[00:01:54] <paulweb515_> rcjsuen: when you get a chance, could you try and replicate ~347857
[00:01:55] <Arbalest> Bug 347857 - https://bugs.eclipse.org/bugs/show_bug.cgi?id=347857 - e4 / UI / unspecified - PC / Linux - NEW / normal / - Assignee: e4.ui-inbox - [Compatibility] A close with 2 WW open causes NPEs
[00:06:30] <rcjsuen> ?
[00:06:44] <rcjsuen> paulweb515_: so just open two windows, open/close editors, leave one open close all other, try exit,y es?
[00:15:17] *** tschindl has quit IRC
[00:23:31] <paulweb515_> open 2 windows, one editor each, restart, close an editor, restart, open an editor in the window with no editor, restart
[00:39:44] <rcjsuen> not realy getting anywhere with my inner
[00:41:16] <paulweb515_> rcjsuen: I'll try and repo it tomorrow, and investigate the stack traces ... maybe this is one of those things where IEventBroker isn't deterministic in handing out the events
[00:45:42] <rcjsuen> Looks like it's trying to create a workbench window on shutdown
[00:50:08] <paulweb515_> rcjsuen: yes, we have code where our first getWW() actually models one ... apparently we don't want that on shutdown :-)
[00:50:23] <paulweb515_> we're setup to run a build today
[01:44:45] *** hmalphettes has joined #eclipse-e4
[02:25:42] *** hmalphettes has quit IRC
[02:33:20] *** borisb6i has joined #eclipse-e4
[02:40:46] *** hmalphettes has joined #eclipse-e4
[03:21:54] *** hmalphettes has quit IRC
[03:37:11] *** hmalphettes has joined #eclipse-e4
[04:06:48] *** borisb6i has quit IRC
[04:39:41] *** rcjsuen has quit IRC
[07:27:58] *** tschindl has joined #eclipse-e4
[08:00:24] *** tschindl has quit IRC
[08:20:02] *** tschindl has joined #eclipse-e4
[10:47:10] *** hmalphettes has quit IRC
[10:49:10] *** hmalphettes has joined #eclipse-e4
[11:33:22] *** rcjsuen has joined #eclipse-e4
[12:11:33] *** hmalphett has joined #eclipse-e4
[12:12:49] *** hmalphettes has quit IRC
[12:40:48] *** hmalphett has quit IRC
[12:52:43] *** hmalphettes has joined #eclipse-e4
[13:02:51] *** rcjsuen has quit IRC
[13:27:17] *** rcjsuen_ is now known as rcjsuen
[13:28:13] *** hmalphettes has quit IRC
[14:11:56] *** hmalphettes has joined #eclipse-e4
[14:29:29] *** hmalphettes has quit IRC
[15:05:32] *** borisb6i has joined #eclipse-e4
[15:49:38] *** borisb6i has quit IRC
[15:52:31] *** hmalphettes has joined #eclipse-e4
[15:53:25] <rcjsuen> Can anyone open the model editor with Ctrl+3?
[15:53:54] <rcjsuen> I can't open it in an inner
[15:56:21] <paulweb515> rcjsuen: what's the command supposed to look like?
[15:56:33] <rcjsuen> Live Application Model
[15:56:35] <rcjsuen> is what the view is called
[15:57:04] <paulweb515> does it open for ALT+SHIFT+F9 in your inner?
[15:57:34] <paulweb515> In my outer, that opens a view in my bottom stack
[15:57:37] <rcjsuen> not really
[15:57:40] <paulweb515> CTRL+3 I mean
[15:57:57] <rcjsuen> outer it works yes
[16:00:06] <rcjsuen> It doesn't seem to think there's a @PC method
[16:02:18] <paulweb515> In my inner, I use CTRL+3 Live and hit enter ... it appears to open the view in my bottom right stack
[16:02:29] <paulweb515> rcjsuen: is that not what it is supposed to do?
