fluid-work IRC Logs-2011-08-03
[08:15:59 CDT(-0500)] <Bosmon> Crikey, you are all such EARLY RISERS
[08:16:08 CDT(-0500)] <Bosmon> So unhealthy
[08:32:59 CDT(-0500)] <huslage> memory leaking everywhere.
[08:33:01 CDT(-0500)] <huslage> what a mess
[08:33:07 CDT(-0500)] <huslage> bits all over the floor
[08:33:20 CDT(-0500)] <Bosmon> Scrape 'em off, Jim
[08:39:03 CDT(-0500)] <huslage> it's memory leak week here at IDRC
[09:50:56 CDT(-0500)] <huslage> man
[09:51:00 CDT(-0500)] <huslage> you guys are making me hungry
[10:58:17 CDT(-0500)] <cindyli> Bosmon: i've sent a pull request on FLUID-4353, options munging. can you have a look? thanks.
[11:56:31 CDT(-0500)] <Bosmon> Thanks, cindyli, I will have a look at it now
[11:56:42 CDT(-0500)] <Bosmon> Did you get to the bottom of the UIEnhancer issue in the end?
[12:40:16 CDT(-0500)] <cindyli> Bosmon: sorry for the late reply. we were having a potluck lunch to celebrate one year anniversary that we move to OCAD
[12:40:31 CDT(-0500)] <cindyli> Bosmon: to ur question - yes
[12:40:58 CDT(-0500)] <Bosmon> Ah, congrats on the anniversary
[12:41:03 CDT(-0500)] <Bosmon> cindyli - what was the problem in the end?
[12:41:21 CDT(-0500)] <cindyli> it was because there's a settingStore sub-component definition in fluid.uiOptions defaults
[12:41:30 CDT(-0500)] <cindyli> which should only acts as a placeholder
[12:41:43 CDT(-0500)] <cindyli> and supplied by caller component
[12:42:10 CDT(-0500)] <cindyli> the fat panel didn't supply it. so it was created on the fly as a new third settingStore
[12:42:43 CDT(-0500)] <cindyli> which contains the very default site settings that store.js defines
[12:43:24 CDT(-0500)] <cindyli> the issue was resolved by, 1. removed this placeholder; 2. supplied settingStore from fat panel
[12:45:44 CDT(-0500)] <cindyli> Bosmon: there's another issue i mentioned a couple of days ago about the template prefix for fat panel
[12:45:47 CDT(-0500)] <cindyli> https://github.com/cindyli/infusion/blob/FLUID-4353/src/webapp/demos/uiOptions/FatPanelUIOptions/js/fatPanelUIOptions.js#L28
[12:46:11 CDT(-0500)] <cindyli> the prefix is a relative path between the template location of iframe
[12:46:25 CDT(-0500)] <Bosmon> Oh, yes
[12:46:45 CDT(-0500)] <cindyli> I thought my code change for options munging broke it, actually it is not
[12:46:55 CDT(-0500)] <Bosmon> That's good
[12:47:35 CDT(-0500)] <jhung> colinclark, I
[12:47:37 CDT(-0500)] <Bosmon> I wonder what the Chinese for "munging" is
[12:47:59 CDT(-0500)] <jhung> i'm testing Decapod and finding that we need a better way of switching between decapod versions.
[12:48:08 CDT(-0500)] <colinclark> How so?
[12:48:33 CDT(-0500)] <cindyli> Bosmon: transform?
[12:48:39 CDT(-0500)] <jhung> Right now I'm using hg to checkout 0.4 or the tip, and installing it using the install script we have.
[12:48:42 CDT(-0500)] <Bosmon> haha
[12:48:48 CDT(-0500)] <cindyli> lol
[12:48:55 CDT(-0500)] <Bosmon> "You tell me what language munging is in, and I'll tell you the Chinese for it"
[12:49:02 CDT(-0500)] <jhung> But I'm finding that any existing Decapod install taints the system for any future installs.
[12:49:08 CDT(-0500)] <Bosmon> So, yes - what policy did you decide on to deal with this relative URL issue?
[12:49:20 CDT(-0500)] <colinclark> jhung: Yes, that's a known issue
[12:49:31 CDT(-0500)] <colinclark> Unfortunately
[12:49:40 CDT(-0500)] <jhung> hrm.
[12:49:59 CDT(-0500)] <cindyli> Bosmon: i think first of all, we need to decide if it's ok that relative path is btw iframe and template
[12:50:37 CDT(-0500)] <jhung> Okay. Makes things interesting for testing. There any way to get things clean again?
[12:50:46 CDT(-0500)] <colinclark> Not that I know of
[12:50:54 CDT(-0500)] <colinclark> The shell scripts are all there in the repo
[12:51:03 CDT(-0500)] <Bosmon> cindyli - I guess this is an issue for the wider group
[12:51:08 CDT(-0500)] <colinclark> It's been long enough that I can't quite remember how they work
[12:51:18 CDT(-0500)] <Bosmon> Perhaps anastasiac, colinclark or michelled have an opinion?
