Page Comparison

[08:47:33 CDT(-0500)] * JoeMoore (89d88117@gateway/web/freenode/ip.137.216.129.23) has joined ##uportal
[08:54:35 CDT(-0500)] * colinclark (~colin@bas2-toronto09-1176130958.dsl.bell.ca) has joined ##uportal
[09:16:50 CDT(-0500)] * EricDalquist (~dalquist@2607:f388:e:0:221:9bff:fe37:e768) has joined ##uportal
[09:32:12 CDT(-0500)] * lfuller (~sparhk@wsip-72-215-204-133.ph.ph.cox.net) has joined ##uportal
[09:53:39 CDT(-0500)] * bsparks (~bsparks@wsip-72-215-204-133.ph.ph.cox.net) has joined ##uportal
[09:54:03 CDT(-0500)] * athena (~athena@c-76-121-97-221.hsd1.wa.comcast.net) has joined ##uportal
[10:21:24 CDT(-0500)] * awills (~awills@wsip-72-215-204-133.ph.ph.cox.net) has joined ##uportal
[10:28:14 CDT(-0500)] <athena> morning EricDalquist
[10:28:19 CDT(-0500)] <EricDalquist> morning athena
[10:28:28 CDT(-0500)] <EricDalquist> had lots of fun doing pluto2 work this weekend
[10:28:31 CDT(-0500)] <athena> i'm trying to untangle some of this update preferences servlet code
[10:28:33 CDT(-0500)] <athena> oh nice!!
[10:28:35 CDT(-0500)] <athena> how's it coming?
[10:28:38 CDT(-0500)] <EricDalquist> good
[10:28:46 CDT(-0500)] <athena> that's great to hear
[10:28:48 CDT(-0500)] <EricDalquist> trying to figure out why actions aren't running
[10:28:54 CDT(-0500)] <EricDalquist> but getting closer to merge
[10:29:00 CDT(-0500)] <EricDalquist> URLs should be all figured out finally
[10:29:02 CDT(-0500)] <EricDalquist> that took forever
[10:29:13 CDT(-0500)] <EricDalquist> I have a few questions too
[10:29:16 CDT(-0500)] <EricDalquist> but ok
[10:29:19 CDT(-0500)] <EricDalquist> update prefs servlet ...
[10:29:22 CDT(-0500)] <EricDalquist> what's up with it?
[10:29:58 CDT(-0500)] <athena> oh, it's just an awful mess
[10:30:07 CDT(-0500)] <athena> tangle of random unplanned methods
[10:30:17 CDT(-0500)] <EricDalquist>
[10:30:23 CDT(-0500)] <athena> finally finished separating it out so it's not all called from one parent method
[10:30:25 CDT(-0500)] <EricDalquist> this is what the JS callbacks use?
[10:30:27 CDT(-0500)] <athena> yeah
[10:30:46 CDT(-0500)] <athena> changed it so each method has its own separate @RequestMapping
[10:30:52 CDT(-0500)] <athena> not sure how much i'll be able to fix
[10:30:57 CDT(-0500)] <EricDalquist> nice
[10:31:11 CDT(-0500)] <athena> part of the problem of course is that the code required to manipulate layouts is just really awkward
[10:31:17 CDT(-0500)] <EricDalquist> yes
[10:31:55 CDT(-0500)] <athena> it looks like it's using the channel manager though, which we need to fix
[10:32:11 CDT(-0500)] <athena> in particular, when you add a new portlet to your page, it calls out to the channel manager to instantiate it in your layout
[10:32:21 CDT(-0500)] <athena> do we need to move that instantiation code somewhere?
[10:32:35 CDT(-0500)] <athena> or will we just not need that in the future?
[10:32:38 CDT(-0500)] <EricDalquist> that's probably simply going to disappear
[10:32:42 CDT(-0500)] <athena> ok
[10:32:48 CDT(-0500)] <EricDalquist> there is no equivalent method in the pluto2 branch
[10:32:51 CDT(-0500)] <athena> gotcha
[10:32:54 CDT(-0500)] <EricDalquist> since portlet's are stateless
[10:32:58 CDT(-0500)] <athena> so should i just leave it until you do the merge?
