fluid-work IRC Logs-2010-06-23

[06:00:27 CDT(-0500)] * justin_o (~jmo@2002:63e1:549a:0:21f:5bff:feca:f969) has joined #fluid-work
[07:37:02 CDT(-0500)] * anastasiac (~team@142.150.154.193) has joined #fluid-work
[08:04:18 CDT(-0500)] * colinclark (~colin@bas2-toronto09-1176132185.dsl.bell.ca) has joined #fluid-work
[08:09:14 CDT(-0500)] * jessm (~Jess@c-71-232-3-151.hsd1.ma.comcast.net) has joined #fluid-work
[08:09:30 CDT(-0500)] <colinclark> Morning, justin_o
[08:09:40 CDT(-0500)] <justin_o> colinclark: hello
[08:09:51 CDT(-0500)] <colinclark> I managed to make some progress last night with pathing issues in the server
[08:10:01 CDT(-0500)] <colinclark> I have some tests that failed initially, and are now passing
[08:10:31 CDT(-0500)] <justin_o> colinclark: that's great...
[08:10:35 CDT(-0500)] <colinclark> In some ways, the code had to take a step backwards first, but the next step will be an improvement, I hope
[08:10:55 CDT(-0500)] <colinclark> If you have a few spare minutes, can you clone decapod-server from my Bitbuck and take it for a spin?
[08:11:21 CDT(-0500)] <colinclark> While I was at it, I took a stab at sorting out the locations where images are captured to
[08:11:41 CDT(-0500)] <colinclark> I ditched the testData directory, and created two directories: mock-images and captured-images
[08:11:53 CDT(-0500)] <colinclark> mock-images is in the repository, containing our sample images
[08:12:08 CDT(-0500)] <colinclark> captured-images is created by the server if it's not present and is used to store, well, captured images.
[08:12:28 CDT(-0500)] <colinclark> The unit tests also clean up after themselves now by deleting everything in captured-images upon tearDown
[08:14:12 CDT(-0500)] <justin_o> colinclark: that all sounds pretty good... i'm firing up linux now to run through the mockserver
[08:14:18 CDT(-0500)] <colinclark> Thanks
[08:14:26 CDT(-0500)] <colinclark> If you look at the code, justin_o, you'll find some fairly ugly stuff in there
[08:14:53 CDT(-0500)] <justin_o> in the mockserver?
[08:14:56 CDT(-0500)] <colinclark> yeah
[08:15:16 CDT(-0500)] <colinclark> The problem was dealing with all kinds of baked-in assumptions about the fact that the web and filesystem paths were the same
[08:15:27 CDT(-0500)] <colinclark> Which is just blatantly incorrect, but not a hard problem to solve
[08:15:40 CDT(-0500)] <colinclark> So as a first step towards green bars, I hacked around it
[08:15:48 CDT(-0500)] <colinclark> My next step will be to refine the ResourceSource so that it handles "virtual relative paths" of the sort we saw in Engage
[08:16:00 CDT(-0500)]

[08:16:26 CDT(-0500)] <colinclark> Those paths can stay the same regardless of whether they ultimately address the file system or a Web URL
[08:16:38 CDT(-0500)] <colinclark> And the ResourceSource can translate them, on demand, to the correct real path
[08:16:49 CDT(-0500)] <colinclark> Does that make any sense, justin_o?
[08:17:03 CDT(-0500)] <colinclark> Hopefully I'm thinking clearly about it
[08:18:36 CDT(-0500)] <justin_o> colinclark: yep i think that makes sense... so the code will never have to change... how will the ResourceSource do the translation?
[08:19:22 CDT(-0500)] * jameswy (~jameswy@pool-173-79-253-178.washdc.fios.verizon.net) has joined #fluid-work
[08:19:27 CDT(-0500)] <colinclark> It already has two central methods
[08:19:32 CDT(-0500)] <colinclark> webURL() and filePath()
[08:20:03 CDT(-0500)] <justin_o> i see
[08:20:13 CDT(-0500)] <colinclark> So, I think a basic implementation will probably tokenize on slashes, parse the first token and replace it with a real value, and then os.path.join() them all back together again
[08:20:34 CDT(-0500)] <colinclark> Code knows when it's trying to access the file system or when it's trying to generate a URL
[08:20:38 CDT(-0500)] <justin_o> that makes sense
[08:20:46 CDT(-0500)] <colinclark> so it can call those methods while passing around paths that are consistent regardless of context
[08:21:17 CDT(-0500)] <colinclark> So the biggest problem I face now is that I've made changes in mockserver.py that I undoubtedly will have to make all over again in the real server
[08:21:31 CDT(-0500)] <colinclark> Which is, as you might imagine, a bit nerve-wracking, since the likelihood of error is pretty high
[08:21:39 CDT(-0500)] <colinclark> I'll probably try to get your Web tests working with the real server
[08:22:42 CDT(-0500)] <justin_o> colinclark: that's a good idea... hopefully it won't be too much of a headache, i think it should just work for the most part
[08:25:20 CDT(-0500)] <colinclark> i think so, too
[08:25:47 CDT(-0500)] * EricDalquist (~dalquist@2607:f388:e:0:221:9bff:fe37:e768) has joined #fluid-work
[08:27:03 CDT(-0500)] * jhung (~Jon@H25.C204.cci.switchworks.net) has joined #fluid-work
[08:27:38 CDT(-0500)] <justin_o> colinclark: hmm... fired up the mockserver and got an error taking the picture.
[08:27:43 CDT(-0500)] <colinclark> awesome
[08:27:46 CDT(-0500)] <colinclark> what's the error?
[08:27:59 CDT(-0500)] <justin_o> well... that was the message that dropped down in the ui
[08:28:08 CDT(-0500)] <justin_o> i'll see if there is an actual error log somewhere
[08:28:15 CDT(-0500)] <colinclark> check the console--there'll be one
[08:28:29 CDT(-0500)] <jhung> justin_o: shut off and restart the cameras.