[12:51:25 CDT(-0500)] <colinclark> or what the limitation is in terms of installing and uninstalling
[12:51:34 CDT(-0500)] <colinclark> Bosmon: What's the question?
[12:51:45 CDT(-0500)] <Bosmon> The question is about template URL configuration for UIOptions
[12:51:46 CDT(-0500)] <colinclark> jhung: Where are you, anyway? Didn't I just see you half an hour ago?
[12:52:15 CDT(-0500)] <Bosmon> So, it happens that in the Fat Panel configuration, some of the URLs need to be supplied based on the URL location within the iframe
[12:52:19 CDT(-0500)] <Bosmon> Rather than that of the parent document
[12:52:57 CDT(-0500)] <Bosmon> The question is whether we should consider it a blocker for this release and try to remove it...
[12:56:07 CDT(-0500)] <colinclark> one sec
[12:56:30 CDT(-0500)] <colinclark> ah
[12:56:36 CDT(-0500)] <colinclark> can you show me the consequences of it?
[12:56:45 CDT(-0500)] <colinclark> what does the config have to look like now?
[12:56:52 CDT(-0500)] <colinclark> and what would it look like after your blocker?
[12:58:34 CDT(-0500)] <colinclark> Wow, I just cloned something with Mercurial for the first time in over a year
[12:58:39 CDT(-0500)] <Bosmon> colinclark - I guess the configuration would not have to be altered unless the user wanted to change the template prefix
[12:58:43 CDT(-0500)] <colinclark> Git has made this so much more complicated
[12:59:05 CDT(-0500)] <Bosmon> In the case they did change it, they would have to issue a different prefix in the case for FatPanel as they would for FullPreview
[12:59:15 CDT(-0500)] <Bosmon> Is that right, cindyli?
[12:59:24 CDT(-0500)] <Bosmon> colinclark - you mean, made mercurial seem much more complicated?
[12:59:48 CDT(-0500)] <cindyli> yes, Bosmon
[13:00:38 CDT(-0500)] <colinclark> Bosmon: Well, it's just a matter of keeping three different version control systems in my head at the same time
[13:01:11 CDT(-0500)] <Bosmon> So in this example cindyli has pasted, the default template prefix configuration is "./"
[13:01:15 CDT(-0500)] <Bosmon> In the case of FatPanel
[13:01:37 CDT(-0500)] <Bosmon> Reflecting the fact that by default, the templates are kept in the same directory as the fat panel iframe source itself
[13:03:00 CDT(-0500)] <Bosmon> Here is the corresponding configuration for "FullPreview"
[13:03:01 CDT(-0500)] <Bosmon> https://github.com/cindyli/infusion/blob/FLUID-4353/src/webapp/demos/uiOptions/FullPreviewUIOptions/html/uiOptions.html
[13:03:12 CDT(-0500)] <Bosmon> Interestingly FatPanel makes things easier in that particular case
[13:03:18 CDT(-0500)] <Bosmon> But I think it is really a bit deceptive
[13:04:18 CDT(-0500)] <Bosmon> So, our options are i) document carefully in the different configurations where the template prefix is considered to be relative to, or ii) salvage some URL-parsing code (from Kettle, perhaps) to do the work of rebasing the URLs so that all configurations of UIOptions use the same style
[13:05:10 CDT(-0500)] <colinclark> How much work do you imagine ii) will take?
[13:05:52 CDT(-0500)] <Bosmon> My guess it would be an hour or so, but you know how I am with estimates
[13:05:59 CDT(-0500)] <Bosmon> Perhaps cindyli would like to give a counter-estimate
[13:08:16 CDT(-0500)] <cindyli> Bosmon: so the decision is to have, fat panell prefix === a relative path between the main html and template
[13:08:35 CDT(-0500)] <cindyli> i need to have a look into kettle url rebasing code
[13:09:00 CDT(-0500)] <cindyli> not sure if one hour is enough
[13:13:31 CDT(-0500)] <cindyli> can someone point me to "Kettle URL-parsing code"?
[13:15:28 CDT(-0500)] <colinclark> So, I still need this boiled down a bit more
[13:15:30 CDT(-0500)] <colinclark> call me stupid
[13:15:33 CDT(-0500)] <colinclark> mostly, I'm just tired
[13:15:34 CDT(-0500)] <colinclark> but
[13:15:39 CDT(-0500)] <colinclark> what's the impact going to be for the user?
[13:15:45 CDT(-0500)] <colinclark> the meaning of a particular option
[13:15:52 CDT(-0500)] <colinclark> will be different between fat and full UIO
[13:15:58 CDT(-0500)] <colinclark> despite sharing the same name?
[13:16:00 CDT(-0500)] <colinclark> is that correct?
