uPortal IRC Logs-2008-08-05

Owned by Apereo Infrastructure
Last updated: Aug 05, 2008Version comment
10 min read

[09:55:17 EDT(-0400)] * michelled (n=team@142.150.154.197) has joined ##uportal
[09:57:00 EDT(-0400)] * anastasiac (n=team@142.150.154.160) has joined ##uportal
[10:04:02 EDT(-0400)] * athena7 (n=athena7@99.145.99.32) has joined ##uportal
[10:23:46 EDT(-0400)] * colinclark (n=colin@142.150.154.101) has joined ##uportal
[10:48:38 EDT(-0400)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined ##uportal
[11:40:27 EDT(-0400)] * holdorph (n=holdorph@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[12:01:57 EDT(-0400)] * michelled (n=team@142.150.154.197) has joined ##uportal
[12:01:57 EDT(-0400)] * anastasiac (n=team@142.150.154.160) has joined ##uportal
[12:02:16 EDT(-0400)] * holdorph (n=holdorph@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[12:02:19 EDT(-0400)] * bulloche (n=bulloche@134.250.4.77) has joined ##uportal
[12:02:35 EDT(-0400)] * ChanServ (ChanServ@services.) has joined ##uportal
[12:02:35 EDT(-0400)] * jessm (n=Jess@c-76-19-199-61.hsd1.ma.comcast.net) has joined ##uportal
[12:02:35 EDT(-0400)] * Tuomaz (n=fredrik@kaffe.umdc.umu.se) has joined ##uportal
[12:02:35 EDT(-0400)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined ##uportal
[12:02:41 EDT(-0400)] * colinclark (n=colin@142.150.154.101) has joined ##uportal
[12:02:41 EDT(-0400)] * higpetter (n=petter@brunte.HIG.SE) has joined ##uportal
[16:08:10 EDT(-0400)] * dstn (n=dstn@unaffiliated/dstn) has joined ##uportal
[16:08:56 EDT(-0400)] <athena7> what happened to spring-mock in spring 2.5?
[16:09:00 EDT(-0400)] <athena7> did it go away?
[16:09:05 EDT(-0400)] <dstn> In portlets when you do <role-link>local.50</role-link> Where is the local coming from?
[16:09:08 EDT(-0400)] <athena7> and spring-portlet
[16:09:15 EDT(-0400)] <athena7> that's the name of a uportal role
[16:09:23 EDT(-0400)] <athena7> so local vs. pags
[16:09:28 EDT(-0400)] <dstn> so the portal puts that into the session or something?
[16:09:30 EDT(-0400)] <athena7> depending on where you want the roles to come from
[16:09:49 EDT(-0400)] <athena7> the portlet connector thing puts those into the portlet request, but only when asked to do so by the portlet.xml
[16:09:59 EDT(-0400)] <athena7> it actually checks to see which properties you're asking for
[16:10:43 EDT(-0400)] <dstn> ok
[16:11:15 EDT(-0400)] <athena7> if you need to find the id for a group, you can look at the groups table
[16:11:20 EDT(-0400)] <holdorph> portlet and web mvc got reorganzied to different modules in spring 2.5
[16:11:23 EDT(-0400)] <dstn> Ya that's where I was just at
[16:11:32 EDT(-0400)] <athena7> mrgh.
[16:11:43 EDT(-0400)] <dstn> was trying to determine if the local.# was uportal specific or something in the portlet spec
[16:11:56 EDT(-0400)] <athena7> spring-webmvc-portlet is what i want i'm guess - thanks cris
[16:12:10 EDT(-0400)] <athena7> the format of it is uportal-specific
[16:12:20 EDT(-0400)] <dstn> yes, 2.5 is spring-webmvc-portlet
[16:12:22 EDT(-0400)] <athena7> there's been talk about giving the roles unique names in uportal
[16:12:25 EDT(-0400)] <holdorph> <dependency>
[16:12:25 EDT(-0400)] <holdorph> <groupId>org.springframework</groupId>
[16:12:25 EDT(-0400)] <holdorph> <artifactId>spring-webmvc-portlet</artifactId>
[16:12:25 EDT(-0400)] <holdorph> <version>2.5.4</version>
[16:12:25 EDT(-0400)] <holdorph> <scope>compile</scope>
[16:12:26 EDT(-0400)] <holdorph> </dependency>
[16:12:36 EDT(-0400)] <athena7> since they'll annoyingly change between uportal deployments now
[16:12:41 EDT(-0400)] <holdorph> is what I used last time
[16:12:56 EDT(-0400)] <athena7> thanks!