[08:28:35 CDT(-0500)] <colinclark> mockserver
[08:28:41 CDT(-0500)] <jhung> oh
[08:28:50 CDT(-0500)] <jhung> ignore me.
[08:28:52 CDT(-0500)] <colinclark> never
[08:28:53 CDT(-0500)] <justin_o>
[08:28:59 CDT(-0500)] <justin_o> it was a 500 on the post
[08:29:12 CDT(-0500)] <colinclark> justin_o: But what does Python say?
[08:29:25 CDT(-0500)] <colinclark> jhung: I'm getting close to pushing my server changes to the master
[08:29:30 CDT(-0500)] <colinclark> justin_o is helping doing some testing
[08:29:33 CDT(-0500)] <justin_o> colinclark: yep... reading through it now
[08:30:07 CDT(-0500)] <jhung> colinclark: okay.
[08:30:18 CDT(-0500)] <justin_o> looks like IOError: [Errno 2] No such file or directory: u'/home/justin/Desktop/decapod tests/ decapod/decapod-server/captured-images/Image0-Image1.png'
[08:30:26 CDT(-0500)] <colinclark> aha
[08:30:31 CDT(-0500)] <colinclark> that's very interesting
[08:30:36 CDT(-0500)] <colinclark> is it correct?
[08:30:44 CDT(-0500)] <colinclark> seems like it's a nice path
[08:30:47 CDT(-0500)] <justin_o> i'll take a look
[08:31:27 CDT(-0500)] <justin_o> i guess the stitching didn't work
[08:31:36 CDT(-0500)] <justin_o> because there are just the two jpg files
[08:31:43 CDT(-0500)] <colinclark> ok, interesting
[08:31:56 CDT(-0500)] <colinclark> I guess I'll start by cloning a clean copy and seeing if I can reproduce
[08:32:54 CDT(-0500)] <justin_o> okay.. let me know if you need me to check anything
[08:33:01 CDT(-0500)] <colinclark> thanks, justin_o
[08:36:08 CDT(-0500)] * michelled (~michelled@CPE001310472ade-CM0011aefd3ca8.cpe.net.cable.rogers.com) has joined #fluid-work
[08:36:11 CDT(-0500)] <colinclark> justin_o: one other quick thing
[08:36:21 CDT(-0500)] <colinclark> I had a real head-scratching experience last night
[08:36:24 CDT(-0500)] <colinclark> to put it politely
[08:36:36 CDT(-0500)] <colinclark> Python has these things you should never call a "constructor"
[08:36:47 CDT(-0500)] <colinclark> but which get called immediately after instantiation
[08:36:59 CDT(-0500)] <colinclark> Here's what mine looked like last night:
[08:37:06 CDT(-0500)] <colinclark> def _init__(self, configPath):
[08:37:15 CDT(-0500)] <colinclark> Depending on your font, you might be able to see the error
[08:37:39 CDT(-0500)] <justin_o> the extra _
[08:38:01 CDT(-0500)] <colinclark> you got it
[08:38:18 CDT(-0500)] <colinclark> By default, at least with Eclipse in Ubuntu, the editor's font doesn't show any gaps between underscores
[08:38:24 CDT(-0500)] <colinclark> In fact, almost any font I use doesn't show gaps
[08:38:58 CDT(-0500)] <colinclark> Took me good chunk of time to finally figure out that it wasn't getting called because my method name was slightly heavier on right side than on the left.
[08:39:00 CDT(-0500)] <justin_o> colinclark: i was confused about the number of those when I was reading the book as it was hard or impossible to tell how many there were until they mentioned it.
[08:39:32 CDT(-0500)] <colinclark> It's just lunacy
[08:39:40 CDT(-0500)] <justin_o> i don't suppose that eclipse gave you a warning or anything eh
[08:39:47 CDT(-0500)] <colinclark> If it did, I didn't notice
[08:39:47 CDT(-0500)] <justin_o> i agree
[08:39:58 CDT(-0500)] <colinclark> Just totally crazy
[08:39:59 CDT(-0500)] <colinclark>
[08:40:50 CDT(-0500)] <justin_o> so now you have to watch out for whitespace and underscores
[08:44:52 CDT(-0500)] <colinclark> I don't mind the whitespace so much
[08:45:12 CDT(-0500)] <colinclark> the colons and indenting seem just fine as reasonable visual delineations of blocks
[08:45:21 CDT(-0500)] <colinclark> But underscores are just, well...
[08:45:31 CDT(-0500)] <colinclark>
[08:46:12 CDT(-0500)] <justin_o>
[08:47:02 CDT(-0500)] <jameswy> justin_o: What's the difference between calibration.html and cameraMessage.html?
[08:47:39 CDT(-0500)] * yura_ (~yura@142.150.154.114) has joined #fluid-work
[08:47:48 CDT(-0500)] <justin_o> jameswy: the callibration.html page will be the one where you do the swapping and rotation.
[08:48:01 CDT(-0500)] <jameswy> justin_o: ahh, right. gotcha.
[08:48:05 CDT(-0500)] <justin_o> the other one is where the error and success messages are printed for when the cameras are tested
[08:49:10 CDT(-0500)] <jameswy> justin_o: also, I think there might a minor typo in one of the folders--decapod-ui/components/callibration --I think you meant calibration
[08:49:50 CDT(-0500)] <justin_o> yes... i think i updated that last night
[08:50:04 CDT(-0500)] <justin_o> you may need to update again... i did a few commits yesterday
[08:50:30 CDT(-0500)] <justin_o> also please let me know if there are more typos.. I'm not very good with the spelling
[08:53:56 CDT(-0500)] * clown (~clown@142.150.154.202) has joined #fluid-work
[09:00:48 CDT(-0500)] <colinclark> jameswy: Flashy bold and underlined text, dude
[09:01:16 CDT(-0500)] <colinclark> I keep thinking "Is that a hyperlink?"
