[00:08:00 EDT(-0400)] * lennar1 (n=sparhk@ip68-98-56-21.ph.ph.cox.net) has joined ##uportal
[01:36:12 EDT(-0400)] * lennar1 (n=sparhk@ip68-98-56-21.ph.ph.cox.net) has left ##uportal
[08:54:45 EDT(-0400)] * jessm (n=Jess@c-76-19-199-61.hsd1.ma.comcast.net) has joined ##uportal
[09:54:39 EDT(-0400)] * michelled (n=team@142.150.154.197) has joined ##uportal
[09:57:13 EDT(-0400)] * athena7 (n=athena7@adsl-99-149-83-32.dsl.wlfrct.sbcglobal.net) has joined ##uportal
[10:07:28 EDT(-0400)] * colinclark (n=colin@bas1-toronto09-1176018330.dsl.bell.ca) has joined ##uportal
[11:42:18 EDT(-0400)] * jessm (n=Jess@c-76-19-199-61.hsd1.ma.comcast.net) has joined ##uportal
[11:44:46 EDT(-0400)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined ##uportal
[11:49:01 EDT(-0400)] <EricDalquist> so I made a fun little discovery last night, hibernate has a very nice object model you can build to describe a database table
[11:49:09 EDT(-0400)] <EricDalquist> which has a method where you just pass it a dialect
[11:49:17 EDT(-0400)] <EricDalquist> and it generates the SQL to create/alter/drop it
[11:49:31 EDT(-0400)] * colinclark (n=colin@142.150.154.101) has joined ##uportal
[11:50:02 EDT(-0400)] <EricDalquist> and put together a little bit of code that can parse a uPortal tables.xml into the OM
[11:50:18 EDT(-0400)] * MarkRogers (n=MarkRoge@addhcp129.cc.umanitoba.ca) has joined ##uportal
[11:50:28 EDT(-0400)] <EricDalquist> so I think we're going to be able to do away with DbLoader and the manual database mapping and just use the hibernate dialect
[11:53:35 EDT(-0400)] <MarkRogers> hibernate is good
[11:58:46 EDT(-0400)] <EricDalquist> not having custom code to create tables would be great
[12:32:43 EDT(-0400)] <athena7> YAY
[12:32:45 EDT(-0400)] <athena7> that'd be awesome
[12:32:47 EDT(-0400)] <EricDalquist> yeah
[12:32:53 EDT(-0400)] * athena7 wishes everyone a happy talk like a pirate day
[12:32:57 EDT(-0400)] <EricDalquist> it is remakably simple
[12:32:59 EDT(-0400)] <EricDalquist> arrr!
[12:33:23 EDT(-0400)] <EricDalquist> I just added support for unique columns and index specification in tables.xml quite easily too
[12:33:26 EDT(-0400)] <EricDalquist> :-D
[12:34:49 EDT(-0400)] <athena7> that's awesome!
[12:35:40 EDT(-0400)] <athena7> happen to remember how user profiles and user agent mappings get added to the db?
[12:36:00 EDT(-0400)] <EricDalquist> nope
[12:36:11 EDT(-0400)] <EricDalquist> well profiles get added when you log in
[12:36:17 EDT(-0400)] <EricDalquist> and then updated when a layout is created for you
[12:36:27 EDT(-0400)] <EricDalquist> user-agent ... I can't remember
[12:36:59 EDT(-0400)] <athena7> the default user profiles though
[12:37:27 EDT(-0400)] <athena7> hm.
[13:15:09 EDT(-0400)] <athena7> what exactly does the -Durl parameter in the maven deploy:deploy-file commnad signify?
[13:15:14 EDT(-0400)] <athena7> i'm not sure what url that's supposed to be
[13:15:25 EDT(-0400)] <EricDalquist> the URL to the repository you're deploying the artifact to
[13:15:32 EDT(-0400)] <athena7> is it just the base url though?