[13:16:05 CDT(-0500)] <cindyli> yes
[13:16:12 CDT(-0500)] <colinclark> ok
[13:16:17 CDT(-0500)] <colinclark> that strikes me as fairly problematic
[13:16:36 CDT(-0500)] <colinclark> time estimates range from 1 hour to "more"
[13:16:40 CDT(-0500)] <colinclark> that strikes me as a bit risky
[13:16:46 CDT(-0500)] <cindyli> lol
[13:16:51 CDT(-0500)] <colinclark> Justin_o: There's your Kingly summary right there
[13:17:00 CDT(-0500)] <colinclark> You are the boss of the freezer
[13:17:44 CDT(-0500)] <colinclark> Bosmon: While Justin_o catches up, can you think aloud a bit more about what you will take from Kettle, and how it will solve the problem?
[13:23:02 CDT(-0500)] <Bosmon> Yes, let me dig into it...
[13:23:15 CDT(-0500)] <Bosmon> Actually exactly this workflow is inside the URL-rewriting code that is there...
[13:23:47 CDT(-0500)] <Justin_o> cindyli, Bosmon, colinclark: I'm a bit on the fence with all this since i feel the none of the solutions are perfect. I think that once I know there is an iframe used for fat panel, i'll be confused about having to make the paths relative to the parent. I think I will be happy with this though if the rebasing is smart enough to allow me to add the path relative to either the parent or the iframe directly, without things getting messed up
[13:24:03 CDT(-0500)] <Justin_o> Is that how it will work or could work?
[13:24:42 CDT(-0500)] <colinclark> We could argue that the iFrame is an implementation detail--something that a "top-level component user" should never even be aware of
[13:24:54 CDT(-0500)] <colinclark> but that can be an awfully slippery slope in our world
[13:24:56 CDT(-0500)] <Bosmon> https://github.com/fluid-project/kettle/blob/master/src/main/webapp/kettle/js/kettle.js#L76-77
[13:25:11 CDT(-0500)] <Bosmon> So, here is the place where the utilities are used
[13:25:27 CDT(-0500)] <Justin_o> colinclark: yes.. because they will have to modify or supply their own iframe if they want to change any dependencies.. i.e. themes and etc.
[13:25:40 CDT(-0500)] <colinclark> can you elaborate, Justin_o?
[13:25:57 CDT(-0500)] <cindyli> thanks, Bosmon
[13:26:01 CDT(-0500)] <Bosmon> https://github.com/fluid-project/kettle/blob/master/src/main/webapp/kettle/js/kettleUtils.js
[13:26:06 CDT(-0500)] <Bosmon> The utilities themselves are in this file
[13:26:36 CDT(-0500)] <colinclark> Ah, I remember these utilities
[13:26:40 CDT(-0500)] <Bosmon> At least some of them
[13:26:44 CDT(-0500)] <Bosmon> Others are in this file
[13:26:45 CDT(-0500)] <Bosmon> https://github.com/fluid-project/kettle/blob/master/src/main/webapp/kettle/js/kettleCore.js
[13:27:14 CDT(-0500)] <Bosmon> I can't quite remember the rationale for splitting them like this, but it might be a matter of which could be used on both the client and the server
[13:27:31 CDT(-0500)] <Bosmon> I remember that when we made the "Screen Navigator", this split changed somewhat
[13:29:01 CDT(-0500)] <Justin_o> colinclark: so lets say that someone wants to supply there own contrast theme, they would have to make sure that the dependency is linked into the iframe or else it would never have an effect on the panel itself... this means that we can expect that people who want to do customization of uio may have to interact with the iframe
[13:30:03 CDT(-0500)] <colinclark> Bosmon: Yes, that's right. There was some intention to share things in kettleCore.js more widely
[13:30:21 CDT(-0500)] <colinclark> thought in fact I think you and I wrote slightly overlapping implementations of some utilities between client and sever
[13:30:31 CDT(-0500)] <Bosmon> Ah yes, I remember that now
[13:31:41 CDT(-0500)] <colinclark> actually
[13:31:47 CDT(-0500)] <colinclark> there are two kettleCore.js files
[13:31:57 CDT(-0500)] <colinclark> my head is spinning
[13:31:58 CDT(-0500)] <colinclark> https://github.com/fluid-project/engage/blob/master/framework/js/kettleCore.js
[13:32:24 CDT(-0500)] <Bosmon> Ah, engage is "dead"
[13:32:27 CDT(-0500)] <Bosmon> At least in the github sense
[13:32:29 CDT(-0500)] <colinclark> yes, here is one of your most famous function names
[13:32:32 CDT(-0500)] <colinclark> https://github.com/fluid-project/engage/blob/master/framework/js/kettleCore.js#L151
[13:32:42 CDT(-0500)] <Bosmon> I migrated everything out of it that would be needed to make Kettle self-contained
[13:32:57 CDT(-0500)] <colinclark> ok
[13:33:07 CDT(-0500)] <colinclark> so then i guess that explains it
[13:33:33 CDT(-0500)] <Justin_o> huslage: i think there's a new version of jira out (v4.4).. i wonder if it will solve some of the issues we are having with jira at the moment
[13:33:35 CDT(-0500)] <colinclark> kettleCore.js probably started its life in "kettle", moved into "engage," and now you've moved portions of it back to kettle
[13:33:43 CDT(-0500)] <colinclark> note my strategic use of quotes
[13:33:57 CDT(-0500)] <huslage> yeah Justin_o i was just thinking that
[13:34:03 CDT(-0500)] <huslage> it just came out the other day
[13:34:12 CDT(-0500)] <huslage> since atlassian CARES SO MUCH
[13:34:13 CDT(-0500)] <huslage> hehe
[13:34:19 CDT(-0500)] <colinclark>
[13:34:29 CDT(-0500)] <colinclark> Bosmon: So Justin_o raises an interesting point
[13:34:39 CDT(-0500)] <colinclark> Do you think it's impossible for a user to be unaware of the iFrame?
