[07:23:19 EST(-0500)] * jayshao (n=jayshao@66.94.87.210) has joined ##uportal
[08:43:39 EST(-0500)] * jayshao (n=jayshao@campuseai.expedient.com) has joined ##uportal
[09:09:47 EST(-0500)] * clown (n=clown@user383.megabit.utoronto.ca) has joined ##uportal
[09:10:22 EST(-0500)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined ##uportal
[09:12:06 EST(-0500)] * jayshao (n=jayshao@campuseai.expedient.com) has joined ##uportal
[09:17:22 EST(-0500)] <EricDalquist> good morning
[09:17:30 EST(-0500)] <EricDalquist> jayshao: you around?
[09:18:56 EST(-0500)] * jayshao_ (n=jayshao@campuseai.expedient.com) has joined ##uportal
[09:31:10 EST(-0500)] * esm (n=esm@asdf.dkc.jhu.edu) has joined ##uportal
[09:35:21 EST(-0500)] <EricDalquist> morning
[09:49:33 EST(-0500)] * athena7 (n=athena7@lumina.its.yale.edu) has joined ##uportal
[10:00:13 EST(-0500)] * colinclark (n=colin@bas1-toronto09-1279534436.dsl.bell.ca) has joined ##uportal
[10:05:45 EST(-0500)] * anastasiac (n=chatzill@142.150.154.149) has joined ##uportal
[10:25:32 EST(-0500)] <EricDalquist> so I'm trying figure out the best way to allow people that download the quickstart to do uPortal development on it and provide patches back to the uPortal developers
[10:27:49 EST(-0500)] <EricDalquist> one thought is when building the quickstart to include the .svn meta-data directories but that adds an extra 30MB or so (uncompressed) to the quickstart size
[10:30:34 EST(-0500)] <EricDalquist> another idea is to provide an ant task in the quickstart build.xml that would download the version of uPortal from SVN that the quickstart was built with
[10:31:07 EST(-0500)] <athena7> hm, interesting
[10:31:41 EST(-0500)] <athena7> for gary and i's work, we really need a way to move changes in both directions, too
[10:32:12 EST(-0500)] <EricDalquist> both directions? like from SVN into your working copy and also back into SVN
[10:33:42 EST(-0500)] <EricDalquist> I'm thinking a 'developers quick start' would be a handy tool ... something that is pretty much a quickstart but with the uPortal code still attached to SVN
[10:34:03 EST(-0500)] <EricDalquist> maybe that is just another download option since that would be easy enough to package up in the current quickstart building scripts
[10:34:11 EST(-0500)] <athena7> yes, i need to get changes from gary, and he also will need to get changes from me
[10:34:39 EST(-0500)] <athena7> i think we're both going to need to make several iterations of changes to the same files in similar places to get everything finished
[10:34:44 EST(-0500)] <EricDalquist> yup
[10:34:55 EST(-0500)] <EricDalquist> so you both need a SVN attached version of the uPortal code to develop on
[10:35:17 EST(-0500)] <EricDalquist> and for Gary in particular the quickstart format is needed to make it easier for him to work with
[10:35:32 EST(-0500)] <athena7> yeah, that sounds right
[10:36:01 EST(-0500)] <athena7> actually, having a strategy to do all this would help us at yale too
[10:36:10 EST(-0500)] <athena7> when we work with our internal designers
[10:38:08 EST(-0500)] <EricDalquist> ok, easy enough
[10:38:32 EST(-0500)] <EricDalquist> I'll have the quickstart build scripts generate a standard quickstart and a developer-quickstart
[10:38:45 EST(-0500)] <EricDalquist> where the developer quickstart will still be attached to SVN
[10:38:57 EST(-0500)] <EricDalquist> to allow commits, updates, etc...
[10:41:12 EST(-0500)] <athena7> sounds great!