[09:06:23 CDT(-0500)] <colinclark> So the Fluid Connect room is experiencing issues this morning
[09:06:29 CDT(-0500)] <colinclark> We may have to do an IRC standup today
[09:07:07 CDT(-0500)] <jameswy> colinclark: argh. it's this client... I had no idea... is this different/better?
[09:07:15 CDT(-0500)] <jameswy> /normal?
[09:07:22 CDT(-0500)] <colinclark> definitely more normal
[09:07:27 CDT(-0500)] <colinclark> I was just teasing you, anyway
[09:07:38 CDT(-0500)] <jameswy> Lol
[09:07:58 CDT(-0500)] <jameswy> The interface to Freenode's Java client is appalling.
[09:11:58 CDT(-0500)] <colinclark> Almost any Java client interface is appalling
[09:12:02 CDT(-0500)] <colinclark> hey jhung
[09:12:04 CDT(-0500)] <michelled> I don't think I can make it to standup today because there is an app layer walkthrough for cspace
[09:12:17 CDT(-0500)] <colinclark> yeah, i'll try to attend that, too
[09:12:29 CDT(-0500)] <jhung> colinclark: yep.
[09:12:33 CDT(-0500)] <colinclark> justin_o and I are in the midst of estimating the last few tasks for the Decapod 0.4 release
[09:12:40 CDT(-0500)] <colinclark> http://wiki.fluidproject.org/display/fluid/Decapod+0.4+Release+Plan
[09:12:55 CDT(-0500)] <colinclark> We are just working on estimates for the first section, Camera Calibration
[09:13:12 CDT(-0500)] <colinclark> Have you managed to find any creative solutions to the problem of uniquely identifying a camera, by chance?
[09:14:12 CDT(-0500)] <jhung> Not really.
[09:14:23 CDT(-0500)] <colinclark> eek
[09:14:25 CDT(-0500)] <jhung> I was thinking that we can make our best attempt and the user can correct.
[09:14:35 CDT(-0500)] <colinclark> How would we make our best attempt?
[09:16:01 CDT(-0500)] <jhung> Perhaps we can get the gphoto2 --show-info for each camera and diff it with subsequent detection / error checks. Sort of finger printing a camera based on that info.
[09:16:33 CDT(-0500)] <jhung> Although there will be two identical camera models, there will be slight variations in camera date/time, # of shots, etc.
[09:17:17 CDT(-0500)] <colinclark> Here's my fear...
[09:17:24 CDT(-0500)] <colinclark> We could write an awful lot of code to do something sort of fuzzy
[09:17:32 CDT(-0500)] <colinclark> Only to find that it isn't particularly reliable in the real world
[09:17:49 CDT(-0500)] <colinclark> I'm thinking we may need to dive deep into this one and see what we can figure out
[09:18:06 CDT(-0500)] <colinclark> Otherwise, it's going to have a real impact on the UX for calibrating cameras, since users will have to do a it a lot more
[09:18:31 CDT(-0500)] <colinclark> I dunno, am I missing something?
[09:18:40 CDT(-0500)] <colinclark> Tell me more about the kinds of information --show-info will give us?
[09:19:10 CDT(-0500)] <colinclark> Actually, jhung, can you pastebin an example of its output for justin_o and I?
[09:19:35 CDT(-0500)] <jhung> colinclark: sure. I'll do that once I fire up my test machine
[09:20:30 CDT(-0500)] <colinclark> thanks
[09:22:29 CDT(-0500)] * jhung1 (~decapod@H25.C204.cci.switchworks.net) has joined #fluid-work
[09:27:26 CDT(-0500)] <jhung-alt> colinclark: sorry, I meant --summary, not --show-info. Here's the summary output for the camera. It's what we saw yesterday. http://pastebin.com/fzeM7iXT
[09:29:52 CDT(-0500)] <colinclark> Ok, as a next step I think I'll have to grab the other G10 and compare the output of them both to see if this is a viable technique
[09:29:59 CDT(-0500)] <colinclark> The risk is that it depends largely on luck
[09:30:01 CDT(-0500)] <colinclark>
[09:31:43 CDT(-0500)] <jhung> I'm thinking perhaps we should revisit the memory card route. The reason we don't use them is because Ubuntu mounts them by default. So we need to unmount in order to use cameras.
[09:32:08 CDT(-0500)] <jhung> But if we can control that in some way and use the memory as a medium to place a decapod config file, then it can be a good solution.
[09:35:49 CDT(-0500)] <colinclark> That might work, yes
[09:38:21 CDT(-0500)] <jhung> justin_o: I can't seem to commit to Master?
[09:38:48 CDT(-0500)] <justin_o> jhung: what error are you getting?
[09:38:53 CDT(-0500)] <jhung> my contributor setting says: "none"
[09:39:48 CDT(-0500)] <jhung> justin_o: "HTTP Error 405: Method Not Allowed"
[09:40:26 CDT(-0500)] <jhung> nevermind. Needed to use https not http.
[09:45:08 CDT(-0500)] <jhung> justin_o, colinclark, jameswy: just pushed some changes up to the master. (finally).
[09:45:14 CDT(-0500)] <colinclark> cool
[09:45:16 CDT(-0500)] <colinclark> congrats
[09:47:20 CDT(-0500)] * zafar (~zafar@2001:638:208:4807:a6ba:dbff:fe02:2266) has joined #fluid-work
[09:48:26 CDT(-0500)] <jhung> colinclark: in the end it was because I was attempting to push changes into a subrepo which was different than the repo used to clone to my local machine.
