uPortal IRC Logs-2009-04-10

Owned by Apereo Infrastructure
Last updated: Apr 10, 2009Version comment

[00:58:47 EDT(-0400)] * Sememmon (n=Sememmon@ip70-190-32-223.ph.ph.cox.net) has joined ##uportal
[01:07:15 EDT(-0400)] * lennard1 (n=sparhk@ip68-98-56-21.ph.ph.cox.net) has joined ##uportal
[01:16:48 EDT(-0400)] * apetro (n=apetro@ip68-3-207-51.ph.ph.cox.net) has joined ##uportal
[08:10:30 EDT(-0400)] * ChanServ (ChanServ@services.) has joined ##uportal
[09:10:19 EDT(-0400)] * athena (n=athena@dhcp128036159138.central.yale.edu) has joined ##uportal
[09:10:22 EDT(-0400)] * awills (n=awills@dhcp128036163050.central.yale.edu) has joined ##uportal
[09:11:04 EDT(-0400)] * tsnfoo (n=tsnfoo@wso-mbp15.test.denison.edu) has joined ##uportal
[10:01:09 EDT(-0400)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined ##uportal
[10:07:43 EDT(-0400)] * lennard1 (n=sparhk@ip68-98-56-21.ph.ph.cox.net) has left ##uportal
[10:25:55 EDT(-0400)] * lennard1 (n=sparhk@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[10:48:02 EDT(-0400)] <EricDalquist> found a nice bug in 3.1 that I accidentatlly introduced
[10:48:40 EDT(-0400)] <EricDalquist> The ProcessingPipe is now static ... by accident
[10:48:49 EDT(-0400)] <EricDalquist> so I'll be fixing that and probably cutting a 3.1 release on Monday
[10:48:52 EDT(-0400)] <EricDalquist> er 3.1.1
[10:54:58 EDT(-0400)] * lennard2 (n=sparhk@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[11:09:04 EDT(-0400)] * apetro (n=apetro@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[11:25:33 EDT(-0400)] <athena> ah
[11:25:42 EDT(-0400)] <athena> is there any other stuff that we should be fixing for 3.1.1?
[12:11:15 EDT(-0400)] <EricDalquist> that layout owners can't see the Customize link (tongue)
[12:12:35 EDT(-0400)] <athena> yeah
[12:12:39 EDT(-0400)] <athena> that does need to get fixed
[12:12:51 EDT(-0400)] <athena> do we know what happened there? i haven't had time to look into it
[12:13:01 EDT(-0400)] <athena> did the link get removed from the theme or something?
[12:13:15 EDT(-0400)] <EricDalquist> I think it is a group membership thing
[12:13:22 EDT(-0400)] <EricDalquist> you only get the link if your in the admin group
[12:13:27 EDT(-0400)] <EricDalquist> and the layout owners are not
[12:17:43 EDT(-0400)] * invisibill (i=80876350@gateway/web/ajax/mibbit.com/x-0be36e1fd07b664e) has joined ##uportal
[12:18:33 EDT(-0400)] * lennard1 (n=sparhk@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[12:20:01 EDT(-0400)] <invisibill> Greeting uPortal devs: We are up and running with the 3.1 build and have a few jQuery dependency issues: when landing on the homepage in Firefox we get this in the error console: Error: jQuery is not defined Source File: https://mydev.uchicago.edu/uPortal/media/skins/universality/common/javascript/uportal/up-channel-browser.min.js Line: 1
[12:20:18 EDT(-0400)] <invisibill> with the same error for all files in that folder.
[12:20:45 EDT(-0400)] <invisibill> not sure how to link in the missing jQuery dependency.
[12:21:21 EDT(-0400)] <EricDalquist> it should be defined by default in the 3.1 theme
[12:23:23 EDT(-0400)] <invisibill> the displaying page looks like a new theme as far as the css and placement of element on the page, but then the jQuery error and the "Add Content" and "Add Tab" links don't fire
[12:24:09 EDT(-0400)] <invisibill> we haven't done any changes to the default theme. in the GA release just build and deploy.
[12:24:29 EDT(-0400)] <EricDalquist> hrm
[12:24:32 EDT(-0400)] <EricDalquist> any ideas athena?
[12:24:37 EDT(-0400)] <invisibill> where is it defined?