[13:15:56 EDT(-0400)] <EricDalquist> no, the full URL
[13:16:01 EDT(-0400)] <athena7> hm, ok
[13:16:14 EDT(-0400)] <EricDalquist> so if you were deploying to the jasig repo it would look like
[13:16:34 EDT(-0400)] <EricDalquist> -Durl=scp://developer.ja-sig.org/filesystem/path/to/maven/repository
[13:16:47 EDT(-0400)] <athena7> ah, interesting
[13:16:57 EDT(-0400)] <EricDalquist> since the jasig repo just uses a flat file system
[13:16:59 EDT(-0400)] <EricDalquist> and ssh
[13:17:03 EDT(-0400)] <athena7> yeah
[13:17:07 EDT(-0400)] <athena7> i'm going against archiva
[13:17:10 EDT(-0400)] <EricDalquist> when we deploy to Nexus here it is a http:// url
[13:17:14 EDT(-0400)] <EricDalquist> since it uses WebDav
[13:17:27 EDT(-0400)] <athena7> i sort of had assumed if i just gave it the repository id it could intuit the right url to webdav to or whatever
[13:17:29 EDT(-0400)] <EricDalquist> so the URL really is just the http URL to the repository
[13:17:41 EDT(-0400)] <EricDalquist> it might if you have the repo in your settings.xml
[13:17:46 EDT(-0400)] <EricDalquist> under the same id
[13:17:50 EDT(-0400)] <EricDalquist> but I don't know
[13:19:14 EDT(-0400)] <athena7> ah
[13:19:49 EDT(-0400)] <athena7> oh ok, i see
[13:43:51 EDT(-0400)] * awills (n=awills@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[13:45:16 EDT(-0400)] <awills> I tested and committed the changes to WPP... how are we looking for a release?
[13:45:44 EDT(-0400)] <EricDalquist> of the portlet?
[13:46:01 EDT(-0400)] <awills> yeah, that's what I'm hoping for
[13:46:28 EDT(-0400)] <awills> i'd like to resolve UP-2134: http://www.ja-sig.org/issues/browse/UP-2134
[13:46:40 EDT(-0400)] <EricDalquist> ah
[13:46:41 EDT(-0400)] <EricDalquist> well
[13:46:46 EDT(-0400)] <awills> I have the overlay and 2 CPD files all ready
[13:46:59 EDT(-0400)] <EricDalquist> if you want to take it up you can cut a release of it
[13:47:05 EDT(-0400)] <EricDalquist> tag it and everything
[13:47:51 EDT(-0400)] <awills> I suspect I can do release:prepare (tagging), but I don't know if i can do release:perform (posting to repo)
[13:48:02 EDT(-0400)] <EricDalquist> ah
[13:48:28 EDT(-0400)] <EricDalquist> I can see about doing that on Monday
[13:49:37 EDT(-0400)] <awills> what if I write a PROPOSAL email to portlet-dev on the subject of releasing WPP 1.0.0 now that we've checked in the work from JHU and UW – good way to proceed?
[13:49:53 EDT(-0400)] <awills> make sure everyone's ready?
[13:50:05 EDT(-0400)] <EricDalquist> sounds good
[13:50:14 EDT(-0400)] <awills> alright I'll do that
[14:54:11 EDT(-0400)] <athena7> EricDalquist: do you think there's likely a rational way to hook into the uportal build, maybe say the deploy ear macro, and do some filtering for environment-specific values?
[15:04:53 EDT(-0400)] * apetro-_ (n=apetro@ip24-56-20-55.ph.ph.cox.net) has joined ##uportal
[15:08:53 EDT(-0400)] * colinclark (n=colin@142.150.154.101) has joined ##uportal
[15:47:34 EDT(-0400)] <EricDalquist> athena7: what are you trying to filter?
[15:47:40 EDT(-0400)] <EricDalquist> maven has filtering capabilities
[15:47:44 EDT(-0400)] <athena7> yes
[15:47:47 EDT(-0400)] <EricDalquist> we even use them for the CAS repackaging
[15:48:01 EDT(-0400)] <athena7> we'd been talking about ways to set stuff up for different environments
[15:48:15 EDT(-0400)] <athena7> so server names, database info, etc.
[15:53:32 EDT(-0400)] * MarkRogers (n=MarkRoge@addhcp129.cc.umanitoba.ca) has joined ##uportal
[15:56:12 EDT(-0400)] <EricDalquist> yeah that should all be doable
[15:56:23 EDT(-0400)] <EricDalquist> you'd probably need to do it on the war/jar level though
[15:56:46 EDT(-0400)] <athena7> what do you mean?
[15:57:13 EDT(-0400)] <EricDalquist> at the EAR level maven never unpacks the WARs
[15:57:13 EDT(-0400)] <awills> i think he means both levels
[15:57:17 EDT(-0400)] <athena7> ah.
[15:57:21 EDT(-0400)] <EricDalquist> so you don't really have the ability to filter things
[15:57:31 EDT(-0400)] <athena7> i thought it did unpack the portlet overlays?