[09:50:47 CDT(-0500)] <jameswy> jhung: awesome! So, I'm a bit confused. Should I be working off of Google's, justin_o, or your code base?
[09:51:03 CDT(-0500)] <colinclark>
[09:51:21 CDT(-0500)] <jhung> jameswy: all of the above.
[09:51:24 CDT(-0500)] <jhung> lol
[09:51:59 CDT(-0500)] <jhung> jameswy: you work off your own copy and update from other people's repos as needed, and push yours out as needed.
[09:52:14 CDT(-0500)] <justin_o> jameswy: at the moment at least the master and my bitbucket are the same
[09:52:27 CDT(-0500)] <jhung> I have 3 different repos: justin's, mine, and master.
[09:52:29 CDT(-0500)] <zafar> hi jonathon, when should i push the code on googlecode repository, I am thinking this Friday
[09:52:34 CDT(-0500)] <colinclark> Meaning, the master is up to date and good, justin_o?
[09:53:08 CDT(-0500)] <jhung> zafar: you can push as often as you like.
[09:53:16 CDT(-0500)] <justin_o> colinclark: it is up-to-date... not exactly good now
[09:53:21 CDT(-0500)] <zafar> ok great!
[09:53:45 CDT(-0500)] <colinclark> zafar: What code in particular are you thinking of pushing?
[09:55:03 CDT(-0500)] <zafar> the calibration code, documentation, and some test data
[09:55:36 CDT(-0500)] <jhung> zafar: excellent!
[09:57:17 CDT(-0500)] * athena (~athena@c-76-121-97-221.hsd1.wa.comcast.net) has joined #fluid-work
[09:57:45 CDT(-0500)] <jhung> zafar: I assume there will be a Readme on how to use it?
[09:58:42 CDT(-0500)] <colinclark> zafar: Cool. Tell us more. What will it do, exactly? Is this for stereo calibration?
[09:58:55 CDT(-0500)] <zafar> Yes i wrote a readme just today, i have included the dependencies there are, to run the code is actually quite simple just python filename
[09:59:56 CDT(-0500)] <zafar> actually i will push the code for both stereo calibration and independent camera calibration, i will also put the checkerboard pattern there
[10:05:17 CDT(-0500)] <colinclark> ok
[10:09:12 CDT(-0500)] <colinclark> cool
[10:09:51 CDT(-0500)] * kasper (~kasper@189.130.61.33) has joined #fluid-work
[10:12:06 CDT(-0500)] * erinyu (~erinyu@bas15-toronto63-1279474358.dsl.bell.ca) has joined #fluid-work
[10:13:34 CDT(-0500)] * jameswy_ (~jameswy@pool-173-79-253-178.washdc.fios.verizon.net) has joined #fluid-work
[10:31:18 CDT(-0500)] <jhung> looks like breeze is down.
[10:31:50 CDT(-0500)] <jessm> yep, down
[10:32:21 CDT(-0500)] <jessm> shall we do standup in here?
[10:32:45 CDT(-0500)] <jameswy> Sure. Should we just spit it all out at once?
[10:35:40 CDT(-0500)] <erinyu> sure. i've updated the wireframes, and have added a few styling touches to make clear what the information groups and sub-groups are in the schema. have run them by jesse and he is pleased. i'm still working on these two design problems - adding quick links to right side for the full list (but do we need a quick link for each of the procedures? etc), and repeating sub-set of fields within a table.
[10:37:29 CDT(-0500)] <jhung> banging away at Mercurial and figuring out its strange code parenting issues (they need some counselling). Currently investigating a scheme for camera calibration that will use camera's memory card to store Decapod configuration information. This will help improve UX for calibration.
[10:38:20 CDT(-0500)] <jameswy> Getting my hands dirty with markup and styling, breaking things, putting them back together (sometimes), etc. Right now, I'm trying to make the toolbar for the capture interface work (the bar that holds the export button for 0.4/0.5, and which will eventually hold most of the actions in the future releases). Also, got in touch with Karl Fogel--meeting with him this Friday to do a Decapod design walkthrough and get feedback
[10:38:20 CDT(-0500)] <jameswy> designs (jhung, if you're free to join for this, that'd be great!).
[10:39:17 CDT(-0500)] <jhung> jameswy: that sounds great!
[10:41:44 CDT(-0500)] <justin_o> Finished tasking the decapod 0.4 release with Colin.
[10:41:44 CDT(-0500)] <justin_o> http://wiki.fluidproject.org/display/fluid/Decapod+0.4+Release+Plan
[10:41:44 CDT(-0500)] <justin_o> Continued working on the doh.robot stuff this morning. Yesterday I had wrapped the keyboard functions, today I'm working on wrapping the mouse functions. I'm currently making the assumption that you will only move to a place that is currently visible on screen. This is okay if you are moving over an element and can pass in a selector, but means that coordinate based moving will be limited. If this is needed we will have to do some wor
[10:41:45 CDT(-0500)] <justin_o> page scrolling.
[10:41:45 CDT(-0500)] <justin_o> I had a bunch of discussion yesterday with Kasper about the inject.js script. For the time being Jamon has provided a work around which will inject the script tag into the checked out code that testswarm runs against. This should tied us over till we can think of a proper solution.
[10:41:45 CDT(-0500)] <justin_o> Will continue working on camera message component for the calibration ui this afternoon.
[10:42:36 CDT(-0500)] <colinclark> For me, green bars on the server.
[10:42:58 CDT(-0500)] <colinclark> Trying to carefully sort out all the spaghetti related to hard-coded path assumptions in the Decapod server.
[10:43:16 CDT(-0500)] <colinclark> Made a good bit of progress last night, and will continue on it today
[10:44:06 CDT(-0500)] <jessm> i've started diving into the next Decapod quarterly report, but JT thinks we might need something more summative since this report is getting wrapped into the transferring of Decapod to OCAD. Also, herding cats on CFI quotes – it's coming along! The personal computer drama continues as I try to figure out what is ailing my machine.
