Page Comparison

[05:14:47 EDT(-0400)] * JASIGLogBot (i=jasigch@jasigch.Princeton.EDU) has joined ##uportal
[05:14:47 EDT(-0400)] * Topic is 'http://uportal.pastebin.com/ - http://www.ja-sig.org/wiki/display/UPC/uportal+IRC+Logs' set by EricDalquist on 2008-02-27 12:32:13 EST(-0500)
[05:37:26 EDT(-0400)] * EiNZTEiN (n=einztein@bl7-230-37.dsl.telepac.pt) has joined ##uportal
[07:53:11 EDT(-0400)] * athena7 (n=athena7@c-24-2-189-230.hsd1.ct.comcast.net) has joined ##uportal
[07:54:50 EDT(-0400)] * michelled (n=team@142.150.154.197) has joined ##uportal
[09:21:02 EDT(-0400)] * esm (n=esm@128.220.212.115) has joined ##uportal
[09:38:30 EDT(-0400)] * colinclark (n=colin@bas1-toronto09-1279543604.dsl.bell.ca) has joined ##uportal
[09:57:04 EDT(-0400)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined ##uportal
[10:06:45 EDT(-0400)] * athena7 cheers for code freezes
[10:07:40 EDT(-0400)] <EricDalquist>
[10:07:47 EDT(-0400)] * EiNZTEiN (n=einztein@bl10-251-105.dsl.telepac.pt) has joined ##uportal
[10:08:33 EDT(-0400)] * MarkRogers (n=MarkRoge@addhcp129.cc.umanitoba.ca) has joined ##uportal
[10:20:31 EDT(-0400)] <athena7> by the way, it turns out i was wrong about the javascript gzipping issues with jquery
[10:20:37 EDT(-0400)] <athena7> i'd misinterpreted their documentation
[10:20:47 EDT(-0400)] <EricDalquist> ah
[10:20:52 EDT(-0400)] <EricDalquist> what is the issue?
[10:21:06 EDT(-0400)] <athena7> they're specifically referring to a "packed" distribution that causes the client to have to load the code via an eval statement
[10:21:14 EDT(-0400)] <athena7> not to gzip'd javascript code
[10:21:26 EDT(-0400)] <athena7> so i think the thing in the future to do is to use the minified version and gzip it
[10:21:37 EDT(-0400)] <athena7> but presumably we'll need a filter or something to set the headers?
[10:21:47 EDT(-0400)] <EricDalquist> yeah
[10:21:58 EDT(-0400)] <EricDalquist> there are existing gzip filters out there
[10:22:10 EDT(-0400)] <EricDalquist> or people can do it via Apache
[10:22:23 EDT(-0400)] <athena7> yeah
[10:22:26 EDT(-0400)] <athena7> also true
[10:22:32 EDT(-0400)] <EricDalquist> a lot of this may best be captured in a manual page on download performance tuning
[10:22:37 EDT(-0400)] <athena7> yeah
[10:22:51 EDT(-0400)] <EricDalquist> pretty much a summary of yslows recommendations translated into uP specific suggestions
[10:22:53 EDT(-0400)] <athena7> and make sure we have the directory structures set such that it's easily accomplishable
[10:22:58 EDT(-0400)] <athena7> yeah
[10:23:24 EDT(-0400)] <EricDalquist> yup
[10:23:37 EDT(-0400)] <EricDalquist> getting the JS moved around into versioned directories will be a help
[10:23:41 EDT(-0400)] <athena7> yes
[10:23:55 EDT(-0400)] <athena7> thinking that might be good for the next release
[10:24:04 EDT(-0400)] <athena7> i'd kind of wanted to hold off because of the css structure issues
[10:24:09 EDT(-0400)] <EricDalquist> yeah, we can schedule that for 3.1
[10:24:38 EDT(-0400)] <athena7> and because we're going to want to upgrade to jquery ui 1.5 anyway, now that it's been released
[10:24:42 EDT(-0400)] <athena7> yeah
[10:24:50 EDT(-0400)] <athena7> when were you thinking of trying to release 3.1?
[10:24:57 EDT(-0400)] <EricDalquist> not sure
[10:25:08 EDT(-0400)] <EricDalquist> perhaps late summer?