[16:02:40] <rcjsuen> paulweb515: sure it's supposed to do that, but it NPEs on my inner ;)
[16:02:56] <paulweb515> rcjsuen: that's less good :-)
[16:03:43] <rcjsuen> And Alt+Shift+F9 gives me an empty shell :)
[16:52:37] *** borisb6i has joined #eclipse-e4
[17:29:26] <emoffatt> paulweb515: the patch is now on ~347940
[17:29:27] <Arbalest> Bug 347940 - https://bugs.eclipse.org/bugs/show_bug.cgi?id=347940 - e4 / UI / unspecified - PC / Windows XP - NEW / critical / - Assignee: e4.ui-inbox - [Compatibility] Timer execs greatly affect XWindow based systems
[17:30:52] <paulweb515> emoffatt: thanx
[17:32:01] <emoffatt> paulweb515: Go ahead and commit if you think it's OK but let me know so I can update the wiki
[17:32:28] <paulweb515> emoffatt: OK
[18:07:59] <rcjsuen> paulweb515: Okay so you just used extract method to make the ISR into a getter
[18:08:04] <rcjsuen> at least, so it seems on my computer
[18:36:17] *** tschindl has quit IRC
[18:40:52] *** hmalphettes has quit IRC
[19:10:49] *** tschindl has joined #eclipse-e4
[19:46:54] <rcjsuen> paulweb515: can you repro that shutdown NPE problem consistently?
[19:47:12] <rcjsuen> i have an idea for a potential fix but...
[19:49:26] <emoffatt> paulweb515: should we mark ~347940 as FIXED ?
[19:49:27] <Arbalest> Bug 347940 - https://bugs.eclipse.org/bugs/show_bug.cgi?id=347940 - e4 / UI / unspecified - PC / Windows XP - NEW / critical / - Assignee: emoffatt - [Compatibility] Timer execs greatly affect XWindow based systems
[19:49:56] <paulweb515> emoffatt: yes, I believe so ... then we should get Car to retest some time at the end of the week
[19:50:28] <emoffatt> yes, she's already agreed to take another pass once I can tell her which build to try...
[19:50:34] <paulweb515> emoffatt: OK, that's it then
[20:09:22] *** susanmccourt has quit IRC
[20:16:32] <paulweb515> rcjsuen: yeah, usually within about 3 session restarts
[20:17:16] <rcjsuen> emoffatt: there was alrdy a bug open actually ~347561
[20:17:18] <Arbalest> Bug 347561 - https://bugs.eclipse.org/bugs/show_bug.cgi?id=347561 - e4 / UI / 1.0 - PC / Windows 7 - NEW / major / - Assignee: e4.ui-inbox - Cannot create new stacks in detached windows
[20:17:42] <emoffatt> kk, sry...mark that one as a dup I guess
[20:18:24] <emoffatt> I'll do it
[20:20:15] <paulweb515> rcjsuen: if you have a potential fix, I can try it
[20:24:51] <rcjsuen> emoffatt: Is it easier to disable outer docking or to enable it? As it requires additional steps to move stacks around.
[20:25:17] <paulweb515> rcjsuen: should we just move windowsClosed = true; to before we call ww.close(false) ?
[20:25:47] <emoffatt> rcjsuen: how do you mean...?
[20:26:01] <rcjsuen> well some ppl ref getAWW during shutdown so that is hazardous
[20:26:16] <rcjsuen> paulweb515: I was thinking of destroying non "top" windows first, then going for the application's selected element last
[20:27:11] <rcjsuen> emoffatt: if you want to make a stack left of two existing stacks stacked vertically you need two actions instead of one
[20:28:57] <emoffatt> yes, that's one of the consequences...I'm sure there's likely some more polish on this TBD in 4.2  :)
[20:29:43] <emoffatt> 'outerDock' is really all about moving between the different levels of MPSC's in the UI
[20:30:01] <rcjsuen> paulweb515: i mean something like this http://pastebin.ca/raw/2073337
[20:34:43] <paulweb515> rcjsuen: but won't that last close simply cause the same problem?