[13:35:05 CDT(-0500)] <colinclark> and if that's true, is it more confusing to express paths not relative to the iFrame's document but to the main document instead?
[13:35:11 CDT(-0500)] <Bosmon> I guess in the cases other than the ones he mentions, yes
[13:35:18 CDT(-0500)] <Justin_o> huslage:
[13:35:25 CDT(-0500)] <huslage> maybe tomorrow
[13:35:27 CDT(-0500)] <Bosmon> How common a case is it for users to want to add new themes, etc.?
[13:35:32 CDT(-0500)] <colinclark> Bosmon: I guess what you're saying is that for the absolute "top level user"...
[13:35:40 CDT(-0500)] <Bosmon> As opposed to customising the mount point of UIOptions
[13:35:41 CDT(-0500)] <colinclark> they will still be oblivious to the iFrame
[13:35:42 CDT(-0500)] <Justin_o> huslage: sure, thanks
[13:35:46 CDT(-0500)] <Bosmon> I imagine that almost EVERY user will do the latter
[13:35:51 CDT(-0500)] <Bosmon> Whereas very few will do the former
[13:36:01 CDT(-0500)] <colinclark> and that all customizers will simply need to be aware that this prefix is relative to the main document, not the iFrame
[13:36:08 CDT(-0500)] <colinclark> Justin_o: Are you any more or less convinced?
[13:36:19 CDT(-0500)] <Justin_o> colinclark: i think i'm about the same
[13:36:31 CDT(-0500)] <colinclark> So, we can agree on one thing, first...
[13:36:32 CDT(-0500)] <colinclark> I think
[13:36:50 CDT(-0500)] <colinclark> We're going to have to carefully document what this prefix option means, either way
[13:37:04 CDT(-0500)] <Justin_o> colinclark: agreed
[13:37:08 CDT(-0500)] <colinclark> so the question is really "should it match the other UI Optionses or not?"
[13:37:24 CDT(-0500)] <colinclark> I guess, if it helps to hear my opinion, I lean towards "yes"
[13:37:31 CDT(-0500)] <Bosmon> Bagginses!!
[13:37:35 CDT(-0500)] <colinclark>
[13:37:53 CDT(-0500)] <colinclark> Though time will really determine the decision
[13:38:15 CDT(-0500)] <colinclark> Bosmon: If Justin_o thinks it's cool, I'd say you should try to dive into solving the problem for a couple hours
[13:38:21 CDT(-0500)] <colinclark> if it goes well, we'll have a pull request to consider
[13:38:46 CDT(-0500)] <Bosmon> Momentarily, two KINGGS were reigning
[13:38:55 CDT(-0500)] <Bosmon> They must have found it uncomfortable fitting onto the same throne
[13:39:02 CDT(-0500)] <Justin_o> Bosmon: i was trying to sit on both sides of the fence
[13:40:15 CDT(-0500)] <Justin_o> Bosmon, colinclark: So i think i'd be fine with exploring the rebasing and would be happier if a user could specify the url to the fat panel as either relative to the parent or the iframe without it making a difference
[13:42:59 CDT(-0500)] <colinclark> can Justin_o have his cake and eat it to, Bosmon?
[13:43:00 CDT(-0500)] <colinclark> too
[13:43:18 CDT(-0500)] <Justin_o> colinclark: i did have "two" pieces of cake today.. so it seems fitting
[13:43:34 CDT(-0500)] <colinclark> The King says "let them eat cake"
[13:43:38 CDT(-0500)] <colinclark> or maybe he's saying
[13:43:42 CDT(-0500)] <colinclark> "let me eat cake"
[13:43:46 CDT(-0500)] <colinclark> I'm so confused
[13:43:49 CDT(-0500)] <colinclark>
[13:43:52 CDT(-0500)] <Justin_o> "let us eat cake"
[13:44:22 CDT(-0500)] <colinclark> "and, if we run out of cake, I baked some really tasty cookies"
[13:44:31 CDT(-0500)] <Justin_o>
[13:44:31 CDT(-0500)] <colinclark> A good King, indeed
[13:44:42 CDT(-0500)] <colinclark> Those were really good cookies, by the way
[13:44:54 CDT(-0500)] <Justin_o> thanks..