[10:25:11 EDT(-0400)] <athena7> cool
[10:25:16 EDT(-0400)] <EricDalquist> but we need to get the Fluid re-orderer in
[10:25:20 EDT(-0400)] <EricDalquist> and the GAPs work done
[10:25:20 EDT(-0400)] <athena7> yes
[10:25:26 EDT(-0400)] <EricDalquist> those are the two promised 3.1 issues
[10:25:29 EDT(-0400)] <athena7> ah ok
[10:25:35 EDT(-0400)] <EricDalquist> and I won't have time to really help on either until august
[10:25:43 EDT(-0400)] <EricDalquist> so actually it may be early fall
[10:25:44 EDT(-0400)] <athena7> yeah
[10:25:53 EDT(-0400)] <athena7> i'm not sure i'm going to have a lot of resources the next couple months
[10:53:23 EDT(-0400)] <colinclark> EricDalquist: Sorry, jumping in late...
[10:53:29 EDT(-0400)] <EricDalquist> ?
[10:53:35 EDT(-0400)] <colinclark> Layout Customizer is all ready to go for uP 3.1
[10:53:40 EDT(-0400)] <EricDalquist> awesome
[10:53:55 EDT(-0400)] <EricDalquist> I did see that Gary had created the jira issue and marked it as in-progress
[10:53:57 EDT(-0400)] <colinclark> I have been amiss in getting in touch to talk about how we can get it in.
[10:54:07 EDT(-0400)] <EricDalquist> I haven't had time to ping him to see what he has done
[10:54:19 EDT(-0400)] <colinclark> It's working pretty nicely. We actually probably sat on it too long, polishing. But we should have probably just lived with a few bugs and gone for it.
[10:54:25 EDT(-0400)] <colinclark> Lesson learned.
[10:54:26 EDT(-0400)] <EricDalquist> well I guess it is just letting us know what you need in the uP3 code to integrate it
[10:54:30 EDT(-0400)] <EricDalquist> eh
[10:54:37 EDT(-0400)] <EricDalquist> it likely wouldn't have made 3.0 anyways
[10:54:42 EDT(-0400)] <EricDalquist> and 3.1 is still a ways off
[10:54:45 EDT(-0400)] <EricDalquist> so you're fine
[10:54:51 EDT(-0400)] <colinclark> Integration is all around the JSON data for layouts and permissions.
[10:54:58 EDT(-0400)] <colinclark> It should be pretty straightforward.
[10:55:07 EDT(-0400)] <EricDalquist> ok
[10:55:10 EDT(-0400)] <colinclark> EricDalquist: Good to know. Thanks for all your support on this one.
[10:55:31 EDT(-0400)] <EricDalquist> so if I understand it all there are a few parts
[10:55:47 EDT(-0400)] <EricDalquist> 1. update the uPortal theme XSL to use the layout customizer
[10:56:01 EDT(-0400)] <EricDalquist> 2. implement the json data structure to communicate layout restrictions
[10:56:15 EDT(-0400)] <EricDalquist> 3. implement the appropriate ajax callback servlets
[10:56:35 EDT(-0400)] <colinclark> That sounds about right. Here's a bit of documentation on the Layout and Permissions data structures. These were the ones you and Antranig worked out here in Toronto, with some variation for simplicity. http://wiki.fluidproject.org/display/fluid/Layout+Customizer+-+Layout+and+Permissions
[10:57:04 EDT(-0400)] <colinclark> Some examples: http://wiki.fluidproject.org/display/fluid/Using+the+Reorderer#UsingtheReorderer-SortingPortletsorothercontentmodules
[10:57:18 EDT(-0400)] <EricDalquist> nice
[10:57:21 EDT(-0400)] <colinclark> And more documentation: http://wiki.fluidproject.org/display/fluid/Layout+Customizer+API
[10:58:21 EDT(-0400)] <EricDalquist> so the servlet work would likely end up being done by Jen, myself or another uP java dev. Would a Fluid person (gary?) be able to do the theme XSL side first?
[10:59:14 EDT(-0400)] <colinclark> EricDalquist: Definitely. Is Gary the best person for the job, then?
[10:59:40 EDT(-0400)] <EricDalquist> well he knows the theme XSL really well since he wrote them
[11:00:09 EDT(-0400)] <athena7> yes
[11:00:19 EDT(-0400)] <colinclark> Okay, great.
[11:00:20 EDT(-0400)] <athena7> i'd probably recommend gary and i collaborating on it?
[11:00:32 EDT(-0400)] <colinclark> I'll talk with Jess and Gary about it. athena7: We'd love to have you collaborate on it.
[11:00:36 EDT(-0400)] <EricDalquist> sounds good
[11:00:42 EDT(-0400)] <EricDalquist> I can probably help too
[11:00:46 EDT(-0400)] <athena7> since between us, we know the theme and current ajax capabilities
[11:00:48 EDT(-0400)] <athena7> awesome
[11:00:49 EDT(-0400)] <EricDalquist> I don't think the servlet side work will be too bad
[11:00:54 EDT(-0400)] <colinclark> This is great.
