uPortal IRC Logs-2010-09-24

[02:57:00 CDT(-0500)] * Topic is 'http://uportal.pastebin.com/ - http://www.ja-sig.org/wiki/display/UPC/uportal+IRC+Logs' set by EricDalquist on 11:32:13 CST(-0600)
[10:36:21 CDT(-0500)] <athena> where do we want to document the resource server?
[10:36:31 CDT(-0500)] <athena> should it just have space in uportal's wiki space?
[10:36:38 CDT(-0500)] <EricDalquist> yeah, that seems fine
[10:36:44 CDT(-0500)] <EricDalquist> I thought it had some stub of a page already in the wiki
[10:42:23 CDT(-0500)] <athena> sounds good
[10:42:28 CDT(-0500)] <athena> i think you do
[10:42:37 CDT(-0500)] <athena> more for our documentation about how to add things to it
[10:42:44 CDT(-0500)] <EricDalquist> https://wiki.jasig.org/display/UPC/Resource+Server
[10:42:49 CDT(-0500)] <EricDalquist> ah yeah
[10:43:05 CDT(-0500)] <athena> i'll just add it as a child page?
[10:44:24 CDT(-0500)] <EricDalquist> sounds good
[10:49:45 CDT(-0500)] <athena> aright
[10:49:45 CDT(-0500)] <athena> https://wiki.jasig.org/display/UPC/Adding+New+Fluid+Releases+to+the+Resource+Server
[10:49:55 CDT(-0500)] <athena> needs more, but at least that's something
[10:50:02 CDT(-0500)] <EricDalquist> looks good
[11:19:26 CDT(-0500)] <EricDalquist> wow ... I just had to debug through a PAGs group resolution
[11:19:43 CDT(-0500)] <EricDalquist> that code code have some serious optimization done to it
[11:27:09 CDT(-0500)] <athena> uhoh
[11:27:27 CDT(-0500)] <EricDalquist> like it walks over the PAGs group as a list
[11:27:39 CDT(-0500)] <EricDalquist> but then for each group it walks up the tree from it making sure the user is a member of all parent groups
[11:27:48 CDT(-0500)] <EricDalquist> and it doesn't appear that any of the lookup results are cached during this process
[11:27:57 CDT(-0500)] <athena> oh
[11:28:00 CDT(-0500)] <EricDalquist> so you hit those parent groups tests A LOT
[11:28:07 CDT(-0500)] <athena> wow yeah
[11:28:11 CDT(-0500)] <EricDalquist> none are all that terribly expensive
[11:28:15 CDT(-0500)] <EricDalquist> but it all adds up
[11:28:19 CDT(-0500)] <athena> yep
[11:28:47 CDT(-0500)] <EricDalquist> well ... maybe once all the current trunk stuff is done and released I can make time to actually do that GAPs cleanup we've wanted
[11:28:53 CDT(-0500)] <EricDalquist> make it all springy
[11:29:07 CDT(-0500)] <EricDalquist> figure out what we want for functional changes
[11:29:24 CDT(-0500)] <athena>
[11:29:29 CDT(-0500)] <athena> would be cool
[11:29:49 CDT(-0500)] <EricDalquist> yeah
[11:29:57 CDT(-0500)] <EricDalquist> just not having static instances anymore will be nice
[11:31:40 CDT(-0500)] <athena> yeah
[11:31:44 CDT(-0500)] <athena> makes it so hard to unit test
[11:32:04 CDT(-0500)] <athena> which makes me very sad
[11:32:24 CDT(-0500)] <EricDalquist> yes
[11:32:33 CDT(-0500)] <EricDalquist> if you touch the GAPs code you load up an entire app context
[11:32:55 CDT(-0500)] <athena> yeah
[11:33:20 CDT(-0500)] <athena> think i'm close to having all the RSS channels ported over to portlets
[11:33:36 CDT(-0500)] <athena> so we should pretty much just have the content portlet stuff left to move
[11:33:41 CDT(-0500)] <athena> and a couple random custom channels
[11:34:21 CDT(-0500)] <EricDalquist> yay
[11:34:31 CDT(-0500)] <EricDalquist> I just found a bug locally in 3.2 with the minimize link
[11:34:46 CDT(-0500)] <EricDalquist> it doesn't override the expiration-cache for portlets
[11:36:38 CDT(-0500)] <athena> oh
[11:36:47 CDT(-0500)] <EricDalquist> so it appears to have no effect
[11:36:54 CDT(-0500)] <EricDalquist> since the portal just re-renders the portlet from cache
[11:36:55 CDT(-0500)] <EricDalquist> oops
[11:38:32 CDT(-0500)] <athena> oh
[11:38:33 CDT(-0500)] <athena> lol
[12:06:36 CDT(-0500)] <athena> hmm
[12:06:47 CDT(-0500)] <athena> looks like in trunk, the news reader portlet only works the first time i access it
[12:06:49 CDT(-0500)] <athena> for each login
[12:06:52 CDT(-0500)] <EricDalquist> weird
[12:06:58 CDT(-0500)] <EricDalquist> what happens after that?
