fluid-work IRC Logs-2011-05-24

Owned by Fluid IRC Log Bot
Last updated: May 24, 2011Version comment
16 min read

[08:02:09 CDT(-0500)] <pulkit> hi michelled
[08:02:20 CDT(-0500)] <michelled> hi pulkit
[08:02:44 CDT(-0500)] <michelled> how's it going?
[08:03:32 CDT(-0500)] <pulkit> its been good. I have started building the first prototype for the component
[08:03:55 CDT(-0500)] <michelled> that's great
[08:04:12 CDT(-0500)] <michelled> have you put it on github/
[08:04:13 CDT(-0500)] <michelled> ?
[08:04:19 CDT(-0500)] <pulkit> I took a little help from the ImageGallery demo for Uploader
[08:04:26 CDT(-0500)] <pulkit> not yet
[08:05:14 CDT(-0500)] <pulkit> Should the ImageEditor component have an integrated uploader?
[08:05:51 CDT(-0500)] <michelled> I would say that the image editor should be concerned only with editing
[08:06:02 CDT(-0500)] <michelled> there may be another component that combines uploading and editing
[08:06:11 CDT(-0500)] <pulkit> ok.
[08:06:55 CDT(-0500)] <michelled> and as you build the image editor I would think that you'll actually end up with lots of little components that will be combined to make up the image editor
[08:06:57 CDT(-0500)] <michelled> does that make sense?
[08:07:36 CDT(-0500)] <pulkit> yes, i think so.
[08:08:33 CDT(-0500)] <pulkit> Yesterday I was thinking of making a menu bar that contains image editing options. That can be one component that can be added to fluid
[08:09:39 CDT(-0500)] <michelled> sounds like
[08:09:41 CDT(-0500)] <michelled> it
[08:09:56 CDT(-0500)] <michelled> pulkit: if you could get your code somewhere public then other people can take a look at it
[08:10:07 CDT(-0500)] <michelled> maybe we can get some design resources on the project too
[08:10:19 CDT(-0500)] <michelled> I'll have a quick chat with James when he gets in
[08:10:28 CDT(-0500)] <pulkit> ok
[08:10:50 CDT(-0500)] <michelled> I think it makes sense for you to work in github if you're comfortable with that
[08:11:25 CDT(-0500)] <pulkit> yes, that would be fine
[08:11:40 CDT(-0500)] <michelled> it's so exciting that you're getting working on this (smile)
[08:12:05 CDT(-0500)] <pulkit> (smile)
[08:15:18 CDT(-0500)] <pulkit> I think its a silly question to ask, but which IDE do you generally use for development on fluid?
[08:20:40 CDT(-0500)] <michelled> pulkit: we use different things. I use eclipse
[08:20:56 CDT(-0500)] <michelled> Justin_o, yura_: what do you use?
[08:21:09 CDT(-0500)] <yura_> eclipse for me
[08:21:19 CDT(-0500)] <Justin_o> i'm mostly using textmate at the moment
[08:21:34 CDT(-0500)] <michelled> cindyli: do you use eclipse too?
[08:21:45 CDT(-0500)] <cindyli> yes, michelled
[08:22:04 CDT(-0500)] <cindyli> michelled: one sec, i use aptana for js, eclipse for php
[08:22:20 CDT(-0500)] <pulkit> ok. I also use eclipse with aptana plugin
[08:25:31 CDT(-0500)] <heidi> hey jameswy ! so cindyli added the control templates in, do you want to update FullNoPreview to use them?
[08:25:49 CDT(-0500)] <heidi> Full with preview is already, and i'm doing fat panel now
[08:27:06 CDT(-0500)] <heidi> Justin_o are you in tomorrow? maybe we can do our pairing work then? finishing up sliding panel tests and wrapper for JQ UI tabs
[08:30:41 CDT(-0500)] <Justin_o> heidi: sure sounds good
[08:33:22 CDT(-0500)] <heidi> cool
[08:39:05 CDT(-0500)] <jameswy> heidi: Sures, I can try to make time for it today or tomorrow. How are the tweaks coming along for fat panel?
[08:39:34 CDT(-0500)] <heidi> jameswy gonna finish up and test on different browsers today! we can look it over tomorrow when i'm in?
