uPortal IRC Logs-2008-06-12

[08:07:33 EDT(-0400)] * dstn (i=dstn@unaffiliated/dstn) has joined ##uportal
[08:30:48 EDT(-0400)] * athena7 (n=athena7@c-24-2-189-230.hsd1.mn.comcast.net) has joined ##uportal
[08:46:52 EDT(-0400)] * MarkRogers (n=MarkRoge@addhcp129.cc.umanitoba.ca) has joined ##uportal
[08:49:06 EDT(-0400)] * esm (n=esm@207-53-192-202.dynamic-dsl.qis.net) has joined ##uportal
[09:18:59 EDT(-0400)] * esm_ (n=esm@207-53-192-202.dynamic-dsl.qis.net) has joined ##uportal
[09:53:51 EDT(-0400)] * colinclark (n=colin@bas1-toronto09-1279543604.dsl.bell.ca) has joined ##uportal
[09:57:30 EDT(-0400)] * esm (n=esm@128.220.212.115) has joined ##uportal
[10:18:40 EDT(-0400)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined ##uportal
[10:22:46 EDT(-0400)] <dstn> EricDalquist, do you have any recommendations for books on JMX?
[10:23:08 EDT(-0400)] <EricDalquist> nope, I don't think I've ever read one
[10:23:52 EDT(-0400)] <dstn> So where did you learn it from?
[10:24:07 EDT(-0400)] <EricDalquist> I don't really do much with it
[10:24:27 EDT(-0400)] <EricDalquist> I figured out how to expose stuff reading the spring docs
[10:27:02 EDT(-0400)] <EricDalquist> and reading it just read the javadocs from the JDK
[10:37:34 EDT(-0400)] <MarkRogers> this may be a dumb question (but I'm not proud) ... where or for what is hibernate used in uportal?
[10:38:22 EDT(-0400)] <EricDalquist> it is used for persistence layer for the portlet object model
[10:38:36 EDT(-0400)] <EricDalquist> and as other DAOs are added or re-written it will be used for those as well
[10:38:46 EDT(-0400)] <EricDalquist> do you have background on what hibernate is/does?
[10:40:10 EDT(-0400)] <MarkRogers> just trying to learn it now ... reading Pro Hibernate 3 ... not sure that I really understand the design pattern
[10:40:51 EDT(-0400)] <EricDalquist> if you're looking for a good book on it I would recommend Java Persistence with Hibernate
[10:40:56 EDT(-0400)] <EricDalquist> it is much closer to what uP3 does
[10:41:07 EDT(-0400)] <MarkRogers> thanks
[10:41:38 EDT(-0400)] <EricDalquist> uPortal in the Java code actually uses JPA to load/store/query persisted objects
[10:41:45 EDT(-0400)] <MarkRogers> http://www.manning.com/bauer2/
[10:42:01 EDT(-0400)] <EricDalquist> yup
[10:42:16 EDT(-0400)] <EricDalquist> but JPA is just an API specification that is part of J2EE 5
[10:42:32 EDT(-0400)] <EricDalquist> you have to use some other object relational mapping tool under it
[10:42:37 EDT(-0400)] <EricDalquist> so we use hibernate for that
[10:43:18 EDT(-0400)] <MarkRogers> ah, I saw the references to jpa while looking for the hibernate.connection information that I expected to see somewhere in the code
[10:43:38 EDT(-0400)] <EricDalquist> yeah
[10:44:11 EDT(-0400)] <EricDalquist> so the goal is that as much as possible the code just references JPA apis (this isn't completely possible)
[10:44:46 EDT(-0400)] <EricDalquist> and uPortal uses the JPA annotations to do the mapping
[10:45:07 EDT(-0400)] <EricDalquist> so an example here is found looking at: PortletDefinitionImpl
[10:46:01 EDT(-0400)] <EricDalquist> that class is annotated to map its properties to a database table
[10:46:21 EDT(-0400)] <EricDalquist> then JpaPortletDefinitionDao does the persistence/loading of the annotated object
[10:49:17 EDT(-0400)] * esm (n=esm@128.220.212.115) has joined ##uportal
[10:51:28 EDT(-0400)] <MarkRogers> I think I ran across that last night
[10:53:54 EDT(-0400)] <MarkRogers> thanks, that helps a lot
[11:00:08 EDT(-0400)] * michelled (n=team@142.150.154.197) has joined ##uportal
[11:30:45 EDT(-0400)] * colinclark (n=colin@142.150.154.101) has joined ##uportal
[11:49:32 EDT(-0400)] * holdorph (n=holdorph@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[11:50:25 EDT(-0400)] * esm (n=esm@128.220.212.115) has joined ##uportal
[11:52:15 EDT(-0400)] <dstn> hey EricDalquist, is it possible to build quickstarts with up3 that are specific to your environment?