[20:34:56] <rcjsuen> paulweb515: the getAWW call will return itself
[20:35:15] <rcjsuen> the problem i was seeing was that a non-selected element's close caused the getAWW call
[20:35:22] <rcjsuen> but the selected element was alrdy destroyed
[20:35:24] <rcjsuen> so it tried to spawn a WW
[20:35:57] <paulweb515> rcjsuen: OK, that makes sense.  In my first 2 cases, I exited from window2 so it was selected
[20:36:07] <paulweb515> rcjsuen: in the last case, I exit from window1 and cause the problme
[20:36:11] <paulweb515> uh, problem
[20:38:07] <rcjsuen> wonder if i can induce the pb again
[20:38:12] <rcjsuen> i haven't had any luck for a while >_>
[20:38:28] <paulweb515> for me it's fairly reliable
[20:43:03] <paulweb515> rcjsuen: could you attach your patch as well, I'm just testing it out now
[20:51:45] <rcjsuen> I think it is working...for some value of "think".
[20:54:17] <rcjsuen> well i repro'd it twice in a row and after the patch it didn't happen again anyway
[20:57:22] <rcjsuen> emoffatt: While you're committing the SDA fix, can you add the finally clause mentioned in ~340358c5, thx
[20:57:29] <rcjsuen> ~340358c5
[20:57:29] <Arbalest> Bug 340358 - https://bugs.eclipse.org/bugs/show_bug.cgi?id=340358#c5 - e4 / UI / unspecified - PC / Windows XP - RESOLVED / FIXED / normal / - Assignee: e4.ui-inbox - Trim can go empty after maximizing a part that was restored after being minimized from a maximized shared area
[20:58:36] <paulweb515> rcjsuen: yes, I can repo reliably, but after the patch I can't repo
[20:59:17] <paulweb515> rcjsuen: is the patch you attached identical to your pastebin?
[20:59:34] <rcjsuen> paulweb515: I'll double check lol
[21:00:08] <rcjsuen> paulweb515: Yes they are identical
[21:00:27] <rcjsuen> well, except the time ;)
[21:00:30] <rcjsuen> +++ Eclipse UI/org/eclipse/ui/internal/Workbench.java	1 Jun 2011 18:40:16 -0000
[21:00:47] <paulweb515> rcjsuen: that I can live with
[21:01:15] <paulweb515> I'll +1 your patch, can you commit and close the bug?
[21:01:24] <paulweb515> I'll announce on e4-dev
[21:01:31] <paulweb515> Does that sound OK?
[21:02:07] <rcjsuen> paulweb515: Okelydokely
[22:11:30] <emoffatt> rcjsuen: is there a bug open for that bogus stack DnD ?
[22:11:35] <rcjsuen> i think so
[22:11:41] <rcjsuen> ~347831
[22:11:42] <Arbalest> Bug 347831 - https://bugs.eclipse.org/bugs/show_bug.cgi?id=347831 - e4 / UI / 1.0 - PC / Windows XP - NEW / normal / - Assignee: e4.ui-inbox - IAE thrown when dropping first view on itself
[22:11:46] <rcjsuen> emoffatt: ^
[22:32:56] *** borisb6i has quit IRC
[22:55:08] *** rcjsuen_ has joined #eclipse-e4
[22:55:54] <rcjsuen_> paulweb515_: Eric delivered for ~347831 but just added it to the wiki page without emailing the list
[22:55:54] <Arbalest> Bug 347831 - https://bugs.eclipse.org/bugs/show_bug.cgi?id=347831 - e4 / UI / 1.0 - PC / Windows XP - NEW / major / - Assignee: e4.ui-inbox - IAE thrown when dropping first view on itself
[22:56:11] <rcjsuen_> i assumed the list was "static" until we update it on thursday meetings
[22:56:20] <rcjsuen_> and thot all other bugs had to be via email
[22:56:37] <rcjsuen_> but i don't think we emailed for ~347858
[22:56:38] <Arbalest> Bug 347858 - https://bugs.eclipse.org/bugs/show_bug.cgi?id=347858 - e4 / UI / unspecified - PC / Linux - NEW / normal / - Assignee: e4.ui-inbox - [Compatibility] a lot of NPEs in timer execs
[22:56:39] <rcjsuen_> so....
[22:56:40] * rcjsuen_ shrugs.
[23:29:54] *** rcjsuen has quit IRC
[23:29:56] *** rcjsuen_ is now known as rcjsuen
[23:30:27] *** rcjsuen_ has joined #eclipse-e4





top