[10:33:03 CDT(-0500)] <EricDalquist> yeah
[10:33:06 CDT(-0500)] <athena> sounds good
[10:33:14 CDT(-0500)] <EricDalquist> documenting shib stuff today
[10:33:20 CDT(-0500)] <athena> going to be sooo glad to get rid of this stuff
[10:33:20 CDT(-0500)] <EricDalquist> then going to look at pluto2 again
[10:33:21 CDT(-0500)] <athena> oh yay
[10:34:45 CDT(-0500)] <EricDalquist> so I'm trying to figure out where to track WindowState
[10:35:24 CDT(-0500)] <EricDalquist> in 3.x it is tracked in the user;s layout (theme parameters) and in the PortletWindow objects (session level data)
[10:35:43 CDT(-0500)] <EricDalquist> and when a new PortletWindow is created the corresponding state is pulled from the layout
[10:35:52 CDT(-0500)] <athena> ah, interesting
[10:35:53 CDT(-0500)] <EricDalquist> I'd really like to only track it in one location
[10:35:57 CDT(-0500)] <athena> yeah
[10:36:02 CDT(-0500)] <EricDalquist> PortletWindow makes the most sense I think
[10:36:12 CDT(-0500)] <athena> yeah that sounds reasonable
[10:36:24 CDT(-0500)] <EricDalquist> I'm just not sure how much work it would be to update the theme parameters based on all windows every render
[10:36:30 CDT(-0500)] <EricDalquist> at least in the current rendering pipe
[10:36:46 CDT(-0500)] <athena> are we likely to replace it w/ the stax pipe anytime soon?
[10:36:57 CDT(-0500)] <EricDalquist> I'd really like to
[10:37:17 CDT(-0500)] <EricDalquist> I'll have a better idea once the pluto2 merge is done
[10:37:27 CDT(-0500)] <EricDalquist> so I think I'm just going to leave the window state syncing out for now
[10:37:35 CDT(-0500)] <EricDalquist> and file it as a known issue in trunk for a bit
[10:38:37 CDT(-0500)] <athena> sounds good
[10:38:44 CDT(-0500)] <athena> will be nice to have everything merged in
[10:38:48 CDT(-0500)] <EricDalquist> yes
[10:38:50 CDT(-0500)] <athena> speaking of layouts
[10:38:55 CDT(-0500)] <athena> do you have a suggestion for replacing this?
[10:38:56 CDT(-0500)] <athena> http://uportal.pastebin.com/aN03B7nN
[10:39:09 CDT(-0500)] <athena> i see the ulm.getUserLayoutDOM() method has been deprecated
[10:39:30 CDT(-0500)] <EricDalquist> no
[10:39:34 CDT(-0500)] <EricDalquist> it was deprecated
[10:39:39 CDT(-0500)] <EricDalquist> but no real replacement was added
[10:39:45 CDT(-0500)] <EricDalquist> that was Peter back in 2.4 I think
[10:40:07 CDT(-0500)] <athena> oh yikes.
[10:40:10 CDT(-0500)] <athena> that's like, forever ago
[10:40:13 CDT(-0500)] <EricDalquist> yeah
[10:40:18 CDT(-0500)] <EricDalquist> so there was intent to replace it
[10:40:26 CDT(-0500)] <EricDalquist> but I think that was planned as part of ALM
[10:40:31 CDT(-0500)] <athena> yeesh
[10:40:46 CDT(-0500)] <EricDalquist> yeah
[10:40:51 CDT(-0500)] <athena> well it does seem reasonable to not have the code manipulate the DOM directly
[10:40:52 CDT(-0500)] <EricDalquist> maybe if the new stax pipeline happens
[10:40:58 CDT(-0500)] <athena> which the new subscribe tab behavior is very guilty of
[10:41:00 CDT(-0500)] <athena> yeah
[10:41:00 CDT(-0500)] <EricDalquist> the next version we can look at refactoring the layout management
[10:41:07 CDT(-0500)] <athena> i think that'll be a really nice mod
[10:41:08 CDT(-0500)] <athena> yeah
[10:41:11 CDT(-0500)] <EricDalquist> since its all really ugly right now
[10:41:17 CDT(-0500)] <athena> no kidding
[10:41:24 CDT(-0500)] <athena> this code is all so awkward