[11:52:48 EDT(-0400)] <EricDalquist> um, if your environment uses HSQL then yes it would be very easy to do
[11:53:07 EDT(-0400)] <EricDalquist> if it isn't using hsql the assembly build.xml would likely need some minor work to deal with the database stuff
[11:53:41 EDT(-0400)] <dstn> we are thinking about the possibility of building quickstart environments for our developers
[11:53:58 EDT(-0400)] <dstn> just wanted to get an idea if it was even remotely possible
[11:54:10 EDT(-0400)] <athena7> i think that's totally the right thing to do
[11:54:23 EDT(-0400)] <EricDalquist> it should be very possible
[11:54:35 EDT(-0400)] <EricDalquist> though how you do the database would be interesting
[11:55:00 EDT(-0400)] <EricDalquist> since what the current build.xml does is start a local HSQL instance, run initportal against it and the local tomcat
[11:55:10 EDT(-0400)] <EricDalquist> then copy the data files from hsql into the package
[11:55:21 EDT(-0400)] <EricDalquist> doing that with a different database might be trickey
[11:55:35 EDT(-0400)] <EricDalquist> since you can't exactly package them in the zip
[11:55:46 EDT(-0400)] <athena7> might want to just either package it with hsql or have it packaged with no database initialization
[11:55:56 EDT(-0400)] <EricDalquist> yeah
[11:56:01 EDT(-0400)] <athena7> since in the kind of environment dstn is probably talkign about you may have more than one person linking to the same db
[11:56:09 EDT(-0400)] <athena7> i think it would be really useful for designers
[11:56:11 EDT(-0400)] <EricDalquist> you can just instruct people to run initdb (once 3.0.1 is out)
[11:56:20 EDT(-0400)] <EricDalquist> or HSQL if your devs aren't doing DB specific work
[13:15:28 EDT(-0400)] * EiNZTEiN (n=einztein@bl7-141-230.dsl.telepac.pt) has joined ##uportal
[14:08:50 EDT(-0400)] <EricDalquist> yay, only one or two more jira issues!
[14:40:40 EDT(-0400)] * colinclark (n=colin@142.150.154.101) has joined ##uportal
[14:49:25 EDT(-0400)] <athena7> ooh
[14:50:10 EDT(-0400)] <athena7> and i think matt's fixed up-2076
[14:50:24 EDT(-0400)] <EricDalquist> nice
[14:50:36 EDT(-0400)] <athena7> and gary's working on 2070
[14:50:39 EDT(-0400)] <athena7> just talked to them
[14:50:42 EDT(-0400)] <athena7> so great
[14:51:13 EDT(-0400)] <EricDalquist> any ideas for http://www.ja-sig.org/issues/browse/UP-2023
[14:51:33 EDT(-0400)] <athena7> you know i'd been wondering about that
[14:51:54 EDT(-0400)] <athena7> is there a java class we can easily get that into?
[14:52:14 EDT(-0400)] <EricDalquist> which ones set the AJAX params and such
[14:53:00 EDT(-0400)] <athena7> the authenticated variable is the only one that' snot set in the xsl now
[14:53:06 EDT(-0400)] <athena7> and that's set in a dlm parameter processor
[14:53:13 EDT(-0400)] <athena7> which works well, but won't work for this (sad)
[14:53:18 EDT(-0400)] <EricDalquist> why not this?
