen__: could you explain what you've seen happening wit the test? If our infrastructure is wrong, I'd like to fix that.
[20:48:48] <pinkerton> rsesek: wtf is up with my new tab button!?
[20:48:56] <phajdan-jr> jochen__: in particular, both old and new ways should be equally solid in theory
[20:49:19] *** Omen_20 has quit IRC
[20:49:22] *** pfeldman has joined #chromium
[20:50:36] <rsesek> pinkerton: talk to Cole
[20:50:45] <pinkerton> grrrr
[20:50:55] <pinkerton> it looks like ass
[20:50:56] <mirandac> mmoss: committed a patch that should fix mini_installer_test breakage.
[20:51:28] *** jwalden has left #chromium
[20:53:02] <jochen__> phajdan-jr: sec
[20:53:58] <pinkerton> rsesek: ok cole sent me a screen shot of his, it has a gray bg. my button background is white (latest ToT)
[20:54:01] <pinkerton> rsesek: is something wrong?
[20:54:28] <rsesek> pinkerton: I dropped in the images he gave me and sent me screen shots and he said LG; forward me his screen shot?
[20:54:38] <rsesek> s/sent me/sent him
[20:54:38] *** satish_ has quit IRC
[20:55:26] *** phajdan-jr is now known as phajdan-jr|afk
[20:55:54] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "unit_tests" on "Mac10.5 Tests (dbg)(1)" from 52707: dhollowa at chromium dot org)
[20:56:50] <pinkerton> rsesek: sent
[20:57:33] <rsesek> pinkerton: that screen shot is what mine looks like too
[20:57:38] <pinkerton> weird
[20:57:47] <rsesek> you using a theme?
[20:57:49] <pinkerton> nope
[20:57:58] <rsesek> huh
[20:58:09] <rsesek> check a waterfall build?
[20:58:09] <pinkerton> can you pull latest ToT continuous build and see if you get the right image?
[20:58:14] <pinkerton> this is a waterfall build
[20:58:56] <rsesek> pinkerton: weird. it's wrong there, too
[20:58:58] <pinkerton> heh
[20:59:03] <pinkerton> maybe it's a packaging bug?
[20:59:05] <rsesek> pinkerton: I wonder if it tried to svn-up it or something
[20:59:10] <rsesek> and treated it as text
[20:59:14] <pinkerton> did the file get renamed?
[20:59:16] <rsesek> nope
[20:59:19] <pinkerton> weird
[20:59:23] <rsesek> just droped in new images over it
[20:59:27] <rsesek> looks fine for me on a localbuild
[20:59:36] <pinkerton> must be packaging then
[20:59:56] <trungl-bot> Tree opened by mmoss at chromium dot org: Tree is open
[21:00:03] <rsesek> pinkerton: how do we troubleshoot? this is not so good
[21:00:30] * pinkerton gazes longingly at markmentovai who understands the extra packaging steps for a real build
[21:00:43] <pinkerton> i'll file a bug
[21:00:47] <rsesek> sg
[21:00:48] <rsesek> thx
[21:03:11] *** Omen_20 has joined #chromium
[21:04:22] <markmentovai> pinkerton, rsesek: looking
[21:04:53] <pinkerton> rsesek: i can't actually file a new bug
[21:04:58] <pinkerton> all i get is "error bad request"
[21:05:09] <rsesek> pinkerton: then it's clearly not a bug. the tracker won't accept it
[21:05:28] <pinkerton> can anyone else file a bug?
[21:05:59] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "unit_tests" on "Vista Tests (dbg)(1)" from 52707: dhollowa at chromium dot org, jam at chromium dot org (:jam2), mirandac at google dot com)
[21:07:04] <shess> rohitrao: just occurred to me that just because the cursor CL was your bug doesn't mean you need to review it - would you prefer I flip it to someone else?  I think it should be clear to most anyone.
[21:07:11] <shess> [well, insofar as it's clear at all.]
[21:07:51] <rohitrao> shess: too late
[21:07:58] <markmentovai> dmaclach: this thing that pinkerton, rsesek and i are talking about. my money?s on pdfsqueeze harming the pdf.
[21:07:58] *** sogar_gofin has joined #chromium
[21:08:01] <shess> rohitrao: even better.
[21:08:03] <rohitrao> shess: what happens when you click on padding between decorations?
[21:08:17] <pinkerton> markmentovai: yay.
[21:08:25] <pinkerton> rsesek: can you try filing a bug with ToT?
[21:08:29] <shess> rohitrao: still broken.  I'm not really sure what to do about it, though.  at least the cursor doesn't flip back and forth.
[21:08:41] <rsesek> pinkerton: I can file bug yes
[21:08:44] <pinkerton> weird
[21:09:00] <trungl-bot> Tree opened by mmoss at chromium dot org: Tree is open
[21:09:01] <shess> I'm mixed as to whether it should ignore the click, or be loose with targeting.
[21:09:15] <dmaclach> markmentovai, pinkerton, rsesek can somebody send me the pdf?
[21:10:03] <pinkerton> rsesek should have the original
[21:10:12] *** Cheery has left #chromium
[21:10:13] <dmaclach> rsesek: better yet, log a bug here: http://code.google.com/p/pdfsqueeze/issues/list
[21:10:16] <pinkerton> oh, i guess it's in the tree, but he'd know the name faster than i would
[21:10:34] <shess> rohitrao: I could perhaps loosen the NSMouseInRect() test in -decorationForEvent:* easily enough.  Just NSInsetRect() by half of the padding.
[21:10:40] <rsesek> dmaclach: //chrome/app/theme/new_tab*.pdf
[21:11:03] <rohitrao> shess: I don't want it to select-all
[21:11:13] <rohitrao> I don't think I care if it triggers the action or not
[21:11:25] <pfeldman> tony^work: you plan to roll today, right?
[21:11:30] <rohitrao> actually, maybe I do care.  I'd like it to be ignored
[21:11:43] <shess> rohitrao: select-all should not happen if it already has focus.
[21:12:12] *** taf2_ is now known as taf2
[21:12:44] *** eseidel has quit IRC
[21:14:38] *** mnissler_ has joined #chromium
[21:14:56] *** chrisccoulson has quit IRC
[21:17:30] *** alyxuk has joined #chromium
[21:18:10] *** BrendenB has quit IRC
[21:18:26] *** mnissler_ has quit IRC
[21:18:26] *** chaitanyag has quit IRC
[21:18:26] *** malavv__ has quit IRC
[21:18:26] *** cbentzel has quit IRC
[21:18:26] *** tfarina has quit IRC
[21:18:26] *** rsimha has quit IRC
[21:18:26] *** jschuh has quit IRC
[21:19:33] <dmaclach> mmentovai,rsesek: just ran new_tab.pdf and new_tab_h.pdf through pdfsqueeze and they look identical to me here.
[21:19:41] <rsesek> dmaclach: on 10.5?
[21:19:55] *** chrisccoulson has joined #chromium
[21:20:08] <dmaclach> rsesek touche. on 10.6. will try 10.5 shortly
[21:20:15] <rsesek> dmaclach: thanks
[21:20:22] *** Herald has quit IRC
[21:20:39] *** Herald has joined #chromium
[21:21:35] *** felipec has joined #chromium
[21:22:22] <felipec> is anyone testing switching from a normal network to a proxied network back and forth?
[21:22:25] *** phajdan-jr|afk is now known as phajdan-jr
[21:22:51] <rohitrao> trungl_afk, thakis: would either of you like to look at my one line of code? :)
[21:23:14] <thakis> rohitrao: layers?
