uPortal IRC Logs-2011-05-17
[10:40:55 CDT(-0500)] <athena> hey EricDalquist - when are you looking to have everything for the RC in by? end of today? tomorrow morning?
[10:41:17 CDT(-0500)] <EricDalquist> was just going to send an email out about that
[10:41:20 CDT(-0500)] <EricDalquist> tomorrow end of the day
[10:41:21 CDT(-0500)] <athena>
[10:41:24 CDT(-0500)] <athena> oh ok
[10:41:27 CDT(-0500)] <athena> so we have a bit of time
[10:41:53 CDT(-0500)] <athena> i think most of my stuff is in though
[10:41:59 CDT(-0500)] <athena> except for the widgets portlet release
[10:42:10 CDT(-0500)] <athena> if you see something i've missed or that you'd like me to look at, definitely let me know though?
[10:42:25 CDT(-0500)] <EricDalquist> will do
[10:42:32 CDT(-0500)] <EricDalquist> digging through config mode stuff right now
[10:42:39 CDT(-0500)] <EricDalquist> I'll let you know as soon as that is ready
[10:44:23 CDT(-0500)] <athena> sounds good, thanks
[10:47:22 CDT(-0500)] <athena> oh by the way - do we want to apply the short-term cache filter to the images in uPortal?
[10:47:35 CDT(-0500)] <EricDalquist> yeah
[10:47:36 CDT(-0500)] <EricDalquist> I think so
[10:47:39 CDT(-0500)] <EricDalquist> maybe 30 minutes?
[10:47:42 CDT(-0500)] <athena> ok, that was my thought as well
[10:47:47 CDT(-0500)] <athena> sounds good
[10:48:10 CDT(-0500)] <athena> also thinking that maybe we might want a portlet filter that can do gzipping of JSON content
[10:48:17 CDT(-0500)] <athena> for the RSS portlet and such
[10:48:33 CDT(-0500)] <EricDalquist> yeah
[10:48:36 CDT(-0500)] <EricDalquist> ehcache-web includes one
[10:48:44 CDT(-0500)] <EricDalquist> though I'm not the most thrilled with it
[10:48:54 CDT(-0500)] <athena> ah
[10:49:02 CDT(-0500)] <athena> we have some modifications to their one for servlets, right?
[10:49:14 CDT(-0500)] <EricDalquist> oh sorry I miss-read that
[10:49:20 CDT(-0500)] <EricDalquist> yeah there is no portlet filter
[10:49:24 CDT(-0500)] <athena>
[10:49:29 CDT(-0500)] <athena> figure we could do something analogous
[10:49:33 CDT(-0500)] <EricDalquist> that would be nice now that portlet fitlers are possible
[10:49:39 CDT(-0500)] <athena> might not get to that today, but seems like we could do that in the near future
[10:49:40 CDT(-0500)] <athena> yeah
[10:49:46 CDT(-0500)] <athena> think it would help performance
[10:49:52 CDT(-0500)] <athena> dropping that 302 redirect makes a big difference
[10:50:02 CDT(-0500)] <EricDalquist> yeah
[12:16:29 CDT(-0500)] <EricDalquist> athena: you around?
[12:16:55 CDT(-0500)] <athena> yep
[12:17:05 CDT(-0500)] * athena hops around
[12:17:19 CDT(-0500)] <EricDalquist> so with some of the structural changes to the registries I complicated things for the delegation support
[12:17:22 CDT(-0500)] <EricDalquist> and need some ideas
[12:17:35 CDT(-0500)] <athena> uhoh what's up?
[12:17:40 CDT(-0500)] <EricDalquist> around how to handle the portlet entity objects for the delegation
[12:17:57 CDT(-0500)] <athena> oh . . . yeah i bet that is more complex than it used to be
[12:18:00 CDT(-0500)] <EricDalquist> so a portlet entity represents a portlet in a users layout
[12:18:33 CDT(-0500)] <EricDalquist> there is support code for persistent entities (entities that have prefs or are minimized) for transient portlets (stuff rendered by fname that doesn't actually exist in the user's layout)
[12:18:57 CDT(-0500)] <EricDalquist> so if you always use the weather portlet by fname
[12:19:02 CDT(-0500)] <EricDalquist> it remembers your settings for example
[12:19:12 CDT(-0500)] <athena> gotcha
[12:19:14 CDT(-0500)] <EricDalquist> what I'm stuck on is how to handle the entity object for delegation
[12:19:36 CDT(-0500)] <EricDalquist> when portlet A delegates to portlet B we need to have an entity and window object for portlet B
[12:19:48 CDT(-0500)] <EricDalquist> since all the portlet support code requires those exist
[12:20:07 CDT(-0500)] <EricDalquist> what I'm stuck on is what to do with that entity object
[12:20:13 CDT(-0500)] <EricDalquist> do I just store it in the session for now
[12:20:15 CDT(-0500)] <athena> uhhhhhh . . . . yeah.