[10:52:23 CDT(-0500)] <colinclark> So let's do a short dev meeting today as soon as this CSpace app layer walkthrough ends
[11:00:39 CDT(-0500)] <colinclark> Ok, CSpace meeting is done
[11:00:54 CDT(-0500)] <colinclark> With Connect down, we'll have to do something short and sweet for a dev meeting here in IRC
[11:02:19 CDT(-0500)] * kasper (~kasper@189.130.61.33) has joined #fluid-work
[11:03:09 CDT(-0500)] <colinclark> michelled, anastasiac, and yura_: do you have any collectionspace issues you want to chat about?
[11:03:17 CDT(-0500)] <colinclark> I know you guys have an impending deadline
[11:06:19 CDT(-0500)] <michelled> we do have an impending deadline - I guess I should give my stand up report first - I made my first csapce commit yesterday
[11:06:34 CDT(-0500)] <anastasiac> yay!
[11:06:46 CDT(-0500)] <michelled> the repeatable component is looking nice for a tiny component - it could probably move to Infusion as sneak peek soon
[11:07:22 CDT(-0500)] <michelled> now I'm going to look at patches that anastasiac and yura_ have pointed me at - more functionality for repeatable - styling and deleting
[11:07:39 CDT(-0500)] <yura_> michelled: and primary
[11:09:37 CDT(-0500)] <michelled> colinclark: as far as talking about cspace I don't think there is anything I wanted help with at the moment. I do think it would be good to talk a bit about markup generating component patterns.
[11:09:57 CDT(-0500)] <colinclark> Ok
[11:10:16 CDT(-0500)] <colinclark> I was pretty excited to see a component that emerged from CollectionSpace that is almost immediately reusable
[11:10:20 CDT(-0500)] <colinclark> Pretty great
[11:10:35 CDT(-0500)] <colinclark> anastasiac and yura_: quick updates?
[11:10:39 CDT(-0500)] <anastasiac> yesterday, I was: helping michelle and yura_; looking into some of the blocker bugs; making sure there's nothing on the mailing lists I've been neglecting. Today, I'm looking at styling the repeatable stuff, and I want to check in on the status of remaining blocker bugs - see if there's anything waiting on the UI
[11:10:55 CDT(-0500)] <anastasiac> I'm trying to apply repeatable to a table row with two fields
[11:10:59 CDT(-0500)] <colinclark> nice
[11:11:02 CDT(-0500)] <anastasiac> not working yet - don't know why
[11:11:15 CDT(-0500)] <anastasiac> the decorator isn't even being added to the renderer decorator queue
[11:11:22 CDT(-0500)] <anastasiac> so I may have done something wrong
[11:11:28 CDT(-0500)] <anastasiac> still investigating
[11:11:39 CDT(-0500)] <yura_> I was working on delete/primary functionality for the repeatable, wrote lots of tests for that and the patch is on the jira
[11:12:05 CDT(-0500)] <anastasiac> yura_, I can sit with you and you can fill me in on what you've got
[11:12:14 CDT(-0500)] <yura_> totally , anastasiac
[11:13:36 CDT(-0500)] <colinclark> I guess for me, the biggest dev-related issue I've been contemplating is the client/server architecture for Decapod as a whole.
[11:14:00 CDT(-0500)] <colinclark> justin_o and I haven't had any trouble mapping any of the common actions in the application to resource-style REST
[11:14:18 CDT(-0500)] <colinclark> But on the other hand, one does have to ask the question "what does the server give us, now and potentially in the future?"
[11:14:37 CDT(-0500)] <colinclark> And there are some interesting answers--the potential for remote process and PDF generation, etc.
[11:14:42 CDT(-0500)] <colinclark> But nothing really clearly compelling
[11:14:58 CDT(-0500)] <colinclark> I want to sit with jameswy today for a bit and talk about some of these issues from the UX perspective
[11:15:13 CDT(-0500)] <colinclark> But I'm beginning to wonder if we shouldn't be using something like Titanium to build a desktop-like experience using Web tools
[11:15:36 CDT(-0500)] <colinclark> http://www.appcelerator.com/products/titanium-desktop-application-development/
[11:15:58 CDT(-0500)] <colinclark> Dunno if anyone has an immediate thoughts on that issue?
[11:17:37 CDT(-0500)] <colinclark> If not, we can move onto michelled's "markup generating component patterns"
[11:18:16 CDT(-0500)] <michelled> colinclark: that's an interesting idea. it makes sense for decapod which has never very well as a web app
[11:18:31 CDT(-0500)] <michelled> never fit
[11:18:34 CDT(-0500)] <colinclark> Arguably, it doesn't exclude the prospect that some processing operations might occur on a server elsewhere
[11:18:55 CDT(-0500)] * laurel (~Laurel@142.150.154.178) has joined #fluid-work
[11:19:04 CDT(-0500)] <colinclark> In that as long as it's well-factored, the same JavaScript code could live on a remote server using Kettle
[11:19:18 CDT(-0500)] <colinclark> It's a very company-dominated product, which makes me a bit nervous
[11:19:25 CDT(-0500)] <colinclark> Though they do advertise it as Apache-licensed
[11:22:10 CDT(-0500)] <justin_o> colinclark: i'm sure jameswy would like the idea...