[21:23:14] <dmaclach> rsesek: seems good on 10.5 as well.
[21:23:22] <rsesek> markmentovai: ^
[21:23:24] <rsesek> dmaclach: weird
[21:23:29] <rohitrao> thakis: http://codereview.chromium.org/1599028/show
[21:23:34] <markmentovai> dmaclach: what we're seeing...
[21:23:41] <markmentovai> pdfsqueezed on 10.5
[21:23:43] <markmentovai> then open it on 10.6
[21:23:45] <markmentovai> it looks different
[21:23:51] <dmaclach> k
[21:23:51] <pinkerton> i'm running on 10.5, fwiw
[21:23:53] <markmentovai> in chromium, the difference isn't really perceptible in preview
[21:24:25] <rohitrao> thakis: on the bright side, I figured out half my layer drawing problems.  pattern phases in layers are relative to the layer, not to the window
[21:24:34] <dmaclach> ah
[21:24:34] <thakis> rohitrao: i parsed your "this" in "this feels safer" the wrong way
[21:25:03] <rohitrao> thakis: sorry, I wrote it post-new-upload, not pre-new-upload :)
[21:25:10] <rohitrao> timestamps are key!
[21:26:23] <thakis> rohitrao: i feel that it's cleaner to put the logic for this in the xplatform parts
[21:26:39] <thakis> if you strongly feel that i'm full of it, then i'm fine with your patch too tho
[21:27:06] <rohitrao> thakis: how do you mean?
[21:27:25] <thakis> rohitrao: i'd put this logic in browser.cc (or whereever fullscreen is triggered)
[21:28:40] *** rsimha has joined #chromium
[21:29:17] <rohitrao> I don't know if there's any sort of xplatform bookmark bubble code though
[21:29:19] <rohitrao> let's see
[21:29:39] *** felipec has quit IRC
[21:29:57] <dmaclach> rsesek: did you get the new_tab pdf's from cole?
[21:30:03] <rsesek> dmaclach: si
[21:30:29] <dmaclach> rsesek, please go smack him around and suggest that UI graphics should be done in RGB and not in CMYK ;-)
[21:30:30] <rsesek> dmaclach: I sent him mail asking him if he did anything "weird" to them
[21:30:35] <rsesek> hahah
[21:30:52] <markmentovai> dmaclach: thanks :)
[21:30:54] <thakis> dmaclach: but what if we want to print them?
[21:30:55] *** fantasticulous has joined #chromium
[21:31:14] <motownavi> dmaclach: but you're the one in MTV
[21:31:20] <thakis> they contain strictly more information in cmyk!
[21:31:34] <dmaclach> motownavi: I haven't been in MTV for a looong time now
[21:31:42] <dmaclach> in fact I bet you've been there more recently than I have ;-)
[21:31:47] <pinkerton> :)
[21:32:10] <evmar> landed my monster xml change, please let me know if anything goes fishy in xml-land
[21:32:20] <phajdan-jr> evmar: now that you've updated bundled libxml, can linux distros switch back to the system one?
[21:32:28] <evmar> phajdan-jr: unfortunately, no
[21:32:44] <leiz> Mac10.5 Tests (dbg)(1) (codf72) will get a reboot after the current test run finishes
[21:32:45] <evmar> (sec)
[21:32:51] <phajdan-jr> evmar: I see. Can I help with upstreaming our patches?
[21:33:17] <evmar> sorry, was gonna link to a patch
[21:33:23] <evmar> ...but the tree hasn't updated yet
[21:33:52] *** fta_ has joined #chromium
[21:34:04] <evmar> phajdan-jr: i had to commit a real hack of a patch
[21:34:14] <evmar> i will send it to you once the servers all sync
[21:35:09] *** fta has quit IRC
[21:35:12] *** fta_ is now known as fta
[21:35:14] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Linux Builder (Views dbg)" from 52722: avi at chromium dot org (:motownavi), estade at chromium dot org, jrg at chromium dot org, lzheng at chromium dot org, mark at chromium dot org (:markmentovai), mattm at chromium dot org (:mattm_g), mirandac at google dot com, nirnimesh at chromium dot org, oshima at chromium dot org, tfarina at chromium dot org, thakis at chromium dot org, tonyg at chromium dot org (:tonyg-cr))
[21:35:27] <markmentovai> not it
[21:35:33] *** pfeldman has quit IRC
[21:35:35] <evmar> anyway, the patch is third_party/libxml/patches/libxml-disable-autodetect
[21:35:39] *** evmar is now known as evmar_afk
[21:35:43] <markmentovai> it's tfarina
[21:36:25] <rohitrao> thakis: looking at Browser::ToggleFullscreenMode(), I don't really think it fits there
[21:36:27] <dmaclach> markmentovai: maybe the builders just don't like references to bad 80s rap?
[21:36:37] *** feldstein has joined #chromium
[21:36:37] *** ChanServ sets mode: +v feldstein
[21:36:57] <rohitrao> thakis: that doesn't do much other than forward the call to the window, and browser.cc doesn't have any pointers to bubbles
[21:37:09] <markmentovai> dmaclach: it was 1990, ok?
[21:37:14] <leiz> stop, revert time?
[21:37:22] <pinkerton> ok, i still cannot submit a bug
[21:37:33] <pinkerton> rsesek: did you try the waterfall build?
[21:37:46] *** tfarina has joined #chromium
[21:37:55] *** nickcarter has joined #chromium
[21:38:09] <thakis> rohitrao: and all the browser window classes don't share any code?
[21:38:12] <thakis> hm look like it
[21:38:14] <thakis> *sigh*
[21:38:15] <rsesek> pinkerton: yes I just dled one. looks like crap on my machine. i sent mail to cole asking for rgb icons
[21:38:16] <thakis> rohitrao: lg then
[21:38:18] <rsesek> s/icons/images
[21:38:30] <pinkerton> rsesek: i mean did you try submitting a bug with the waterfall build
[21:38:31] *** jshin has quit IRC
[21:38:35] <rsesek> pinkerton: oh no
[21:38:40] <pinkerton> can you?
[21:39:00] <markmentovai> dmaclach: and, for the record, 1990 also brought us "ice ice baby." which would you prefer?
[21:39:29] <rohitrao> ooh
[21:39:30] <dmaclach> tfarina
[21:39:53] <rsesek> pinkerton: worked. cced you on the bug
[21:40:02] <pinkerton> weird
[21:40:06] <pinkerton> i simply cannot file any bugs
[21:40:18] <trungl-bot> Tree closed by mmoss at chromium dot org: Tree is closed - Linux Builder (Views dbg) -> tfarina
[21:40:19] <tfarina> dmaclach: I'm here
[21:40:32] <rohitrao> markmentovai: wait a second, this is just like under pressure
[21:40:39] <markmentovai> i know, what a rip!
[21:40:40] <tfarina> dmaclach: it passed on the linux trybot though
[21:41:01] <sogar_gofin> hey all
[21:41:02] <eglaysher> tfarina: the linux trybot or the linux_views trybot?
[21:41:16] <oshima> tfarina: linux and linux_view are different
[21:41:37] <oshima> tfarina: run it on linux_view
[21:41:38] <sogar_gofin> general coder question: if I wanted to start to get an understanding of how Chromium does allocation hijacking, in release, is there a particular spot in the directory structure i should focus on?