[13:45:01 CDT(-0500)] <Bosmon2> Sorry, my turn to get disconnected
[13:45:06 CDT(-0500)] <Bosmon2> I asked how they could do either
[13:45:20 CDT(-0500)] <michelled> fluid-everyone: since we didn't have the meeting that directly precedes dev meeting I completely forgot
[13:45:28 CDT(-0500)] <Bosmon2> Oh gosh
[13:45:32 CDT(-0500)] <Bosmon2> Is it Wednesday already
[13:45:41 CDT(-0500)] <Justin_o> Bosmon2: i can't quite remember how the rebasing worked, but i figure it should know that it doesn't have to change anything
[13:45:59 CDT(-0500)] <michelled> fluid-everyone: the meeting room is taken today so perhaps we can just have a quick meeting here
[13:46:11 CDT(-0500)] <colinclark> yeah, i think we're probably on top of the meeting right here
[13:46:18 CDT(-0500)] <colinclark> Can I take an unconventional approach
[13:46:27 CDT(-0500)] <colinclark> and summarize some of the interesting things I think people are doing?
[13:46:37 CDT(-0500)] <colinclark> And then we can add to it all the things I'm forgetting?
[13:46:38 CDT(-0500)] <michelled> sounds good
[13:46:49 CDT(-0500)] <Bosmon2> The "auto-Socratic" method
[13:47:09 CDT(-0500)] <colinclark> Okay, so michelled is feeling the pressure to get image URL rewriting added to our CSS Generator for UIO's themes.
[13:47:18 CDT(-0500)] <colinclark> And feeling incredibly inspired and productive as a result
[13:48:59 CDT(-0500)] <colinclark> harriswong has pretty much moved off to work on ATutor, since they've got a pending release. But his work includes integrating Infusion 1.4 components such as the Uploader
[13:49:15 CDT(-0500)] <colinclark> mlam is off pair programming with yura on CollectionSpace now, which is awesome
[13:49:24 CDT(-0500)] <colinclark> I've noticed some giggling in the back there
[13:50:04 CDT(-0500)] <colinclark> jhung is suffering through poor shell scripts written by yours truly for Decapod, while trying to run comprehensive tests on Decapod's PDF generation capabilities
[13:50:36 CDT(-0500)] <colinclark> we've been doing some roadmapping for that project recently, which you can see in the wiki http://wiki.fluidproject.org/display/fluid/Decapod+Roadmap
[13:51:01 CDT(-0500)] <colinclark> lahabana is making awesome new features for the video player, currently focusing on keyboard navigation
[13:51:30 CDT(-0500)] <colinclark> anastasiac is laying the foundation for our Infusion 1.4 documentation, especially UI Options, and getting us reading for our impending doc sprint
[13:51:47 CDT(-0500)] <colinclark> heidi_'s making a new IDI website with WordPress, the FSS, and UI Options
[13:51:54 CDT(-0500)] <colinclark> jameswy's planning our future
[13:52:09 CDT(-0500)] <colinclark> jessm's bouncing from meeting to meeting in anticipation of some time off
[13:52:17 CDT(-0500)] <colinclark> How's that?
[13:52:21 CDT(-0500)] <colinclark> Did I even vaguely get it right?
[13:52:24 CDT(-0500)] <colinclark> Am I missing anyone?
[13:53:24 CDT(-0500)] <anastasiac> Justin_o?
[13:54:08 CDT(-0500)] <michelled> the KING is asking people to please get their tasks done rather then sending them to the dungeons
[13:54:21 CDT(-0500)] <colinclark>
[13:55:38 CDT(-0500)] <colinclark> What is on your list at the moment, Justin_o?
[13:56:28 CDT(-0500)] <Justin_o> colinclark: i'm trying to work my way through the styling issues for UIO that are blockers
[13:56:46 CDT(-0500)] <Justin_o> currently i'm trying to fix the tab bar in fat panel when the contrast themes are applied
[13:57:01 CDT(-0500)] <colinclark> cool
[13:57:09 CDT(-0500)] <colinclark> cindyli: Did I get your work right?
[13:57:19 CDT(-0500)] <cindyli> yes, colinclark
[13:57:30 CDT(-0500)] <colinclark> anastasiac: I can imagine I was a bit fuzzy on yours, so feel free to fill any interesting details
[13:57:32 CDT(-0500)] <Justin_o> at the moment i'm struggling with how to get everything, all the icons and etc all styled properly without making the theme know about it too much
[13:58:49 CDT(-0500)] <anastasiac> colinclark, you've got it right. Planning for the 1.4 docs sprint is at http://wiki.fluidproject.org/display/fluid/Docs+Sprint+August+2011. I'm also working on a GSoC Docs Camp proposal.
[13:59:12 CDT(-0500)] <colinclark> Bosmon2: Rumor has it you were doing some Kettle thinking recently. Anything interesting to summarize?