[11:22:28 CDT(-0500)] <michelled> seems like it's worth looking into - if you have time for that
[11:22:40 CDT(-0500)] <colinclark> eventually
[11:22:46 CDT(-0500)] <colinclark> I want to get this next release out the door first
[11:22:59 CDT(-0500)] <colinclark> But I do think that it's likely we're going to have to rewrite everything in the server before long
[11:23:06 CDT(-0500)] <colinclark> Luckily it's only about 300 lines of code so far
[11:23:08 CDT(-0500)] <colinclark> but a dense 300 lines
[11:23:10 CDT(-0500)] <colinclark>
[11:23:42 CDT(-0500)] <michelled> I haven't been watching the code base but when I last saw it - it all needed to be rewritten
[11:24:08 CDT(-0500)] <colinclark> So tell us about markup-generating components, michelled
[11:25:10 CDT(-0500)] <michelled> I guess there are a couple patterns that we should follow whenever we are generating markup. perhaps we should encapsulate the pattern in framework but for now it would be good if we all keep them in mind
[11:25:29 CDT(-0500)] <michelled> firstly, we should always check to see if the markup we are interested in is already in the DOM
[11:25:51 CDT(-0500)] <michelled> meaning we'd have selectors for the things we are planning to generate in our options block
[11:26:24 CDT(-0500)] <michelled> if the thing isn't in the DOM we go ahead and generate the markup and put the default selector class on the node
[11:27:02 CDT(-0500)] <michelled> this forces a contract with the user of the component - if they don't provide the markup then they can't override the default selector for that 'thing'
[11:27:14 CDT(-0500)] <colinclark> Yep, which makes sense, I think
[11:27:28 CDT(-0500)] <colinclark> If you don't even have the markup, why do you care what selector it will be given?
[11:27:43 CDT(-0500)] <michelled> yes, you shouldn't
[11:28:41 CDT(-0500)] <michelled> I also think we should provide a means to override the default markup generation. I'm not so certain how we should do that. for example, in the repeatable component I'm generating markup for 4 things: add, remove, primary and content container
[11:28:59 CDT(-0500)] <michelled> should I provide four function names that could be overridden?
[11:29:14 CDT(-0500)] <michelled> or should I put my basic markup in my options block?
[11:29:34 CDT(-0500)] <michelled> or should I provide a single function name that could be overridden?
[11:29:47 CDT(-0500)] <michelled> I'm currently doing the third but I'm not convinced it is the best way
[11:29:54 CDT(-0500)] <colinclark> In the past, we've typically put markup generation code into a single function
[11:30:29 CDT(-0500)] <colinclark> I think it's probably simplest for users who want something completely different, but might be a bit annoying for someone who wants to subtly refine the existing implementation
[11:30:59 CDT(-0500)] <michelled> ya - which was why I was thinking of putting the markup string in the defaults.
[11:31:15 CDT(-0500)] <michelled> I guess I could do both 1 and 3
[11:31:22 CDT(-0500)] <michelled> or 2 and 3
[11:32:15 CDT(-0500)] <michelled> wow - I've really managed to confuse this - I should have put numbers beside the points above. what I'm trying to say is that I could put the markup strings in to the defaults and then either have 1 or 4 function names in the default as well
[11:33:15 CDT(-0500)] <colinclark> michelled: I suspect that you'll find putting markup into the defaults is inconvenient
[11:33:30 CDT(-0500)] <colinclark> Because in many case you'll be writing code to generate it
[11:33:40 CDT(-0500)] <colinclark> And you'd have to fall back to old-style string templating or something
[11:34:04 CDT(-0500)] <colinclark> I don't think there's anything terribly wrong about providing a function pointer that the user can override, or several if it's appropriate
[11:34:37 CDT(-0500)] <michelled> ya, you're probably right. it's only because repeatable is so simple at this point that it would work
[11:35:20 CDT(-0500)] <michelled> colinclark: do you think it would be better to provide a single function or a function for each thing my creating that is specified in the selectors
[11:36:00 CDT(-0500)] <colinclark> lemme check the code
[11:38:32 CDT(-0500)] <colinclark> I might be tempted to split the two different functions out
[11:38:44 CDT(-0500)] <colinclark> One for wrapping and one for the chrome
[11:38:51 CDT(-0500)] <colinclark> But the only way to really know is to try it
[11:39:01 CDT(-0500)] <colinclark> Have someone customize the markup, like we did with InlineEdit
[11:39:05 CDT(-0500)] <colinclark> It was enlightening
[11:39:27 CDT(-0500)] <michelled> hence the sneak peek designation
[11:39:35 CDT(-0500)] <colinclark> yep
[11:40:03 CDT(-0500)] <michelled> I can't think of other general things to say about markup generating components
[11:40:07 CDT(-0500)] <michelled> anyone else?
[11:40:24 CDT(-0500)] <anastasiac> michelled, what you've laid out sound like reasonable guidelines
[11:44:21 CDT(-0500)] <michelled> anything else anyone wants to talk about?
[11:45:45 CDT(-0500)] <colinclark> I think I'm cool
[11:46:00 CDT(-0500)] <anastasiac> FYI, that problem I was having applying the makeRepeatable to a table row seems to have been "user error
[11:46:11 CDT(-0500)] <colinclark> If anyone is curious about Decapod, justin_o and I had a chat about things this morning, you can check the logs
[11:46:27 CDT(-0500)] <colinclark> And our tentative plan for the next release, 0.4, is here: http://wiki.fluidproject.org/display/fluid/Decapod+0.4+Release+Plan
[11:46:39 CDT(-0500)] <anastasiac> now I'm butting my head up again against what I am becoming convinced is a bug in getBeanValue, so I'm darn-well going to write a test, and fix it!
[11:48:22 CDT(-0500)] <colinclark> What's the suspected bug?