[10:41:43 CDT(-0500)] <athena> and every method seems to need the person, ulm, and upm objects to manipulate layouts
[10:41:56 CDT(-0500)] <EricDalquist> yup
[10:43:15 CDT(-0500)] <EricDalquist> oi
[10:43:21 CDT(-0500)] <EricDalquist> so pages like this are no good: https://wiki.jasig.org/display/UPM31/03+Shibboleth
[10:43:31 CDT(-0500)] <EricDalquist> we don't need feature history in a manual
[10:47:00 CDT(-0500)] <athena> yeah i've been removing that stuff
[10:47:27 CDT(-0500)] <athena> some of the documentation was just long treatises on the history of the feature rationale and who contributed
[10:47:32 CDT(-0500)] <athena> with no info about like, how to configure it
[10:47:40 CDT(-0500)] <EricDalquist> yeah
[10:47:46 CDT(-0500)] <EricDalquist> well I'll get that updated today
[10:47:52 CDT(-0500)] <athena> i'm all for giving people credit, but we need the manual to be useful
[10:47:53 CDT(-0500)] <athena> awesome
[10:52:31 CDT(-0500)] <EricDalquist> like on that page ... should it even mention setting up Shibboleth? Or should I just add a link to the shib docs saying "you need to have a functional Shib IdP before reading this page"
[10:53:18 CDT(-0500)] <athena> well, a link is a good start
[10:53:36 CDT(-0500)] <athena> it seems like we have people that are asking us to document things like tomcat, maven, etc. without having to go to other websites
[10:53:49 CDT(-0500)] <athena> but shib is really complicated and is pretty early in its development
[10:53:52 CDT(-0500)] <EricDalquist> I guess I figure that setting up and IdP and configuring the apache module are things that should be read on the shib docs
[10:53:57 CDT(-0500)] <athena> yeah
[10:54:06 CDT(-0500)] <athena> i kind of hate to try and duplicate those
[10:54:39 CDT(-0500)] <EricDalquist> yeah
[11:07:18 CDT(-0500)] * athena (~athena@c-76-121-97-221.hsd1.wa.comcast.net) has joined ##uportal
[11:11:15 CDT(-0500)] <EricDalquist> also wondering what the appropriate use of comments on a manual page is
[11:12:37 CDT(-0500)] <athena> good question
[11:13:27 CDT(-0500)] <awills> i had occationally thought that requests for clarification or missing info would be suitable comments, but moreso if they're subsequently pruned
[11:13:39 CDT(-0500)] <EricDalquist> yeah
[11:17:10 CDT(-0500)] <athena> sounds reasonable to me
[11:17:35 CDT(-0500)] * colinclark (~colin@142.150.154.130) has joined ##uportal
[11:18:16 CDT(-0500)] <awills> i just added a watch to the 3.2 manual space
[11:22:08 CDT(-0500)] <EricDalquist> https://wiki.jasig.org/display/UPM31/03+Shibboleth
[11:22:10 CDT(-0500)] <EricDalquist> there we go
[11:25:14 CDT(-0500)] <athena> looks better
[11:25:42 CDT(-0500)] <EricDalquist> on to the delegation docs
[11:44:53 CDT(-0500)] <athena> hmm, you know, we could cut a lot of the stupid out of this code if we dealt w/ the whole active/default tab issue
[11:45:04 CDT(-0500)] <EricDalquist> ah yeah
[11:45:11 CDT(-0500)] <EricDalquist> I had a note to deal with that eventually
[11:45:25 CDT(-0500)] <EricDalquist> really we just need to not overload one structure parameter
[11:45:53 CDT(-0500)] <athena> yeah
[11:45:54 CDT(-0500)] <EricDalquist> like have a 'defaultTab' that is in the structure def
[11:45:58 CDT(-0500)] <EricDalquist> so it is persisted
[11:46:02 CDT(-0500)] <athena> yes
[11:46:11 CDT(-0500)] <EricDalquist> then use activeTab in the XSL
[11:46:13 CDT(-0500)] <athena> do we want to persist the active tab at all?
[11:46:23 CDT(-0500)] <athena> or should that just be a session variable?