[12:07:30 CDT(-0500)] <athena> getting Caused by: javax.portlet.PortletException: [org.jasig.portlet.newsreader.mvc.controller.AjaxSingleFeedNewsController] does not handle render requests
[12:08:18 CDT(-0500)] <EricDalquist> is it doing AJAX requests?
[12:08:28 CDT(-0500)] <EricDalquist> I haven't tested that yet
[12:08:43 CDT(-0500)] <athena> yes
[12:09:08 CDT(-0500)] <EricDalquist> can you see anything obviously wrong watching in firebug?
[12:09:24 CDT(-0500)] <athena> it doesn't even make it to firebug - that error's from the error chanenl
[12:09:35 CDT(-0500)] <athena> first request looks great
[12:09:38 CDT(-0500)] <EricDalquist> what about on the first request
[12:09:41 CDT(-0500)] <EricDalquist> no exceptions anywhere?
[12:10:40 CDT(-0500)] <athena> doesn't look like it
[12:10:47 CDT(-0500)] <EricDalquist> huh
[12:11:02 CDT(-0500)] <EricDalquist> sounds like another bug in the portlet integration
[12:13:13 CDT(-0500)] <athena> sort of interesting - seems like the problem might not occur in focus mode
[12:23:00 CDT(-0500)] <athena> yeah
[12:23:02 CDT(-0500)] <athena> ok, that's really weird
[12:23:37 CDT(-0500)] <athena> i don't think it should be ever asking the ajax controller to handle ajax requests
[12:24:37 CDT(-0500)] <athena> though it shouldn't really be trying to call that controller at all in the main page
[13:11:45 CDT(-0500)] <athena> EricDalquist: do you think we want to keep having a flyout menu on the active tab?
[13:12:12 CDT(-0500)] <EricDalquist> not sure, what are the pros/cons?
[13:12:15 CDT(-0500)] <EricDalquist> I know we use it here
[13:14:39 CDT(-0500)] <athena> ah ok
[13:14:53 CDT(-0500)] <athena> for one thing, just kind of wondering if it's confusing w/ the new tab stuff
[13:15:08 CDT(-0500)] <athena> since the active tab has an inline edit, a delete button, and a drag handle
[13:15:10 CDT(-0500)] <EricDalquist> well it can always be off by default
[13:15:15 CDT(-0500)] <mpolizzotti> I was just wondering the same thing.
[13:15:18 CDT(-0500)] <athena> kind fo weird to have the flyout menu come out when you're editing it
[13:15:41 CDT(-0500)] <athena> but matt's also run into some conflicts between the flyout nav and the drag and drop
[13:15:48 CDT(-0500)] <EricDalquist> you could always disable it when using any of the other controls too
[13:15:49 CDT(-0500)] <athena> only in IE, of course
[13:16:31 CDT(-0500)] <athena> what do you mean by disable it when using the other controls?