[13:59:46 CDT(-0500)] <Bosmon2> Well, nothing terribly interesting
[14:00:07 CDT(-0500)] <Bosmon2> I think I am basically clear now how the request cycle will operate via IoC
[14:00:15 CDT(-0500)] <Bosmon2> But it will be quite a different system to "old Kettle"
[14:00:25 CDT(-0500)] <Bosmon2> With its "rats nest of function pointers" model
[14:00:35 CDT(-0500)] <Bosmon2> Otherwise known as "functional programming", I guess
[14:01:38 CDT(-0500)] <Bosmon2> The idea is similar to the way CollectionSpace constructs multiple "tabs"
[14:01:52 CDT(-0500)] <Bosmon2> With one subtree representing an instance of a tab and its subcomponents
[14:02:04 CDT(-0500)] <Bosmon2> Whereas in "New Kettle", this will represent a "request"
[14:07:05 CDT(-0500)] <colinclark> Anyone else have anything they want to cover?
[14:08:19 CDT(-0500)] <colinclark> Bosmon2: Are you going to go ahead and see if you can work out a URL rebasing scheme for UIO?
[14:08:30 CDT(-0500)] <colinclark> maybe time box it for tomorrow a.m. at the latest?
[14:08:49 CDT(-0500)] <Bosmon2> Ok, sounds reasonable
[14:17:53 CDT(-0500)] <huslage> anastasiac: you around?
[14:18:08 CDT(-0500)] <anastasiac> huslage, yep, I'm here
[14:18:24 CDT(-0500)] <huslage> are you doing anything on the fluid docs mediawiki
[14:18:26 CDT(-0500)] <huslage> ?
[14:18:29 CDT(-0500)] <huslage> is that live?
[14:19:43 CDT(-0500)] <anastasiac> huslage, I haven't been working on it recently - focus has been on the impending Infusion release, but after the release, I hope to go back to working on that setup (mediawiki) as well as the wordpress, mindtouch and drupal instances. They are all test-beds for evaluation, to help choose a system for our docs
[14:19:54 CDT(-0500)] <huslage> ok.
[14:20:01 CDT(-0500)] <huslage> i'm going to turn it off for the moment.
[14:20:09 CDT(-0500)] <huslage> it's making tons of db connections for some reason
[14:20:12 CDT(-0500)] <huslage> and crashing the server
[14:20:15 CDT(-0500)] <anastasiac> ?!
[14:20:18 CDT(-0500)] <anastasiac> oh dear
[14:20:24 CDT(-0500)] <anastasiac> odd
[14:20:31 CDT(-0500)] <huslage> i'll debug it, but to make it easier i'll turn it off first
[14:20:40 CDT(-0500)] <anastasiac> np, thanks for the heads-up, huslage
[14:24:14 CDT(-0500)] <jhung> colinclark: most ocropus components have either a uninstall script or a make clean command (which is essentially a rm on the installed directories).
[14:24:24 CDT(-0500)] <colinclark> great
[14:24:33 CDT(-0500)] <colinclark> should be possible to modify our install script accordingly
[14:24:50 CDT(-0500)] <colinclark> can you make a list, for each dependency, on which command(s) need to be run to remove Ocropus?
[14:24:55 CDT(-0500)] <jhung> the only exception is ocropy. But I hope those are a bit less sensitive to what is already installed
[14:25:09 CDT(-0500)] <huslage> ugh
[14:25:13 CDT(-0500)] <huslage> crashy crashy
[14:25:26 CDT(-0500)] <jhung> I'm currently testing to see if those uninstall scripts are enough for a clean install of decapod colinclark.
[14:25:34 CDT(-0500)] <colinclark> great
[14:29:34 CDT(-0500)] <cindyli> Bosmon2: i was looking at kettle url rebasing code and thinking how to calculate the actual prefix in fat panel
[14:29:48 CDT(-0500)] <cindyli> the integrator will provide us 2 relative urls
[14:30:11 CDT(-0500)] <cindyli> one is from the main html to iframe html, the other is from the main to templates
[14:30:24 CDT(-0500)] <cindyli> both of them could be from anywhere
[14:31:07 CDT(-0500)] <cindyli> for instance, if iframe html is at "../../a/b/iframe.html", templates prefix is "c/"
[14:32:02 CDT(-0500)] <cindyli> first, we need to reverse the relative from main to iframe, which is into - the relative from iframe to main
[14:32:32 CDT(-0500)] <cindyli> "../../a/b/iframe.html" needs to turn into "../../components/uiOptions"
[14:33:03 CDT(-0500)] <cindyli> then, concat it with the 2nd relative path "c/"
[14:33:34 CDT(-0500)] <cindyli> my problem is at the first conversion, how to find out where my main html is located
[14:34:05 CDT(-0500)] <cindyli> in other words, comes up with the part of "....components/uiOptions"
[14:36:13 CDT(-0500)] <cindyli> Bosmon2: any idea?
