uPortal IRC Logs-2011-02-09

Owned by Apereo Infrastructure
Last updated: Feb 09, 2011Version comment
6 min read

[11:41:29 CST(-0600)] <EricDalquist> anyone here play with the JPA2 CriteraBuilder before?
[11:53:46 CST(-0600)] <athena> not i
[12:30:20 CST(-0600)] <DrewWills> Hey folks... I could spend some time integrating the subtabs feature from Illinois in the very near future (https://wiki.jasig.org/display/UPC/Tab-SubTab+Functionality)...
[12:30:32 CST(-0600)] <EricDalquist> great
[12:30:42 CST(-0600)] <EricDalquist> I know we're missing some features in trunk they need for integration
[12:30:57 CST(-0600)] <EricDalquist> namely the URL provider code doesn't yet handle 'layout' parameters
[12:31:02 CST(-0600)] <DrewWills> it seems like the reaction from the community was positive... is there new news or thoughts about how that would ork in rel-3-2 or trunk?
[12:31:06 CST(-0600)] <DrewWills> like what features?
[12:31:11 CST(-0600)] <EricDalquist> and the structure and theme prefs stores are not yet refactored
[12:31:39 CST(-0600)] <EricDalquist> they need to be re-written so that there is a clear deliniation between data coming from the request, the session and the db
[12:31:54 CST(-0600)] <EricDalquist> how near is very near future?
[12:32:09 CST(-0600)] <DrewWills> ummm... starting today?
[12:32:13 CST(-0600)] <EricDalquist> (smile)
[12:32:18 CST(-0600)] <DrewWills> or in a few...
[12:32:37 CST(-0600)] <DrewWills> it's percolated high enough in priority that I could dig in, basically
[12:32:52 CST(-0600)] <EricDalquist> I can see if I can get into the URL/prefs stuff on Sunday
[12:35:15 CST(-0600)] <DrewWills> I could warn up on rel-3-2... tackle trunk after your changes... I could also maybe put in a hand on the struct/theme user prefs refactor
[12:35:26 CST(-0600)] <EricDalquist> sounds good
[12:35:31 CST(-0600)] <DrewWills> excellent
[13:12:20 CST(-0600)] <athena> so for the subtabs in trunk
[13:12:35 CST(-0600)] <athena> we'd really need some investigation about how that would fit into the user interface
[13:15:47 CST(-0600)] <athena> EricDalquist: initportal no longer works in trunk
[13:15:58 CST(-0600)] <EricDalquist> yeah I think I just found that too
[13:16:05 CST(-0600)] <athena> http://uportal.pastebin.com/wSTBJqZ8
[13:16:05 CST(-0600)] <EricDalquist> I'll get a fix in justa few minutes
[13:16:07 CST(-0600)] <athena> ah, ok
[13:16:08 CST(-0600)] <athena> thanks (smile)
[13:21:42 CST(-0600)] <EricDalquist> actually athena something just broke here
[13:21:48 CST(-0600)] <EricDalquist> in PortletEntityRegistryImpl
[13:21:49 CST(-0600)] <EricDalquist> in getPortletEntityLock
[13:21:57 CST(-0600)] <EricDalquist> replace "return null;" with "return NoopLock.INSTANCE;"
[13:22:03 CST(-0600)] <EricDalquist> sorry
[13:22:12 CST(-0600)] <athena> ok (smile)
[13:22:18 CST(-0600)] <athena> had initially thought it was my changes
[13:22:39 CST(-0600)] <EricDalquist> no it was the portlet entity sync patch I just did
[13:22:44 CST(-0600)] <EricDalquist> 3.2-patches needs the same fix
[13:23:21 CST(-0600)] <athena> figured it might be (smile)
[13:23:24 CST(-0600)] <athena> no worries!
[13:23:30 CST(-0600)] <athena> thanks for the fix
[13:25:07 CST(-0600)] <athena> close to having that portlet descriptor refactoring thing done
[13:25:22 CST(-0600)] <athena> though i haven't changed the import format yet
[13:25:49 CST(-0600)] <athena> shouldn't be hard now that it's based on JAXB though (smile)
[13:28:28 CST(-0600)] <rickard> DrewWills: Subtab functionality would be a great addition to uPortal, but in the long run, I think a more permanent solution would have to be added. To realize our navigation needs, we need sub-sub-sub-tabs. Don't ask me, I didn't make up the navigation scheme (big grin)
[13:32:52 CST(-0600)] <rickard> athena: Any progress on the content portlet? (wink) I think we'll have to come up with something on our own for the time being. I'm considering writing a simple portlet for editing all content elements from one location. And then use SimpleContentPortlet for the frontend only.
[13:33:17 CST(-0600)] <athena> hey rickard - i'm really sorry, but no
[13:33:43 CST(-0600)] <athena> i unfortunately don't really have time for stuff that's not work-sponsored right now
[13:34:22 CST(-0600)] <athena> my schedule will probably open up for a while in a few weeks, but until then i don't have much time for coding outside of work
[13:34:49 CST(-0600)] <rickard> Ok. Good to know. Then I'll probably go ahead with the admin portlet I mentioned.
[13:36:35 CST(-0600)] <athena> sounds justified (smile)
[13:36:56 CST(-0600)] <athena> i won't forget about it though - hopefully we can get a nice i18n UI sometime soonish
[13:38:54 CST(-0600)] <athena> EricDalquist: just realized that this portlet descriptor key change means i have to edit every .portlet file again (tongue)
[14:16:33 CST(-0600)] <rickard> Referenced file contains errors (http://java.sun.com/xml/ns/portlet/portlet-app_1_0.xsd). For more information, right click on the message in the Problems View and select "Show Details..." portlet.xml /StudentSearchPortlet/WebContent/WEB-INF line 1 XML Problem
[14:16:42 CST(-0600)] <rickard> Anyone else encounter this error in eclipse?