[16:12:58 EDT(-0400)] <dstn> What do you mean change?
[16:13:08 EDT(-0400)] <athena7> so in uportal 2.6 and below
[16:13:17 EDT(-0400)] <athena7> the group ids were hard-coded
[16:13:38 EDT(-0400)] <athena7> now the init scripts just import the relevant data and it gets assigned whatever the next id in the sequence is
[16:13:53 EDT(-0400)] <athena7> so you can't count on the student group always having the ID 14, or whatever it used to be
[16:14:06 EDT(-0400)] <athena7> also, if you do an export and import, all the group, channel, etc. ids will chagne
[16:14:21 EDT(-0400)] <dstn> i c
[16:14:30 EDT(-0400)] <athena7> so if you deploy your portlet in development and production, the group might be "local.50" in one, and "local.30" in the other
[16:14:32 EDT(-0400)] <dstn> so from uportal version to version, assuming you migrated data
[16:14:35 EDT(-0400)] <athena7> yes
[16:14:37 EDT(-0400)] <athena7> that too
[16:14:59 EDT(-0400)] <dstn> or ya, exporting from production and into another database
[16:15:02 EDT(-0400)] <dstn> ok, make sense
[16:15:06 EDT(-0400)] <athena7> yep
[16:15:07 EDT(-0400)] <dstn> makes*
[16:15:09 EDT(-0400)] <athena7> this is a pain
[16:15:16 EDT(-0400)] <athena7> so it'd be cool if you could do local.admin
[16:15:18 EDT(-0400)] <athena7> or whatever
[16:15:29 EDT(-0400)] <athena7> sort of like having an fname for groups
[16:16:35 EDT(-0400)] <athena7> this is also an issue for things like CAnnouncements that index data by channel subscribe id
[16:16:40 EDT(-0400)] <athena7> or the yale version of the calendar portlet
[16:17:38 EDT(-0400)] <dstn> i c
[16:17:49 EDT(-0400)] <dstn> holdorph: have you used the spring 2.5 annotations yet?
[16:18:15 EDT(-0400)] <dstn> portlet-webmvc annotations specifically
[16:18:16 EDT(-0400)] <holdorph> john lewis and I did a pre-conference seminar in st. paul. and one hour was on them
[16:18:28 EDT(-0400)] <holdorph> more him, then me, but I was there (wink)
[16:19:09 EDT(-0400)] <holdorph> they're pretty trippy if you ask me
[16:19:22 EDT(-0400)] <dstn> agreed, some of them didn't make sense
[16:19:40 EDT(-0400)] <dstn> like you return a String but the model is set in the method and still found
[16:20:27 EDT(-0400)] <holdorph> hmm... not sure I either follow that or believe it.
[16:23:21 EDT(-0400)] <dstn> like this: http://uportal.pastebin.com/d39adebe7
[16:24:37 EDT(-0400)] <holdorph> the model gets passed, in. so, i don't know what you mean by being surprised it's found.
[16:24:47 EDT(-0400)] <holdorph> the spring code created the model and let your method simply use it.
[16:25:05 EDT(-0400)] <dstn> but you added something too it and it wasn't returned
[16:25:18 EDT(-0400)] <holdorph> but you mutated the object
[16:25:34 EDT(-0400)] <holdorph> objects are pass by reference not by value
[16:26:33 EDT(-0400)] <dstn> I was under the impression it was the other way around
[16:26:42 EDT(-0400)] <dstn> no such thing as pass by reference in java
[16:26:53 EDT(-0400)] <holdorph> primitives are pass by value, objects are pass by reference
[16:27:01 EDT(-0400)] <holdorph> there is no way to do anything besides that in java
[16:28:34 EDT(-0400)] * apetro-_ (n=apetro@CPE-65-30-162-97.wi.res.rr.com) has joined ##uportal
[16:29:21 EDT(-0400)] <dstn> lol: http://javadude.com/articles/passbyvalue.htm
[16:30:49 EDT(-0400)] <holdorph> wow, nothing like going over the top and becoming the poster child for 'pedantic'
[16:32:36 EDT(-0400)] <dstn> lol
[16:32:40 EDT(-0400)] <athena7> lol
[16:33:04 EDT(-0400)] <EricDalquist> athena7: http://erics-notes.blogspot.com/2007/12/spring-25-module-contents.html
[16:33:12 EDT(-0400)] <athena7> thanks (smile)
[16:33:13 EDT(-0400)] <holdorph> dstn: http://uportal.pastebin.com/m3080f248
[16:33:17 EDT(-0400)] <EricDalquist> spring mock is now in spring-test
[16:33:21 EDT(-0400)] <holdorph> if you want to convince yourself any more.