[12:20:20 CDT(-0500)] <EricDalquist> and punt on the persistence?
[12:20:24 CDT(-0500)] <athena> hmm
[12:20:32 CDT(-0500)] <athena> not sure we really need persistence
[12:20:39 CDT(-0500)] <EricDalquist> yeah we don't for config mode
[12:20:42 CDT(-0500)] <athena> because all the persistence already assumes it's doing portlet config instead, right?
[12:20:43 CDT(-0500)] <athena> ok
[12:20:48 CDT(-0500)] <EricDalquist> which is why I think I'll just punt
[12:20:50 CDT(-0500)] <athena> you're thinking of other use cases?
[12:20:51 CDT(-0500)] <athena> yeah.
[12:20:55 CDT(-0500)] <athena> i think that's totally fair for now
[12:21:11 CDT(-0500)] <EricDalquist> so this stuff is written in a generic manner
[12:21:20 CDT(-0500)] <EricDalquist> so config is just one use of the delegation
[12:21:30 CDT(-0500)] <EricDalquist> but since that is our only use for now
[12:21:38 CDT(-0500)] <EricDalquist> punting seems ok and just not worry about persistence
[12:21:40 CDT(-0500)] <athena> yeah
[12:21:45 CDT(-0500)] <EricDalquist> ok
[12:21:48 CDT(-0500)] <EricDalquist> thanks
[12:21:51 CDT(-0500)] <athena> we can always revisit that later as we think about additional use cases
[12:21:55 CDT(-0500)] <athena> sure, that was easy enough
[12:22:24 CDT(-0500)] <EricDalquist> also something I realized, and I don't really know how big of a deal this is, is portlets that switch to the detached window state won't ever see portlet events
[12:23:15 CDT(-0500)] <athena> hmm.
[12:23:21 CDT(-0500)] <athena> that's probably not so good
[12:23:28 CDT(-0500)] <athena> but we're probably also not actually using detached mode right now?
[12:23:43 CDT(-0500)] <EricDalquist> no, but I know portlets here that use it a lot
[12:23:46 CDT(-0500)] <athena> ahhh
[12:24:00 CDT(-0500)] <EricDalquist> the problem is that window state was designed to be stateless portal side
[12:24:05 CDT(-0500)] <athena> what kind of stuff are you guys using it for?
[12:24:12 CDT(-0500)] <EricDalquist> like popups from portlets
[12:24:17 CDT(-0500)] <athena> makes sense
[12:24:20 CDT(-0500)] <EricDalquist> there is no portlet window object in memory for detached windows
[12:24:25 CDT(-0500)] <athena> ah
[12:24:25 CDT(-0500)] <EricDalquist> the entire window state is on the url
[12:24:32 CDT(-0500)] <EricDalquist> which means you can keep poping them up
[12:24:40 CDT(-0500)] <EricDalquist> and there are no nav issues
[12:24:55 CDT(-0500)] <EricDalquist> the problem is if there is no server side state for them we have no idea that they exist to send events to
[12:25:04 CDT(-0500)] <EricDalquist> I'm just going to punt on that as well
[12:25:14 CDT(-0500)] <EricDalquist> since I have no idea how we'd track new browser windows
[12:25:15 CDT(-0500)] <athena> makes sense
[12:25:17 CDT(-0500)] <athena> lol
[12:25:17 CDT(-0500)] <athena> yeah
[12:45:10 CDT(-0500)] <athena> hmmm.
[12:45:18 CDT(-0500)] <athena> we should do something about the weather portlet
[12:45:45 CDT(-0500)] <athena> think we talked about making it use yahoo by default, since we can get weather w/o a key? was that right?
