[00:30:08] *** briandealwis has quit IRC
[01:48:18] *** rcjsuen has quit IRC
[01:48:53] *** rcjsuen has joined #eclipse-e4
[02:01:11] *** rcjsuen__ has joined #eclipse-e4
[02:03:51] *** rcjsuen has quit IRC
[02:04:05] *** rcjsuen has joined #eclipse-e4
[02:06:42] *** rcjsuen__ has quit IRC
[02:12:18] *** rcjsuen__ has joined #eclipse-e4
[02:12:19] *** rcjsuen has quit IRC
[02:33:52] *** rcjsuen has joined #eclipse-e4
[02:36:10] *** rcjsuen__ has quit IRC
[02:47:31] *** rcjsuen__ has joined #eclipse-e4
[02:50:44] *** rcjsuen has quit IRC
[03:25:41] *** rcjsuen has joined #eclipse-e4
[03:27:56] *** rcjsuen__ has quit IRC
[03:31:38] *** rcjsuen__ has joined #eclipse-e4
[03:34:31] *** rcjsuen has quit IRC
[03:48:09] *** rcjsuen__ has quit IRC
[05:53:22] *** magnet has quit IRC
[05:55:54] *** magnet has joined #eclipse-e4
[13:30:57] *** rcjsuen_ is now known as rcjsuen
[13:55:39] *** briandealwis has joined #eclipse-e4
[14:14:12] *** briandealwis has quit IRC
[14:16:49] *** briandealwis has joined #eclipse-e4
[14:45:34] *** yarthorn has joined #eclipse-e4
[15:17:57] <emoffatt> briandealwis: no problem...the email was to elicit this type of info about pending commits
[15:18:56] <emoffatt> it certainly isn't a limitation on whether a defect is a candidate for 4.1.1, just making sure we get the timing correct so that folks use the correct branch to commit on
[15:19:23] <briandealwis> emoffatt: always a good thing :)
[15:19:24] <briandealwis> amoff
[15:19:54] <briandealwis> oops, I meant to say: I'm a bit confused as to the difference in the commits on master vs R4_development
[15:23:33] <emoffatt> I *think* master is verbotten
[15:24:05] <briandealwis> oh!
[15:24:34] <emoffatt> so we should only be committing to R4_development (and cherry picking for now over to R3_development for fixes in 'common' bundles
[16:24:11] *** tschindl has joined #eclipse-e4
[16:26:49] <tschindl> emoffatt: is it somehow possible in e3 to make the system not hide a view stack one the last view in there is closed?
[16:27:48] <emoffatt> tschindl: the control for that is in the CleanupAddon...it has a few assumptions in it that are there to support the compat layer
[16:28:08] <tschindl> well in e4 it there but in e3?
[16:28:43] <emoffatt> ooooh...in e3 I don't think so unless you write a complete presentation
[16:29:07] <tschindl> well i have my own presentation already
[16:29:27] <emoffatt> BTW, are my abstracts OK now ?
[16:29:37] <tschindl> ... and it might make a good addition to e4 so that one can flag a stack to be not removed when the last entry is gone
[16:30:25] <emoffatt> +1, tags are our friend here...we could just have a 'No Close if empty tag or somesuch
[16:30:52] <tschindl> in rcp it is quite a common thing to have one area with views who act similar than the editor area but the areas content is different from perspective to perspective
[16:31:00] <tschindl> let me file a bug for this
[16:31:15] <rcjsuen> interesting
[16:31:19] <rcjsuen> that thought has never crossed my mind
[16:31:21] <rcjsuen> to make a fake editor area
[16:31:24] <emoffatt> there are two cases actually...'no hide'  and 'no remove'  one would leave the stack there but set TBR to false, the other would actually remove it
[16:31:48] <tschindl> well rcjsuen in RCP this is quite a common thing
[16:32:04] <tschindl> this is why most applications there are using ViewPart + ISaveablePart2
[16:32:32] <emoffatt> yes, like the mail 'preview' area in Notes...empty but still there if there's no mail selected...
[16:33:42] <emoffatt> tschindl: does ISaveablePart2 somehow control the stack's management ?
[16:34:05] <tschindl> nope it only makes the view getting part of save+restore lifecycle
[16:34:15] <tschindl> sorry save+dirty
[16:34:22] <tschindl> not restore ;-)
[16:35:21] <emoffatt> got it...there's an example of this type of 'tag tweaking' in the DnD addon...the various 'dragAgents' check for the 'No Move' tag and won't start a drag if it's there
[16:36:03] <emoffatt> this makes that tag implicit API which should be in the docs somewhere but AFAIK isn't
[16:37:38] <emoffatt> note that for folks that have even more special requirements they can just replace the addon, that's what they're for :)
[16:38:09] <tschindl> emoffatt: in e3 the disposal of the stack is done by the perspective
[16:38:27] <emoffatt> but I agree we'll get more mileage if we make the 'regular' one cover the more general cases
[16:39:33] <emoffatt> yep and can't be accessed...that's 3.x's 'renderer' I guess
[16:41:37] <emoffatt> please do open a defect about making the CleanupAddon more general, perhaps we can even split it so that there's a general base class version and the yucky 'editor area' handling can be put into on specific for the compat layer ?
