uPortal IRC Logs-2009-05-11

Owned by Apereo Infrastructure
Last updated: May 11, 2009Version comment
7 min read

[02:07:00 EDT(-0400)] * Sememmon (n=Sememmon@ip70-190-32-223.ph.ph.cox.net) has joined ##uportal
[08:54:52 EDT(-0400)] * dstn (n=dstn@unaffiliated/dstn) has joined ##uportal
[09:10:52 EDT(-0400)] * fj4000 (n=Jacob@142.150.154.106) has joined ##uportal
[09:24:58 EDT(-0400)] * lennard1 (n=sparhk@ip68-98-56-21.ph.ph.cox.net) has joined ##uportal
[09:25:13 EDT(-0400)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined ##uportal
[09:29:17 EDT(-0400)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined ##uportal
[09:31:54 EDT(-0400)] * [jlee] (n=jlee@adsl-074-184-125-241.sip.asm.bellsouth.net) has joined ##uportal
[10:02:12 EDT(-0400)] * lennard1 (n=sparhk@ip68-98-56-21.ph.ph.cox.net) has left ##uportal
[10:21:34 EDT(-0400)] * lennard1 (n=sparhk@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[10:28:36 EDT(-0400)] * michelled (n=team@142.150.154.193) has joined ##uportal
[11:22:39 EDT(-0400)] * awills (n=awills@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[11:32:25 EDT(-0400)] * colinclark (n=colin@bas2-toronto09-1176131435.dsl.bell.ca) has joined ##uportal
[11:36:13 EDT(-0400)] * holdorph (n=holdorph@uni1.unicon.net) has joined ##uportal
[11:42:51 EDT(-0400)] * jessm (n=Jess@c-71-232-1-65.hsd1.ma.comcast.net) has joined ##uportal
[11:45:54 EDT(-0400)] * lennard1 (n=sparhk@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[12:06:46 EDT(-0400)] * lennard1 (n=sparhk@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[12:08:55 EDT(-0400)] * Sememmon (n=Sememmon@uni1.unicon.net) has joined ##uportal
[12:25:24 EDT(-0400)] * fj4000 (n=Jacob@142.150.154.106) has joined ##uportal
[13:12:12 EDT(-0400)] <awills> morning people
[13:12:58 EDT(-0400)] <awills> I'm doing some work on the plan for channel definition persistence posted by Drew Mazurek to the dev list recently
[13:13:41 EDT(-0400)] <awills> Part of what I'm looking at is a JPA-backed impl of IChannelRegistryStore
[13:14:22 EDT(-0400)] <awills> I'm reviewing some refactoring opportunities, and I wonder if anyone caan look with me at this: http://uportal.pastebin.com/d17645e45
[13:14:59 EDT(-0400)] <awills> there are several methods that follow this sort of pattern
[13:15:36 EDT(-0400)] <awills> i'd like to change "List channelDefs" to "Set<ChannelDefinition> channelDefs"...
[13:16:00 EDT(-0400)] <EricDalquist> makes sense to me
[13:16:18 EDT(-0400)] <EricDalquist> also lets not throw SQLException everywhere (smile)
[13:16:33 EDT(-0400)] <awills> and I'm wondering if others feel that is safe, or too much risk that there are logically different ChannelDefinition objects that nevertheless return equals() == true
[13:16:39 EDT(-0400)] <awills> sounds good
[13:17:15 EDT(-0400)] <EricDalquist> I think that sounds safe
[13:17:31 EDT(-0400)] <EricDalquist> I'm assuming you'll make sure the ChannelDefinition has a good equals and hashCode impl{color}
[13:17:39 EDT(-0400)] <awills> great... that's how i feel... just getting a second opinion
[13:17:55 EDT(-0400)] <awills> i'll look at that
[13:19:39 EDT(-0400)] <awills> though it wouldn't shock me terribly if somewhere, somehow, someone has uP data dating back to 2.1 or somesuch that actually behaves differently with this change... but if that's the case, it's data that needs looking at anyway
[13:19:51 EDT(-0400)] <EricDalquist> yup
[13:19:59 EDT(-0400)] <EricDalquist> oh and during your work
[13:20:26 EDT(-0400)] <EricDalquist> we need to start enforcing the following for fnames: [0-9a-zA-Z_-]
[13:20:33 EDT(-0400)] <EricDalquist> only those characters should be allowed