[21:41:39] <tfarina> arg, sorry about that
[21:42:00] <leiz> tfarina: when refactoring the gtk stuff, it's good to run it through the views trybot too
[21:42:02] <sogar_gofin> for Windows?
[21:42:04] <tfarina> should I revert so?
[21:42:21] <eglaysher> tfarina: yes.
[21:42:31] <leiz> markmentovai: codf72 is still not back up, do macs take this long to reboot?
[21:42:33] <tfarina> ok will do, just a second
[21:42:35] <jamesr> sogar_gofin: i think you might want http://src.chromium.org/viewvc/chrome/trunk/src/base/allocator/
[21:42:37] <pinkerton> rsesek: weird, i can't even make changes to existing bugs
[21:42:51] <rsesek> pinkerton:  you just filed a bug? (different browser?)
[21:42:55] <markmentovai> leiz: 60 seconds is normal for a mini to be back on-net, another 30 til it's connected to the master is normal
[21:43:00] <pinkerton> rsesek: camino
[21:43:07] <sogar_gofin> ah, thanks james!
[21:43:12] <leiz> markmentovai: it's been a few minutes, I'll ask bevc to take a look
[21:43:25] <markmentovai> leiz: we can check the kvm or do a power-cycle if we have to
[21:43:31] <phajdan-jr> maruel: 20+ jobs pending on both mac and windows trybots :(
[21:43:31] <rsesek> pinkerton: did you try clearing cache and clearing host cache in about:net-internals
[21:43:34] <eglaysher> so there's a lot of windows red, too.
[21:43:36] <tfarina> erg: running drover right now
[21:43:39] *** jshin has joined #chromium
[21:43:39] *** ChanServ sets mode: +v jshin
[21:43:41] <pinkerton> rsesek: i'll try now
[21:43:54] <jochen__> phajdan-jr: ping
[21:43:55] <eglaysher> either evan or rsimha's patch?
[21:43:56] <mmoss> eglaysher: I think windows red is still from grd changes
[21:44:08] <eglaysher> mmoss: grds cause compile failures?
[21:44:16] <mmoss> eglaysher: err, nope, sorry
[21:44:17] *** chronarion has joined #chromium
[21:44:22] <leiz> markmentovai: can you access the kvm?
[21:44:29] <markmentovai> if there is a problem, yo, i'll solve it
[21:44:45] <pinkerton> rsesek: didn't help
[21:44:51] <phajdan-jr> jochen__: pong
[21:44:56] <rsesek> pinkerton: have you tried a fresh profile?
[21:45:01] <jochen__> phajdan-jr: -> msg
[21:45:06] <pinkerton> i'll try that now
[21:45:30] <phajdan-jr> jochen__: didn't get any, my irc client might be broken/misconfigured. chat?
[21:46:03] <jochen__> sure
[21:46:20] <tfarina> erg: http://codereview.chromium.org/2835043/show
[21:46:25] <jochen__> use irssi :)
[21:46:46] <pinkerton> rsesek: weird, that worked
[21:47:15] *** estes has joined #chromium
[21:47:21] <pinkerton> and putting my old profile back no longer works
[21:47:48] <phajdan-jr> jochen__: I am, maybe I have misconfigured it
[21:48:06] <jochen__> you should see a [Act: 3] or something in the status bar?
[21:48:14] <rsesek> dmaclach:  "They are all RGB as far as illustrator is concerned. :/" -Cole
[21:48:19] <markmentovai> leiz: i can't get into the kvm, it's already at the max number of connections. try bevc.
[21:48:47] <markmentovai> leiz: forget it, it seems to be up now
[21:48:51] <jochen__> and esc-3 would change to that window
[21:48:51] <leiz> markmentovai: ya, it's back
[21:48:55] <markmentovai> must have gotten stuck doing a long fsck
[21:49:30] <mmoss> anybody got an idea on the Windows compiles?
[21:49:33] <tfarina> leiz: how I fix the revert? just patch in the old cl and fix whatever broke the tree?
[21:49:39] *** davirtavares has quit IRC
[21:49:43] <leiz> tfarina: I usually use drover
[21:49:55] <leiz> tfarina: you can ask a sheriff to do it too
[21:50:03] <tfarina> leiz: I already reverted :)
[21:50:08] *** ebraminio has joined #chromium
[21:50:08] *** senjin has joined #chromium
[21:50:25] <trungl-bot> Tree closed by mmoss at chromium dot org: Tree is closed - Linux Builder (Views dbg) -> tfarina reverted; Windows compile -> ???
[21:50:31] <pinkerton> rsesek: even cleared cookies, didn't help
[21:50:34] <markmentovai> tfarina: you can drover into a tree
[21:50:38] * dmaclach virtually smacking cole around for rsesek
[21:50:58] <markmentovai> mmoss: windows is xml. evmar_afk?
[21:51:00] <tfarina> markmentovai: what you mean?
[21:51:30] <markmentovai> tfarina: you can tell drover to do a revert into a checked-out working copy, instead of doing its own thing and committing it. you can use that to back out your back-out locally
[21:51:33] <jamesr> windows is xml?
[21:51:40] <leiz> tfarina: depot_tools/drover
[21:51:42] <jamesr> explains a lot, actually
[21:51:56] <markmentovai> jamesr: maybe those are just warnings? i see a lot of xml at the tail of the compile log.
[21:52:30] <mmoss> markmentovai: the link failures are due to xml?
[21:52:38] *** fantasticulous has quit IRC
[21:52:44] <markmentovai> mmoss: not sure.
[21:52:55] <markmentovai> there are also errors about V8DerivedSources
[21:53:29] <mmoss> markmentovai: yeah, neither am I. I'm going to see what the current compile does.
[21:53:38] <markmentovai> a subsequent compile on one of the windows builders also failed
[21:53:50] <tfarina> markmentovai: revert the revert? I don't want to do everything again, I just want to upload the patch again, with the fix for the linux_view.
[21:54:10] <rsimha> mmoss: I looked at the build errors -- they seem unrelated to my patch
[21:54:21] <mmoss> rsimha: agreed
[21:54:34] <markmentovai> tfarina: you can use drover to automatically back out your backout in your working copy, and then you can make changes to fix linux_views. then you can upload that.
[21:55:10] <rsesek> dmaclach: you talking to Cole?
[21:55:16] <dmaclach> rsesek: yep
[21:55:19] <rsesek> thanks
[21:55:40] <tfarina> markmentovai: how I can do that?
[21:56:14] <markmentovai> tfarina: use --local with drover. type drover with no arguments and it will print some help.
[21:56:29] <mmoss> hmmm, the xml changes worked on the trybot though
[21:56:34] <mmoss> maybe I'll try a clobber
[21:57:27] <tfarina> markmentovai: you mean to merge a specific revision into my working branch?
[21:57:41] *** johnny_g has quit IRC
[21:57:42] <tfarina> markmentovai: drover --merge #rev --local?
[21:57:52] <markmentovai> can you use --revert with --local? that's what i mean.
[21:58:58] <tfarina> markmentovai: not sure, the examples doesn't show that.
[21:59:16] <oshima> tfarina: aren't you using git? git revert <commit>?
[21:59:39] <tfarina> oshima: yep, using git.
[21:59:51] <eglaysher> tfarina: then just cherry-pick your previous patch
[22:00:16] <oshima> tfarina: find your commit using git log, then git revert <commit>
[22:00:35] <tfarina> eglaysher: I never used cherry-pick before, put I will take a look.