[10:42:59 EST(-0500)] <EricDalquist> ant really needs to release 1.7.1
[10:43:15 EST(-0500)] <EricDalquist> there are some features I really want to use for the uPortal build scripts but can't use 1.7.0 
[10:44:24 EST(-0500)] <jayshao_> I've been itching to do a dev VMWare build that included Eclipse and a JDK as well
[10:58:54 EST(-0500)] <athena7> i'd like to use ant from eclipse again
[10:59:30 EST(-0500)] <EricDalquist> yeah ... it is really annoying doing uP3 dev and not being able to use ant in eclipse
[10:59:46 EST(-0500)] <EricDalquist> it looks like they have a release manager finally and actually have a beta of 1.7.1
[11:00:14 EST(-0500)] <EricDalquist> the problem is the whole way the ant script hooks back into maven breaks in 1.7.0 (but works in 1.7.1 
)
[11:22:07 EST(-0500)] <athena7> yeah
[11:28:02 EST(-0500)] <athena7> does anyone have any insight into the entity types tables in the uportal database?
[11:28:20 EST(-0500)] <athena7> i'd like to start the import/export work for them, but i'm not really sure what i'm looking at
[11:28:36 EST(-0500)] <EricDalquist> hrm
[11:28:47 EST(-0500)] <EricDalquist> let me fire up a squirellsql and look
[11:29:24 EST(-0500)] <athena7> in particular, i wasn't sure if any other tables reference the up_entity_type table
[11:32:29 EST(-0500)] <EricDalquist> hrm ... so what that defines all the 'valid' IBasicEntity types?
[11:33:28 EST(-0500)] <EricDalquist> the only place in uPortal I see that table even being used in RDBMUserIdentityStore ....
[11:33:50 EST(-0500)] <athena7> i have no idea
[11:33:54 EST(-0500)] <athena7> i don't even know what they are
[11:34:07 EST(-0500)] <athena7> or how they're used
[11:35:14 EST(-0500)] <EricDalquist> so the only two classes I see that use that table are EntityTypes and RDBMUserIdentityStore
[11:36:37 EST(-0500)] <EricDalquist> and it does look like EntityTypes is used in quite a few places ... I'd start hunting down those usages to see what exactly is being done with that data
[11:37:34 EST(-0500)] <athena7> thanks
[11:37:39 EST(-0500)] <EricDalquist> yup
[11:37:41 EST(-0500)] <EricDalquist> good luck
[11:37:43 EST(-0500)] <EricDalquist>
[11:37:46 EST(-0500)] <athena7> i want to make sure if there are any other dependencies they get exported correctly
[11:38:01 EST(-0500)] <athena7> i mean i can just export the table data, but i don't know if there's anything that needs to go along with it
[11:38:39 EST(-0500)] <EricDalquist> so I think all the group info relies on those keys
[11:38:50 EST(-0500)] <EricDalquist> like UP_GROUP.ENTITY_TYPE_ID
[11:39:18 EST(-0500)] <EricDalquist> that specifies the types of entities that are in the group as described in UP_ENTITY_TYPE
[11:40:45 EST(-0500)] <athena7> does it actually make sense to import/export entity types? or are these things that are hard-coded and integral to uportal's inner workings?
[11:41:54 EST(-0500)] <EricDalquist> looking at the code I don't think they are hard coded
[11:42:07 EST(-0500)] <EricDalquist> all the groups code goes through that EntityTypes class to determine descriptions and types and such
[11:43:44 EST(-0500)] <athena7> ok
[11:43:54 EST(-0500)] <athena7> so which field do you think we should use as the identifier?
[11:43:59 EST(-0500)] <athena7> the type name? descriptive name?
[11:44:07 EST(-0500)] <EricDalquist> type name
[11:44:11 EST(-0500)] <athena7> ok
[11:44:15 EST(-0500)] <EricDalquist> that would need to be unique I think
[11:44:19 EST(-0500)] <athena7> that'll stay constant between versions?
[11:44:20 EST(-0500)] <athena7> yeah
[11:44:49 EST(-0500)] <EricDalquist> yeah, that is referring to the actual Java class
[11:45:14 EST(-0500)] <athena7> ok
[11:49:45 EST(-0500)] <athena7> lunchtime, back later
[11:50:31 EST(-0500)] * dstn (n=dstn@unaffiliated/dstn) has joined ##uportal
[12:02:24 EST(-0500)] * colinclark (n=colin@142.150.154.101) has joined ##uportal
[12:13:53 EST(-0500)] <EricDalquist> oi .. testing changes to the quickstart build process is like watching paint dry
[12:26:29 EST(-0500)] <dstn> that fun!?