[13:20:41 EDT(-0400)] <EricDalquist> and that column in the DB should be unique
[13:21:03 EDT(-0400)] <awills> sounds good... very compelling in fact
[13:21:05 EDT(-0400)] <EricDalquist> that charset restriction is needed for the URL work I'm doing
[13:22:08 EDT(-0400)] <awills> the new URL work is a popular initiative... it seems to come up a lot
[13:22:22 EDT(-0400)] <EricDalquist> yeah
[13:22:34 EDT(-0400)] <EricDalquist> it will be good to get done
[13:33:23 EDT(-0400)] <athena> that all sounds fantastic to me
[13:33:38 EDT(-0400)] <athena> i might suggest we think about exceptions in general during this work too
[13:33:48 EDT(-0400)] <EricDalquist> yes
[13:33:50 EDT(-0400)] <dstn> fname restriction would be very nice
[13:33:52 EDT(-0400)] <athena> like currently if you try and create a new channel with an already-in-use fname, you don't ever get a warning
[13:33:59 EDT(-0400)] <EricDalquist> yup
[13:34:02 EDT(-0400)] <athena> adn then it sends a non-sensical exception at the end
[13:34:04 EDT(-0400)] <dstn> just found some issues with the fnames and the flyout navs
[13:34:12 EDT(-0400)] <EricDalquist> oh yeah
[13:34:14 EDT(-0400)] <athena> and you lose your in-progress portlet and are very sad
[13:34:19 EDT(-0400)] <EricDalquist> we had an fname break our UI a few days ago
[13:34:21 EDT(-0400)] <athena> what's the issue dstn?
[13:34:25 EDT(-0400)] <EricDalquist> since the layout includes them as CSS class names
[13:34:31 EDT(-0400)] <athena> ah
[13:34:33 EDT(-0400)] <EricDalquist> bad things happen if an fname has a " in it (smile)
[13:34:39 EDT(-0400)] <athena> yeah
[13:34:47 EDT(-0400)] <athena> we need to address that stuff
[13:34:53 EDT(-0400)] <EricDalquist> eh
[13:35:01 EDT(-0400)] <EricDalquist> really it is all fixed by restricting the fname charset
[13:35:06 EDT(-0400)] <athena> i went through the portlet admin portlet recently and wrapped all user-input string data in appropriate tags
[13:35:12 EDT(-0400)] <athena> yeah, that will definitely help a lot
[13:35:21 EDT(-0400)] <dstn> a, 2 issues, the flayout navs use the ID attribute so there is potential for multiple IDs with the same name and the other issue is if it has an invalid character in it, like the portlet manager's fname has a slash
[13:36:01 EDT(-0400)] <athena> the flyout nav uses the fname as the element ID?
[13:36:10 EDT(-0400)] <dstn> <li class="portal-subnav first" id="uPfname_MyWeather">
[13:36:18 EDT(-0400)] <athena> oh ick :/
[13:36:28 EDT(-0400)] <athena> i don't know why it's using that rather than the actual id
[13:36:32 EDT(-0400)] <athena> that's strange
[13:36:45 EDT(-0400)] <athena> hm.
[13:36:49 EDT(-0400)] <dstn> I was just filing jiras for em
[13:37:01 EDT(-0400)] <athena> thanks for filing
[13:37:11 EDT(-0400)] <dstn> haven't had a chance to look at it, found them troubleshooting another prob
[13:37:17 EDT(-0400)] <athena> gotcha
[13:37:26 EDT(-0400)] <athena> i checked in a fix for the columns last night
[13:38:10 EDT(-0400)] <athena> argh i keep having problems w/ my eclipse svn getting corrupted
[13:38:17 EDT(-0400)] <athena> it's fine on disk but eclipse loses track of things (sad)
[13:38:28 EDT(-0400)] <athena> and i can't figure out how to fix it, aside from doing a fresh checkout
[13:38:47 EDT(-0400)] <dstn> uh oh
[13:38:51 EDT(-0400)] <dstn> (sad)
[13:42:23 EDT(-0400)] <dstn> so I filed UP-2438 for the ID attributes and the restriction of the fnames should solve the other one
[13:42:39 EDT(-0400)] <athena> awesome
[14:11:43 EDT(-0400)] * awills1 (n=awills@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[14:14:31 EDT(-0400)] <lennard1> Eric?