[11:46:36 CDT(-0500)] <EricDalquist> nope, just a session variable
[11:46:46 CDT(-0500)] <EricDalquist> well structure parameter
[11:46:56 CDT(-0500)] <EricDalquist> but if it isn't listed in the structure def it won't be persisted
[11:47:05 CDT(-0500)] <EricDalquist> then we just need to expose a UI for setting it
[11:47:14 CDT(-0500)] <athena> that should be easy enough
[11:47:20 CDT(-0500)] <EricDalquist> essentially a 'make this tab default' button
[11:47:27 CDT(-0500)] <athena> yeah
[11:47:37 CDT(-0500)] <athena> thinking maybe a checkbox or button on the page layout dialog
[11:47:40 CDT(-0500)] <EricDalquist> yeah
[11:48:09 CDT(-0500)] <EricDalquist> so I think what needs to be done is update uportal-war/src/main/resources/layout/structure/columns/columns.sdf
[11:48:17 CDT(-0500)] <EricDalquist> replace activeTab with defaultTab
[11:48:40 CDT(-0500)] <EricDalquist> and then figure out places in code and xsl where the differentiation is needed
[11:49:01 CDT(-0500)] <athena> and i guess we'd need to have a way to migrate user layouts, in case someone's really using activeTab
[11:49:14 CDT(-0500)] <EricDalquist> yeah
[11:49:30 CDT(-0500)] <EricDalquist> that could be handled in the 3.2->3.3 layout scripts
[11:49:35 CDT(-0500)] <EricDalquist> that's an easy fix
[11:49:49 CDT(-0500)] <athena> yep
[11:50:08 CDT(-0500)] <athena> well maybe i'll take a look at that today
[11:50:15 CDT(-0500)] <EricDalquist>
[11:50:23 CDT(-0500)] <athena> cleaning up this stuff would make the new DLM permission changes a lot easier
[11:50:48 CDT(-0500)] <EricDalquist> yeah
[11:51:01 CDT(-0500)] <EricDalquist> and it would make 'default tab selection' a functional feature
[11:54:00 CDT(-0500)] <athena> yeah
[11:54:34 CDT(-0500)] <athena> so presumably we need to initialize the in-session activeTab variable to the default tab at some point
[11:55:39 CDT(-0500)] <EricDalquist> or update the XSL to handle both
[11:55:53 CDT(-0500)] <EricDalquist> the structure XSL I believe has the logic to handle the activeTab value
[11:56:05 CDT(-0500)] <EricDalquist> and I think it could deal with it being unset if modified appropriately
[11:56:07 CDT(-0500)] <athena> sort of, but it's in StaticRenderingPipeline before that
[11:56:12 CDT(-0500)] <EricDalquist> ah
[11:56:33 CDT(-0500)] <athena> that's where it transforms the int value to an actual node, it looks like
[11:57:10 CDT(-0500)] <athena> so is the right thing to do to set it in the structure stylesheet rather than the session directly?
[11:57:26 CDT(-0500)] <EricDalquist> right, the structure style sheet parameters are session scoped
[11:57:29 CDT(-0500)] * jessm (~Jess@c-68-40-84-116.hsd1.mi.comcast.net) has joined ##uportal
[11:57:32 CDT(-0500)] <EricDalquist> and that's where that info is currently tracked
[11:58:03 CDT(-0500)] <athena> gotcha
[11:58:04 CDT(-0500)] <athena> ok
[11:58:07 CDT(-0500)] <athena> i think i see how to do that
[11:58:07 CDT(-0500)] <EricDalquist> and actually when I merge the pluto2 branch the activeTab will be set on every request
[11:58:07 CDT(-0500)] <athena>
[11:58:14 CDT(-0500)] <athena> oh?
[11:58:17 CDT(-0500)] <EricDalquist> since tab will be part of the URL
[11:58:19 CDT(-0500)] <EricDalquist> yeah
[11:58:26 CDT(-0500)] <EricDalquist> the pluto2 branch has the new consistent URL work as well
[11:58:30 CDT(-0500)] <athena> oh right!
[13:30:03 CDT(-0500)] <EricDalquist> yay for espn online streams + extra monitor
[13:36:49 CDT(-0500)] <athena> nice
[13:36:56 CDT(-0500)] * athena ponders creeping downstairs
[13:37:14 CDT(-0500)] <athena> man i can't believe how clear the mountains are today
[13:46:07 CDT(-0500)] <athena> EricDalquist: what's structuredLayout.dtd?
[13:46:15 CDT(-0500)] <athena> looks like whatever it is is super outdated
[13:46:25 CDT(-0500)] <EricDalquist> probably a very old attempt at creating a DTD for the layout XML
[13:46:28 CDT(-0500)] <athena> ah
[13:46:33 CDT(-0500)] <athena> there's like, colors
[13:47:13 CDT(-0500)] <EricDalquist> lol
[13:47:40 CDT(-0500)] <athena> guess i'll just ignore that one for now
[13:47:54 CDT(-0500)] <EricDalquist> yeah
[13:47:57 CDT(-0500)] <EricDalquist> or delete it
[13:48:07 CDT(-0500)] <athena> woah
[13:48:22 CDT(-0500)] <athena> game's off to quite a start
[13:48:26 CDT(-0500)] <EricDalquist> yeah
[13:48:46 CDT(-0500)] <EricDalquist> and I have to spend the next hour in a meeting now
[13:48:51 CDT(-0500)] <athena> no fair!