[11:49:42 CDT(-0500)] <anastasiac> well, I suspect that actually writing the test case will clarify it in my own mind (translation: I'm not entirely sure yet), but it thows an error in certain cases when the model being asked for doesn't exist
[11:49:47 CDT(-0500)] <anastasiac> at the very least, it should be more graceful
[11:50:03 CDT(-0500)] <anastasiac> it has to do with arrays in the model
[11:52:29 CDT(-0500)] <colinclark> ok
[12:24:51 CDT(-0500)] * kasper (~kasper@189.130.61.33) has joined #fluid-work
[12:37:32 CDT(-0500)] * elicochran (~elicochra@adsl-70-137-148-4.dsl.snfc21.sbcglobal.net) has joined #fluid-work
[12:40:46 CDT(-0500)] * justin_o (~jmo@CPE001b63f2cc0e-CM0011aec4b062.cpe.net.cable.rogers.com) has joined #fluid-work
[12:44:43 CDT(-0500)] <michelled> oh my goodness - the earth just moved!
[12:45:41 CDT(-0500)] * jhung (~Jon@H25.C204.cci.switchworks.net) has joined #fluid-work
[12:46:22 CDT(-0500)] <yura_> yes
[12:46:24 CDT(-0500)] <jameswy> michelled: twitter is abuzz with it... earthquake in Toronto!
[12:46:41 CDT(-0500)] <jhung> I just felt it.
[12:46:43 CDT(-0500)] <jhung> Crazy.
[12:46:48 CDT(-0500)] <jameswy> I'm missing out
[12:47:19 CDT(-0500)] <jhung> I thought it was washing machine going crazy again, but realized all my laundry is done.
[12:49:22 CDT(-0500)] <michelled> I'm glad nothing fell - I remember Eli being freaked out by my stacking of glasses on top of cabinets
[12:51:09 CDT(-0500)] <michelled> even heidi felt it - that's a pretty big range
[12:54:12 CDT(-0500)] <justin_o> that's so funny, my girlfriend just called me about it because she felt it, but i didn't feel anything.
[12:56:19 CDT(-0500)] <jameswy> Haha, justin_o
[12:56:26 CDT(-0500)] <jameswy> justin_o: how high up are you in your condo?
[12:56:34 CDT(-0500)] <justin_o> only the 5th floor
[12:57:44 CDT(-0500)] <jameswy> @GlobalNational is reporting that the earthquake was felt as far as Rochester, NY.
[12:58:06 CDT(-0500)] <jhung> Ottawa, Montreal, Vermont...
[12:59:21 CDT(-0500)] * kasper (~kasper@189.130.61.33) has joined #fluid-work
[13:01:52 CDT(-0500)] <jameswy> 5.5
[13:03:20 CDT(-0500)] <michelled> elicochran: is that fairly common for your neck of the woods?
[13:04:10 CDT(-0500)] <elicochran> very
[13:10:48 CDT(-0500)] * kasper (~kasper@189.130.61.33) has joined #fluid-work
[13:19:29 CDT(-0500)] <elicochran> although not as often as the media makes it out
[13:23:33 CDT(-0500)] <elicochran> http://earthquake.usgs.gov/earthquakes/recenteqsww/Quakes/us2010xwa7.php#details
[13:33:55 CDT(-0500)] <jameswy> jhung: we'll be meeting with Karl at 11 AM EDT on Friday.
[13:34:26 CDT(-0500)] <jhung> ok thanks!
[13:39:21 CDT(-0500)] * laurel (~Laurel@142.150.154.178) has left #fluid-work
[13:40:09 CDT(-0500)] * colinclark (~colin@bas2-toronto09-1176132185.dsl.bell.ca) has joined #fluid-work
[13:40:16 CDT(-0500)] <colinclark> wowza
[13:40:23 CDT(-0500)] <colinclark> I went to go get the second Decapod camera
[13:40:25 CDT(-0500)] <michelled> thanks for the link elicochran - interesting
[13:40:33 CDT(-0500)] <colinclark> And the whole library was closed due to the Earthquake
[13:40:44 CDT(-0500)] <michelled> did you manage to get into the building?
[13:40:53 CDT(-0500)] <colinclark> Nope, fire trucks, cops everything
[13:41:04 CDT(-0500)] <colinclark> They evacuated the whole building because Bissel "wobbled"
[13:41:27 CDT(-0500)] <colinclark> So I guess I'm out of luck for testing the real server this week, but I think I can get by
[13:41:33 CDT(-0500)] <colinclark> Did anyone else feel the earthquake?
[13:41:35 CDT(-0500)] <michelled> anastasiac: did they let you back in?
[13:41:58 CDT(-0500)] <anastasiac> michelled, yes, we're back.
[13:42:00 CDT(-0500)] <michelled> colinclark: yep - we felt it
[13:42:14 CDT(-0500)] <colinclark> I guess I should have stuck around, then
[13:42:16 CDT(-0500)] <colinclark> oh well
[13:42:17 CDT(-0500)] <michelled> I think you just missed the link eli put into the channel
[13:42:17 CDT(-0500)] <michelled> http://earthquake.usgs.gov/earthquakes/recenteqsww/Quakes/us2010xwa7.php#details
[13:43:19 CDT(-0500)] <colinclark> so cool
[13:43:31 CDT(-0500)] <colinclark> my rack of glasses also made elicochran nervous
[13:43:33 CDT(-0500)] <colinclark> but the seem unmoved
[13:44:32 CDT(-0500)] <colinclark> so jameswy, are you up for chatting web vs. native in Decapod?
[13:44:39 CDT(-0500)] <colinclark> everyone else is welcome, too
[13:44:49 CDT(-0500)] <jameswy> colinclark: Yep, definitely.
[13:45:16 CDT(-0500)] <jhung> jameswy, colinclark: can I listen in?
[13:45:25 CDT(-0500)] <colinclark> of course!
[13:45:43 CDT(-0500)] <justin_o> colinclark: where are you guys chatting?
