[10:10:24 CST(-0600)] <athena> morning EricDalquist
[10:10:31 CST(-0600)] <athena> how goes the stats work?
[10:10:46 CST(-0600)] <EricDalquist> good morning
[10:10:54 CST(-0600)] <EricDalquist> heh I haven't done anything on it since before the conf
[10:10:59 CST(-0600)] <athena> 
[10:11:09 CST(-0600)] <athena> so the other thing i wanted to mention
[10:11:13 CST(-0600)] <EricDalquist> with the travel issues last week
[10:11:18 CST(-0600)] <athena> yeah, totally get it
[10:11:42 CST(-0600)] <athena> been playing with creating a reusable library like what we use to generate forms from CPD files
[10:11:59 CST(-0600)] <EricDalquist> brb in ~ 5 minutes
[10:12:02 CST(-0600)] <athena> not sure we want to keep the JPA aspect of it - figure mostly we'll want to spring-configure the fields
[10:12:03 CST(-0600)] <athena> oh, sure
[10:49:16 CST(-0600)] <EricDalquist> ok sorry
[10:49:17 CST(-0600)] <EricDalquist> back
[10:49:27 CST(-0600)] <athena> no worries
[10:49:36 CST(-0600)] <EricDalquist> so what is the use case here?
[10:49:43 CST(-0600)] <athena> so specifically the calendar and news portlets
[10:49:49 CST(-0600)] <athena> right now the administrative interface is pretty lame
[10:50:01 CST(-0600)] <EricDalquist> ah ok
[10:50:06 CST(-0600)] <athena> you have to know the bean name of the type of calendar/news item you'd liek to create
[10:50:10 CST(-0600)] <EricDalquist> so a generic library for putting together config mode UIs
[10:50:12 CST(-0600)] <EricDalquist> right?
[10:50:13 CST(-0600)] <athena> then know the expected parameters and add each one
[10:50:14 CST(-0600)] <athena> yeah
[10:50:35 CST(-0600)] <athena> my thought is the calendar/news item types are already configured in spring
[10:50:39 CST(-0600)] <EricDalquist> I wish we had time to just make it possible to include CPDs in META-INF
[10:50:44 CST(-0600)] <EricDalquist> and have uPortal just find them
[10:51:05 CST(-0600)] <athena> well, CPDs wouldn't actually be enough for this use case - we need something more dynamic
[10:51:12 CST(-0600)] <EricDalquist> like extend the PPD format enough to also include the info that is in portlet-definition.xml
[10:51:55 CST(-0600)] <athena> anyway, thinking we could enhance those bean registrations with a list of valid parameters and all the type of data we associate w/ parameters in the PPD - message key, description, field type, etc.
[10:52:40 CST(-0600)] <athena> this all goes in custom database tables, so i don't think we'd want to uportal to manage it via something in META-INF, even if we could
[10:54:34 CST(-0600)] <athena> in my initial attempt i stripped out all the JAXB stuff
[10:54:46 CST(-0600)] <EricDalquist> ah right
[10:54:53 CST(-0600)] <athena> it's nice for the portal, but if we're not actually creating things from XML, some of the structure makes it harder to configure those elements in spring
[10:54:57 CST(-0600)] <EricDalquist> so the META-INF option would just be for a PPD clone
[10:55:00 CST(-0600)] <athena> since you wind up with weird jaxb wrappers around things
[10:55:00 CST(-0600)] <EricDalquist> where it is all portlet preferences
[10:55:03 CST(-0600)] <athena> yeah
[10:55:31 CST(-0600)] <athena> would be cool to pick up stuff in META-INF, but it's not too hard to drop the file in - and you need to register the type in the database to use it anyway
[10:55:37 CST(-0600)] <athena> so don't think that's too critical
[10:56:05 CST(-0600)] <EricDalquist> right, well that would be the change I'd like to make
[10:56:17 CST(-0600)] <EricDalquist> remove the need for portlet-descriptor registration
[10:56:22 CST(-0600)] <athena> ahh, gotcha
[10:56:25 CST(-0600)] <athena> that'd be cool
[10:56:26 CST(-0600)] <EricDalquist> and just discover PPD.xml files via META-INF
[10:56:29 CST(-0600)] <athena> yeah
[10:56:32 CST(-0600)] <athena> that'd be pretty neat
[10:56:36 CST(-0600)] <EricDalquist> then when you publish a new portlet .... tada ..
[10:56:37 CST(-0600)] <EricDalquist> yeah
[10:56:39 CST(-0600)] <EricDalquist> all very doable
[10:56:41 CST(-0600)] <athena> yep
[10:56:45 CST(-0600)] <EricDalquist> just need a free ~ week to do it
[10:56:48 CST(-0600)] <athena>
[10:57:24 CST(-0600)] <athena> so for this use case, do you have strong feelings about keeping it jaxb-ish?