[12:26:43 EST(-0500)] <dstn> I didn't even know... :-D
[12:59:10 EST(-0500)] * michelled (n=team@142.150.154.199) has joined ##uportal
[13:41:57 EST(-0500)] * wilj (n=wilj@wljohnson.ais.fsu.edu) has joined ##uportal
[13:42:59 EST(-0500)] * wilj (n=wilj@wljohnson.ais.fsu.edu) has left ##uportal
[13:43:30 EST(-0500)] * wilj (n=wilj@wljohnson.ais.fsu.edu) has joined ##uportal
[13:47:20 EST(-0500)] <wilj> Hi Eric, I'm Wil Johnson. I'm with the Kuali Student project. I was told by Colin Clark and Jens Hasseur that I could direct any questions about uPortal to you via IRC.
[13:49:37 EST(-0500)] <athena7> looks like EricDalquist might be idle
[13:49:42 EST(-0500)] <athena7> but hello
[13:49:46 EST(-0500)] <wilj> hi there
[13:49:49 EST(-0500)] <EricDalquist> hi
[13:49:52 EST(-0500)] <EricDalquist> I'm back
[13:49:54 EST(-0500)] <athena7> there he is!
[13:49:58 EST(-0500)] <athena7> hurray
[13:50:00 EST(-0500)] <EricDalquist>
[13:50:49 EST(-0500)] <wilj> I'm working on the UI/UX framework for Kuali Student, and we're evaluating AJAX tools in a portal environment, so some of the things we're coming up against aren't explicitly covered in the JSR.
[13:51:11 EST(-0500)] <EricDalquist> ok
[13:51:22 EST(-0500)] <EricDalquist> so you're talking about AJAX tools in a JSR-168 portlet?
[13:52:00 EST(-0500)] <athena7> what kind of ajax tools?
[13:52:04 EST(-0500)] <wilj> yes, and we are specifically looking at GWT at the moment, since it handles a lot of our other issues (i18n, accessibility, etc) automatically
[13:52:34 EST(-0500)] <wilj> we've also been looking at jQuery and FLUID, but we're probably going to go with GWT since we've been able to integrate it with jQuery to get "the best of both worlds"
[13:53:20 EST(-0500)] <athena7> jquery is nice - i'm in the middle of integrating it into the up3 theme
[13:53:34 EST(-0500)] <athena7> GWT does some other kinds of things, though, from what i understand
[13:53:37 EST(-0500)] <EricDalquist> I'm doing design with someone else right now so my answers may have delays but I am checking on IRC regularly
[13:53:46 EST(-0500)] <wilj> ok
[13:54:18 EST(-0500)] <wilj> one of our big questions right now handling the difference between state management in an AJAX app vs. in a traditional portlet
[13:55:02 EST(-0500)] <athena7> session type stuff?
[13:55:04 EST(-0500)] <wilj> such as when someone selects the "edit" option on a portlet, if there is some clean way of handling that client-side via an event to the AJAX app, instead of going round-trip to the portlet
[13:55:37 EST(-0500)] <athena7> ah.
[13:55:49 EST(-0500)] <wilj> with GWT, most of your application is running client-side already, and if the page reloads to react to the edit event, then we have to push our state to the server, then reload it on redraw
[13:56:35 EST(-0500)] <wilj> we've been looking at some other AJAX enabled "portals" (not necessarily JSR168) that might allow us to hook into a JS event of sorts when that happens
[13:56:48 EST(-0500)] <EricDalquist> no there isn't in JSR-168
[13:56:51 EST(-0500)] <wilj> but we'd like to stay as close to JSR168/286 compliance as possible
[13:57:35 EST(-0500)] <EricDalquist> sorry ... brb again
[13:59:44 EST(-0500)] <wilj> our fall-back approach is to have a state management service, and to hook into the browser navigation events
[14:00:10 EST(-0500)] * dstn (n=dstn@unaffiliated/dstn) has left ##uportal
[14:00:12 EST(-0500)] <wilj> on page reload, we could push our state to the service before the navigate away, and then on the render we could pull our state from the service
[14:00:18 EST(-0500)] <wilj> but, it seems a bit messy
[14:01:05 EST(-0500)] <wilj> we're in the R&D phase of our project, so we're very open to alternate approaches
[14:06:13 EST(-0500)] <EricDalquist> ok
[14:06:17 EST(-0500)] <EricDalquist> reading ...