[14:53:21 EDT(-0400)] <athena7> because those don't get applied to guest layouts
[14:53:25 EDT(-0400)] <EricDalquist> ahhh
[14:53:27 EDT(-0400)] <EricDalquist> darn guests
[14:53:27 EDT(-0400)] <athena7> so it would only work for logged in users
[14:53:34 EDT(-0400)] <athena7> i think it's because the guest layout is immutable
[14:54:21 EDT(-0400)] <EricDalquist> yeah
[14:54:34 EDT(-0400)] <EricDalquist> I'm going to look into doing it via a Xalan handler
[14:55:37 EDT(-0400)] <EricDalquist> I think that will be the last issue then
[14:56:19 EDT(-0400)] <athena7> ok
[14:56:22 EDT(-0400)] <athena7> neat (smile)
[14:57:52 EDT(-0400)] <EricDalquist> so its looking more and more like we're going to be implementing 'virtual portals' eventually
[14:58:06 EDT(-0400)] <athena7> oh cool
[14:58:10 EDT(-0400)] <EricDalquist> multiple domain names pointed at one uPortal instance, each getting its own skin
[14:58:26 EDT(-0400)] <EricDalquist> and of course targeted content base on who the user is and where they are coming from
[14:59:07 EDT(-0400)] <athena7> neat
[14:59:19 EDT(-0400)] <athena7> oh no my pet ticket fell out of the bucket!
[14:59:26 EDT(-0400)] * athena7 hovers sadly over her ticket
[14:59:28 EDT(-0400)] <athena7> haha
[14:59:50 EDT(-0400)] <EricDalquist> ?
[15:00:00 EDT(-0400)] <athena7> the import/export postgres one
[15:00:05 EDT(-0400)] <EricDalquist> ah
[15:00:07 EDT(-0400)] <EricDalquist> lol
[15:00:10 EDT(-0400)] <EricDalquist> yeah sorry
[15:00:28 EDT(-0400)] <EricDalquist> though I did update the hibernate jars a patch revision or two each
[15:00:33 EDT(-0400)] <EricDalquist> so maybe you can try again with the patches branch
[15:00:37 EDT(-0400)] <EricDalquist> and see if it still breaks :/
[15:02:44 EDT(-0400)] <athena7> yeah i'll try and see
[15:03:00 EDT(-0400)] <athena7> no need to apologize, i don't know how to fix it either (smile)
[15:03:34 EDT(-0400)] <EricDalquist> well I just wish we even knew what the heck was going wrong!@
[15:03:41 EDT(-0400)] <athena7> i know
[15:03:46 EDT(-0400)] <athena7> it's weird
[15:03:50 EDT(-0400)] <athena7> i don't usually have pet tickets
[15:03:55 EDT(-0400)] <athena7> usually i try and do something about them
[15:03:58 EDT(-0400)] <athena7> but this one is beyond me
[15:08:01 EDT(-0400)] <EricDalquist> hrm ...
[15:08:15 EDT(-0400)] <EricDalquist> in the rendering pipeline code there is:
[15:08:16 EDT(-0400)] <EricDalquist> tst.setParameter("uP_productAndVersion", versionString);
[15:08:45 EDT(-0400)] <EricDalquist> perhaps just change uP_productAndVersion to UP_VERSION?
[15:08:53 EDT(-0400)] <EricDalquist> trying it now ...
[15:09:08 EDT(-0400)] <athena7> oh
[15:09:19 EDT(-0400)] <athena7> was this broken because the parameter name got changed in the xsl?
[15:09:39 EDT(-0400)] <EricDalquist> it may be
[15:09:43 EDT(-0400)] <athena7> that may be the case - there were a few variable names gary changed that didn't line up wtih the java code
[15:09:53 EDT(-0400)] <athena7> for one it was too hard to change
[15:10:07 EDT(-0400)] <athena7> so i put in a statement assigning the java code variable to the theme variable
[15:10:16 EDT(-0400)] <EricDalquist> yeah
[15:13:50 EDT(-0400)] <EricDalquist> lol that was it
[15:16:23 EDT(-0400)] <EricDalquist> so
[15:16:30 EDT(-0400)] <EricDalquist> do I just rename the variable?