[14:40:10 CDT(-0500)] <harriswong> colinclark: the fluid.uploader demo mode doesn't actually upload anything right?
[14:40:20 CDT(-0500)] <colinclark> nope
[14:40:44 CDT(-0500)] <colinclark> It just fakes it to show the features of the Uploader without requiring us to worry about where the files should go
[14:40:52 CDT(-0500)] <colinclark> But cindyli's Image Gallery demo is the real deal
[14:41:01 CDT(-0500)] <colinclark> and includes PHP code as an example of how you might like to handle the server side, too
[14:44:19 CDT(-0500)] <harriswong> thanks!
[15:47:25 CDT(-0500)] <azeckoski> hello fluiders
[15:47:35 CDT(-0500)] <azeckoski> is that right? fluiders?
[15:47:46 CDT(-0500)] <azeckoski> fluidators?
[15:48:12 CDT(-0500)] <azeckoski> either way I guess
[15:48:37 CDT(-0500)] <azeckoski> so, I have a boring question about the pager which I hope is easy for an expert
[15:49:14 CDT(-0500)] <azeckoski> let's say some data changes in the items in the pager
[15:49:20 CDT(-0500)] <azeckoski> and I want to cause it to refresh
[15:49:32 CDT(-0500)] <azeckoski> I used to do this:
[15:49:36 CDT(-0500)] <azeckoski> fluid.clear(that.pager.options.dataModel);
[15:49:36 CDT(-0500)] <azeckoski> fluid.model.copyModel(that.pager.options.dataModel, items);
[15:49:36 CDT(-0500)] <azeckoski> var newModel = fluid.copy(that.pager.model);
[15:49:37 CDT(-0500)] <azeckoski> newModel.totalRange = items.length;
[15:49:37 CDT(-0500)] <azeckoski> newModel.sortKey = that.state.sortKey;
[15:49:37 CDT(-0500)] <azeckoski> newModel.sortDir = that.state.sortDir;
[15:49:37 CDT(-0500)] <azeckoski> newModel.pageCount = Math.max(1, Math.floor((newModel.totalRange - 1)/ newModel.pageSize) + 1);
[15:49:38 CDT(-0500)] <azeckoski> fluid.model.copyModel(that.pager.model, newModel);
[15:49:38 CDT(-0500)] <azeckoski> that.pager.events.onModelChange.fire(newModel, that.pager.model, that.pager);
[15:49:52 CDT(-0500)] <azeckoski> that's a mess but the code is here:
[15:50:03 CDT(-0500)] <azeckoski> https://source.sakaiproject.org/svn/msub/unicon.net/kaltura/trunk/src/main/webapp/javascript/KalturaMediaSelector.js
[15:50:13 CDT(-0500)] <azeckoski> around line 160
[15:50:32 CDT(-0500)] <harriswong> fluid-everyone ^
[15:50:37 CDT(-0500)] <azeckoski> but either way, I used to do that but now it seems this does not actually refresh the pager anymore
[15:51:03 CDT(-0500)] <azeckoski> is this a bug in the pager or a bug in me?
[15:51:04 CDT(-0500)] <azeckoski>
[15:51:28 CDT(-0500)] <azeckoski> what's fluid-everyone caret?
[15:51:40 CDT(-0500)] <colinclark> a great way to annoy people
[15:51:45 CDT(-0500)] <azeckoski> ah
[15:51:47 CDT(-0500)] <colinclark> right, harriswong?
[15:52:00 CDT(-0500)] <colinclark> harriswong: ^
[15:52:01 CDT(-0500)] <colinclark> harriswong: ^
[15:52:02 CDT(-0500)] <azeckoski> it pings everyone?
[15:52:03 CDT(-0500)] <colinclark> hey harriswong
[15:52:07 CDT(-0500)] <colinclark> harriswong: hey
[15:52:10 CDT(-0500)] <azeckoski> fun
[15:52:23 CDT(-0500)] <colinclark> azeckoski: Yeah, several of us set up an extra ping
[15:52:27 CDT(-0500)] <colinclark> nice way to get people's attention
[15:52:35 CDT(-0500)] <colinclark> also vaguely annoying
[15:52:36 CDT(-0500)] <colinclark>
[15:52:37 CDT(-0500)] <azeckoski> I will try to avoid that
[15:52:43 CDT(-0500)] <azeckoski> I figure this is asynchronous
[15:52:57 CDT(-0500)] <colinclark> Well, if your leg is on fire and we can help, it might be a useful tactic
[15:53:11 CDT(-0500)] <azeckoski> nah, just my left eyebrow
[15:53:18 CDT(-0500)] <azeckoski> and I can live with that
[15:53:23 CDT(-0500)] <colinclark> You should be able to put an eyebrow out yourself
[15:53:29 CDT(-0500)] <azeckoski> you would think
[15:53:44 CDT(-0500)] <colinclark> What is it that you think is asynchronous?
[15:53:53 CDT(-0500)] <colinclark> your code or IRC?