[14:07:57 EST(-0500)] <EricDalquist> so if you want to rely on Portlet state/mode features the only way to use those in JSR-168 is to do full portlet re-loads using URLs generated by the portlet API itself (generating URLs in JavaScript is not OK)
[14:08:29 EST(-0500)] <EricDalquist> we've talked about this with uPortal and the best solution we've talked about is providing a method to render just a portlet by itself
[14:09:26 EST(-0500)] <EricDalquist> this would allow AJAX code to re-render a specific portlet without re-rendering the entire page
[14:10:20 EST(-0500)] <EricDalquist> but with the current JSR-168 APIs there isn't a way to access any portlet specific data or change and portlet specific state information without doing a full portlet render
[14:10:22 EST(-0500)] <EricDalquist> :/
[14:11:41 EST(-0500)] <athena7> having a way to render a portlet by itself would actually be extremely valuable even beyond this issue
[14:11:48 EST(-0500)] <EricDalquist> yup
[14:11:51 EST(-0500)] <EricDalquist> it would
[14:11:51 EST(-0500)] <athena7> we could do ajax loading of portlets in general
[14:11:57 EST(-0500)] <EricDalquist> and it is something we should look at post 3.0
[14:12:02 EST(-0500)] <athena7> yeah
[14:12:04 EST(-0500)] <EricDalquist> I don't think it would be all that difficult
[14:12:12 EST(-0500)] <athena7> i'd really wanted to do it previously and just couldn't quite figure it out
[14:12:16 EST(-0500)] <wilj> so, this isn't something that would be present in the initial 3.0 release then, correct?
[14:12:21 EST(-0500)] <EricDalquist> your other option is to just not use the PortletMode/WindowState constructs from the portal and just track all that info yourself
[14:12:24 EST(-0500)] <athena7> but i think it's completely possible with some help from people who know core code
[14:12:41 EST(-0500)] <EricDalquist> wilj: probably not but it would likely be easy to add in a 3.0.X release
[14:13:00 EST(-0500)] <EricDalquist> also there will be a 3.1 not too long after 3.0 to incorporate some Fluid work
[14:13:07 EST(-0500)] <athena7> yeah i'd be very interested in doing something like this for post 3.0
[14:13:13 EST(-0500)] <wilj> so far what we've done is to write a simple GwtContainerPortlet class that can wrap any GWT application
[14:13:23 EST(-0500)] <athena7> i'm really busy for the next few weeks, but after that i could see putting some time in
[14:13:30 EST(-0500)] <wilj> and in the GWT application we detect if we're being rendered in portlet, or fullscreen, and render accordingly
[14:13:39 EST(-0500)] <wilj> and manage all events client-side
[14:13:41 EST(-0500)] <EricDalquist> cool, does it deal with namespacing?
[14:14:14 EST(-0500)] <wilj> yeah, we namespace based on the portlet instance
[14:14:34 EST(-0500)] <wilj> we take the portlet session id and use that as part of a composite ID for div that the portlet renders
[14:14:41 EST(-0500)] <wilj> then the GWT app binds itself to the div
[14:14:58 EST(-0500)] <wilj> so you can have n number of instances of the same GWT portlet in single page
[14:15:02 EST(-0500)] <wilj> or mix and match them
[14:15:03 EST(-0500)] <EricDalquist> that isn't good enough ... depending on the container the portlet session id is shared
[14:15:19 EST(-0500)] <EricDalquist> if you want to follow spec there is a portlet namespace string available at render time
[14:15:33 EST(-0500)] <EricDalquist> which the portal generates, will be unique on the page and is javascript/css safe
[14:15:34 EST(-0500)] <wilj> ok, that's a 1-liner to switch out, thanks for the heads-up on it
[14:15:41 EST(-0500)] <EricDalquist> no problem
[14:18:19 EST(-0500)] <wilj> athena, you said that you're integrating jQuery with uPortal
[14:18:35 EST(-0500)] <wilj> is this for AJAX enabled portlets? or for look and feel of the portal itself?