[14:14:36 EDT(-0400)] <EricDalquist> hello
[14:14:50 EDT(-0400)] <lennard1> Are you aware of anyone that has ran uPortal 2.6 with tomcat 6?
[14:15:11 EDT(-0400)] * Sememmon (n=Sememmon@uni1.unicon.net) has joined ##uportal
[14:15:19 EDT(-0400)] <lennard1> in a prod type env?
[14:15:24 EDT(-0400)] <EricDalquist> no
[14:15:36 EDT(-0400)] <EricDalquist> but my guess is there would be no issue
[14:16:05 EDT(-0400)] <lennard1> I found something in an email list where a user or two had some issues and you were helping them out
[14:16:33 EDT(-0400)] <lennard1> Personally... I would prefer to wait for the tomcat upgrade until we roll to uPortal 3
[14:17:21 EDT(-0400)] <lennard1> but if you think it will be 'easy' that will likely result in us moving to tomcat 6 sooner.
[14:17:28 EDT(-0400)] <EricDalquist> so
[14:17:42 EDT(-0400)] <EricDalquist> I think it should work
[14:17:53 EDT(-0400)] <EricDalquist> once you do the same things to it you have to do for 3
[14:17:57 EDT(-0400)] <EricDalquist> primarily setup shared/lib
[14:18:31 EDT(-0400)] <lennard1> would imagine the ant tasks will need to change a bit and yeah, I am aware that tomcat 6 has a different structure
[14:19:22 EDT(-0400)] <EricDalquist> it doesn't really though
[14:19:31 EDT(-0400)] <EricDalquist> there is just no shared/lib there by default
[14:19:38 EDT(-0400)] <EricDalquist> once you add that the structure should be the same
[14:19:56 EDT(-0400)] <lennard1> k
[14:20:47 EDT(-0400)] <lennard1> so you would classify the change as easy but you don't know of anyone who is currently doing that in production today, right?
[14:21:25 EDT(-0400)] <EricDalquist> right
[14:21:40 EDT(-0400)] <lennard1> thanks!
[14:21:55 EDT(-0400)] <EricDalquist> no problem
[15:20:28 EDT(-0400)] * fj4000 (n=Jacob@142.150.154.106) has joined ##uportal
[15:44:58 EDT(-0400)] <dstn> EricDalquist, are you guys running 3.1 or 3.0?
[15:48:11 EDT(-0400)] <EricDalquist> 3.1
[15:48:18 EDT(-0400)] <EricDalquist> just updated a week a go
[15:49:21 EDT(-0400)] <dstn> Have you seen any issues with dragging and dropping portlets that have jQuery within them?
[15:49:52 EDT(-0400)] <EricDalquist> we don't have DnD enabled yet
[15:50:00 EDT(-0400)] <EricDalquist> yay for politics (tongue)
[15:50:03 EDT(-0400)] <dstn> We are getting some portlets that on dragging them, spit out "unexpected call to method or property"
[15:50:05 EDT(-0400)] <dstn> LOL
[15:50:08 EDT(-0400)] <EricDalquist> our 3.1 portal looks exactly like our 2.5 portal did
[15:50:34 EDT(-0400)] <dstn> ya, I am wishing our portal didn't at this point
[16:16:24 EDT(-0400)] <athena> dstn: have you filed a jira for that issue? does it happen in the vanillla uportal distribution?
[17:15:11 EDT(-0400)] <dstn> I don't have a vanilla uportal handy so I haven't tried yet
[17:15:39 EDT(-0400)] <dstn> I'm waiting on filing a jira before determining if its a custom issue or a general issue
[17:26:06 EDT(-0400)] * holdorph (n=holdorph@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[17:27:53 EDT(-0400)] <awills> so inconnection with IChannelRegistryStore not throwing SQLException – can we go as far as not declaring checked exceptions on any of the methods? currently 100% of them say "throws Exception"
[17:28:21 EDT(-0400)] <EricDalquist> YES
[17:28:35 EDT(-0400)] <awills> i'd be happy to refactor to appropriate RuntimeException instances
[17:28:38 EDT(-0400)] <awills> lol, cool
[17:28:58 EDT(-0400)] <awills> i'm glaf you feel strongly wnough to use CAPITAL LETTERS
[17:29:01 EDT(-0400)] <EricDalquist> +∞
[17:29:11 EDT(-0400)] <EricDalquist> is that less subtle?