[15:53:54 CDT(-0500)] <colinclark>
[15:53:55 CDT(-0500)] <azeckoski> this channel of communication
[15:53:58 CDT(-0500)] <colinclark> ah, yes
[15:54:07 CDT(-0500)] <colinclark> Your code looks pretty synchronous to me
[15:54:13 CDT(-0500)] <azeckoski> yeah, it's lame
[15:54:21 CDT(-0500)] <azeckoski> I blame it's mother for that
[15:54:27 CDT(-0500)] <azeckoski> I wanted to raise it all asynchronous
[15:54:46 CDT(-0500)] <azeckoski> but she wanted a balanced approach or some such
[15:55:22 CDT(-0500)] <colinclark> lol
[15:55:32 CDT(-0500)] <azeckoski>
[15:55:50 CDT(-0500)] <colinclark> i'm just looking at your code
[15:55:54 CDT(-0500)] <azeckoski> cool
[15:56:21 CDT(-0500)] <azeckoski> full disclosure that Jen B wrote this code so I am sorta trying to understand it at the moment
[15:56:53 CDT(-0500)] <azeckoski> I think I can say what it is supposed to do but not so much why it was written this way (if that ends up being a question)
[15:57:27 CDT(-0500)] <colinclark> Well, the Pager sometimes requires code to be written in such a way
[15:57:38 CDT(-0500)] <colinclark> it's a little buggy, and athena has become the master of workarounds
[15:57:54 CDT(-0500)] <azeckoski> sweet
[15:58:57 CDT(-0500)] <azeckoski> it IS still in preview/beta I guess
[15:59:33 CDT(-0500)] <colinclark> it is like that, unfortunately
[16:03:57 CDT(-0500)] <colinclark> azeckoski: I don't see anything superficially wrong with your code
[16:04:00 CDT(-0500)] <colinclark> It's insane
[16:04:11 CDT(-0500)] <colinclark> clearly a workaround for limitations in the Pager
[16:04:12 CDT(-0500)] <azeckoski> you mean, except that it is insane?
[16:04:17 CDT(-0500)] <colinclark> well, no
[16:04:19 CDT(-0500)] <azeckoski>
[16:04:34 CDT(-0500)] <colinclark> its insaneness is well within operating parameters
[16:04:38 CDT(-0500)] <azeckoski> nice
[16:04:49 CDT(-0500)] <colinclark> it's just not updating itself?
[16:04:56 CDT(-0500)] <azeckoski> it renders all blanks
[16:05:06 CDT(-0500)] <colinclark> wow
[16:05:10 CDT(-0500)] <azeckoski> it seems to work fine the first time around
[16:05:15 CDT(-0500)] <colinclark> hmm
[16:05:16 CDT(-0500)] <colinclark> fascinating
[16:05:19 CDT(-0500)] <azeckoski> and when we call the refresh method everything vanishes
[16:05:27 CDT(-0500)] <azeckoski> but I can see the data in there
[16:05:37 CDT(-0500)] <azeckoski> so I figured it was something in our hackish reset method
[16:05:46 CDT(-0500)] <azeckoski> but don't know what
[16:05:50 CDT(-0500)] <colinclark> I'll have to look at it a bit more closely, and I'm just heading out the door
[16:05:56 CDT(-0500)] <colinclark> lemme get back to you tomorrow
[16:06:02 CDT(-0500)] <azeckoski> thanks, that would be great
[16:06:04 CDT(-0500)] <azeckoski>
[16:06:50 CDT(-0500)] <colinclark> did it work in 1.1, azeckoski?
[16:06:56 CDT(-0500)] <azeckoski> yes
[16:07:06 CDT(-0500)] <azeckoski> that code there is unchanged from the 1.1 stuff
[16:07:23 CDT(-0500)] <colinclark> ah, it's quite likely that we may have fixed the need for this workaround
[16:07:23 CDT(-0500)] <azeckoski> we changed some other stuff but mostly only superfically
[16:07:29 CDT(-0500)] <azeckoski> oh sweet
[16:07:36 CDT(-0500)] <azeckoski> so maybe I just need to do it the right way
[16:07:37 CDT(-0500)] <azeckoski>
[16:07:53 CDT(-0500)] <colinclark> If you feel like the detective work, you could poke around our JIRA and see what changed in Pager between 1.1 and 1.3.1
[16:07:58 CDT(-0500)] <colinclark> or we can look at it in the morning
[16:08:11 CDT(-0500)] <colinclark> so much for backwards compatibilty
[16:08:23 CDT(-0500)] <azeckoski> meh, that's for losers anyway
[16:08:52 CDT(-0500)] <azeckoski> ping me on AIM or whatever you prefer if I happen to not be in here already
[16:35:51 CDT(-0500)] <athena> hmm, dunno if i want to be seen as the mother of that code
[17:49:16 CDT(-0500)] <azeckoski> :-
[17:49:19 CDT(-0500)] <azeckoski>