[16:33:27 EDT(-0400)] * dstn needs to refer to dictionary.com for holdorph's vocab
[16:33:56 EDT(-0400)] <athena7> by the way eric, i think it's possible by JPA problem from before was just something weird that happened to the db
[16:34:26 EDT(-0400)] <EricDalquist> ah ok
[16:34:31 EDT(-0400)] <EricDalquist> I was meaning to ask you about that
[16:35:46 EDT(-0400)] <dstn> holdorph: indeed it does work, suppose I'm not used to doing things like that...used to returning it
[16:36:52 EDT(-0400)] <holdorph> well immutable objects like String, and as a general design pattern, also make it pretty rare
[16:37:27 EDT(-0400)] <dstn> i c i c
[16:43:37 EDT(-0400)] * michelled (n=team@142.150.154.197) has left ##uportal
[16:53:11 EDT(-0400)] <athena7> EricDalquist: i'll let you know if it comes back, but i haven't seen it since i did an initportal
[16:53:19 EDT(-0400)] <EricDalquist> sounds good
[16:53:31 EDT(-0400)] <EricDalquist> fyi, I'm cutting a 3.1-M1 tag today
[16:53:43 EDT(-0400)] <apetro-_> congratulations on the M1
[16:53:45 EDT(-0400)] <EricDalquist> figured 70 some issues would be good to have something people can more easily test
[16:53:50 EDT(-0400)] <athena7> woo!
[16:55:51 EDT(-0400)] <dstn> sweetness
[17:00:52 EDT(-0400)] <EricDalquist> once the pluto 1.1.6 vote passes I can do a 3.0.2 release too
[17:04:38 EDT(-0400)] <holdorph> whats pluto 1.1.6 mean to uportal (i could guess it must be some bug fix that uportal needs, but dont remember you saying something)
[17:04:55 EDT(-0400)] <EricDalquist> a performance related bug fix
[17:05:16 EDT(-0400)] <EricDalquist> 1.1.5 internally calls store on the portlet preferences each time they are created
[17:06:06 EDT(-0400)] <EricDalquist> http://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=10560&amp;styleName=Html&amp;version=12313075
[17:06:14 EDT(-0400)] <EricDalquist> that's the full summary of 1.1.6 changes
[17:07:32 EDT(-0400)] <holdorph> nod cool, the more perf improvements in 3.0.2 the better.
[17:08:10 EDT(-0400)] <EricDalquist> yeah figure since all the stuff I found with Lennard will make it in we should try to get that in there too
[17:10:07 EDT(-0400)] <EricDalquist> still need to get the portlet css fixed :/
[17:13:00 EDT(-0400)] <athena7> which portlet css?
[17:13:08 EDT(-0400)] <athena7> is this the select menu background thing?
[17:13:26 EDT(-0400)] <EricDalquist> the jsr168 spec css added for 3.0.1
[17:13:30 EDT(-0400)] <EricDalquist> makes form fields black
[17:13:33 EDT(-0400)] <athena7> yeah
[17:13:36 EDT(-0400)] <athena7> that should be easy to fix
[17:13:38 EDT(-0400)] <athena7> one-liner
[17:13:56 EDT(-0400)] <EricDalquist> would you mind?
[17:13:59 EDT(-0400)] <EricDalquist> CSS scares me (tongue)
[17:14:12 EDT(-0400)] <EricDalquist> http://www.ja-sig.org/issues/browse/UP-2113
[17:15:17 EDT(-0400)] <athena7> sure
[17:15:20 EDT(-0400)] <athena7> can you assign it to me?
[17:15:24 EDT(-0400)] <EricDalquist> yup
[17:15:27 EDT(-0400)] <athena7> thanks!
[17:15:45 EDT(-0400)] <athena7> i think i finally have the duke mail portlet more or less correctly persisting mail accounts to the database
[17:15:52 EDT(-0400)] <EricDalquist> nice
[17:15:56 EDT(-0400)] <athena7> i ran into some issues
[17:16:08 EDT(-0400)] <athena7> one of which eventually turned out to have nothing to do with my changes
[17:16:26 EDT(-0400)] <athena7> i hope it's ok to check in - i never heard back from gary
[17:18:07 EDT(-0400)] <dstn> "CSS scares me (tongue)" <--- me too
[17:18:28 EDT(-0400)] <athena7> maybe i'll try emailing him again
[17:18:29 EDT(-0400)] <holdorph> CSS and I mutually ignore each other.