[22:00:45] <csilv> dmaclach, mmoss, i have a low-priority grd change ready to go.  i can wait to piggyback on someone else's grd change if helpful.
[22:01:19] <markmentovai> mmoss: just clobbering the one?
[22:01:29] <mmoss> markmentovai: for now to test
[22:01:40] <rohitrao> rsesek: are there mocks for the wrench menu update notification?
[22:01:44] <dmaclach> csilv: if you can piggyback that would be great. Anybody else got grd changes lined up?
[22:01:45] * pinkerton finds more regressions
[22:01:46] <markmentovai> mmmkay
[22:01:47] <mmoss> csilv: sorry, need to wait a bit
[22:01:50] <pinkerton> wow ToT is ass
[22:01:56] <rohitrao> rsesek: we just saw it on linux and couldn't figure out what the yellow bubble meant :)
[22:02:00] <tfarina> oshima: inside the branch I used to do the patch?
[22:02:06] <csilv> mmoss: np... in no rush.
[22:02:10] <rsesek> rohitrao: yes. that little bubble clearly means "update available
[22:02:28] <markmentovai> this is exactly why i didn't want to do the ui work on this one
[22:02:30] <oshima> oshima: create new branch, rebase/pull
[22:02:33] <rohitrao> my sarcasm detector is failing
[22:02:33] <rsesek> rohitrao: it's supposed to pulse, too, but I don't think that's gonna make M6
[22:02:36] <rohitrao> or maybe not
[22:02:40] <rsesek> the ball is gonna get it
[22:02:47] <oshima> new branch should have your commit.
[22:03:07] <rohitrao> rsesek: I think I would've gotten it faster if the menu item had the same bubble
[22:03:10] <oshima> or the one you were working on, but i usually create new branch.
[22:03:12] <tfarina> oshima: oh I see, pull, checkout new branch, the revert to the commit ?
[22:03:19] <oshima> yes
[22:03:20] <rsesek> rohitrao: the menu item has this ltitle checkbox thing
[22:03:23] <rsesek> er exclamation thing
[22:03:27] <markmentovai> rohitrao: i think on linux it gets the orange ! icon
[22:03:33] <rsesek> rohitrao: http://cl.ly/87272d74c840a39a3c2c
[22:03:39] <markmentovai> yeah, that
[22:03:45] *** Murray_B has joined #chromium
[22:03:48] <rohitrao> ah, that's not so bad then
[22:03:59] <oshima> tfarina: you should be able to use the branch you were on, but just in case, use clean branch.
[22:03:59] <tfarina> oshima: thanks I will try that. Then I need to get a new review from the original reviewer?
[22:04:20] <oshima> tbr is fine.
[22:04:32] <tfarina> oshima: yeah, I will use a clean branch.
[22:04:32] <Murray_B> Hi I am looking for a bookmark-menu without the bookmark-bar but couldn't find it - how can I do this in chromium?
[22:04:57] <pinkerton> Murray_B: unless you're on mac, you're out of luck
[22:05:07] <tfarina> oshima: just making sure the trybots passes, then tbr will be fine?
[22:06:17] <rsesek> Murray_B: there's an extension that replicates the functionality
[22:06:18] <Murray_B> pinkerton: You are kidding, aren't you? They don't provide a real menu to save space but you are forced to enable the bookmark-bar to get the bookmark-button?
[22:06:20] <oshima> depends on your change, but for revert case, people just do tbr. or just paste URL here and ask sheriff to review.
[22:07:14] *** chocobo__ has quit IRC
[22:08:56] *** motownavi has quit IRC
[22:09:01] *** evmar_afk is now known as evmar
[22:09:04] <evmar> did i break stuff?
[22:09:28] <markmentovai> uncertain.
[22:09:55] <evmar> i see, lots of libxml warnings, perhaps
[22:10:01] <evmar> but i think i may have seen those before?
[22:10:16] <markmentovai> yeah, i'm not sure if they?re ?new? or even ?broken?
[22:10:26] <evmar> it looks the real failure is this v8bindings stuff
[22:10:35] <markmentovai> mmoss has clobbered a win builder to see if that helps
[22:11:25] *** nickcarter has quit IRC
[22:11:44] *** shreyas_ has joined #chromium
[22:12:22] <evmar> phajdan-jr: http://src.chromium.org/viewvc/chrome/trunk/src/third_party/libxml/patches/libxml-disable-autodetect?view=markup&pathrev=52725
[22:12:25] <evmar> phajdan-jr: is the patch i was talking about
[22:12:34] <evmar> i don't think upstream would like it?
[22:12:39] <mmoss> evmar: I think the trybot had those warnings too, but it compiled
[22:12:46] <mmoss> we'll see what the clobber does
[22:13:17] <tfarina> oshima: git revert <commit-ish> didn't make anything as far as I can tell. I just said "Finished one revert. "
[22:13:46] <phajdan-jr> evmar: I see. But there are other patches, right? Have we sent them upstream?
[22:14:09] *** rohan has joined #chromium
[22:14:30] <rohan> the latest dev channel release for linux still doesn't enable consolidated menu by default?
[22:14:34] <oshima> it's already commited
[22:14:43] <oshima> try git diff
[22:14:45] <oshima> or just upload it
[22:14:56] *** AaronMT has quit IRC
[22:15:01] <eglaysher> rohan: IIRC this is the second dev channel with it on by default.
[22:15:31] <oshima> tfarina: i mean, it's committed to your branch
[22:15:35] <phajdan-jr> evmar: oh, and btw the webkit bug it points to seems to be inactive. :( can we move forward at least with that webkit patch?
[22:15:44] <rohan> eglaysher: oh, well the googlechromereleases blogspot did not mention it, so i thought it's not enabled for linux yet
[22:16:06] *** AaronMT has joined #chromium
[22:16:06] *** motownavi has joined #chromium
[22:16:06] *** ChanServ sets mode: +v motownavi
[22:16:17] <phajdan-jr> sheriffs, the tree is closed suspiciously long. Can I help?
[22:16:22] <phajdan-jr> dmaclach: ^
[22:16:25] <phajdan-jr> mmoss: ^
[22:16:28] <phajdan-jr> tyoshino: ^
[22:16:42] <mmoss> phajdan-jr: waiting on windows clobber
[22:16:56] <markmentovai> tree status should reflect
[22:16:56] <phajdan-jr> mmoss: ok, could you update the status then?
[22:17:11] <mmoss> done
[22:17:20] <markmentovai> trungl-bot: show me the money!
[22:17:21] <trungl-bot> markmentovai: Error: "show" is not a valid command.
[22:17:41] <tfarina> oshima: nope, that didn't work, no git diff output, and git cl upload didn't work either.
[22:17:41] <trungl-bot> Tree closed by mmoss at chromium dot org: Tree is closed - Linux Builder (Views dbg) -> tfarina reverted; Windows compile -> trying clobber
[22:17:49] <evmar> it's not clear to me why any of those changes would affect win
[22:18:17] <evmar> http://build.chromium.org/buildbot/waterfall/builders/Chromium%20Builder%20(dbg)/builds/22353
[22:18:23] <evmar> the stdio there is rather clear
[22:18:24] <oshima> tfarina: hold on...
[22:19:14] <oshima> tfarina: i just created cl: http://codereview.chromium.org/2836056
[22:19:21] <oshima> does this look good?