[11:00:58 EDT(-0400)] <colinclark> You guys rule.
[11:01:07 EDT(-0400)] <EricDalquist> athena7: one thing I'd like to do is to replace the current ajax servlets with Spring MVC controllers
[11:01:15 EDT(-0400)] <athena7> YES
[11:01:23 EDT(-0400)] <EricDalquist> I think this is a good opportunity to do that
[11:01:25 EDT(-0400)] <athena7> no the servlet side shouldn't be bad
[11:01:27 EDT(-0400)] <athena7> yep
[11:01:31 EDT(-0400)] <athena7> that shouldn't be hard anyway
[11:01:42 EDT(-0400)] <EricDalquist> yup
[11:01:45 EDT(-0400)] <athena7> but colin yes, please keep me up to date
[11:01:59 EDT(-0400)] <athena7> by the way, what is the reorderer using? is it using the jqueryui 1.5 release?
[11:02:52 EDT(-0400)] * michelled (n=team@142.150.154.197) has left ##uportal
[11:02:55 EDT(-0400)] <colinclark> athena7: Sure, totally. As I say, I've been a bit slack in my communication recently. Now that Jess is onboard and taking some weight off my plate, I'm back to real development work (which is awesome). So don't hesitate to ping me if you want to know what's going on.
[11:03:00 EDT(-0400)] <colinclark> As for our jQuery UI version...
[11:03:10 EDT(-0400)] <colinclark> We're actually using a late prerelease version.
[11:03:15 EDT(-0400)] <colinclark> Somewhere between the beta and the release.
[11:03:30 EDT(-0400)] <colinclark> We'll upgrade to the final 1.5 release shortly.
[11:03:38 EDT(-0400)] <colinclark> First week in July, actually.
[11:03:41 EDT(-0400)] <athena7> sounds great
[11:03:47 EDT(-0400)] <athena7> we need to update uportal to use 1.5
[11:04:01 EDT(-0400)] <colinclark> I think we can help with that if you're short on cycles.
[11:04:27 EDT(-0400)] <athena7> cool
[11:04:32 EDT(-0400)] <athena7> i think i will indeed be short on cycles
[11:52:07 EDT(-0400)] * holdorph (n=holdorph@ip72-201-251-192.ph.ph.cox.net) has joined ##uportal
[11:57:12 EDT(-0400)] <EricDalquist> so ... I know this is last minute ... but I'd like to get a quick survey of if we just just remove services.xml from 3.0.1
[11:57:21 EDT(-0400)] <EricDalquist> I really wish I had thought to do that for 3.0.0
[11:57:32 EDT(-0400)] <EricDalquist> and was reminded of that by Drew's email
[11:57:35 EDT(-0400)] <athena7> what does that do?
[11:57:56 EDT(-0400)] <EricDalquist> you can list some java classes with start/stop methods
[11:58:02 EDT(-0400)] <EricDalquist> and it starts/stops them with the portal
[11:58:19 EDT(-0400)] <EricDalquist> aka a very minimal version of what the spring app-context can do
[12:00:18 EDT(-0400)] <athena7> oh
[12:00:26 EDT(-0400)] <athena7> i didn't even know that was there
[12:00:33 EDT(-0400)] <athena7> hmm
[12:00:51 EDT(-0400)] <EricDalquist> yeah there is nothing in it by default
[12:00:56 EDT(-0400)] <athena7> do some of the toro things use that?
[12:01:00 EDT(-0400)] <EricDalquist> not sure
[12:01:11 EDT(-0400)] <EricDalquist> are they deployed in the uPortal context?
[12:01:29 EDT(-0400)] <athena7> channels yes, portlets no
[12:02:16 EDT(-0400)] <athena7> ok yes, they do
[12:02:34 EDT(-0400)] <EricDalquist> the channels do?
[12:02:41 EDT(-0400)] <athena7> portlets, apparently
[12:02:47 EDT(-0400)] <EricDalquist> really?