[12:48:04 CDT(-0500)] <EricDalquist> right
[12:48:08 CDT(-0500)] <EricDalquist> we just can't search
[12:48:11 CDT(-0500)] <athena> yeah
[12:48:13 CDT(-0500)] <EricDalquist> if I remember correctly
[12:48:17 CDT(-0500)] <athena> yeah
[12:48:17 CDT(-0500)] <EricDalquist> which seems reasonble
[12:48:20 CDT(-0500)] <athena> it does
[12:48:30 CDT(-0500)] <athena> probably could just update the config w/ an overlay
[12:48:33 CDT(-0500)] <athena> should be able to do that today
[12:48:47 CDT(-0500)] <EricDalquist> I wonder how hard it would be to enahnce the portlet to show an error in the edit view saying "you need to go get an api key"
[12:49:45 CDT(-0500)] <athena> not very hard
[12:49:51 CDT(-0500)] <athena> but not something we'll probably get to this week
[12:49:54 CDT(-0500)] <athena> or at least i won't get to it
[12:49:57 CDT(-0500)] <EricDalquist> sounds good
[12:50:02 CDT(-0500)] <athena> ok
[12:51:34 CDT(-0500)] <EricDalquist> also I think we should be able to have resource urls work in config mode
[12:51:39 CDT(-0500)] <EricDalquist> so delegate portlets can do ajax
[12:55:07 CDT(-0500)] <athena> oh! that'd be fantastic
[12:55:18 CDT(-0500)] <athena> what unit is the cache control stuff in? seconds? millis?
[12:55:53 CDT(-0500)] <EricDalquist> um I think the cache setting in portlet.xml was seconds
[12:56:04 CDT(-0500)] <athena> er, sorry, i'm being really vague
[12:56:06 CDT(-0500)] <EricDalquist> so maybe seconds again for the new cache control apis?
[12:56:11 CDT(-0500)] <athena> for the PathBasedCacheExpirationFilter
[12:56:22 CDT(-0500)] <athena> we haven't actually implemented the cache control stuff, right?
[12:56:22 CDT(-0500)] <EricDalquist> oh ... probably millis
[12:56:27 CDT(-0500)] <EricDalquist> since that is what ehcache uses
[12:56:31 CDT(-0500)] <EricDalquist> not yet
[12:56:39 CDT(-0500)] <EricDalquist> so everything just renders all the time
[12:57:38 CDT(-0500)] <athena> looking forward to having that eventually
[12:57:44 CDT(-0500)] <athena> might be pretty interesting w/ some of our ajax features
[12:58:45 CDT(-0500)] <EricDalquist> yes
[12:58:47 CDT(-0500)] <EricDalquist> it will be
[12:59:03 CDT(-0500)] <EricDalquist> and getting portlets updated to use the new features will be really nice
[12:59:11 CDT(-0500)] <EricDalquist> like web proxy
[12:59:17 CDT(-0500)] <EricDalquist> there are a lot of nice things we can do there
[13:00:15 CDT(-0500)] <EricDalquist> also header support is in place right now
[13:00:28 CDT(-0500)] <EricDalquist> the only problem is the spring 3 portlet code doesn't support it
[13:00:41 CDT(-0500)] <EricDalquist> so to use it you have to implement GenericPortlet directly
[13:00:48 CDT(-0500)] <EricDalquist> one of the todos is to fix that for them
[13:00:51 CDT(-0500)] <EricDalquist> and submit a patch
[13:00:52 CDT(-0500)] <athena> ah
[13:00:53 CDT(-0500)] <athena> yeah
[14:55:56 CDT(-0500)] <athena> well, we're slowly getting better ratings w/ yslow
[14:56:03 CDT(-0500)] <EricDalquist> yay
[14:56:05 CDT(-0500)] <EricDalquist> that's good
[14:56:06 CDT(-0500)] <athena> yeah
[14:56:11 CDT(-0500)] <athena> up to a B now
[14:56:53 CDT(-0500)] <athena> probably the biggest thing we could do would be to address the use of images in our theme
[14:57:00 CDT(-0500)] <EricDalquist> oh
[14:57:04 CDT(-0500)] <EricDalquist> do more sprite stuff?
[14:57:10 CDT(-0500)] <athena> yeah, i think that'd probably help a lot
[14:57:23 CDT(-0500)] <athena> using CSS/CSS3 instead of images where possible
[14:57:39 CDT(-0500)] <athena> and if we could find a way to version the image names, of course that'd help too
[14:57:42 CDT(-0500)] <athena> but that wouldn't be trivial
[14:58:24 CDT(-0500)] <EricDalquist> I wonder if SASS can do any sort of css/sprite generation?
[14:58:30 CDT(-0500)] <athena> dunno - could ask gary
[14:58:34 CDT(-0500)] <athena> what version of tomcat are we going to use w/ the quickstart?