[22:19:46] *** pinkerton has quit IRC
[22:19:52] <markmentovai> evmar, mmoss: looks like it successfully linked chrome.dll after the clobber
[22:19:56] *** Murray_B has quit IRC
[22:20:07] <markmentovai> evmar: your xml warnings are still present, but none of those V8DerivedSources errors.
[22:20:15] <oshima> tfarina: i can submit if you don't mind.
[22:20:35] *** Venom_bbiab is now known as Venom_X
[22:20:50] <tfarina> oshima: I'm not seeing the fix for linux_view though.
[22:20:59] <phajdan-jr> markmentovai: do you have a while to help me write correct gyp solution?
[22:21:10] <markmentovai> a correct gyp solution for what?
[22:21:28] <phajdan-jr> markmentovai: for using a system library when we have multiple targets
[22:21:36] <phajdan-jr> markmentovai: I'm trying to come up with the right "design"
[22:21:44] <markmentovai> phajdan-jr: sounds better suited to email
[22:22:01] <oshima> tfarina: oh, sorry may be my misunderstanding.
[22:22:36] <oshima> tfarina: i thought you want to revert 52721, which you already did.
[22:22:48] <phajdan-jr> markmentovai: okay, so just some basic info for now. Could you look at third_party/libpng/libpng.gyp and see how it handles use_system_libpng?
[22:23:15] <markmentovai> yes, i'm aware of how that works
[22:23:16] <phajdan-jr> markmentovai: I'd like to do something similar in third_party/icu/icu.gyp, but it has more targets. Do you have any suggestions where to start?
[22:24:04] <markmentovai> phajdan-jr: does anyone ever depend on the individual icu component libraries singly, or does everyone always depend on the same single "monster" that brings in all of icu?
[22:24:13] *** Viclame has left #chromium
[22:24:43] <trungl-bot> Tree opened by mmoss at chromium dot org: Tree is open - Linux Builder (Views dbg) -> tfarina reverted
[22:25:38] <phajdan-jr> markmentovai: others depend on selected targets inside icu
[22:26:04] *** shreyas_ has quit IRC
[22:27:58] <markmentovai> phajdan-jr: and do those map correctly to the set of libraries available with system icu?
[22:29:53] *** Omen_20 has quit IRC
[22:30:17] <phajdan-jr> markmentovai: seems to be so, but I'd probably have to hardcode some linker flags rather than using icu-config
[22:30:37] <markmentovai> phajdan-jr: not necessarily. what i'm getting at is...
[22:30:43] *** fta_ has joined #chromium
[22:30:59] <markmentovai> if use_system_libicu, provide a none-type target for each library that chrome normally exposes via icu.gyp
[22:31:14] <markmentovai> phajdan-jr: have each of those depend on a single target that does the pkg-config stuff
[22:31:22] *** shreyas_ has joined #chromium
[22:31:38] <markmentovai> phajdan-jr: if it doesn't work with none-type targets, make them dummy staticlibs, and provide a single empty .c file in each.
[22:32:10] *** shreyas_ has quit IRC
[22:32:18] *** fta has quit IRC
[22:32:27] *** fta_ is now known as fta
[22:32:39] <phajdan-jr> markmentovai: okay, should that single target be in build/linux/system.gyp or third_party/icu/icu.gyp? Where should I define use_system_icu variable?
[22:33:48] <markmentovai> phajdan-jr: same as was done with libpng: icu.gyp is the answer to both questions
[22:34:17] <akalin> yay open tree
[22:34:23] <davidben> So, git-cl seemed to get slightly confused about issues so my latest commit has completely the wrong message. Is there a standard way of dealing with this?
[22:34:33] <davidben> s/issues/CLs/
[22:34:36] <markmentovai> phajdan-jr: talk to jshin about this proposal, too. in the past, there were reasons to use our own copy of icu as opposed to the system's.
[22:35:02] *** eseidel has joined #chromium
[22:35:02] *** ChanServ sets mode: +v eseidel
[22:35:44] <phajdan-jr> markmentovai: thanks. I know that the bundled icu has some fixes, but distros really want the system one. I think I can also help with upstreaming our custom patches if needed/possible.
[22:35:46] <oshima> davidben: i never had issue. which cl/commit it is?
[22:36:10] <jochen__> mmoss: btw, i'm still working on the two failing tests
[22:36:12] <markmentovai> phajdan-jr: ok, then i suspect we won't make this the *default* like we did with libpng, or at least not for branded builds.
[22:36:35] <evmar> davidben: fixing the bug seems the normal course of action
[22:37:01] <phajdan-jr> markmentovai: sure, I don't expect that to happen. :)
[22:37:25] <jshin> phajdan-jr: smailler pieces are mostly upstreamed. the biggest chunk is Chinese/Japanese segmentation support.
[22:37:38] <davidben> oshima: r52735. I think what happened was that I was testing someone's patch at some point on the wrong branch, so it's not necessary a git-cl bug.  My question is rather if there's a standard place to note that "this commit makes sense, but it's for the wrong CL".
[22:37:39] <jshin> phajdan-jr: I'm going to do that for this quarter
[22:38:00] <evmar> davidben: i think you can edit the commit message in svn
[22:38:05] <jshin> phajdan-jr: however, that won't be picked up by Linux distros soon enough
[22:38:08] <evmar> davidben: requires admin access to the server tho
[22:38:48] <phajdan-jr> jshin: depends which distros. :) Gentoo will probably get it pretty fast. Anyway, I think I can work on an option to build with system icu, and then each distro will make its own decision.
[22:39:33] <evmar> phajdan-jr: it makes me a little sad that people will end up with broken chromes and not know it's the fault of gentoo and not us
[22:39:45] <davidben> evmar: svn lets you do that? Huh. Whom do I poke to change that?
[22:40:07] <evmar> davidben: not sure, i don't think we've ever done it.  i would poke around on google and then track down an admin if you really wanted
[22:40:13] <oshima> hmm, i'd create new branch from it.
[22:40:17] <evmar> davidben: it prompts you before committing with the commit message, was that not enough?
[22:40:30] <jamesr> it's possible in SVN but basically never worth it
[22:40:38] *** atwilson has quit IRC
[22:40:46] <oshima> davidben: then you can assign it to different issue or create new issue.
[22:40:58] <akalin> jamesr: what's the cost
[22:41:16] <davidben> evmar: Yeah, I apparently didn't notice that until later. :-/ I did verify the diffstat to make sure that was correct. Should remember to check the message too in future...
[22:41:37] <evmar> i wonder how to improve the ui.  i think it's a ui failure rather than an operator error
[22:41:41] <oshima> davidben: oh, you wan to change the one already submitted. sorry, i misunderstood.
[22:41:45] <phajdan-jr> evmar: hmm, what kind of brokenness it is? I'd like both chromium and distros to be happy.
[22:41:50] <evmar> perhaps make it something like:
[22:41:58] <jshin> phajdan-jr: there are always some small pieces that we're ahead of ICU release (mostly locale data - word/line/char break iterators, various names). So even if Gentoo picks the latest release, they'll be behind a bit
[22:41:59] <evmar> commit the change "<first line of commit message here>"? Y/n
[22:42:17] <jamesr> akalin: breaks everything that depends on SVN logs not changing. for example, it'd break the git mirror
[22:42:23] <phajdan-jr> jshin: this applies to the general issue of bundled vs. system libraries
[22:42:34] *** shepazu has quit IRC
[22:42:40] *** maikmerten_ has quit IRC
[22:42:43] <akalin> jamesr: yeah that sucks
[22:43:04] <akalin> but who uses git anyway??!