[13:49:07 CDT(-0500)] <EricDalquist> not at all
[15:37:53 CDT(-0500)] <EricDalquist> oh athena
[15:38:06 CDT(-0500)] <EricDalquist> when you're working on the JS for the user preferences
[15:38:20 CDT(-0500)] <EricDalquist> can you switch it to use absolute URLs
[15:38:57 CDT(-0500)] <EricDalquist> when I merge the pluto2 branch /uPortal/ will no longer be the only folder in the URL
[15:39:54 CDT(-0500)] <athena> ah
[15:40:04 CDT(-0500)] <EricDalquist> its on my list of things to do
[15:40:11 CDT(-0500)] <EricDalquist> but if you're digging around in that code right now ...
[16:11:01 CDT(-0500)] <athena> no that's a good thing to do
[16:11:03 CDT(-0500)] <athena> i'll take a look
[16:11:21 CDT(-0500)] * colinclark (~colin@142.150.154.130) has joined ##uportal
[16:11:22 CDT(-0500)] <athena> i guess what we need to do is figure out how to inject the context name of the portal into the theme?
[16:11:35 CDT(-0500)] <EricDalquist> yeah, already doing that in pluto2 it may already be in trunk
[16:11:37 CDT(-0500)] <EricDalquist> let me check
[16:12:09 CDT(-0500)] <athena> oh!
[16:12:11 CDT(-0500)] <athena> neato
[16:12:14 CDT(-0500)] <athena> that makes it really easy then
[16:12:26 CDT(-0500)] <EricDalquist> we're passing in CONTEXT_PATH as a param in the pluto2 branch
[16:12:31 CDT(-0500)] <EricDalquist> but I don't remember adding that
[16:12:36 CDT(-0500)] <EricDalquist> so it may have already been there
[16:12:43 CDT(-0500)] <athena> gotcha
[16:12:45 CDT(-0500)] <athena> i'll look
[16:13:03 CDT(-0500)] <EricDalquist> looks like it isn't
[16:13:25 CDT(-0500)] <athena> hm, guess not
[16:13:36 CDT(-0500)] <athena> maybe easier to just wait until the pluto 2.0 merge
[16:13:41 CDT(-0500)] <athena> shoudln't be a big deal
[16:13:49 CDT(-0500)] <EricDalquist> ok
[16:13:51 CDT(-0500)] <EricDalquist> sounds good
[16:14:02 CDT(-0500)] <EricDalquist> I figured it would be a fairly easy fix
[16:14:28 CDT(-0500)] <athena> ya
[16:14:44 CDT(-0500)] <athena> should just be injecting that into the default skin location
[16:15:02 CDT(-0500)] <EricDalquist> yeah
[16:15:09 CDT(-0500)] <EricDalquist> that works for the css/js files
[16:15:14 CDT(-0500)] <EricDalquist> but the callbacks were still horked
[16:15:17 CDT(-0500)] <EricDalquist> at least in the branch
[16:15:23 CDT(-0500)] <EricDalquist> we'll see the state of things post-merge
[16:15:30 CDT(-0500)] <athena> ah, gotcha
[16:15:39 CDT(-0500)] <athena> well those are actually overridable defaults already
[16:15:39 CDT(-0500)] <athena> s
[16:15:39 CDT(-0500)] <athena> o
[16:15:47 CDT(-0500)] <athena> so it won't be a big deal to to just set them in the XSL
[16:16:03 CDT(-0500)] <EricDalquist> ok
[16:27:55 CDT(-0500)] * awills (~awills@wsip-72-215-204-133.ph.ph.cox.net) has joined ##uportal
[16:27:55 CDT(-0500)] * bsparks (~bsparks@wsip-72-215-204-133.ph.ph.cox.net) has joined ##uportal
[17:37:03 CDT(-0500)] * bsparks (~bsparks@wsip-72-215-204-133.ph.ph.cox.net) has joined ##uportal