[14:58:47 CDT(-0500)] <EricDalquist> 6.0.x
[14:58:53 CDT(-0500)] <athena> just the current version?
[14:58:57 CDT(-0500)] <EricDalquist> yeah
[14:59:01 CDT(-0500)] <athena> cool
[14:59:09 CDT(-0500)] <EricDalquist> you have a request for a more specific version"
[14:59:17 CDT(-0500)] <EricDalquist> I might also give tc7 a try
[14:59:20 CDT(-0500)] <athena> just discovered the reason disableURLRewriting refused to work on my local install was because i had a slightly older version (6.0.29)
[14:59:21 CDT(-0500)] <EricDalquist> just to see how things work
[14:59:22 CDT(-0500)] <athena> so not that one
[14:59:28 CDT(-0500)] <EricDalquist> ah
[14:59:29 CDT(-0500)] <EricDalquist> yeah
[14:59:31 CDT(-0500)] <athena> just dropped in the current version, seems fine
[14:59:37 CDT(-0500)] <EricDalquist> I committed that context.xml file
[14:59:47 CDT(-0500)] <EricDalquist> I also disabled dynamic redeploy of the uportal webapp
[15:00:23 CDT(-0500)] <EricDalquist> since the current version of pluto does not play nice with redeploying the portal
[15:00:39 CDT(-0500)] <athena> ah
[17:33:46 CDT(-0500)] <holdorph> athena / EricDalquist : I need to prepare for one of my Jasig seminars (wed afternoon) and want to test out some of my portlet exercises. Is this something I could try right now with uPortal trunk and be reasonably safe they'll work the same way as of Sunday? Or should I wait until Eric's done with the RC cut (because things will be changing before then)?
[17:34:07 CDT(-0500)] <EricDalquist> yes, that's reasonable holdorph
[17:34:14 CDT(-0500)] <EricDalquist> I think the only big thing being fixed right now is config mode
[17:34:35 CDT(-0500)] <EricDalquist> though while you're testing if you find issues (and I'm sure you will) please stick them in jira
[17:35:59 CDT(-0500)] <holdorph> ok. I'll try to start tonight. (of course this is all dependent on my puppy not demanding my attention, which is unlikely 90% of the time.)
[17:36:08 CDT(-0500)] <EricDalquist> lol
[17:36:28 CDT(-0500)] <EricDalquist> yeah I "lucked out" in that wife and child are visiting her family in LA for 2 weeks
[17:36:31 CDT(-0500)] <EricDalquist> so I have lots of time
[17:36:39 CDT(-0500)] <EricDalquist> but its never fun being away
[17:36:57 CDT(-0500)] <holdorph> nod
[17:37:14 CDT(-0500)] <athena> holdorph: i have pretty much all my stuff in
[17:37:25 CDT(-0500)] <athena> just doing testing and bug fixing before cutting a couple portlet releases
[17:37:36 CDT(-0500)] <athena> and i hear that about the puppy attention demands
[17:37:44 CDT(-0500)] <athena> mine is not happy with my work schedule of late
[17:38:02 CDT(-0500)] <EricDalquist> oh athena I just remembered something
[17:38:26 CDT(-0500)] <athena> not allowed
[17:38:29 CDT(-0500)] <athena>
[17:38:33 CDT(-0500)] <EricDalquist> it appears that the portlet publishing UI in the portlet manager still has the fields that allow people to manually edit the portlet webapp name and the portlet name
[17:38:47 CDT(-0500)] <EricDalquist> I don't believe we need that at all anymore
[17:38:55 CDT(-0500)] <athena> hmm
[17:39:08 CDT(-0500)] <athena> i think those were only supposed to show up when the values weren't already set?