[14:21:25 EST(-0500)] <athena7> for the portal preferences editing interface
[14:21:31 EST(-0500)] <athena7> up2.6 has a version with dojo
[14:21:45 EST(-0500)] <athena7> i've decided to switch to jquery rather than upgrading dojo
[14:21:58 EST(-0500)] <athena7> it seems like a better library, and it will be convenient when we do the fluid integration
[14:22:12 EST(-0500)] <athena7> at yale we'll also convert all our portlets over to jquery
[14:22:31 EST(-0500)] <athena7> which means that eventually some of the portlets in the ja-sig repository will likely make use of the jquery libraries as well
[14:22:47 EST(-0500)] <athena7> of course, it will always be possible to include a portlet with another javascript toolkit
[14:23:11 EST(-0500)] <athena7> and jquery is supposed to be good about not conflicting with other libraries, which is one of the reasons it seems like a good choice for the portal itself to use
[14:23:50 EST(-0500)] <wilj> yeah, we just did an evaluation of a few JS libraries, and jQuery was the best in our opinion
[14:24:10 EST(-0500)] <athena7> yeah i've been very impressed by it
[14:24:27 EST(-0500)] <athena7> i took a look at it and it took about half an hours to decide that yes, it really is all that!
[14:26:11 EST(-0500)] <wilj> we built a proof-of-concept app, and then rebuilt it as standalone and as a portlet using each framework, and compared the results
[14:26:55 EST(-0500)] <wilj> GWT still won, in our opinion, but jQuery was the best of the pure-JS libraries, so we plan on doing GWT but supporting jQuery
[14:31:11 EST(-0500)] <EricDalquist> athena7: I think I have a -dev version of the quickstart now
[14:31:33 EST(-0500)] <EricDalquist> I'm doing a full build of it for the trunk then I'll merge the changes over to the theme branch and build a qs for that too
[14:32:00 EST(-0500)] <athena7> awesome!
[14:32:11 EST(-0500)] <athena7> sounds great
[14:32:23 EST(-0500)] <EricDalquist> it looks just like a 'normal' quick start
[14:32:32 EST(-0500)] <EricDalquist> but the uPortal code still has all of the .svn directories
[14:32:38 EST(-0500)] <EricDalquist> so you can still run svn commands against it
[14:33:53 EST(-0500)] <athena7> that's fantastic
[14:33:58 EST(-0500)] <athena7> i'll have to try it out too
[14:34:12 EST(-0500)] * SusanBramhall (i=Susan@dhcp128036196205.central.yale.edu) has joined ##uportal
[14:34:19 EST(-0500)] <EricDalquist> I'll email out links when they're up
[14:34:34 EST(-0500)] <EricDalquist> it takes like 15 mintues for my work pc to generate it 
[14:34:39 EST(-0500)] <athena7> oh sad
[14:34:44 EST(-0500)] <wilj> one last quick question: all of the 3.0 releases on the download page look pretty dated (2006 and earlier)
[14:34:55 EST(-0500)] <athena7> although i guess once it's run you dont' have to run it again? you can just update?
[14:34:57 EST(-0500)] <wilj> is there a better place to get milestone releases of up3?
[14:35:08 EST(-0500)] <EricDalquist> wilj: yes and they are actually based on a different code base than what the upcoming 3.0 release will be
[14:35:21 EST(-0500)] <EricDalquist> are you on the uportal-dev email list by chance?
[14:35:27 EST(-0500)] <wilj> or should I just check it out of svn and build it myself?
[14:35:32 EST(-0500)] <wilj> no, I'm not on the dev email list
[14:35:45 EST(-0500)] <EricDalquist> ok, well if you idle in here I'll post the links here too
[14:35:55 EST(-0500)] <wilj> ah, ok
[14:36:00 EST(-0500)] <wilj> thanks!