[14:16:54 CST(-0600)] <rickard> I doubt there's an error in portlet-app_1_0.xsd
[14:17:40 CST(-0600)] <DrewWills> i believe I have encountered that problem in eclipse too
[14:21:55 CST(-0600)] <DrewWills> does it make sense to call the Illinois subtabs impl "supertabs?"
[14:22:18 CST(-0600)] <rickard> Isn't it really tab groups?
[14:22:19 CST(-0600)] <DrewWills> it looks like subtabs, but the ones that are added are above in the hierarchy
[14:22:33 CST(-0600)] <DrewWills> yes, illinois calls them tabTags
[14:22:51 CST(-0600)] <DrewWills> but I wonder if another name could offer more clarity
[14:23:02 CST(-0600)] <DrewWills> tabGroup sounds better too
[14:26:09 CST(-0600)] <athena> yeah, you're right that it's not typical tagging behavior
[14:26:29 CST(-0600)] <athena> grouping is probably a bit more accurate
[14:26:38 CST(-0600)] <DrewWills> well, i agree it uses a tag... but that seems like an implementation detail
[14:26:51 CST(-0600)] <athena> right
[14:27:02 CST(-0600)] <athena> from a user experience standpoint it's more of a group than a tag
[14:27:36 CST(-0600)] <DrewWills> basically there has to be some config + documentation somewhere that describes turning this feature on... I want the name to be meaningful and clear
[14:27:44 CST(-0600)] <athena> i think illinois's implimentation lets youmove tabs around within a group, but not between groups
[14:27:51 CST(-0600)] <athena> yeah
[14:28:19 CST(-0600)] <DrewWills> ok, for now I'm going to call the feature "Tab Groups"
[14:28:41 CST(-0600)] <DrewWills> if the community would prefer another name, it can send it along (wink)
[14:28:49 CST(-0600)] <athena> sounds fair (smile)
[15:05:54 CST(-0600)] <holdorph> Multi-level tabs?
[15:06:08 CST(-0600)] <holdorph> nah, guess that's not much better
[15:15:41 CST(-0600)] <rickard> When using the Singleton patter, are there any advantages to creating the object on the first call to getInstance? That is, apart from just creating it during instantiation.
[15:15:48 CST(-0600)] <rickard> pattern even
[15:16:24 CST(-0600)] <holdorph> it can depend on circumstances
[15:16:47 CST(-0600)] <holdorph> like if it takes a while (seconds) to create it, you might not want the first request to have to pay that price, you might want it up before the first request
[15:17:14 CST(-0600)] <holdorph> in other cases, if it's a big memory hog and it doesn't get used all the time, maybe you wait to create it
[15:18:32 CST(-0600)] <rickard> Ok. In this case it's used pretty frequently and isn't much of a memory hog.
[15:18:43 CST(-0600)] <rickard> So whichever then? (smile)
[15:36:40 CST(-0600)] <athena> do we want to start putting <?xml> stuff at the top of our entity files?
[17:00:38 CST(-0600)] <DrewWills> rickard i would say prefer creating a singleton instance as (private) static member right in the declaration... like "private static final FOO = new Foo();"... unless you have a reason not to... avoids many concurrency considerations
[17:03:01 CST(-0600)] <EricDalquist> athena: I think we should
[17:03:09 CST(-0600)] <athena> ok
[17:03:12 CST(-0600)] <athena> sounds good to me (smile)
[17:03:14 CST(-0600)] <EricDalquist> and as we move to JAXB for some of it we should also be referencing schemas
[17:03:31 CST(-0600)] <EricDalquist> https://source.jasig.org/schemas/ exists as a place for final versioned XSDs to end up
[17:03:37 CST(-0600)] <athena> ah hmmm.
[17:03:47 CST(-0600)] <EricDalquist> for example: https://source.jasig.org/schemas/maven-notice-plugin/license-lookup/license-lookup-v1.0.xsd
[17:03:54 CST(-0600)] <athena> do we want to do this stuff as XSDs that get compiled using a maven plugin rather than as java annotations?
[17:04:05 CST(-0600)] <EricDalquist> well you can go the other way too
[17:04:12 CST(-0600)] <EricDalquist> you can generate XSDs from the annotations
[17:04:53 CST(-0600)] <EricDalquist> http://download.oracle.com/docs/cd/E17802_01/webservices/webservices/docs/2.0/jaxb/schemagen.html
[17:05:00 CST(-0600)] <EricDalquist> but having XSDs would be nice
[17:05:09 CST(-0600)] <athena> yeah
[17:05:15 CST(-0600)] <EricDalquist> because then you can use an XML editor and have a clue as to the appropriate elements in the XML document
[17:05:16 CST(-0600)] <athena> just wondering what we actually want to do
[17:05:18 CST(-0600)] <athena> yeah
[17:05:37 CST(-0600)] <athena> there's some advantage to either generating the code from the XSD or vice versa so that we don't have to edit things both places
[17:05:48 CST(-0600)] <EricDalquist> we could add a step as part of the minor releases … generate the entity schemas
[17:07:14 CST(-0600)] <rickard> DrewWills: Ok. Thanks. That's what I have today.