[17:39:19 CDT(-0500)] <athena> like if you wanted to pre-register a portlet you hadn't installed yet
[17:39:21 CDT(-0500)] <EricDalquist> right but I think they may still be looking at the parameter values
[17:39:21 CDT(-0500)] <athena> not sure
[17:39:36 CDT(-0500)] <EricDalquist> and not the on-object properties that we are now using
[17:39:38 CDT(-0500)] <athena> yeah it's probably looking at the wrong thing i guess
[17:39:38 CDT(-0500)] <athena> yeah
[17:40:28 CDT(-0500)] <EricDalquist> iff you happen to have time and are looking for something to do
[17:40:35 CDT(-0500)] <EricDalquist> otherwise it can just get left until after the rc
[17:40:38 CDT(-0500)] <EricDalquist> I'll create a jira issue for it
[17:41:17 CDT(-0500)] <holdorph> maelstrom probably votes until after the RC i'd guess
[17:41:37 CDT(-0500)] <athena> my homework also probably votes as such, unfortunately
[17:41:46 CDT(-0500)] <athena> but i might get to it tomorrow if it turns out to be trivial
[17:41:52 CDT(-0500)] <EricDalquist> no problem
[17:41:56 CDT(-0500)] * athena somehow has to summon the energy to finish homework tonight, ugh
[17:42:04 CDT(-0500)] <EricDalquist> just had the random thought about it
[17:42:06 CDT(-0500)] <EricDalquist> :/
[17:42:14 CDT(-0500)] <athena> no i'm glad you brought it up
[17:42:20 CDT(-0500)] <athena> need to at least get this stuff tracked
[17:55:46 CDT(-0500)] <athena> yay, have code that can create user files for all the conference attendees
[17:55:53 CDT(-0500)] <athena> feeling like things are in decent shape
[17:55:57 CDT(-0500)] <EricDalquist> neat!
[17:56:10 CDT(-0500)] <athena> ya
[17:56:20 CDT(-0500)] <athena> pretty easy to create from a tab-delimited file
[17:56:26 CDT(-0500)] <athena> even auto-created a group file
[18:43:54 CDT(-0500)] <EricDalquist> athena: you still there?
[18:50:54 CDT(-0500)] <athena> in between dog walking and running
[18:51:22 CDT(-0500)] <EricDalquist> so it looks like we don't have explicit fields in the cpds for portletAppName, portletName and frameworkPortlet flag
[18:51:34 CDT(-0500)] <EricDalquist> do you see any problems with me adding those to the CPD XSD?
[18:51:43 CDT(-0500)] <EricDalquist> and updating the CPDs accordingly
[18:51:45 CDT(-0500)] <athena> sounds like a great idea to me
[18:51:48 CDT(-0500)] <EricDalquist> ok
[18:52:01 CDT(-0500)] <EricDalquist> I think that issue I was complaining about earlier I'm going to have to address
[18:52:14 CDT(-0500)] <athena> hmm
[18:52:16 CDT(-0500)] <athena> i see this?
[18:52:17 CDT(-0500)] <athena> <portlet-descriptor>
[18:52:17 CDT(-0500)] <athena> <webAppName>/CalendarPortlet</webAppName>
[18:52:17 CDT(-0500)] <athena> <portletName>calendar</portletName>
[18:52:17 CDT(-0500)] <athena> </portlet-descriptor>
[18:52:27 CDT(-0500)] <athena> is that not what we're looking for?
[18:52:34 CDT(-0500)] <EricDalquist> in the .cpd.xml?
[18:52:39 CDT(-0500)] <athena> oh oh oh
[18:52:40 CDT(-0500)] <athena> right.
[18:52:42 CDT(-0500)] <athena> CPD
[18:52:43 CDT(-0500)] <EricDalquist> the config mode stuff is failing because the portlet manager is in this weird state of flux
[18:52:46 CDT(-0500)] <athena> yes, adding those would be awesome
[18:53:01 CDT(-0500)] <EricDalquist> where some code is looking in portlet parameters and others is looking at the explicit fields on the portlet definition form
[18:53:13 CDT(-0500)] <EricDalquist> so I'm going to see if I can get it completely off of overloading the parameters
[18:53:33 CDT(-0500)] <athena> ahh
[18:53:37 CDT(-0500)] <athena> i think that's a good idea
[18:53:41 CDT(-0500)] <athena> i've foudn that sorta confusing
[18:53:47 CDT(-0500)] <athena> ok, i better get going or i won't find any parking
[18:53:59 CDT(-0500)] <athena> first sunny in day in ages, entire city is going to be outside trying to replace vitamin d
[18:54:09 CDT(-0500)] <athena> but will be back in an hour or two and around for the rest of the night
[18:54:20 CDT(-0500)] <EricDalquist> lol
[18:54:22 CDT(-0500)] <EricDalquist> have fun
[18:54:26 CDT(-0500)] <EricDalquist> I'm still in the office
[18:54:29 CDT(-0500)] <EricDalquist> fewer distractions here
[18:54:37 CDT(-0500)] <EricDalquist> just gotta make sure there is still a bus home when I get done
[18:58:05 CDT(-0500)] <athena> hmm yeah that is key
[18:58:07 CDT(-0500)] <athena> ok, bbiab