[13:16:46 CDT(-0500)] <EricDalquist> like when you start editing with the inline edit
[13:16:57 CDT(-0500)] <EricDalquist> make a JS call that disables the flyout until the edit is complete
[13:17:15 CDT(-0500)] <athena> dunno, i worry that that might actually be just as confusing
[13:17:18 CDT(-0500)] <EricDalquist> yeah
[13:17:22 CDT(-0500)] <EricDalquist> just an idea
[13:17:28 CDT(-0500)] <athena> since then the menu would open on hover, close when you clicked, then open again when you finsihed
[13:17:28 CDT(-0500)] <athena> yeah
[13:23:54 CDT(-0500)] <athena> think matt's going to try some debugging
[13:24:53 CDT(-0500)] <mpolizzotti> Is anyone aware that the "Return to dashboard" isn't working. Is that apart of the new url provider functionality.
[13:27:23 CDT(-0500)] <athena> yes
[13:27:46 CDT(-0500)] <athena> and it's the new url schema / theme rather than our new js component
[13:27:52 CDT(-0500)] <athena> i think i can probably get that fixed though
[13:48:16 CDT(-0500)] <EricDalquist> uhg
[13:48:32 CDT(-0500)] <EricDalquist> Spring's @RequestMapping for portlets has no concept of WindowState support
[13:48:43 CDT(-0500)] <EricDalquist> and so by default it doesn't honor minimization
[13:49:05 CDT(-0500)] <EricDalquist> it looks like the only option is either a handlerInterceptor to return early if minimized
[13:49:16 CDT(-0500)] <EricDalquist> or do a short-circuit in every annotated method
[13:49:27 CDT(-0500)] <EricDalquist> that's really crappy
[13:50:28 CDT(-0500)] <athena> ick
[13:50:43 CDT(-0500)] <athena> seems like a real failure
[13:50:49 CDT(-0500)] <athena> hm
[13:50:58 CDT(-0500)] <EricDalquist> yes
[13:51:02 CDT(-0500)] <EricDalquist> it is
[13:51:17 CDT(-0500)] <EricDalquist> especially for a portal that follows the spec and lets portlets render content when minimized
[13:51:59 CDT(-0500)] <athena> what do you mean by do a short-circuit? like have logic specifically for handling minimized state?
[13:52:08 CDT(-0500)] <EricDalquist> right
[13:52:21 CDT(-0500)] <EricDalquist> if(state=minimized) return "doNothingView"
[13:52:36 CDT(-0500)] <EricDalquist> you would have to add that to the top of ever RequestMapping method
[13:52:51 CDT(-0500)] <athena> or return the view you actually want, yes
[13:52:59 CDT(-0500)] <EricDalquist> also it makes it hard to handle max vs norm states
[13:53:08 CDT(-0500)] <EricDalquist> like if I want my search results to display in maximized
[13:53:15 CDT(-0500)] <athena> since it's plausible you might want to render a small view of the same model or whatever
[13:53:20 CDT(-0500)] <EricDalquist> but then the user through a portal control makes my portlet normal sized
[13:53:27 CDT(-0500)] <EricDalquist> I want to specify that in my annotation
[13:53:32 CDT(-0500)] <athena> yeah
[13:53:36 CDT(-0500)] <athena> which is a reasonable thing to want
[13:53:41 CDT(-0500)] <EricDalquist> grrr
[13:53:53 CDT(-0500)] <athena> maybe enter that as a feature request?
[13:53:57 CDT(-0500)] <EricDalquist> yeah
[13:53:58 CDT(-0500)] <EricDalquist> I will
[13:54:11 CDT(-0500)] <athena> seems pretty reasonable, since you can filter on portlet mode, headers, etc.
