Page Comparison

[00:28:48 EDT(-0400)] * lennar1 (n=sparhk@ip68-98-56-21.ph.ph.cox.net) has joined ##uportal
[00:57:29 EDT(-0400)] * lennar1 (n=sparhk@ip68-98-56-21.ph.ph.cox.net) has left ##uportal
[08:41:52 EDT(-0400)] * mad (n=chatzill@pcit-8752.hig.se) has joined ##uportal
[08:42:59 EDT(-0400)] * anastasiac (n=team@142.150.154.160) has joined ##uportal
[09:02:44 EDT(-0400)] * jessm (n=Jess@c-76-19-199-61.hsd1.ma.comcast.net) has joined ##uportal
[09:19:15 EDT(-0400)] * colinclark (n=colin@209-254-26-66.ip.mcleodusa.net) has joined ##uportal
[09:19:57 EDT(-0400)] * athena7 (n=athena7@99.149.83.32) has joined ##uportal
[09:54:23 EDT(-0400)] * michelled (n=team@142.150.154.197) has joined ##uportal
[10:17:18 EDT(-0400)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined ##uportal
[10:19:33 EDT(-0400)] <EricDalquist> athena7: you were looking into multiple-profile support right?
[10:20:02 EDT(-0400)] <athena7> not really - more just lamenting that it wasn't supported in the import/export
[10:20:07 EDT(-0400)] <EricDalquist> ah
[10:20:23 EDT(-0400)] <EricDalquist> so in going through the import scripts this mornign trying to find all the int issues
[10:20:26 EDT(-0400)] <EricDalquist> I was looking at that
[10:20:28 EDT(-0400)] <athena7> ah
[10:20:30 EDT(-0400)] <athena7> cool
[10:20:41 EDT(-0400)] <EricDalquist> I'm not sure I even understand looking at the data model the interrelation of everything
[10:20:50 EDT(-0400)] <EricDalquist> from what I understand:
[10:20:56 EDT(-0400)] <EricDalquist> a user can have N profiles
[10:21:10 EDT(-0400)] <EricDalquist> a profile points to a structure xsl, theme xsl and layout
[10:21:22 EDT(-0400)] <EricDalquist> multiple profiles can point to the same layout
[10:21:38 EDT(-0400)] <EricDalquist> layouts have structure attributes which are scoped to a particular profile
[10:22:03 EDT(-0400)] <EricDalquist> does that all sound right?
[10:25:30 EDT(-0400)] <athena7> yes, i think so
[10:25:34 EDT(-0400)] <athena7> i'm not sure i understand it all either
[10:25:46 EDT(-0400)] <EricDalquist> so if that is the case there is a fair bit of refactoring needed in the scripts
[10:25:55 EDT(-0400)] <EricDalquist> profiles would need to be broken out into their own files
[10:26:00 EDT(-0400)] <athena7>
[10:26:09 EDT(-0400)] <EricDalquist> and we would need to come up with unique identifiers for both profiles and layouts
[10:26:16 EDT(-0400)] <athena7> be nice if we could just nest them sigh
[10:26:20 EDT(-0400)] <EricDalquist> we could
[10:26:29 EDT(-0400)] <EricDalquist> but that will add a restriction
[10:26:35 EDT(-0400)] <EricDalquist> well
[10:26:36 EDT(-0400)] <EricDalquist> hrm
[10:27:01 EDT(-0400)] <EricDalquist> I suppose a .layout could just include all of the profiles that point to it
[10:27:35 EDT(-0400)] <EricDalquist> then we just need a way, in that .layout to reference a specific <profile> entry from a <structure-attribute>
[10:27:41 EDT(-0400)] <athena7> i guess we really need to also add the user agent mappings, which point to the profiles
[10:27:43 EDT(-0400)] <EricDalquist> that's still a bit of work
[10:27:48 EDT(-0400)] <EricDalquist> yeah
[10:27:54 EDT(-0400)] <EricDalquist> those are completely ignored right now
[10:28:10 EDT(-0400)] <EricDalquist> it sucks not having FKs in the non-hibernate managed tables right now
[10:28:17 EDT(-0400)] <EricDalquist> makes it really hard to decypher the tables
[10:28:36 EDT(-0400)] <EricDalquist> we could add them with the new dbloader but I'd be supprised if the code does inserts/updates/deletes in the right order to deal with them
[10:28:50 EDT(-0400)] <EricDalquist> oh and I just committed a batch of fixes for the import scripts in trunk
[10:28:59 EDT(-0400)] <EricDalquist> I think I found all of those numeric sql type issues
[10:29:45 EDT(-0400)] <athena7> oh yay
[10:30:27 EDT(-0400)] <EricDalquist> the scripts still need work
[10:30:47 EDT(-0400)] <EricDalquist> like importing 2.6 data may not work in trunk right now
[10:30:58 EDT(-0400)] <EricDalquist> I'm planning on working on that tomorrow afternoon
[10:31:01 EDT(-0400)] <EricDalquist> as well as the export scripts
[10:32:07 EDT(-0400)] <athena7> ok
[10:32:23 EDT(-0400)] <EricDalquist> I'll keep the profile stuff in mind too
[10:34:17 EDT(-0400)] <athena7> that'd be great
[10:34:58 EDT(-0400)] <EricDalquist> ok and it looks like the UP_USER_UA_MAP does the user agent to profile mapping?
[10:36:39 EDT(-0400)] <EricDalquist> so that .... hrm
[10:36:47 EDT(-0400)] <athena7> yes
[10:36:49 EDT(-0400)] <EricDalquist> that could get dumped as part of the <profile> entries in a layout
[10:37:01 EDT(-0400)] <EricDalquist> I just want to try and group this data into useful chunks
[10:37:11 EDT(-0400)] <EricDalquist> like that is why we moved profile into layout for 3.0
[10:37:21 EDT(-0400)] <EricDalquist> since you can have a user without a profile or layout
[10:37:30 EDT(-0400)] <EricDalquist> but you need a layout for a profile
[10:37:40 EDT(-0400)] <EricDalquist> and the UA mapping would be the same
[10:37:45 EDT(-0400)] <EricDalquist> useless without the profile
[10:37:54 EDT(-0400)] <athena7> right
[10:38:04 EDT(-0400)] <athena7> and we don't really want it to be torturous for people to create new users
[10:38:13 EDT(-0400)] <EricDalquist> yeah
[10:38:29 EDT(-0400)] <EricDalquist> ok, I think that all makes sense
[10:40:01 EDT(-0400)] <athena7> by the way eric, have you gotten the os x java update?
[10:40:09 EDT(-0400)] <athena7> it doesn't seem to be friendly with cernunnos
[10:40:26 EDT(-0400)] <EricDalquist> the JDK6 update?
[10:40:37 EDT(-0400)] <athena7> no, that was last spring right?
[10:40:45 EDT(-0400)] <athena7> actually i wonder whether it undid whatever we did to fix that
[10:40:47 EDT(-0400)] <EricDalquist> yeah
[10:40:48 EDT(-0400)] <EricDalquist> http://erics-notes.blogspot.com/2008/06/apples-jdk6-broke-my-scripting.html
[10:40:50 EDT(-0400)] <EricDalquist> probably
[10:40:57 EDT(-0400)] <athena7> yeah, probably the sam ething
[10:41:06 EDT(-0400)] <athena7> noticed the fragment portlet died
[10:41:07 EDT(-0400)] <EricDalquist> probably stuck that jdk6 compiled jar back in a global folder