[17:29:36 EDT(-0400)] <awills> nice... + many from me as well then
[17:30:28 EDT(-0400)] * athena cheers and adds several more ++++++s
[17:30:49 EDT(-0400)] <EricDalquist> yeah, as I said in that email. I think as we approach these refactorings we should feel free to create completely new DAO interfaces and even data objects as needed
[17:30:58 EDT(-0400)] <athena> fantastic
[17:31:05 EDT(-0400)] <athena> the current stuff is mean and painful
[17:31:09 EDT(-0400)] <EricDalquist> now that channels are a smaller and smaller part of uPortal we can be more free with the internal APIs
[17:31:20 EDT(-0400)] <awills> great
[17:32:40 EDT(-0400)] <athena> yeah that's a good point
[17:33:05 EDT(-0400)] <athena> by the way, how is the quicklinks stuff currently defined?
[17:33:17 EDT(-0400)] <athena> did that get added directly to the channel manager UI?
[17:33:39 EDT(-0400)] <EricDalquist> I think it is just a manual channel param right now
[17:33:45 EDT(-0400)] <athena> ah, ok
[17:35:10 EDT(-0400)] <athena> we may want to think about havign some standard set of options that apply to all channels
[17:35:19 EDT(-0400)] <athena> like that, the chrome options, whatever
[17:36:07 EDT(-0400)] <EricDalquist> yeah
[17:36:22 EDT(-0400)] <EricDalquist> I think it would be good to have two sources of UI config for the portlet manager
[17:36:25 EDT(-0400)] <awills> yeah we need the global CPD (smile)
[17:36:31 EDT(-0400)] <awills> or something like
[17:36:32 EDT(-0400)] <EricDalquist> yeah
[17:36:47 EDT(-0400)] <EricDalquist> a portal publishing definition (could just be hard-coded JSP for all I care)
[17:36:55 EDT(-0400)] <EricDalquist> plus the type specific CPD
[17:37:07 EDT(-0400)] <EricDalquist> hopefully now that this is JSP based editing the UI will be a bit more sane
[17:37:27 EDT(-0400)] <athena> yeah
[17:37:45 EDT(-0400)] <athena> right now it's actually parsing the CPD and providing parameters based on that
[17:38:10 EDT(-0400)] <EricDalquist> great
[17:38:23 EDT(-0400)] <EricDalquist> so one thing that is kind-of hacked right now is portlet preference support in CPDs
[17:38:32 EDT(-0400)] <athena> so one question i have is whether the old channel will continue to work on portlets created w/ the new tool
[17:38:33 EDT(-0400)] <athena> ah
[17:38:42 EDT(-0400)] <EricDalquist> in 3.1 channel manager was fixed to really handle portlet preferences
[17:38:55 EDT(-0400)] <EricDalquist> but CPDs still use the PORTLET. prefixed channel parameters
[17:38:58 EDT(-0400)] <EricDalquist> which works
[17:39:00 EDT(-0400)] <EricDalquist> but is limited
[17:39:07 EDT(-0400)] <EricDalquist> since preferences don't match parameters exactly
[17:39:28 EDT(-0400)] <EricDalquist> athena: honestly if the code is up to snuff I vote to delete the old channel
[17:39:36 EDT(-0400)] <athena> hey i'd be fine with that
[17:39:57 EDT(-0400)] <athena> are people going to have an issue w/ a portlet that requires ajax though?
[17:40:23 EDT(-0400)] <EricDalquist> Hopefully not
[17:40:26 EDT(-0400)] <EricDalquist> it is just an admin portlet
[17:40:29 EDT(-0400)] <athena> ok
[17:40:36 EDT(-0400)] <athena> in that case maybe that's something to consider
[17:44:25 EDT(-0400)] <awills> is it ok if I use PortalException commonly (which extends RuntimeException)? Looks like it passes problems info to JMX, which is probably desireable, unless that's already done at another level of the call stack or somesuch
[17:44:41 EDT(-0400)] <EricDalquist> sounds good to me
[17:44:44 EDT(-0400)] <awills> cool
[17:44:53 EDT(-0400)] <EricDalquist> I'd also look at Springs data access exception hierarchy
[17:45:03 EDT(-0400)] <EricDalquist> they have some well typed exceptions too
[18:42:52 EDT(-0400)] * lennard1 (n=sparhk@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[19:05:18 EDT(-0400)] * lennard1 (n=sparhk@wsip-98-174-242-39.ph.ph.cox.net) has left ##uportal