[08:40:32 CDT(-0500)] <jameswy> heidi: Yes, let's do that.
[08:40:37 CDT(-0500)] <heidi> cool
[08:54:59 CDT(-0500)] <heidi> Justin_o hey justin, not sure if you're still working on the live preview, but UIO shouldn't be included in the update, just the content below it.
[08:55:11 CDT(-0500)] <heidi> the changes get applied to UIO once the panel is closed and reopened
[08:55:52 CDT(-0500)] <Justin_o> heidi: actually i've put it aside for a bit.. waiting for the ui enhancer stuff to finish.. i figure that's more a ui enhancer issue, but could be wrong
[08:56:05 CDT(-0500)] <Justin_o> michelled: what do you think? ^
[08:56:22 CDT(-0500)] <heidi> what enhancer stuff?
[08:58:48 CDT(-0500)] <michelled> Justin_o: the more challenging part will be applying styles to the whole page except for UIO
[08:59:04 CDT(-0500)] <michelled> that's why we made the preview in an iframe
[08:59:50 CDT(-0500)] <michelled> Justin_o: I should probably have done a pull request by now but there are several little things I wanted to fix up first
[09:00:04 CDT(-0500)] <michelled> I'll try to get that done now and do a pull request this afternoon
[09:00:11 CDT(-0500)] <Justin_o> michelled: thanks
[09:07:14 CDT(-0500)] <jhung> jessm: do you have time to chat this morn?
[09:07:27 CDT(-0500)] <jessm> jhung: sure – 10:30?
[09:07:41 CDT(-0500)] <jhung> jessm: thinking the exact same thing.
[09:07:47 CDT(-0500)] <jhung> jessm: ping you then.
[09:34:12 CDT(-0500)] <jhung> jessm: ping
[09:34:46 CDT(-0500)] <jessm> jhung: hang on a sec...
[09:46:30 CDT(-0500)] <mlam> Justin_o: is there anything you want me to look at outside of the uploader?
[09:49:48 CDT(-0500)] <Justin_o> mlam: did you want to take a look at the bug parade report and see if anything interests you?
[09:50:00 CDT(-0500)] <mlam> sure , I'll take a look
[09:50:13 CDT(-0500)] <Justin_o> mlam: http://issues.fluidproject.org/secure/ConfigureReport.jspa?atl_token=hLyns-VDmh&amp;filterid=10128&amp;mapper=components&amp;selectedProjectId=10001&amp;reportKey=com.atlassian.jira.plugin.system.reports%3Asinglelevelgroupby&amp;Next=Next
[09:55:38 CDT(-0500)] <Justin_o> mlam: if you're still having trouble finding something.. there's a ton of ui options work left..
[09:56:03 CDT(-0500)] <Justin_o> if you want to do something for that, we can coordinate with heidi, cindyli and michelled
[09:56:09 CDT(-0500)] <mlam> ok, cool. I'm just working my way down the list
[09:56:29 CDT(-0500)] <heidi> would love yr help mlam!
[09:57:08 CDT(-0500)] <mlam> Justin_o: heidi. Ok, what can I do to help with UI options?
[09:58:05 CDT(-0500)] <heidi> better to ask cindyli where she's at. one thing that has to happen is to add an option to auto-save when a control is changed (vs. clicking save button, which fat panel doesn't have)
[09:58:24 CDT(-0500)] <heidi> but anything from that list is prob fine!
[10:00:45 CDT(-0500)] <cindyli> mlam: the jira for auto save is 4221, whatever u will pick from that list is fine
[10:02:17 CDT(-0500)] <cindyli> mlam: we also have a wiki road map you may want to have a look and keep it updated besides the jira update: http://wiki.fluidproject.org/display/fluid/UI+Options+1.4+Roadmap. thx a lot, mlam, welcome aboard. (smile)
[10:05:28 CDT(-0500)] <heidi> mlam a lot of UIO options issues depend on one another... so you might have to do some merging to get started on the issue you pick, make sure you have all the updated bits that aren't yet in trunk. cindyli can prob help you with that too
[10:06:00 CDT(-0500)] <cindyli> anytime
[10:06:21 CDT(-0500)] <mlam> ok, cool
[10:07:37 CDT(-0500)] <cindyli> Justin_o: r u working on FLUID-4209 - Table of contents: Replace rsf:ids in template of the table of contents component with flc classes
[10:08:24 CDT(-0500)] <cindyli> wanna update the roadmap
[10:09:58 CDT(-0500)] <Justin_o> cindyli: yes
[10:10:03 CDT(-0500)] <cindyli> thx
[10:23:57 CDT(-0500)] <mlam> cindyli: maybe we can talk after stand up to see where i can help out?