[16:41:55] <tschindl> right the magic happens inside PerspectiveHelper#removePart
[16:42:50] <emoffatt> lol, I'm not even sure what we'd have to do to define the API for this in 3.x
[16:44:48] <tschindl> ha there's API
[16:44:55] <tschindl> ViewStack#setDurable()
[16:45:01] <emoffatt> in 4.x it's 'add a tag (API) and check it in the CleanupAddon' in 3.x it's...what...some sort of extension point that declares a particular stack (by id?) exempt from closing ?
[16:45:15] <emoffatt> really...cool
[16:45:52] <emoffatt> ah...seems to me that you can even declare this in a perspective extension (perspective?)
[16:46:53] <emoffatt> is it available for EditorStack as well ?
[16:47:47] <tschindl> wow it gets even better
[16:47:55] <emoffatt> ...looks like we're going to have to fix the cleanup addon in any case then to support the API (I'm pretty sure we don't now)
[16:48:03] <emoffatt> oh oh
[16:48:32] <tschindl> the durability is requested through WorkbenchWindowAdvisor#isDurableFolder
[16:48:46] <emoffatt> what's the args ?
[16:48:56] <emoffatt> the stack id ?
[16:48:57] <tschindl> String perspectiveId, String folderId
[16:49:54] <emoffatt> who wins between the ViewStack API and the advisor ?
[16:50:07] <tschindl> well ViewStack is internal
[16:50:24] <tschindl> when a ViewStack instance is created the advisor is asked if the ViewStack is durable
[16:50:32] <emoffatt> ah, likely it queries the adisor when making the stack
[16:51:06] <tschindl> right
[16:51:26] <tschindl> PageLayout around line 480
[16:51:43] <emoffatt> got it...so for the compat layer I need to add that code to use the new tags that we define in the CleanupAddon defect
[16:51:58] <tschindl> cool it works!!!
[16:52:24] <emoffatt> I'm banging around in that code now in any case since I'm starting to do the Capabilities/Activities stuff
[16:52:51] <emoffatt> wow, you can do anything in 3.x once you find the API :)
[16:52:59] <tschindl> need to write a blog entry about this API
[16:53:16] <tschindl> you learn something new every day
[16:53:31] <emoffatt> try to :)
[16:53:33] <emoffatt> btw, do you need anything else from me on the EclipseCon front ?
[16:53:36] <tschindl> who the hell would have thought one can define such a thing in the window-advisor
[16:53:43] <tschindl> no not yet
[16:54:03] <tschindl> let's see that we get some of our talks through
[16:54:18] <tschindl> and then I think it is best if we team up together
[16:54:27] <tschindl> on the talks
[16:54:30] <emoffatt> the RCP advisors seem to be a catch-all (sort of equivalent to E4's tags I think
[16:55:31] <emoffatt> sounds excellent, just let me know if I need to do anything...I'm covering for both Paul and McQ now so I'm all over the map
[16:56:35] <emoffatt> right now I have to start a sametime with Pradeep, I'm starting to get the 'manager' feeling (ick!) I've done nothing since I came in except this and sametime
[16:57:41] <emoffatt> oh well...thanks tom, open the defect and post it here please mention the e3 api in it so I won't forget I need to do some compat work
[17:23:02] <emoffatt> rcjsuen: is your sametime working...mine dropped and I can't reconnect....
[17:23:23] <rcjsuen> emoffatt: I am still online as far as I can tell.
[17:23:31] <emoffatt> thanks
[17:27:30] <tschindl> well the advisor-api is not working in a save-restore scenario :-(
[17:32:34] *** magnet has quit IRC
[18:19:36] *** magnet has joined #eclipse-e4
[18:58:09] *** magnet has quit IRC
[19:39:36] *** magnet has joined #eclipse-e4
[20:04:22] *** tschindl has quit IRC
[22:18:27] *** tschindl has joined #eclipse-e4
[22:26:22] *** rcjsuen is now known as rcjsune_
[22:26:24] *** rcjsune_ is now known as rcjsuen_
[22:33:02] <tschindl> emoffatt: ~355763
[22:33:03] <Arbalest> Bug 355763 - https://bugs.eclipse.org/bugs/show_bug.cgi?id=355763 - Platform / UI / 4.1 - PC / Mac OS X - NEW / normal / - Assignee: Platform-UI-Inbox - Provide tag which prohibits a PartStack to collapse when empty
[22:41:25] *** yarthorn has quit IRC
[23:03:32] *** rcjsuen has joined #eclipse-e4





top