[15:16:37 EDT(-0400)] <EricDalquist> or do an alias in the .xsl
[15:16:42 EDT(-0400)] <EricDalquist> if I rename it old themes will break
[15:17:19 EDT(-0400)] <athena7> i'd do an alias
[15:17:28 EDT(-0400)] <athena7> or really, it's probably used in one place in the theme
[15:17:53 EDT(-0400)] <athena7> so in theory you could just chagne that back
[15:17:55 EDT(-0400)] <EricDalquist> yeah
[15:35:23 EDT(-0400)] <EricDalquist> ok
[15:35:29 EDT(-0400)] <EricDalquist> version info in the xsl is fixed
[15:36:58 EDT(-0400)] <EricDalquist> so only those two CSS issues left!
[15:37:21 EDT(-0400)] <EricDalquist> I'm going to be on vacation tomorrow and out of town for the weekend but I'll plan on starting the work to cut 3.0.1 on Monday
[15:37:36 EDT(-0400)] <EricDalquist> athena7: could you check with Gary Friday some time for the status on those last two issues?
[16:00:53 EDT(-0400)] <athena7> yep
[16:00:58 EDT(-0400)] <EricDalquist> thanks (smile)
[16:15:54 EDT(-0400)] <EricDalquist> actually for that virtual portals project we may also be seriously looking into WSRP (tongue)
[16:16:57 EDT(-0400)] <athena7> haha
[16:17:19 EDT(-0400)] <EricDalquist> the project initiator is a system wide portal to provide HR info
[16:17:27 EDT(-0400)] <EricDalquist> which will come from peoplesoft
[16:17:31 EDT(-0400)] <EricDalquist> which does actually produce WSRP
[16:17:53 EDT(-0400)] <EricDalquist> so esm ... I may be digging into WSRP4j and using it in uP3 in late summer / fall
[16:41:52 EDT(-0400)] * michelled (n=team@142.150.154.197) has left ##uportal
[17:20:30 EDT(-0400)] * awills (n=awills@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[17:20:42 EDT(-0400)] <awills> ok, here we go
[17:20:57 EDT(-0400)] <awills> (2:15:43 PM) dalquist: if the building of the uportal-impl artifact fails the whole target should fail-fast with the error message, are you not seeing that when it doesn't build?
[17:21:50 EDT(-0400)] <EricDalquist> yeah
[17:21:53 EDT(-0400)] <awills> to me, it appears as though the uportal-impl artifact gets set to a "default" version, or some other version than the one that didn't build
[17:22:22 EDT(-0400)]

<EricDalquist> where does $

Unknown macro: {uportal-impl.artifact}

point to in that case?


[17:22:33 EDT(-0400)] <awills> the reason is this: when it doesn't build, my SmartLdapGroupStore classes are miossing, and I get NoClassDefFoundError
[17:22:50 EDT(-0400)] <awills> haven't done an <echo>, don't know for certain
[17:22:58 EDT(-0400)] <EricDalquist> that would be very helpful
[17:23:13 EDT(-0400)] <EricDalquist> make sure it isn't pulling in a version of the artifact from the jasig repository or some such
[17:23:18 EDT(-0400)] <awills> i'm only going off info I had before you made your reccomendation
[17:23:37 EDT(-0400)] <awills> yeah that's what I thought it might be doing
[17:23:40 EDT(-0400)] <EricDalquist> so 'when it doesn't build' what does that mean?
[17:24:03 EDT(-0400)] <EricDalquist> are the specific symptoms, error message, test failures that you see?
[17:24:29 EDT(-0400)] <awills> i think so... i think i was seeing the FileSystemGroupStore unit test errors at that time
[17:24:50 EDT(-0400)] <awills> so the tests would fail, but the import/export would still (attempt to) run
[17:24:59 EDT(-0400)] <awills> this is during an 'initportal'
[17:25:11 EDT(-0400)] <EricDalquist> hrm
[17:25:16 EDT(-0400)] <EricDalquist> are you on windows?
[17:25:20 EDT(-0400)] <awills> yep
[17:25:46 EDT(-0400)] <EricDalquist> hrm, I know there was a problem with the 3.0 RCs with the mvn call not failing-fast on windows ... I wonder if this is another manifestation of that
[17:25:52 EDT(-0400)] <EricDalquist> what version of Maven are you using?
[17:26:12 EDT(-0400)] <awills> 2.0.9
[17:27:02 EDT(-0400)] <awills> here's something else i discovered, and i want to get a fix in...