[12:25:06 EDT(-0400)] <athena> hey
[12:25:25 EDT(-0400)] <athena> hm
[12:25:38 EDT(-0400)] <athena> can you take a look at the URL the theme is tryign to import? is it available?
[12:25:48 EDT(-0400)] <athena> it should be coming from /ResourceServingWebapp
[12:25:53 EDT(-0400)] <EricDalquist> line 334 of uportal-war/src/main/resources/layout/theme/universality/universality.xsl
[12:26:03 EDT(-0400)] <invisibill> I'll check
[12:26:10 EDT(-0400)] <EricDalquist> ah that could be it too ... there is a new webapp that is included with 3.1
[12:26:13 EDT(-0400)] <EricDalquist> that the portal depends on
[12:26:31 EDT(-0400)] <invisibill> WebProxy?
[12:26:45 EDT(-0400)] <EricDalquist> ResourceServingWebapp
[12:27:20 EDT(-0400)] <EricDalquist> it provides static, aggressively cached, resources such as CSS, JS and icons that uPortal uses to help reduce load times for end users.
[12:27:57 EDT(-0400)] <athena> yeah, if you haven't run deploy-ear, that might be missing?
[12:28:20 EDT(-0400)] <invisibill> we deploy the wars manaully, I'm loading it in now...
[12:29:43 EDT(-0400)] * apetro (n=apetro@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[12:31:02 EDT(-0400)] * holdorph (n=holdorph@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[12:31:13 EDT(-0400)] <EricDalquist> so I'm working on ideas for: http://www.ja-sig.org/wiki/display/UPC/Consistent+Portal+URLs
[12:31:36 EDT(-0400)] <EricDalquist> I'm trying to scrap together some ideas for portal URLs that are fairly readable
[12:31:53 EDT(-0400)] <EricDalquist> unique to a particular tab or maximized portlet
[12:32:14 EDT(-0400)] <EricDalquist> I'm not so worried about the how side yet, just the ideas for what URLs should look like
[12:32:18 EDT(-0400)] <invisibill> that seems to be it (smile) Thanks for your help athena and EricDalquist
[12:32:25 EDT(-0400)] <EricDalquist> glad we could help invisibill
[12:32:32 EDT(-0400)] <athena> great, i'm glad (smile)
[12:32:39 EDT(-0400)] <athena> ooh eric that'd be fantastic
[12:33:18 EDT(-0400)] <EricDalquist> the other 'fun' requirement that is further out in the future is the ability for something like GoogleBot to index uPortal
[12:33:25 EDT(-0400)] <EricDalquist> which means a guest user with no session at all
[12:33:36 EDT(-0400)] <athena> oooh yay
[12:33:37 EDT(-0400)] <EricDalquist> brb in ~ 10
[12:33:40 EDT(-0400)] <athena> ok
[12:34:25 EDT(-0400)] * Sememmon (n=Sememmon@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[12:39:27 EDT(-0400)] * lennard1 (n=sparhk@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[12:42:14 EDT(-0400)] <EricDalquist> so I'm thinking I'd enforce a unique fname policy for tabs
[12:42:25 EDT(-0400)] <EricDalquist> and more generically folders
[12:42:43 EDT(-0400)] <EricDalquist> so you can have /uPortal/folderFname/folderFname/.....
[12:43:16 EDT(-0400)] <athena> that should take care of the nested folder requirement
[12:43:27 EDT(-0400)] <EricDalquist> for channels
[12:43:32 EDT(-0400)] <EricDalquist> I'd like to use fnames
[12:43:38 EDT(-0400)] <EricDalquist> but you have the problem of multiple subscriptions
[12:43:45 EDT(-0400)] <athena> yeah i was about to ask about that (smile)
[12:43:46 EDT(-0400)] <EricDalquist> so I'm not sure what to do there
[12:44:02 EDT(-0400)] <athena> what if we did /uPortal/folderFname/folderFnmae/channelFname?
[12:44:04 EDT(-0400)] <EricDalquist> maybe /uPortal/folderFname/chanFname.subscribeId.uP
[12:44:10 EDT(-0400)] <athena> yeah
[12:44:18 EDT(-0400)] <EricDalquist> for maximized portlets
[12:44:19 EDT(-0400)] <athena> i think that also makes sense from a user perspective
[12:44:30 EDT(-0400)] <athena> since you think of a channel as being an item on a page
[12:44:30 EDT(-0400)] <athena> yeah
[12:44:35 EDT(-0400)] <EricDalquist> though I also need to come up with similar URLs for detached & exclusive
[12:44:59 EDT(-0400)] <athena> and perhaps /uPortal/chanFname.subscribeId.uP could do what the fname call does now?