[10:57:56 CST(-0600)] <EricDalquist> nope
[10:58:02 CST(-0600)] <athena> ok
[10:58:13 CST(-0600)] <EricDalquist> my only real preference for JAXB is that its easier than parsing XML by hand
[10:58:16 CST(-0600)] <athena> yeah
[10:58:34 CST(-0600)] <athena> thinking if we need that use case at some point it might be easier to just create a whole separate package of JAXB-based objects and have some code that maps them back to the more vanilla version
[10:58:50 CST(-0600)] <EricDalquist> also you shouldn't have to deal with the JAXBElement objects much
[10:58:51 CST(-0600)] <athena> then we could serialize/deserialize if we needed to, but still have java objects that are a bit easier to work with
[10:59:00 CST(-0600)] <EricDalquist> unless you're interacting with elements that are not top level
[10:59:03 CST(-0600)] <athena> yeah
[11:00:17 CST(-0600)] <athena> if you take a look at userContext.xml you can see some of the awkwardness
[11:03:28 CST(-0600)] <athena> so what's the best place to put this library?
[11:03:58 CST(-0600)] <EricDalquist> that portlet-utils git project probably
[11:04:01 CST(-0600)] <EricDalquist> just add a new module
[11:04:05 CST(-0600)] <athena> ok
[11:04:08 CST(-0600)] <athena> sounds good to me
[11:05:52 CST(-0600)] <athena> oh by the way, did you ever get a chance to talk to the board about the google translate api? do you want me to send an email?
[11:06:33 CST(-0600)] <EricDalquist> I did not
[11:06:37 CST(-0600)] <EricDalquist> if you can that would be great
[11:06:40 CST(-0600)] <athena> sure thing
[11:38:14 CST(-0600)] <b-rock> Greetings uPortal devs: good afternoon. I'm looking for tips or info in debugging this type of import error in the 4.0.2 build http://pastebin.com/2wTeVd2W. outside of these layouts, we can get all of the real data into this build.
[11:38:56 CST(-0600)] <b-rock> not sure if you plan on doing more stuff w/ imports this month.
[11:40:32 CST(-0600)] <EricDalquist> can you pastebin jschraegle.layout
[11:43:28 CST(-0600)] <b-rock> yeah its at the bottom of that paste
[11:43:59 CST(-0600)] <b-rock> I'll do a new pastebin w/ just that...
[11:44:12 CST(-0600)] <EricDalquist> my guess is this line is causing a problem:
[11:44:12 CST(-0600)] <EricDalquist> <dlm:position hidden="false" immutable="false" name="uocHome-lo:null" type="dlm:position" unremovable="false"/>
[11:45:57 CST(-0600)] <b-rock> ok thanks EricDalquist . didn't notice that
[11:46:06 CST(-0600)] <EricDalquist> well the import should be more graceful with that
[11:46:09 CST(-0600)] <EricDalquist> so try removing that line
[11:46:12 CST(-0600)] <EricDalquist> and importing that one file
[11:46:20 CST(-0600)] <EricDalquist> if it works then file a jira issue
[11:46:26 CST(-0600)] <b-rock> sure thing...
[11:46:27 CST(-0600)] <EricDalquist> and post the stack trace and the layout file
[11:46:32 CST(-0600)] <b-rock> ok
[11:46:32 CST(-0600)] <EricDalquist> thanks
[11:46:34 CST(-0600)] <b-rock> np
[11:46:37 CST(-0600)] <EricDalquist> hopefully that is it
[11:46:56 CST(-0600)] <b-rock> we'll know momentarily ...
[11:49:43 CST(-0600)] <EricDalquist> another thing you could do
[11:50:19 CST(-0600)] <EricDalquist> is in RDBMDistributedLayoutStore line 1770
[11:50:38 CST(-0600)] <EricDalquist> add a if (name == null) check
[11:50:56 CST(-0600)] <EricDalquist> to get a better error message that might actually point us to the correct part of the layout
[11:54:17 CST(-0600)] <b-rock> ok. same type of error I think. http://pastebin.com/pQ5EV7Yi I'll try adding in the debug info now and give it another run...
[11:55:50 CST(-0600)] <b-rock> actually i've got a mtg @12 running the whole afternoon so I'm not able to re-run right now.
[11:55:56 CST(-0600)] <EricDalquist> no problem
[11:56:08 CST(-0600)] <b-rock> do you want me to re-run it w/ the added debug and send that in the jira?
[11:56:12 CST(-0600)] <EricDalquist> if nothing else when you get a minute create a jira issue with the stack trace and the layout file
[11:56:20 CST(-0600)] <EricDalquist> just so we don't lose track of the bug
[11:56:33 CST(-0600)] <b-rock> ok. sure I've got time to do that now. thanks for looking at this.
[11:58:28 CST(-0600)] * Daskreech may have to sadly go with svn
[12:02:11 CST(-0600)] <b-rock> ok thanks again EricDalquist. here's the issue. https://issues.jasig.org/browse/UP-3251
[15:28:46 CST(-0600)] <EricDalquist> apparently the locally installable github enterprise edition supports both LDAP and CAS for authn 
[15:29:36 CST(-0600)] <athena> oh neat!
[15:29:38 CST(-0600)] <athena> that's pretty cool
[15:32:22 CST(-0600)] <athena> so i'm writing javadocs
[15:32:29 CST(-0600)] <athena> and i keep trying to end sentences with semicolons
[15:32:33 CST(-0600)] <EricDalquist> lol
[15:32:42 CST(-0600)] <athena> apparently i've been writing too much code lately
[15:34:32 CST(-0600)] <Daskreech>