[12:03:00 EDT(-0400)] <athena7> looks like
[12:03:08 EDT(-0400)] <athena7> <service>
[12:03:08 EDT(-0400)] <athena7> <name>uPortal Academus Facade</name>
[12:03:08 EDT(-0400)] <athena7> <class>net.unicon.academus.api.uportal.UPortalAcademusFacade</class>
[12:03:09 EDT(-0400)] <athena7> <method type="static">register</method>
[12:03:09 EDT(-0400)] <athena7> </service>
[12:03:19 EDT(-0400)] <EricDalquist> ah
[12:03:37 EDT(-0400)] <EricDalquist> yeah see that would be easy to do as a MethodInvokingFactoryBean in Spring
[12:04:04 EDT(-0400)] <EricDalquist> well maybe for 3.0.1 I just add a giant deprecated warning in the file
[12:04:09 EDT(-0400)] <EricDalquist> and we can look at removing it for 3.1
[12:05:00 EDT(-0400)] <athena7> that sounds like a good idea
[12:05:06 EDT(-0400)] <athena7> i don't think the toro portlets are up3-ready yet
[12:05:15 EDT(-0400)] <athena7> but just in case anyone else is using that stuff
[12:05:27 EDT(-0400)] <EricDalquist> yeah
[12:05:45 EDT(-0400)] <athena7> and in the meantime, a note to nick bolton might be appreciated, so that he knows that's going away
[12:06:03 EDT(-0400)] <athena7> since i think he's looking at some other toro migration issues
[12:06:18 EDT(-0400)] <EricDalquist> ok
[13:50:50 EDT(-0400)] * michelled (n=team@142.150.154.197) has joined ##uportal
[14:03:10 EDT(-0400)] * colinclark (n=colin@bas1-toronto09-1279543604.dsl.bell.ca) has joined ##uportal
[14:06:20 EDT(-0400)] * jessm (n=Jess@cpe-069-134-127-060.nc.res.rr.com) has joined ##uportal
[14:53:21 EDT(-0400)] * apetro-_ (n=apetro@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[14:53:23 EDT(-0400)] * EiNZTEiN (n=einztein@bl11-74-31.dsl.telepac.pt) has joined ##uportal
[14:54:04 EDT(-0400)] <apetro-_> the paths-too-long-for-under-featured-Windows-Zip-extraction problem will afflict uPortal 2.6.1.1 as it does 3.0.1, right?
[14:54:27 EDT(-0400)] <apetro-_> such that 2.6.1 should be released as a not-zip?
[14:54:33 EDT(-0400)] <EricDalquist> no
[14:54:37 EDT(-0400)] <EricDalquist> it won't
[14:54:50 EDT(-0400)] <apetro-_> oh. Ok. that's easier then. .zip it is.
[14:55:47 EDT(-0400)] <holdorph> yes
[14:55:53 EDT(-0400)] <holdorph> it affects all uportal releases
[14:55:56 EDT(-0400)] <holdorph> i've ever worked with
[14:56:08 EDT(-0400)] <holdorph> this comes from many a portlet training class experience
[14:56:44 EDT(-0400)] <EricDalquist> oh
[14:56:48 EDT(-0400)] <EricDalquist> ignore me then
[14:56:49 EDT(-0400)] <apetro-_> ok. Thinking this through further, I guess the thing to prioritize in 2.6.1.1 is making it as available as possible, so I should probably post both formats.
[14:57:23 EDT(-0400)] <apetro-_> and then for 2.6.2 it is probably more appropriate to adopt the solution used in 3.0.1 for avoiding the Zip-extraction-broken issue
[14:57:24 EDT(-0400)] <apetro-_> ?
[14:58:55 EDT(-0400)] <holdorph> i guess to me, it would depend on the timing of the releases. I think it would be strange to put out two releases, regardless of the 'point' value, at the same time, and treat them differently with respect to packaging
[14:59:26 EDT(-0400)] <holdorph> so if 2.6.1.1 and 3.0.1 came out at the same time, but one had a .zip and one did not, that would feel somehow odd to me. but I'm just one voice among the many.
[15:01:36 EDT(-0400)] * esm (n=esm@pool-71-248-84-117.bltmmd.east.verizon.net) has joined ##uportal
[15:29:57 EDT(-0400)] <MarkRogers> on the topic of 2.6.1.1, is there anything that needs to be announced ?
[16:04:13 EDT(-0400)] * awills (n=awills@mtw160-5.ippl.jhu.edu) has joined ##uportal
[16:17:10 EDT(-0400)] <awills> hey folks... anyone still on?
[16:17:15 EDT(-0400)] <EricDalquist> yup
[16:17:32 EDT(-0400)] <awills> Here's a StackOverflowError I just ran into: http://uportal.pastebin.com/m5ae35ec4
[16:17:47 EDT(-0400)] <awills> if you configure RemoteUserPersonManager
[16:18:17 EDT(-0400)] <EricDalquist> hah
[16:18:22 EDT(-0400)] <EricDalquist> yeah
[16:18:25 EDT(-0400)] <EricDalquist> that would be a bug
[16:18:43 EDT(-0400)] * EricDalquist just finished doing 3.0.1
[16:19:41 EDT(-0400)] * awills congratulates EricDalquist on his fine achievement
[16:19:55 EDT(-0400)] <EricDalquist> so yeah ... that's not good ... hrm
[16:20:13 EDT(-0400)] <awills> do we need this PortalHttpServletRequest class? are we using it?