[12:45:15 EDT(-0400)] <athena> where it just tries to resolve some instance you already have, or just get a temporary one
[12:45:24 EDT(-0400)] <EricDalquist> so chanFname.uP would do that
[12:45:33 EDT(-0400)] <EricDalquist> chanFname.subid.uP would already have the instance
[12:45:39 EDT(-0400)] <athena> ah i see
[12:45:50 EDT(-0400)] <athena> makes sense
[12:46:06 EDT(-0400)] <EricDalquist> so when rendering normal pages something like: /uPortal/root/folderFname/render.uP?targetChan=chanSubId&params....
[12:46:32 EDT(-0400)] <EricDalquist> /uPortal/max/folderFname/chanFname.chanSubId.uP or /uPortal/max/folderFname/chanFname.uP for maximized
[12:46:49 EDT(-0400)] <EricDalquist> /uPortal/det/folderFname/chanFname.chanSubId.uP or /uPortal/det/folderFname/chanFname.uP for detached
[12:46:59 EDT(-0400)] <EricDalquist> same pattern for exclusive would work
[12:47:24 EDT(-0400)] <athena> sounds pretty good to me
[12:47:30 EDT(-0400)] <EricDalquist> what would be really good is if a url like "/uPortal/det/folderFname/chanFname.uP" did a 302 to /uPortal/det/folderFname/chanFname.subId.uP after doing that fname subid lookup
[12:47:39 EDT(-0400)] <athena> yeah
[12:47:43 EDT(-0400)] <EricDalquist> so the URL for that content is always the same
[12:47:44 EDT(-0400)] <athena> i was sort of thinking the same thing (smile)
[12:48:11 EDT(-0400)] <EricDalquist> URLs rendered by channels will likely still be semi-ugly, but there isn't much I can do about that
[12:48:19 EDT(-0400)] <athena> as a general note, we'll have to start using absolute URLs for our skin resources, which we don't do now
[12:48:20 EDT(-0400)] <EricDalquist> luckily the number of channels is decreasing rapidly
[12:48:23 EDT(-0400)] <EricDalquist> yes
[12:48:33 EDT(-0400)] <EricDalquist> I'd have a fair bit of work to do in the URLs generated by the XSL
[12:48:53 EDT(-0400)] <EricDalquist> including maybe riting a xsl taglib to generate the portal URLs
[12:49:02 EDT(-0400)] <athena> well, we could just make them absolute urls
[12:49:03 EDT(-0400)] <EricDalquist> riting=writing
[12:49:10 EDT(-0400)] <EricDalquist> yaeh
[12:49:11 EDT(-0400)] <EricDalquist> yeah
[12:49:15 EDT(-0400)] <EricDalquist> but you need to know the context name
[12:49:15 EDT(-0400)] <athena> we'd just need to have the context name available
[12:49:18 EDT(-0400)] <athena> right
[12:49:19 EDT(-0400)] <EricDalquist> so that would need to be injected
[12:49:22 EDT(-0400)] <athena> that shouldn't be that bad though
[12:49:22 EDT(-0400)] <athena> yep
[12:49:36 EDT(-0400)] <athena> once that was available we'd be ok i think
[12:49:42 EDT(-0400)] <EricDalquist> there should be enough info in the layout XML to render the tab & channel fnames & subids
[12:49:56 EDT(-0400)] <athena> and then there'd be some cleanup in the older channels as well to do the same thing
[12:50:05 EDT(-0400)] <EricDalquist> well
[12:50:13 EDT(-0400)] <EricDalquist> I might give channels a special baseactionurl
[12:50:27 EDT(-0400)] <EricDalquist> that a filter in front of the portal would always rewrite into 302s
[12:50:27 EDT(-0400)] <athena> i meant for the skinning resources, not the links, sorry
[12:50:31 EDT(-0400)] <EricDalquist> ah
[12:50:34 EDT(-0400)] <EricDalquist> yeah
[12:50:42 EDT(-0400)] <EricDalquist> true
[12:50:54 EDT(-0400)] * EricDalquist can't wait for channels to go away completely
[12:51:30 EDT(-0400)] <athena> no kidding!