[15:57:34 EDT(-0400)] <EricDalquist> you have to add the fitlers in maven to the artifact that directly contains the files you want filtered
[15:57:45 EDT(-0400)] <EricDalquist> the individual overlay projects do
[15:57:52 EDT(-0400)] <athena7> ah ok
[15:58:05 EDT(-0400)] <athena7> because i can imagine we'd probably need to do filtering on those too
[15:58:14 EDT(-0400)] <EricDalquist> yaeh
[15:58:15 EDT(-0400)] <athena7> since a lot of those have database configurations
[15:58:16 EDT(-0400)] <EricDalquist> yeah
[15:58:21 EDT(-0400)] <EricDalquist> so not one central place to do that
[15:58:34 EDT(-0400)] <EricDalquist> our eventual goal here is to do all of our configuration via overlays
[15:58:52 EDT(-0400)] <EricDalquist> by having a skeleton Maven project (just war overlays with config files and an ear project)
[15:59:00 EDT(-0400)] <EricDalquist> all our portal code would be in a local maven repository
[15:59:17 EDT(-0400)] <EricDalquist> and then to build an environment you just package the skeleton EAR
[15:59:29 EDT(-0400)] <athena7> i worry some about taking that approach, since when configuration files changed we'd potentially need to edit several similar files to make sure the change actually wound up int he code
[15:59:33 EDT(-0400)] <EricDalquist> which would pull in all of your code, do the overlays/filtering, repackage and package into a central EAR
[16:00:31 EDT(-0400)] <EricDalquist> yeah, that of course requires a good change management approach
[16:00:56 EDT(-0400)] <EricDalquist> the hope would be you'd have all your environment config factored out into specific .properties files
[16:01:04 EDT(-0400)] <EricDalquist> so you're not mucking with spring configs or such
[16:02:16 EDT(-0400)] <athena7> yeah i guess that might be possible for some stuff
[16:02:32 EDT(-0400)] <EricDalquist> do you have specific concerns?
[16:02:37 EDT(-0400)] <EricDalquist> just curious
[16:02:38 EDT(-0400)] <athena7> roles in portlet.xml files
[16:02:41 EDT(-0400)] <athena7> for one thing
[16:02:44 EDT(-0400)] <EricDalquist> you may have thought more than me about it
[16:02:53 EDT(-0400)] <EricDalquist> so I really think you can refer to the proper group name in uP3
[16:02:57 EDT(-0400)] <athena7> although at least those files are less likely to change frequently
[16:03:05 EDT(-0400)] <athena7> i'll try it out
[16:03:15 EDT(-0400)] <EricDalquist> if not
[16:03:18 EDT(-0400)] <EricDalquist> we should add that
[16:03:28 EDT(-0400)] <EricDalquist> it would be as simple change in PortalHttpServletRequest to support it
[16:03:29 EDT(-0400)] <athena7> it's still not really ideal though
[16:03:34 EDT(-0400)] <EricDalquist> and would make uPortal much more sane for that
[16:03:36 EDT(-0400)] <EricDalquist> true
[16:03:56 EDT(-0400)] <athena7> because then we'd need to provide some configuration mapping in all the portlets because groups wouldn't have a predictable name
[16:04:15 EDT(-0400)] <athena7> so we'd need to factor out all the role names into a configuration file or something
[16:04:55 EDT(-0400)] <EricDalquist>
[16:05:00 EDT(-0400)] <EricDalquist> portlet roles are a pain
[16:05:03 EDT(-0400)] <athena7> yes, they are
[16:05:17 EDT(-0400)] <EricDalquist> at that point my best approach would be what you just said I think
[16:05:23 EDT(-0400)] <athena7> if we actually implemented the group fname feature we talked about i think the problem would be solved
[16:05:32 EDT(-0400)] <EricDalquist> in your portlet.xml files have stuff like @portal.admin@
[16:05:40 EDT(-0400)] <athena7> but barring that, we could pull out all the names into properties files
[16:05:44 EDT(-0400)] <EricDalquist> then have a .properties file per environment with those mappings
[16:06:08 EDT(-0400)] <athena7> will portlet.xml pick up values from a properties file?
[16:06:30 EDT(-0400)] <EricDalquist> no
[16:06:37 EDT(-0400)] <athena7> ok, i didn't think so
[16:06:41 EDT(-0400)] <EricDalquist> you'd have to have the overlay do that filtering
[16:06:44 EDT(-0400)] <athena7> yeah
[16:10:18 EDT(-0400)] <athena7> back soon
[16:17:33 EDT(-0400)] * awills (n=awills@wsip-98-174-242-39.ph.ph.cox.net) has left ##uportal
Page Comparison
General
Content
Integrations