[17:18:30 EDT(-0400)] <athena7> aw
[17:18:34 EDT(-0400)] <dstn> lol
[17:18:36 EDT(-0400)] <athena7> CSS is fun and tasty
[17:18:53 EDT(-0400)] <athena7> i mean it can be irritating when you get into cross-browser quirks
[17:18:59 EDT(-0400)] <athena7> but CSS is much better than no CSS
[17:19:18 EDT(-0400)] <athena7> by the way dstn, would you mind if i at some point updated the weather portlet to use jsr168 styles?
[17:19:59 EDT(-0400)] <dstn> hmm, so what would that change?
[17:20:11 EDT(-0400)] <dstn> I've honestly never dealt with the styles before
[17:20:57 EDT(-0400)] <athena7> the jsr 168 css classes aren't great
[17:21:07 EDT(-0400)] <athena7> but if you have them, it'll automatically pick up the styles from the portlet
[17:21:18 EDT(-0400)] <athena7> which means you don't have to skin the portlet for your portal
[17:21:31 EDT(-0400)] <athena7> and if users change skins, the portlet will change appearances with it
[17:21:44 EDT(-0400)] <athena7> if you don't, the portlet will have the same colors regardless of the skin the user picks, which may not be good
[17:22:02 EDT(-0400)] <athena7> you can always have additional weather portlet-specific styles on top of the jsr168 ones
[17:22:17 EDT(-0400)] <holdorph> and would work that way cross-portal software platforms as well
[17:22:28 EDT(-0400)] <athena7> but it's good to have the portlet spec ones there too - right now it doesn't pick up the default button styles, link styles, etc.
[17:22:31 EDT(-0400)] <athena7> yes
[17:22:40 EDT(-0400)] <dstn> hmm, I suppose as long as it doesn't mess up the look of the current
[17:23:25 EDT(-0400)] <athena7> no, it should just make it match portals that don't look like whatever you develop on better
[17:25:44 EDT(-0400)] <dstn> k, sure
[17:28:44 EDT(-0400)] <dstn> being pedantic and all, I noticed you committed <link > to the CSS in the view & edit jsp's...I'd prefer they not be in there and the CSS comes from the portal
[17:29:11 EDT(-0400)] <dstn> i.e. <link rel="stylesheet" href="<c:url value="/css/weather.css"/>" type="text/css" />
[17:29:29 EDT(-0400)] <athena7> oh
[17:29:47 EDT(-0400)] <athena7> sure, i hadn't realized that was the intention
[17:29:59 EDT(-0400)] <athena7> i thought it was just missing, sorry (smile)
[17:30:09 EDT(-0400)] <athena7> i have to run, but can we pick this up tomorrow?
[17:30:17 EDT(-0400)] <dstn> sure
[17:30:20 EDT(-0400)] <dstn> see ya tomorrow
[17:30:32 EDT(-0400)] <EricDalquist> dstn: problem is that is there is no easy way to ensure the CSS does come from the portal
[17:30:57 EDT(-0400)] <EricDalquist> not until we support 286
[17:31:28 EDT(-0400)] <dstn> EricDalquist: I've thought this over and thought that was the best solution vs having <link>'s in the middle of the HTML when they are only supposed to be in the <head>
[17:31:55 EDT(-0400)] <EricDalquist> yeah, it is from a markup point of view
[17:32:00 EDT(-0400)] <EricDalquist> but it makes deploying the thing a pita
[17:32:15 EDT(-0400)] <EricDalquist> since you now have to do non-standard hacking of your portal to make it work
[17:32:51 EDT(-0400)] <dstn> EricDalquist: true, I'm still not convinced its a good solution although I prefer compliant html more so
[17:33:24 EDT(-0400)] <dstn> Its the one configuration item you need to do. Its well documented on the wiki for uPortal...I don't know other frameworks though
[17:33:31 EDT(-0400)] <holdorph> it could be made optional on a per deployment basis
[17:33:48 EDT(-0400)] <EricDalquist> perhaps we control it via a preference
[17:33:51 EDT(-0400)] <holdorph> pretty easy with a portlet.xml init param and jsp if tag magic
[17:33:55 EDT(-0400)] <EricDalquist> or init param
[17:34:06 EDT(-0400)] <EricDalquist> default it to being there so it is easier
[17:34:27 EDT(-0400)] <EricDalquist> but just an init param flag to disable the inline link
[17:36:06 EDT(-0400)] * dstn has never used a portlet inital param in jsp
[17:36:33 EDT(-0400)]

<dstn> is it just <c:if test=$

Unknown macro: {paramNam}

> in the jsp?