[12:51:48 EDT(-0400)] <athena> that'd be excellent
[12:51:49 EDT(-0400)] <EricDalquist> the more we go to portlets the more it opens up things we can do in the framework
[12:54:12 EDT(-0400)] <athena> (smile)
[13:29:11 EDT(-0400)] * apetro_ (n=apetro@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[13:31:55 EDT(-0400)] * apetro- (n=apetro@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[13:39:06 EDT(-0400)] * athena (n=athena@99.129.100.66) has joined ##uportal
[13:54:22 EDT(-0400)] <athena> sorry i disappeared before eric
[13:54:27 EDT(-0400)] <athena> urls looking happy? (smile)
[13:54:35 EDT(-0400)] <EricDalquist> I think I've got some good idea
[13:54:37 EDT(-0400)] <EricDalquist> ideas
[13:54:45 EDT(-0400)] <EricDalquist> going to go get lunch and update the wiki page
[13:55:08 EDT(-0400)] <athena> cool (smile)
[13:55:20 EDT(-0400)] <athena> by the way, i'd like to update uportal to fluid 1.0, if people don't object
[13:55:29 EDT(-0400)] <athena> i think we'd need to make that uportal 3.1+ only
[13:55:38 EDT(-0400)] <athena> as there are some associated css class name changes
[14:00:22 EDT(-0400)] <EricDalquist> so you mean 3.2?
[14:00:37 EDT(-0400)] <athena> oops, yes!
[14:00:38 EDT(-0400)] <athena> that
[14:00:41 EDT(-0400)] <EricDalquist> ok (smile)
[14:00:42 EDT(-0400)] <EricDalquist> yeah
[14:00:44 EDT(-0400)] <EricDalquist> go ahead
[14:00:47 EDT(-0400)] <athena> ok
[14:00:58 EDT(-0400)] <athena> i'm thinking we'll be better off in the long run if we try and update as they do
[14:01:12 EDT(-0400)] <athena> since we'll be more likely to catch any changes that affect us in time to deal with them
[14:01:20 EDT(-0400)] <EricDalquist> of course
[14:01:23 EDT(-0400)] <athena> and also have plenty of time to integrate new functionality
[14:01:30 EDT(-0400)] <EricDalquist> I update all library versions with each new uPortal version
[14:01:34 EDT(-0400)] <athena> yeah
[14:01:44 EDT(-0400)] <EricDalquist> feel free to just do things like that on trunk
[14:01:57 EDT(-0400)] <EricDalquist> if you're worried at all just do a lazy-consensus poll on the dev list
[14:02:02 EDT(-0400)] <athena> well and i think updating our fluid libraries several times while 3.1 was in trunk helped us this release - we weren't troubleshooting much reorderer stuff at the end
[14:02:04 EDT(-0400)] <athena> sure
[14:32:27 EDT(-0400)] * apetro_ (n=apetro@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[15:49:10 EDT(-0400)] <EricDalquist> more work on the URLs page: http://www.ja-sig.org/wiki/display/UPC/Consistent+Portal+URLs
[15:53:12 EDT(-0400)] <holdorph> you use 'normal' when describing state in the URL design section, but use 'norm' when showing example urls
[15:53:22 EDT(-0400)] <EricDalquist> ah good catch
[15:54:25 EDT(-0400)] <holdorph> it might be useful to separate channelName from subscribeId and better describe each of those parts separately. Not sure it matters too much though. But otherwise very readable, makes sense. Looks great to me.
[15:54:50 EDT(-0400)] <EricDalquist> like a / instead of a .?
[15:55:22 EDT(-0400)] <holdorph> no, i meant describe each of the concepts separately
[15:55:27 EDT(-0400)] <EricDalquist> ah
[15:55:35 EDT(-0400)] <holdorph> channelName = the functional name of the portlet/channel
[15:55:53 EDT(-0400)] <holdorph> subscribeId = the id associated with the particular subscription
[15:57:13 EDT(-0400)] <EricDalquist> just updated again
[15:57:48 EDT(-0400)] <EricDalquist> I'll be posting this page to the dev list shortly
[15:57:57 EDT(-0400)] <holdorph> yup, looks good to me.