[17:27:50 EDT(-0400)] <awills> the entity-type imports run first (for dependencies), but they only update the db – they don't actually create EntityType objects in the JVM
[17:28:01 EDT(-0400)] <EricDalquist> next time you run the build and it happens, could you try mvn 2.0.8 just to see if it happens there too?
[17:28:07 EDT(-0400)] <awills> so if something else, later, tries to reference them, they won't be there
[17:28:23 EDT(-0400)] <EricDalquist> why wouldn't the something else just load them from the DB?
[17:28:23 EDT(-0400)] <awills> i'll make a note of that
[17:29:05 EDT(-0400)] <awills> i don't think EntityType.java is that smart... i think it only loads it's collection once, at startup
[17:29:49 EDT(-0400)] <EricDalquist> ah yeah I see that
[17:29:53 EDT(-0400)] <EricDalquist> where are you running into this problem?
[17:29:58 EDT(-0400)] <awills> it's not a probelm for any technology or subsystem that is configured for uP3 by default...
[17:30:12 EDT(-0400)] <EricDalquist> is it because of a groups unit test?
[17:30:14 EDT(-0400)] <awills> i'm hitting it w/ SmartLdapGroupStore
[17:30:26 EDT(-0400)] <EricDalquist> ah during the load
[17:30:38 EDT(-0400)] <awills> (and yes, i think it's related to the FileSystemGroupStore tests)
[17:30:52 EDT(-0400)] <EricDalquist> yeah I think I may just exclude that test
[17:30:57 EDT(-0400)] <EricDalquist> it doesn't really qualify as a unit test
[17:31:02 EDT(-0400)] <awills> i can add 3 lines of XML to make the EntityTypes exist in the JVM too
[17:31:05 EDT(-0400)] <EricDalquist> since it loads the whole freaking groups system
[17:31:15 EDT(-0400)] <awills> yep, it does
[17:31:16 EDT(-0400)] <EricDalquist> ok
[17:31:21 EDT(-0400)] <EricDalquist> that sounds good
[17:31:26 EDT(-0400)] <EricDalquist> create a jira and post a patch today
[17:31:29 EDT(-0400)] <EricDalquist> I'm on vacation again tomorrow
[17:31:36 EDT(-0400)] <awills> but i also hit the issue on channel import... since that loads the groups system as well
[17:31:49 EDT(-0400)] <EricDalquist> if I don't get to it today can you ask athena7 to take a look and commit it?
[17:31:54 EDT(-0400)] <EricDalquist> or do you have commit access?
[17:31:57 EDT(-0400)] <EricDalquist> cause you could just do it
[17:32:02 EDT(-0400)] <EricDalquist> yeah
[17:32:03 EDT(-0400)] <awills> for uP3? i should
[17:32:17 EDT(-0400)] <athena7> sorry, i haven't been following - what's up?
[17:32:18 EDT(-0400)] <EricDalquist> you need to merge it into the 3.0-patches branch too
[17:32:32 EDT(-0400)] <awills> i'll run it out on the list, and commit quickly when there are no objecttions
[17:32:33 EDT(-0400)] <EricDalquist> n/p athena7, just a last minute fix for 3.0.1 from drew
[17:32:38 EDT(-0400)] <athena7> oh ok
[17:32:43 EDT(-0400)] <EricDalquist> sounds good
[17:32:50 EDT(-0400)] <athena7> whatever it is, i don't have objections
[17:32:51 EDT(-0400)] <athena7> (wink)
[17:32:52 EDT(-0400)] <EricDalquist> if it was there when I get in on Monday morning that would be great
[17:32:56 EDT(-0400)] <awills> i remember the relevent branches eric (wink)
[17:33:06 EDT(-0400)] <awills> should be
[17:33:09 EDT(-0400)] <EricDalquist> I'm planning on starting the release process then
[17:33:32 EDT(-0400)] <EricDalquist> yeah
[17:33:33 EDT(-0400)] <EricDalquist> so
[17:33:42 EDT(-0400)] <EricDalquist> I just looked at the mvn.bat in 2.0.9
[17:33:44 EDT(-0400)] <awills> i believe i should also commit the mods i did for Tuy (i think that's who)
[17:33:52 EDT(-0400)] <EricDalquist> and I think the introduced a bug (tongue)
[17:34:46 EDT(-0400)] <awills> mods to export-preferences.crn
[17:35:05 EDT(-0400)] <EricDalquist> ok
[17:36:14 EDT(-0400)] <awills> here: http://uportal.pastebin.com/m22d363fd
[17:36:43 EDT(-0400)] <awills> this was discussed a week or so ago, seems to work for everyone... and no objections were voiced
[17:37:06 EDT(-0400)] <EricDalquist> that's a lot of + and - ... it looks like it is mostly indentation changes though ...