[22:43:11] <jamesr> akalin: lusers
[22:43:17] <akalin> srsly
[22:43:21] <jshin> phajdan-jr: distros can apply our patches before packaging..
[22:43:59] <phajdan-jr> jshin: if they're going to appear in the next upstream release, they probably will decide not to do so
[22:44:19] <davidben> jamesr: Good point. I can make a dummy commit that says "that commit message should have been ...". Or I could just drop it, and we'll have an amusing svn log. :-)
[22:44:40] <jamesr> davidben: does the commit you checked in point to a chromium bug?
[22:45:04] <jamesr> davidben: you could also update the bug entry it points to to say "if you got here from the SVN log for r2398202, you probably want to actually look at HERE"
[22:45:07] <davidben> jamesr: Yes. Oh, ew... it probably updated the bug too. I reverted the changes to the CL.
[22:47:14] <markmentovai> davidben: are you a ssl generalist now? the sort of guy we might be able to assign mac ssl bugs to as they arise?
[22:48:28] *** ROBOd has quit IRC
[22:48:36] <tfarina> why this line is always shown when make runs? webkit/test_shell_common.target.mk:8: warning: overriding commands for target
[22:49:00] <trungl-bot> Tree closed by dmaclach at chromium dot org: Tree is closed - letting builders catch up
[22:49:06] <akalin> yeah so much gray
[22:49:22] <akalin> D:
[22:50:29] <jshin> phajdan-jr: yeah... Ubuntu Lucid has ICU 4.2.1 (the same as Chrome 5.x). My upstream change will not show up until ICU 4.6 (currently, the latest stable is ICU 4.4.x) So if they can apply important patches (most important one being CJ segmentation patch), it'll be good.
[22:51:04] <markmentovai> jshin: were you successful in getting the segmentation patch upstream?
[22:51:14] <mmoss> evmar: fixing libxml perms? http://buildbot.jail.google.com/buildbot/chromium/builders/Chromium%20Linux/builds/13929/steps/check%20perms/logs/stdio
[22:51:44] <jamesr> can you tell it's close to feature freeze?
[22:51:44] *** tlrobinson has joined #chromium
[22:51:59] <phajdan-jr> jshin: here's the state of icu in Gentoo: http://znurt.org/dev-libs/icu ; I'd call it pretty good. We only have chromium for amd64, ~arm and x86, so we're using 4.4 in all cases.
[22:52:01] <timsteele> dmaclach: i have a teeny grd change that needs to land, when do you think would be the best time?
[22:52:03] <markmentovai> mmoss: what's there to fix? those should have the x-bit!
[22:52:39] <phajdan-jr> jshin: if you have some patches that are important and not in the distro package, please just let me know and I'll see what we can do with it
[22:52:40] <jshin> markmentovai: they're willing to take patch (at least likes the idea), but i haven't managed to work on that in Q2, but I 'must' do that in Q3 (now)
[22:52:41] <mmoss> markmentovai: I dunno, but "check perms" disagrees
[22:52:55] <thakis> does someone here know how to commit to the git-cl repo
[22:52:56] <thakis> ?
[22:53:02] <dmaclach> timsteele... once we've caught up a bit. Can you find anybody else with grd changes?
[22:53:02] <markmentovai> check perms' whitelist needs to be kicked
[22:53:10] <mmoss> that works too
[22:53:18] *** rniwa has joined #chromium
[22:53:18] *** ChanServ sets mode: +v rniwa
[22:53:18] <phajdan-jr> markmentovai: is there 'dependent_settings' option in gyp similar to 'direct_dependent_settings'?
[22:53:32] <markmentovai> phajdan-jr: don't use it.
[22:53:41] <markmentovai> phajdan-jr: there's probably a better way to accomplish what you want.
[22:54:13] <markmentovai> phajdan-jr: use export_dependent_settings. see http://code.google.com/p/gyp/wiki/InputFormatReference
[22:54:15] <phajdan-jr> markmentovai: ok. I'm adding this artificial target for icu, that real ones will depend on. I'd like USE_SYSTEM_ICU preprocessor symbol to be defined in .cc files that depend on icu.
[22:54:50] <timsteele> dmaclach: hmm.. is there a good way to do that besides  ####################### CALLING ALL GRD CHANGES #######################
[22:55:28] <timsteele> :)
[22:55:36] <phajdan-jr> markmentovai: I don't see from that doc how to use export_dependent_settings. :(
[22:56:05] <jam2> does anyone know how webkit_strings.h is generated?  I can't find the grd file
[22:56:29] <markmentovai> phajdan-jr: if your icuuc target depends on a (made-up for this example) system_icu target, you would put in icuuc: 'export_dependent_settings': ['system_icu'] to further export whatever system_icu declared as a direct_dependent_setting.
[22:57:17] <thakis> brettw: ping
[22:57:37] <phajdan-jr> markmentovai: cool, and you even guessed the system_icu target name correctly :)
[22:57:42] <markmentovai> phajdan-jr: there are several examples of export_dependent_settings in the chromium .gyp files
[22:57:43] <jam2> ah nm, i see webkit_strings.grd now
[22:57:45] <markmentovai> phajdan-jr: excellent
[22:58:06] *** juanefren has joined #chromium
[22:59:55] <csilv> i have a grd change, should i go for it?
[23:00:13] <akalin> csilv: timsteele has one, too
[23:00:24] *** shepazu has joined #chromium
[23:02:17] <markmentovai> mmoss: have you clobbered webkit builder (dbg) yet. doesn't look like it's been clobbered prior to the build it's currently chewing on, but i don't want to press the button more than once in case you've already pressed it since.
[23:02:23] *** fta_ has joined #chromium
[23:02:45] <mmoss> markmentovai: thought I got them all, let me check
[23:02:57] <csilv> timsteele, ping me when you're about to put your grd change in, i'll piggyback.
[23:03:17] *** hrna has quit IRC
[23:03:23] <rohitrao> shess: is there anything else, or am I free for the weekend?
[23:03:43] *** fta has quit IRC
[23:03:56] *** fta_ is now known as fta
[23:04:13] <shess> rohitrao: I want to get the keyword-hint change up, yet.  what's your time-frame?  i've been debugging why my try jobs haven't been working (lkgr looks old).
[23:04:18] <timsteele> csilv: i'm ready to go, just waiting for the green light from sherrifs
[23:04:26] <timsteele> we could wait for the builders to catch up or land now and watch closely, which might be the fastest way to green
[23:05:05] <rohan> what is the timeframe when beta is going to be updated to 6 branch?
[23:05:15] <dmaclach> timsteele, do you promise to watch closely?
[23:05:25] <shess> rohitrao: mostly would want this one so that I have room to break something by Monday :-).
[23:05:28] <dmaclach> it could be a while...
[23:06:09] <rohitrao> shess: was gonna head home in 15m or so.  I can make time tonight or tomorrow if you'd like
[23:06:34] <evmar> rohan: wikipedia has our release history, we tend to release every few months
[23:07:03] <shess> rohitrao:  ok, that's fine.  as if your fiance cares whether I'm in agreement :-).
[23:07:12] *** dave_levin has joined #chromium
[23:07:12] *** ChanServ sets mode: +v dave_levin
[23:07:15] <rohitrao> heh
[23:07:19] <rohan> yes, but this time i think 6 has remained in dev only for too long. it should have been in -beta by now, evmar
[23:07:39] <shess> rohitrao: the keyword-hint things is 95% deleting code anyhow.