[17:36:41 EDT(-0400)] <holdorph> that's the idea
[17:36:54 EDT(-0400)] <dstn> and <name>paraName</name> in <init-param>
[17:37:12 EDT(-0400)] <holdorph> can't remember if the init params are put into the same flat namespace or not.
[17:37:44 EDT(-0400)] <holdorph> you can use an xml editor in eclipse to get the init-param syntax right in the portlet.xml file
[17:38:44 EDT(-0400)] <EricDalquist> holdorph: I don't think they are
[17:38:53 EDT(-0400)] <EricDalquist> but it would be easy enough to add it to the model in the controller
[17:39:02 EDT(-0400)] <EricDalquist> or if you have lots of controllers in an interceptor
[17:39:18 EDT(-0400)] <holdorph> <init-param>
[17:39:18 EDT(-0400)] <holdorph> <name>contextConfigLocation</name>
[17:39:18 EDT(-0400)] <holdorph> <value>/WEB-INF/context/portlet/mode.xml</value>
[17:39:18 EDT(-0400)] <holdorph> </init-param>
[17:39:25 EDT(-0400)] <holdorph> is the portlet.xml syntax, fwiw
[17:40:58 EDT(-0400)] <dstn> interceptor looks interesting
[17:41:20 EDT(-0400)] <dstn> I've actually got another parameter that needs setting too, and that's isJqueryEnabled
[17:41:33 EDT(-0400)] <dstn> currently just using message.properties with a name=value in there
[17:42:58 EDT(-0400)] <EricDalquist> inti-param or portlet preference would be a good place for that too
[17:43:31 EDT(-0400)] <holdorph> defineObjects tag, makes the PortletConfig object available to the jsp
[17:43:37 EDT(-0400)] <holdorph> the init param should be in there.
[17:43:52 EDT(-0400)] <dstn> I'm still a bit confused on the init-param, what class would it be too?
[17:44:11 EDT(-0400)] <holdorph> it's not in a class, it's in the portlet.xml file
[17:44:49 EDT(-0400)] <dstn> oh, I was assuming it applied to <portlet-class>
[17:45:17 EDT(-0400)] <holdorph> it's an optional element found within the <portlet> element
[17:45:45 EDT(-0400)] <holdorph> comes after porlet-class before supports
[17:47:05 EDT(-0400)] <holdorph> have to run, for today, see you all later.
[17:54:13 EDT(-0400)] <dstn> so is fisheye ever going to be back up? it's always showing unavailable
[17:54:26 EDT(-0400)] <EricDalquist> yeah ... scott and I got a bit side tracked
[17:54:38 EDT(-0400)] <EricDalquist> I might have some time this weekend to dig into it more
[17:55:33 EDT(-0400)] <EricDalquist> jasig needs to find a p/t sysadmin (tongue)
[17:55:42 EDT(-0400)] <EricDalquist> that isn't also one of their project leads (tongue)
[17:55:48 EDT(-0400)] <dstn> ha
[17:55:50 EDT(-0400)] <dstn> !
[18:00:51 EDT(-0400)] <dstn> potentially I could use inline script elements for css imports
[18:24:59 EDT(-0400)] <dstn> EricDaliquist: have you used defineobjects before? If so, how do you get access to portletConfig? I tried portletConfig.paramName but no go?
[18:25:41 EDT(-0400)] <EricDalquist> http://portals.apache.org/pluto/portlet-1.0-apidocs/javax/portlet/PortletConfig.html
[18:25:45 EDT(-0400)] <EricDalquist> that's the javadoc for Portletconfig
[18:26:00 EDT(-0400)] <EricDalquist> doesn't look like there will be a nice JSTL wait to get at an init-param though
[18:26:21 EDT(-0400)] <dstn> lovely
[18:26:28 EDT(-0400)] <EricDalquist> <%=portletConfig.getInitParameter("foo")%>
[18:26:29 EDT(-0400)] <EricDalquist> (sad)
[18:27:07 EDT(-0400)] <dstn> well, another day, time to go home
[18:27:11 EDT(-0400)] <dstn> later
[18:27:38 EDT(-0400)] * ChanServ (ChanServ@services.) has joined ##uportal
[18:38:16 EDT(-0400)] * ChanServ (ChanServ@services.) has joined ##uportal
[19:55:08 EDT(-0400)] * holdorph (n=holdorph@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[22:26:34 EDT(-0400)] * EiNZTEiN (n=einztein@di220.opencominc.com) has joined ##uportal