[13:54:14 CDT(-0500)] <EricDalquist> just not something I want to spend time figuring out
[13:54:16 CDT(-0500)] <athena> yeah
[13:54:19 CDT(-0500)] <athena> no kidding
[13:54:26 CDT(-0500)] <EricDalquist> well maybe we can shoe-horn it into headers
[13:54:43 CDT(-0500)] <EricDalquist> for portlets only modes & parameters are supported right now
[13:55:01 CDT(-0500)] <EricDalquist> so maybe make some faux headers for the state/mode & properties
[13:55:13 CDT(-0500)] <EricDalquist> that the annotation can then reference
[13:55:18 CDT(-0500)] <athena> makes sense
[13:56:36 CDT(-0500)] <EricDalquist> I'm going to have words with johnl at the unconf about this
[13:56:45 CDT(-0500)] <EricDalquist> since his name is on a lot of this code
[13:59:55 CDT(-0500)] <EricDalquist> looks like webflow doesn't handle minimized well either
[14:05:42 CDT(-0500)] <athena> ick
[14:05:51 CDT(-0500)] <athena> it seems like some of the portlet stuff got kind of thrown together in spring 3
[14:06:01 CDT(-0500)] <athena> the documentation was never updated whatsoever, which is really frustrating
[14:06:02 CDT(-0500)] <EricDalquist> this is all 2.5
[14:06:08 CDT(-0500)] <athena> yeah
[14:06:13 CDT(-0500)] <EricDalquist> looks like I'm writing a spring interceptor to handle minimized
[14:06:26 CDT(-0500)] <athena> the portlet support sort of feels like the bastard stepchild
[14:06:38 CDT(-0500)] <EricDalquist> yup
[14:06:49 CDT(-0500)] <EricDalquist> maybe we can contribute a bit back to make it better
[14:07:06 CDT(-0500)] <athena> i'm sure that'd be appreciated
[14:07:11 CDT(-0500)] <athena> by us if by no one else
[14:07:26 CDT(-0500)] <athena> do you want me to file a JIRA or something for the ajax support issue?
[14:07:30 CDT(-0500)] <EricDalquist> yeah
[14:07:33 CDT(-0500)] <athena> ok
[14:07:34 CDT(-0500)] <EricDalquist> just to catalog it
[14:07:40 CDT(-0500)] <EricDalquist> don't want to lose track of those
[14:07:45 CDT(-0500)] <athena> yeah
[14:07:58 CDT(-0500)] <athena> kind of surprised we hadn't run into that yet
[14:08:07 CDT(-0500)] <athena> and it's weird because it works fine in focus mode
[14:08:53 CDT(-0500)] <athena> i've checked in my news reader portlet work so you'll be able to use those portlets to see the issue
[14:08:59 CDT(-0500)] <EricDalquist> great
[14:09:01 CDT(-0500)] <EricDalquist> thanks
[14:10:54 CDT(-0500)] <athena> ok
[14:10:55 CDT(-0500)] <athena> created https://issues.jasig.org/browse/UP-2826
[15:50:34 CDT(-0500)] <athena> hmm
[15:50:54 CDT(-0500)] <athena> seems that dlm attributes for channels get dropped on export (things like moveAllowed)
[15:50:57 CDT(-0500)] <athena> is that a known issue?
[15:51:11 CDT(-0500)] <EricDalquist> in trunk?
[15:51:14 CDT(-0500)] <athena> 3.2
[15:51:26 CDT(-0500)] <EricDalquist> is it exporting to the new layout format?
[15:51:34 CDT(-0500)] <athena> not sure?
[15:51:46 CDT(-0500)] <EricDalquist> does the XML just have folders/channels
[15:51:52 CDT(-0500)] <athena> yeah
[15:51:56 CDT(-0500)] <EricDalquist> or does it have <tab>/<column>
[15:51:59 CDT(-0500)] <EricDalquist> hmr
[15:51:59 CDT(-0500)] <athena> folders
[15:52:02 CDT(-0500)] <athena> is that new or old?
[15:52:06 CDT(-0500)] <EricDalquist> that is the new format then
[15:52:15 CDT(-0500)] <EricDalquist> I thought all that was handled
[15:53:00 CDT(-0500)] <athena> looks like the tab gets handled ok
[15:53:11 CDT(-0500)] <EricDalquist> hrm
[15:53:21 CDT(-0500)] <EricDalquist> sounds like a question for awills
[15:53:33 CDT(-0500)] <athena> yeah
[15:53:37 CDT(-0500)] <athena> i'll have to ask him next week
[15:53:47 CDT(-0500)] <athena> he's traveling this week
[15:57:59 CDT(-0500)] <athena> interesting - looks like it works correctly in trunk