[14:36:07 EST(-0500)] <EricDalquist> the quickstart we're talking about is a uportal build with uportal, tomcat, hsql that is already to go
[14:36:14 EST(-0500)] <EricDalquist> and I'm building two versions right now
[14:36:32 EST(-0500)] <EricDalquist> one is the trunk ... the latest code that is very close to being 3.0.0-RC2
[14:36:45 EST(-0500)] <EricDalquist> and the second one that will be a bit later today is for the new theme branch
[14:36:56 EST(-0500)] <EricDalquist> which athena7 and another person are working on
[14:37:11 EST(-0500)] <EricDalquist> to switch to a new default theme, using jquery and such
[14:38:10 EST(-0500)] <wilj> cool, well I'll pick those up when they are ready and start working against them
[14:39:01 EST(-0500)] <wilj> we're fine with developing against prerelease stuff
[14:39:14 EST(-0500)] <EricDalquist> cool, it is pretty stable code
[14:40:11 EST(-0500)] <athena7> maybe once 3.0 is out the door we can take a look at some of the ajax rendering stuff
[14:40:17 EST(-0500)] <EricDalquist> yeah
[14:40:28 EST(-0500)] <EricDalquist> peter and I had talked about that quite a bit with the old sandbox code
[14:40:42 EST(-0500)] <EricDalquist> have the rendering pipeline not bother with channel rendering at all
[14:40:54 EST(-0500)] <EricDalquist> and have the channels rendered with AJAX callbacks
[14:41:21 EST(-0500)] <EricDalquist> breaks uPortal's concept of rendering groups
[14:41:26 EST(-0500)] <EricDalquist> but that's probably ok for most things
[14:45:53 EST(-0500)] <athena7> cool
[14:50:15 EST(-0500)] <EricDalquist> yay ... 15 minutes 55 seconds to run a quickstart build
[14:51:17 EST(-0500)] <athena7> hehe
[14:51:42 EST(-0500)] <athena7> i think i just finished the import/export stuff for the four entity types we wanted to add support for
[14:57:19 EST(-0500)] <EricDalquist> cool
[14:57:24 EST(-0500)] <athena7> yeah
[14:57:54 EST(-0500)] <EricDalquist> https://mywebspace.wisc.edu/dalquist/web/JA-SIG/uPortal/uPortal-3.0.0-RC2-SNAPSHOT-quick-start/
[14:58:03 EST(-0500)] <EricDalquist> quickstarts of the latest code
[14:58:11 EST(-0500)] <EricDalquist> including -dev quickstarts
[14:58:22 EST(-0500)] <EricDalquist> I'll get the changed merged over to the theme branch and re-build that
[14:58:45 EST(-0500)] <athena7> ok
[14:59:01 EST(-0500)] <athena7> maybe i'll go through this stuff and check in the jquery stuff i've got, so the one you generate will have it?
[14:59:13 EST(-0500)] <EricDalquist> sounds good
[15:10:51 EST(-0500)] <athena7> ooh!
[15:11:09 EST(-0500)] <athena7> i got the jquery interface.js drag and drop working that parker emailed
[15:11:18 EST(-0500)] <EricDalquist> cool
[15:12:57 EST(-0500)] <athena7> very
[15:20:26 EST(-0500)] <athena7> i'm almost done with this part
[15:20:38 EST(-0500)] <athena7> i'll check it in in a few
[15:20:54 EST(-0500)] <EricDalquist> sounds good
[15:21:01 EST(-0500)] <EricDalquist> I'm going to start my merging from trunk in a few as well
[15:22:05 EST(-0500)] <athena7> oh man, this works so much better!
[15:23:27 EST(-0500)] * lescour (n=lescour@129.244.24.211) has joined ##uportal
[15:24:40 EST(-0500)] <athena7> i'm going to leave all the dojo resources in svn for the moment
[15:24:48 EST(-0500)] <athena7> but i anticipate deleting them soon, once everything is set
[15:25:01 EST(-0500)] <EricDalquist> cool
[15:25:16 EST(-0500)] <EricDalquist> it will be good to shed that and the 2.6 theme and corresponding files
[15:25:24 EST(-0500)] <athena7> absolutely
[15:25:31 EST(-0500)] <athena7> i'll get rid of that stuff soon too
[15:26:06 EST(-0500)] <EricDalquist> is it ok if I start commiting my changes to the theme branch?
[15:26:50 EST(-0500)] <athena7> sure
[15:27:01 EST(-0500)] <EricDalquist> ok
[15:32:50 EST(-0500)] <athena7> we'll need to do something about the menus though
[15:32:53 EST(-0500)] <athena7> the theme is ugly.
[15:33:10 EST(-0500)] <athena7> i'll have to look and see if we can recycle any of the old css
[15:33:20 EST(-0500)] <EricDalquist> ok
[15:33:57 EST(-0500)] <athena7> ok, all the jquery stuff is in
[15:34:04 EST(-0500)] <athena7> and working on checking in my changes now
General
Content
Integrations