[23:07:45] <rohitrao> woot
[23:07:47] <evmar> rohan: really? how late are we
[23:08:01] <timsteele> I do, I'll be around
[23:08:43] *** skrul has quit IRC
[23:08:59] <rohan> evmar: i don't claim to know better than the devs :) it was just my random musing.
[23:09:06] <juanefren> using Chrome Inspector, is possible to configure it to show errors in ajax calls the same way than errors in the script?
[23:09:23] <dmaclach> timsteele, csilv: go ahead with grd changes please
[23:09:26] <rohan> i meant to ask if there is some sort of "roadmap" when beta will "become" 6
[23:09:58] <timsteele> dmaclach, csilv: landed.
[23:10:06] *** hsuh has quit IRC
[23:10:33] <evmar> rohan: no, only developers here
[23:11:02] *** MichealH has quit IRC
[23:11:07] <jcastro> evmar: howdy, do you have time next week to talk about appmenu-gtk and friends?
[23:11:10] *** eseidel has quit IRC
[23:11:48] *** michaeln has joined #chromium
[23:11:49] *** ChanServ sets mode: +v michaeln
[23:11:51] <csilv> dmaclach, timsteele: landed. 52747
[23:12:08] *** kermit has quit IRC
[23:13:06] *** MikeSmithW3C^ has joined #chromium
[23:14:54] <evmar> jcastro: sure
[23:15:07] *** loislo has quit IRC
[23:15:18] *** rafaelw_wfh has left #chromium
[23:16:09] *** rohan has quit IRC
[23:16:23] *** MikeSmithW3C has quit IRC
[23:18:21] *** ebraminio has quit IRC
[23:18:23] <estade> dmaclach: can I commit a simple patch that fixes a p0 bug: http://codereview.chromium.org/3007004/show
[23:18:29] <tommi> what's up with the trybots?  I'm getting patch failures on all of them for a simple one liner
[23:18:40] <tommi> and I'm fully synced
[23:18:46] <estade> tommi: maybe you're too synced
[23:18:59] <estade> tommi: they use LKGR, which is probably like, 3 weeks ago
[23:19:05] <estade> try it with --revision
[23:19:14] *** loislo has joined #chromium
[23:19:30] <tommi> ah, that could be the case, in which I'll have to wait for them to catch up since I'm modifying fairly new code
[23:19:44] <tommi> thanks
[23:19:46] <estade> tommi: no, you can tell them to use a newer revision with --revision
[23:19:46] <dmaclach> estade: go ahead
[23:19:50] <estade> dmaclach: thanks
[23:20:01] <tommi> oh, I see.  will try. thanks.
[23:21:54] *** General13372 has joined #chromium
[23:24:25] <jam2> sheriffs: i have a grd change that i need to get in somewhat soon, but i dont think it'll mess up the tree since the strings are at the end: http://codereview.chromium.org/3035005
[23:24:30] *** General1337 has quit IRC
[23:25:23] <dmaclach> jam2... you just missed the last grd train
[23:26:39] *** juanefren has left #chromium
[23:26:59] <jam2> dmaclach: if the strings are at the end, this shouldnt cause tests to fail tho, no?
[23:27:07] <mirandac> dmaclach: jam2's change should cause no problems, because he added the strings at the end, so the ordering won't be messed up
[23:27:17] <mirandac> dmaclach: ...which is what causes test failure
[23:28:19] <phajdan-jr> jshin: ping
[23:28:37] <dmaclach> jam2: go ahead
[23:29:03] *** loislo_ has joined #chromium
[23:30:31] <jshin> phajdan-jr: pong
[23:30:54] <phajdan-jr> jshin: how should we #include ICU libraries? For now I see three conventions in the codebase:
[23:30:58] <jam2> dmaclach: thanks, i'll wait for trybot first and also for the reviewer
[23:31:01] <phajdan-jr> jshin: a) #include "unicode/..."
[23:31:11] <phajdan-jr> jshin: b) #include <unicode/...>
[23:31:22] <phajdan-jr> jshin: c) #include "third_party/icu/..."
[23:31:58] *** mnissler_ has joined #chromium
[23:32:00] *** chrisccoulson has quit IRC
[23:32:05] *** loislo has quit IRC
[23:32:05] *** loislo_ is now known as loislo
[23:32:31] <jshin> phajdan-jr: a and b don't make any difference, do they? for 3rd, I don't understand why we do that (i heard something before, but forgot)
[23:34:14] <phajdan-jr> jshin: might be there a difference when we want to be sure to use system vs. bundled libraries? Should we do a) when we use bundled, and b) when we use system?
[23:34:20] <tfarina> jshin: the second search in the installed path no?
[23:35:10] *** chrisccoulson has joined #chromium
[23:36:22] <rsesek> is anyone else getting: WARNING trychange.py(326): ('http protocol error', 0, 'got a bad status line', None)
[23:36:23] <rsesek> ?
[23:36:32] *** mnissler_ has quit IRC
[23:37:32] <phajdan-jr> sheriffs, may I reserve a commit slot when the tree re-opens? I have a patch that re-shuffles some download code files a lot, and would like to land it to avoid possible merge conflicts later.
[23:37:36] <phajdan-jr> dmaclach: ^
[23:37:38] <tfarina> mmoss: may I land http://codereview.chromium.org/2812056/show?
[23:37:38] <phajdan-jr> mmoss: ^
[23:37:41] <trungl-bot> Tree opened by mmoss at chromium dot org: Tree is open
[23:37:53] *** estes has quit IRC
[23:37:54] <mmoss> phajdan-jr: I'm gonna go with yes :)
[23:37:55] <phajdan-jr> yay! commit, commit, commit!
[23:38:08] <dmaclach> be kind... it is Friday afternoon...
[23:38:34] *** magn3ts has quit IRC
[23:38:38] <estade> feature deadline is coming up... unleash hell!
[23:38:38] <phajdan-jr> watching the tree
[23:38:40] <jshin> tfarina: i guess you're right. for phajdan-jr's issue at hand (bundled vs system), I guess it doesn't matter
[23:39:16] <tfarina> dmaclach: permission to land http://codereview.chromium.org/2812056?
[23:39:31] <dmaclach> tfarina: tree is open
[23:40:08] <tfarina> dmaclach: just in case I need a explicit LGTM, ok. I guess it's safe so.
[23:40:56] <phajdan-jr> tfarina: could you please file a FlakyTest bug for the HighResNow flake you got from the win trybot and send the bug link to me?
[23:41:27] *** lnostdal has quit IRC
[23:42:05] *** Precea is now known as Precea[BNC]
[23:43:01] <rsesek> thakis: is my cl lg, pending trybot happiness?
[23:43:27] <thakis> rsesek: not too happy that we don't understand why this doesn't work with the model change for bounds
[23:43:47] <tfarina> dmaclach: committed
[23:43:53] <rsesek> thakis: it has no visual affect afaict
[23:43:56] <jshin> phajdan-jr: for important patches to ask distros to apply, I'll give the list offline (it's a bit involved because some of them are purely to cut down the build size, which is not necessary for distros. other patches have dual-purpose. still others add features/fix bugs and there are still other types). btw, converter patch is another important one.
[23:44:14] <phajdan-jr> dmaclach: we start seeing some test failures from past running tests, I suggest closing the tree for investigation
[23:44:22] <dcheng> How come I have a horizontal scrollbar in Gmail in Linux Chrome no matter how wide I make my window?