[17:37:12 EDT(-0400)] <EricDalquist> is this that BigInteger issue?
[17:37:19 EDT(-0400)] <awills> yeah, unfortunately
[17:37:26 EDT(-0400)] <EricDalquist> ah yeah
[17:37:27 EDT(-0400)] <EricDalquist> that's fine
[17:38:11 EDT(-0400)] <EricDalquist> one thing
[17:38:26 EDT(-0400)] <EricDalquist> wasn't it going to use the global APP_CONTEXT variable?
[17:38:41 EDT(-0400)] <EricDalquist> I think the root import, export, and delete scripts all define it
[17:38:41 EDT(-0400)] <awills> nick b. astutely points out that maybe it's not ok if tuy's dbloader settings are creating a column for USER_ID that's natively a BigDecimal, but i'm happy to make this chance since it works for all of the above
[17:39:04 EDT(-0400)] <awills> i'll review the correspondence, thought it did that already
[17:39:11 EDT(-0400)] <awills> i think so too
[17:39:16 EDT(-0400)] <EricDalquist> well USER_ID is just a 'really big number'
[17:39:34 EDT(-0400)] <EricDalquist> and the Spring JDBC code in cernunnos will likely use the smallest type it can for the range of the column
[17:39:40 EDT(-0400)] <awills> yeah, all my DBs have some integer type for it
[17:39:41 EDT(-0400)] <EricDalquist> that column must be bigger than 2^64
[17:39:55 EDT(-0400)] <EricDalquist> or it is 2^64 unsigned
[17:40:04 EDT(-0400)] <EricDalquist> which would require BigDecimal in Java
[17:40:40 EDT(-0400)] <awills> what about a Long? that's not big enough
[17:41:05 EDT(-0400)] <EricDalquist> that's -2^63 to 2^63-1
[17:41:16 EDT(-0400)] <EricDalquist> so if you have a 64bit unsigned database field a long can't hold it
[17:41:41 EDT(-0400)] <EricDalquist> and actually since it did a BigDecimal that implies it is floating point too (tongue)
[17:41:47 EDT(-0400)] <EricDalquist> oh well
[17:41:49 EDT(-0400)] <awills> what about BigInteger?
[17:42:04 EDT(-0400)] <EricDalquist> the joys of databases
[17:42:11 EDT(-0400)] <awills> heh
[17:51:25 EDT(-0400)] * awills (n=awills@wsip-98-174-242-39.ph.ph.cox.net) has left ##uportal
[17:59:02 EDT(-0400)] <EricDalquist> athena7: you still around?
[18:00:55 EDT(-0400)] <athena7> sort of
[18:00:56 EDT(-0400)] <athena7> what's up?
[18:01:04 EDT(-0400)] <EricDalquist> the HttpProxyServlet
[18:01:13 EDT(-0400)] <EricDalquist> so I did verify it is only used by the ProxyWriter
[18:01:17 EDT(-0400)] <EricDalquist> which is disabled by default
[18:01:31 EDT(-0400)] <athena7> yeah
[18:01:38 EDT(-0400)] <athena7> so when enabled, it should be used by everything
[18:01:40 EDT(-0400)] <EricDalquist> so I'm thinking I can comment out the servlet and servlet-mapping in web.xml and add a comment to portal.properties with instructions on how to enable it
[18:01:41 EDT(-0400)] <athena7> including portlets
[18:01:46 EDT(-0400)] <athena7> not specifically the proxy channel
[18:01:49 EDT(-0400)] <EricDalquist> yup
[18:01:56 EDT(-0400)] <athena7> sounds reasonable
[19:48:31 EDT(-0400)] * johnalewis (n=quassel@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[21:27:06 EDT(-0400)] * esm (n=esm@207-53-192-243.dynamic-dsl.qis.net) has joined ##uportal