[16:20:18 EDT(-0400)] <EricDalquist> yes
[16:20:20 EDT(-0400)] <EricDalquist> very very much so
[16:20:57 EDT(-0400)] <EricDalquist> the idea was to wrap the standard request to provide uPortal's version of its methods and the ability to change request parameters in the request processing chain ...
[16:21:18 EDT(-0400)] <EricDalquist> and that everything that needed the original request would be complete by then
[16:21:29 EDT(-0400)] <EricDalquist> hrm
[16:21:40 EDT(-0400)] <EricDalquist> I actually have a little more to do for 3.0.1
[16:21:42 EDT(-0400)] <awills> do we need to override getRemoteUser()?
[16:21:42 EDT(-0400)] <EricDalquist> then I can take a look
[16:21:46 EDT(-0400)] <EricDalquist> yes
[16:22:03 EDT(-0400)] <EricDalquist> either there or in another wrapper added not much later
[16:24:28 EDT(-0400)] <awills> off the top of my head... i could use a Filter impl to "wormhole" the actual remoteUser value to the RemoteUserPersonManager only
[16:25:13 EDT(-0400)] <EricDalquist> or we can make RemoteUserPersonManager aware of the wrapper and unwrap it
[16:25:55 EDT(-0400)] <awills> pass it to what?
[16:26:13 EDT(-0400)] <awills> does the PortalHttpServletRequest invoke RemoteUserPersonManager?
[16:26:30 EDT(-0400)] <awills> i suppose it does
[16:26:54 EDT(-0400)] <EricDalquist> yeah, that's why you get a stack overflow
[16:26:56 EDT(-0400)] * awills checks his own pastebin for clarification
[16:27:06 EDT(-0400)] <EricDalquist> and that's what causes the recursion
[16:27:23 EDT(-0400)] * michelled (n=team@142.150.154.197) has left ##uportal
[16:27:24 EDT(-0400)] <awills> yes, actually, that would probably work
[16:27:51 EDT(-0400)] <awills> pass the wrapped req to the limited situation of invoking getPerson()
[16:28:02 EDT(-0400)] <EricDalquist> yeah, need to double check to make sure the other customized bits of the request are OK to ignore for the personManager
[16:28:33 EDT(-0400)] <awills> i'm going to try it myself here
[16:28:42 EDT(-0400)] <EricDalquist> ok
[16:54:41 EDT(-0400)] <EricDalquist> awills: can you create a jira issue for this?
[17:13:48 EDT(-0400)] <awills> sure... just saw this note
[17:14:36 EDT(-0400)] * MarkRogers (n=MarkRoge@addhcp129.cc.umanitoba.ca) has joined ##uportal
[17:23:32 EDT(-0400)] <awills> http://www.ja-sig.org/issues/browse/UP-2092
[17:39:35 EDT(-0400)] <holdorph> athena7 you still around?
[17:42:08 EDT(-0400)] <athena7> sort of
[17:42:09 EDT(-0400)] <athena7> what's up?
[17:42:38 EDT(-0400)] <holdorph> was wondering if you did much with the maven jspc plugin, i saw it in the calendar portlet pom.xml
[17:42:56 EDT(-0400)] <athena7> no i copy/pasted from eric's bookmarks portlet and went "yea! it worked!"
[17:43:04 EDT(-0400)] <holdorph> lol
[17:43:13 EDT(-0400)] <holdorph> ok. i was wondering why i'm getting
[17:43:14 EDT(-0400)] <athena7> and then went back to writing java code
[17:43:21 EDT(-0400)] <athena7> so uh, that's about as much as i know about it
[17:43:24 EDT(-0400)] <holdorph> [WARN] Unknown element (deferred-value) in attribute
[17:43:24 EDT(-0400)] <holdorph> [WARN] Unknown element (deferred-value) in attribute
[17:43:24 EDT(-0400)] <holdorph> [WARN] Unknown element (deferred-value) in attribute
[17:43:30 EDT(-0400)] <holdorph> from it
[17:43:52 EDT(-0400)] <athena7> wow
[17:43:53 EDT(-0400)] <athena7> no idea
[17:43:58 EDT(-0400)] <holdorph> normally I wouldn't care. but I also see in when "running" the portlet in up3 as well.