[23:44:36] <thakis> rsesek: "i know why this works" > "seems to work"
[23:44:45] <phajdan-jr> jshin: sure. Feel free to use either my @chromium or @gentoo address, or even file a Gentoo bug and send the link to me.
[23:44:52] <rohitrao> what cl is this?
[23:45:12] <phajdan-jr> tfarina: "TBR:brettw" is not the best commit message :(
[23:45:15] <tfarina> phajdan-jr: you want me to file a bug for the flakyness of HighResNow?
[23:45:27] <phajdan-jr> tfarina: yes please, I'll take care of it
[23:45:46] *** lnostdal has joined #chromium
[23:46:09] <tfarina> phajdan-jr: oh, that was not what I wanted
[23:46:11] <phajdan-jr> jshin: by the way, could you take a look at http://codereview.chromium.org/3031006/show ?
[23:46:13] <tfarina> :(
[23:46:31] <phajdan-jr> jshin: and what would be the steps to land it? Will I need to bump a rev number in some DEPS file?
[23:47:08] <phajdan-jr> tfarina: don't worry too much about the one you committed, but please try to write better commit messages next time (and it should have a codereview link)
[23:47:39] <rohitrao> rsesek: I've been killing animations with the delegate method instead.  -actionForLayer:forKey:
[23:47:52] <trungl-bot> Tree closed by dmaclach at chromium dot org: Tree is closed - Too much redness
[23:48:01] <dmaclach> phajdan-jr: what specifically are you looking at?
[23:48:10] <tfarina> phajdan-jr: sure, I just tried to used --tbr with dcommit. I thought the TBR line it will be for the email, and then it would include the cl description. I just didn't know.
[23:48:16] <akalin> I get an "error: invalid preprocessing directive #FIXME" error when compiling webcore_bindings_sources
[23:48:22] *** mrossetti has joined #chromium
[23:48:34] <akalin> anyonw know where that came from?
[23:48:35] <phajdan-jr> dmaclach: the Mac failures
[23:48:45] <phajdan-jr> dmaclach: on the console view it looks like a lot of redness
[23:48:54] <phajdan-jr> dmaclach: ui_tests and browser_tests
[23:48:56] <dmaclach> phajdan-jr: agreed looking
[23:49:13] <rsesek> rohitrao: as opposed to animationDidStop:finished:?
[23:49:42] <jshin> phajdan-jr: you folded everything inside condition use_system_icu=0 and added what's necessary for use_system_icu !=0, right?
[23:49:46] <rohitrao> rsesek: er, we might be talking about different things.  This was to get rid of implicit animations
[23:49:46] *** estes has joined #chromium
[23:49:54] <rohitrao> although if you're explicitly animating, that might not apply
[23:50:12] <phajdan-jr> jshin: yeah, no code changes inside use_system_icu=0 branch. Sorry it's non-obvious from the diff output.
[23:50:15] <rsesek> rohitrao: the issue we were having is that the explicit animations would take and then flash back to the original state b/c the model wasn't updated right before the animation window went away
[23:50:28] *** tonikitoo has quit IRC
[23:50:50] <markmentovai> what we need for this would be a diff -w for rietveld
[23:50:53] <rohitrao> huh
[23:50:58] <rohitrao> i hate animations :(
[23:51:32] <rsesek> the implicit+explicit trick works for everything but bounds
[23:51:35] <markmentovai> evmar: excellent work with the FilePathing!
[23:52:05] <evmar> markmentovai: thanks!  it took a really damn long time
[23:52:08] <jshin> phajdan-jr: windows will never system_icu (neither will Mac althouhg Mac OS X does have icu internally, but it's not exposed at all). we can treat Mac/Win separately from Linux, but there's not much merit in doing so (or is there? )
[23:52:12] <evmar> tfarina and others helped too
[23:52:12] <markmentovai> tell me about it :(
[23:52:27] <rohitrao> rsesek: does bounds work the way frame does?
[23:52:34] *** loislo_ has joined #chromium
[23:52:34] <evmar> markmentovai: i cheated at the end, they're all behind #ifdef OS_WIN now.  but at least that way people stop adding more code
[23:52:37] <rohitrao> I know frame is really made of frameOrigin and frameSize
[23:52:47] <tfarina> evmar: yep :)
[23:52:51] <markmentovai> evmar: the #ifdef at least keeps us pure on platforms where it matters
[23:53:11] <rohitrao> rsesek: try doing the trick, but override boundsOrigin and boundsSize
[23:53:15] <evmar> well...
[23:53:15] <evmar> $ git gs WStringHack | wc -l
[23:53:16] <evmar> 386
[23:53:18] <rsesek> rohitrao: bounds works with anchorPoint and position (we're also animating position)
[23:53:22] *** pkasting has quit IRC
[23:53:24] <evmar> still some work to be done :)
[23:53:25] <markmentovai> those must not matter :)
[23:53:39] <rsesek> rohitrao: there is no boundsOrigin or boundsSize
[23:53:48] *** loislo_ has quit IRC
[23:54:10] <rohitrao> rsesek: oh, they exist on NSViews :)
[23:54:19] <rsesek> arargasdgfasdf
[23:54:21] <rsesek> apple
[23:54:31] <phajdan-jr> jshin: the default is still 0, so I think there's no benefit to put some code to make it impossible to set use_system_icu on Windows
[23:54:40] <phajdan-jr> jshin: it's similar with other use_system_ libraries
[23:54:44] <jamesr> rsesek: sounds like you need a boundsOriginBumper
[23:55:08] <rsesek> indeed
[23:55:11] *** loislo has quit IRC
[23:55:34] <Nola> hi -- why does Chrome look so totally out of place in my Os?
[23:55:37] <Nola> its not themed
[23:56:12] *** pkasting has joined #chromium
[23:56:12] *** ChanServ sets mode: +v pkasting
[23:56:12] <jibot> pkasting is a well-known rabble-rouser and works on Google Chrome
[23:56:14] <Nola> i had to switch back to Firefox cause it just sticks out like a sore thumb
[23:56:50] <Nola> and none of the Chrome Themes look anything like my Os theme
[23:56:58] <trungl-bot> Tree closed by tony at chromium dot org (:tony^work): tree is closed - (layout tests -> ojan, browser_tests -> tony)
[23:57:02] <Nola> why is Chrome the *only* browser that cant do this?
[23:57:05] <tony^work> anyone want to look into the check perms failures?
[23:57:15] <tfarina> Nola: please see the topic
[23:57:16] <markmentovai> check perms -> evmar, update the perm checker whitelist
[23:57:23] <Nola> there literally is no other browser for windows *besides chrome* that is unable to do simple win32 theme rendering
[23:57:28] <jshin> phajdan-jr: i agree !
[23:58:28] *** senjin has quit IRC
[23:58:28] <phajdan-jr> jshin: cool. Now about the landing: when I commit it, I'll need to update the DEPS file to pull more recent version from the third_party repo, right?
[23:59:19] <Nola> i dont want bullshit, I dont want fruity colors, i dont want paperclips, i dont want soccerballs, I dont want nature, I dont want DJ tiesto, I dont want mariah cary, I dont want smiley anime kiddies, I dont want oriental rug pattern designs on my browser
[23:59:23] <Nola> i just want it to look like fucking windows
[23:59:30] <Nola> is that too much to ask of Chrome?
[23:59:53] <rohitrao> Nola: I don't know if it's too much to ask, but this is not the place to ask it.  Please see the channel topic





top