[13:45:49 CDT(-0500)] <colinclark> i guess skype
[13:45:59 CDT(-0500)] <colinclark> darcie's just using the blender, i'll call when she's done
[13:45:59 CDT(-0500)] <justin_o> mind if i listen in too
[13:46:04 CDT(-0500)] <colinclark> ah, working from home
[13:46:10 CDT(-0500)] <colinclark> justin_o: for sure
[13:46:12 CDT(-0500)] <colinclark> everyone can join!
[13:47:48 CDT(-0500)] <justin_o> sounds like a party
[13:48:04 CDT(-0500)] <jhung> With Decapod, it always is.
[13:48:16 CDT(-0500)] <justin_o>
[14:04:15 CDT(-0500)] <anastasiac> colinclark and/or Bosmon2: moment for a quick question about EL paths?
[14:04:27 CDT(-0500)] <colinclark> I'm on a call, but can be available a bit later
[14:05:08 CDT(-0500)] <anastasiac> if an EL path has an integer in it (e.g. "foo.0.bat"), can it be assumed that the integer refers to an index into an array, or might it be possible that someone actually wants a property with the name "0"? (asking anyway, in case Bosmon2 is around)
[14:06:03 CDT(-0500)] <Bosmon2> anastasiac - you cannot "firmly" determine that the path refers to a number
[14:06:22 CDT(-0500)] <Bosmon2> This sounds like the beginnings of some scheme to try to avoid having to construct a schema
[14:06:30 CDT(-0500)] <Bosmon2> All I can remark is "that it will end in tears"
[14:07:05 CDT(-0500)] <anastasiac> Bosmon2, so what if a renderer autobinding EL path has "foo.0.bat" but the model is empty?
[14:20:23 CDT(-0500)] <athena> crazy about your earthquake
[14:20:48 CDT(-0500)] <athena> i get nervous every time i get stuck on the drawbridge out of my neighborhood
[14:22:57 CDT(-0500)] <Bosmon2> anastasiac: at present, it will create an object, {}
[14:23:24 CDT(-0500)] <anastasiac> Bosmon2, yes, but what if I want an array?
[14:23:41 CDT(-0500)] <Bosmon2> anastasiac: Then you will need to make one yourself
[14:23:42 CDT(-0500)] <anastasiac> it will create an object with a property called "0"
[14:23:47 CDT(-0500)] <anastasiac> :-P
[14:44:42 CDT(-0500)] * jhung-alt (~decapod@H25.C204.cci.switchworks.net) has left #fluid-work
[15:19:13 CDT(-0500)] <colinclark> so, anastasiac, that all trailed off with a lot of tongue sticking
[15:19:21 CDT(-0500)] <colinclark> But did you get any good answers to your questoins
[15:19:32 CDT(-0500)] <colinclark> or need any help in terms of next steps?
[15:22:13 CDT(-0500)] * justin_o (~jmo@CPE001b63f2cc0e-CM0011aec4b062.cpe.net.cable.rogers.com) has joined #fluid-work
[15:28:02 CDT(-0500)] <colinclark> justin_o: I can't reproduce the problem you had earlier today with the mockserver changes I made
[15:28:20 CDT(-0500)] <colinclark> I checked out a fresh copy of decapod-default, and the replaced server with a clone of my bitbucket repo
[15:28:25 CDT(-0500)] <colinclark> Seemed to take some decent pictures
[15:29:01 CDT(-0500)] <colinclark> Any more advice on how I can recreate the problem?
[15:30:26 CDT(-0500)] <justin_o> colinclark: am i still supposed to start the mockserver by doing "python mockserver.py"?
[15:31:03 CDT(-0500)] <colinclark> yep
[15:31:28 CDT(-0500)] <justin_o> hmm... not sure
[15:31:39 CDT(-0500)] <colinclark> Did you do the same thing I did?
[15:31:40 CDT(-0500)] <justin_o> did you update any of the dependencies (libraries)
[15:31:42 CDT(-0500)] <justin_o> i did
[15:31:45 CDT(-0500)] <colinclark> hmm
[15:31:46 CDT(-0500)] <colinclark> odd
[15:31:50 CDT(-0500)] <colinclark> I don't think I updated anything
[15:31:58 CDT(-0500)] <colinclark> should we screenshare for a minute?
[15:32:03 CDT(-0500)] <colinclark> Can you even screenshare on Linux?
[15:32:04 CDT(-0500)] <anastasiac> colinclark: Bosmon's did give me an answer
[15:32:17 CDT(-0500)] <anastasiac> I might still post a patch to the list and ask for comment
[15:32:20 CDT(-0500)] <anastasiac> but i have to head out now
[15:32:23 CDT(-0500)] <colinclark> ok
[15:32:49 CDT(-0500)] <justin_o> colinclark: not sure
[15:32:56 CDT(-0500)] <justin_o> i'll try to install skype there and see if it work
[15:32:58 CDT(-0500)] <justin_o> works
[15:33:09 CDT(-0500)] <colinclark> we can worry about it later if you're in the midst of coding
[15:33:34 CDT(-0500)] <justin_o> i think i can spare a few minutes
[15:37:00 CDT(-0500)] * anastasiac (~team@142.150.154.193) has left #fluid-work
[16:45:58 CDT(-0500)] * jameswy (~jameswy@pool-173-79-253-178.washdc.fios.verizon.net) has joined #fluid-work
[17:07:40 CDT(-0500)] * justin_o (~jmo@2002:63e1:549a:0:21f:5bff:feca:f969) has joined #fluid-work
[19:45:48 CDT(-0500)] * jhung (~Jon@H25.C204.cci.switchworks.net) has left #fluid-work
[21:06:33 CDT(-0500)] * kasper (~kasper@189.130.61.33) has joined #fluid-work
[22:16:29 CDT(-0500)] * elicochran (~elicochra@adsl-70-137-148-4.dsl.snfc21.sbcglobal.net) has joined #fluid-work