[15:58:00 EDT(-0400)] <EricDalquist> I'm really hoping that I'm not forgetting any edge cases in the design
[15:58:20 EDT(-0400)] <holdorph> well it's a nice base document, even if more cases come up
[15:58:26 EDT(-0400)] <holdorph> what's there right now makes sense at least.
[15:59:02 EDT(-0400)] <EricDalquist> sounds good
[15:59:14 EDT(-0400)] <EricDalquist> hopefully the code that supports this will be fairly flexible
[15:59:33 EDT(-0400)] <EricDalquist> so we can change the syntax in the future without having to change more than the generation/parsing implementations
[16:00:16 EDT(-0400)] <holdorph> what about portlet render parameters
[16:01:12 EDT(-0400)] <EricDalquist> those would all be added after the ?
[16:01:46 EDT(-0400)] <EricDalquist> and some things will still be stateful, like the last set of render parameters sent to each portlet
[16:02:21 EDT(-0400)] <EricDalquist> channel, layout, portlet parameters would still be encoded as standard URL parameters I think
[16:02:45 EDT(-0400)] <EricDalquist> the base URL part would be addressing a particular 'location' in the layout ... if that is the right word to use
[16:03:01 EDT(-0400)] <holdorph> i guess that's where I see a possible explanation being helpful. if you're displaying just a tab (folder) and that tab has multiple portlets on it. how do render parameters show up, on the url, or not on the URL.
[16:03:26 EDT(-0400)] <holdorph> some portals they are totally url based and you can 'hack' your URL to change the render paramters, without interacting with the portlet to do it.
[16:03:49 EDT(-0400)] <EricDalquist> ah ok
[16:03:51 EDT(-0400)] <holdorph> i'm not saying that's right or wrong, but it would be good to know how it will work in uPortal under a new URL design
[16:04:59 EDT(-0400)] <EricDalquist> yup
[16:18:49 EDT(-0400)] <EricDalquist> ok, updated with a bit about parameters
[16:18:54 EDT(-0400)] <EricDalquist> I'll put in some examples too
[16:21:30 EDT(-0400)] <athena> ooh it's a pretty rainbow
[16:21:39 EDT(-0400)] <EricDalquist> lol
[16:22:05 EDT(-0400)] <EricDalquist> I think I even did it in the right order ... unintentionally
[16:22:14 EDT(-0400)] <EricDalquist> roygbiv right?
[16:22:22 EDT(-0400)] <athena> (smile)
[16:22:38 EDT(-0400)] * athena awards EricDalquist a kindergarten diploma
[16:22:43 EDT(-0400)] <EricDalquist> yay!
[16:22:52 EDT(-0400)] <athena> hehe
[16:23:17 EDT(-0400)] <EricDalquist> hey that means I've remembered something for 24 years ... that's impressive! (wink)
[16:23:21 EDT(-0400)] <athena> it always seemed weird that indigo was another color
[16:23:23 EDT(-0400)] <athena> lol
[16:23:27 EDT(-0400)] <athena> it is
[16:23:36 EDT(-0400)] <athena> i still remember the song "do your ears hang low"
[16:23:39 EDT(-0400)] <athena> so i guess we're both doing ok
[16:23:41 EDT(-0400)] <EricDalquist> lol
[16:24:00 EDT(-0400)] <EricDalquist> so the more I think about it I think I can get channels to play nice with this too
[16:24:07 EDT(-0400)] <EricDalquist> well
[16:24:08 EDT(-0400)] <EricDalquist> hrm
[16:24:10 EDT(-0400)] <EricDalquist> maybe not
[16:24:24 EDT(-0400)] <EricDalquist> because they would render the incorrect URL syntax for switching to different states and such
[16:24:35 EDT(-0400)] <EricDalquist> guess I need to document a 'legacy' syntax too (tongue)
[16:35:28 EDT(-0400)] * lennard1 (n=sparhk@wsip-98-174-242-39.ph.ph.cox.net) has left ##uportal
[16:37:32 EDT(-0400)] <EricDalquist> ok added a bit about legacy URLs
[19:23:20 EDT(-0400)] * Sememmon (n=Sememmon@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[19:36:34 EDT(-0400)] * apetro (n=apetro@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[23:31:38 EDT(-0400)] * tsnfoo (n=tsnfoo@cpe-65-24-108-125.columbus.res.rr.com) has joined ##uportal