[10:24:10 CDT(-0500)] <cindyli> sure. mlam. thx
[11:09:06 CDT(-0500)] <michelled> Justin_o: is it known that the project repo master has broken integration examples for UIOptions and UIEnhancer?
[11:09:17 CDT(-0500)] <michelled> both the sakai mockup and the uportal mockup
[11:10:21 CDT(-0500)] <Justin_o> michelled: http://issues.fluidproject.org/browse/FLUID-4239
[11:10:24 CDT(-0500)] <Justin_o> that's for ui options
[11:10:30 CDT(-0500)] <Justin_o> and it's on bug parade,
[11:10:36 CDT(-0500)] <Justin_o> don't know about UI Enhancer though
[11:10:56 CDT(-0500)] <michelled> the uportal mockup has 4 links which use UI enhancer
[11:11:00 CDT(-0500)] <michelled> they are now throwing an error
[11:11:24 CDT(-0500)] <michelled> Justin_o: they are in the 'customize my portal' list
[11:11:43 CDT(-0500)] <michelled> thanks for the info on the other JIRA (smile)
[11:12:04 CDT(-0500)] <Justin_o> michelled: thanks.. will you file a jira for that one.. and we can add it to bug parade as well..
[11:12:18 CDT(-0500)] <Justin_o> i'm hoping it's just a missing dependency or something else that is simple
[11:12:41 CDT(-0500)] <michelled> I think it's probably an API change
[11:12:58 CDT(-0500)] <michelled> yes, I'll file a JIRA after this meeting
[11:17:13 CDT(-0500)] <heidi> Justin_o any guesses why the HC and HCI themes change slider margins ? i can't imagine why...
[11:18:21 CDT(-0500)] <Justin_o> heidi: no idea... maybe because they remove background images..so the text equivalents may have change something
[11:18:56 CDT(-0500)] <heidi> Justin_o terrible headache for UIO sliders. hmm
[11:20:21 CDT(-0500)] <Justin_o> heidi: sorry to hear that..
[11:20:24 CDT(-0500)] <heidi> Justin_o but yeah i think you're right, it gets rid of the slider img and replaces with blob that has h/w/bg colour
[11:20:32 CDT(-0500)] <heidi> thanks!
[11:21:10 CDT(-0500)] <Justin_o> heidi: np .. i think jameswy has taken the stance that background images should be allowed but should be appropriate
[11:21:35 CDT(-0500)] <heidi> Justin_o yeah i'm not really sure why this would be more helpful than an img... hm
[11:21:50 CDT(-0500)] <heidi> easier to see/control colour of it i guess?
[11:22:16 CDT(-0500)] <heidi> which is prob important and worth keeping
[11:24:55 CDT(-0500)] <Justin_o> heidi: but you could just make your jquery ui theme with the correct colour.. harder to overide though
[11:25:39 CDT(-0500)] <michelled> Justin_o: I have a process question - if I want code review and I'm fairly sure my branch isn't ready to be integrated do I still do a pull request?
[11:25:44 CDT(-0500)] <heidi> jameswy what do you think? specific bg images for slider images or stick with coloured blob
[11:25:50 CDT(-0500)] <heidi> for diff themes
[11:27:15 CDT(-0500)] <Justin_o> michelled: good question..
[11:27:27 CDT(-0500)] <Justin_o> michelled: i'm not sure.. i guess you could just request it directly
[11:27:28 CDT(-0500)] <heidi> Justin_o it seems the HC/HCI themes have a lot of jquery over-rides in them, instead of using the jquery theme roller styles. thoughts?
[11:29:59 CDT(-0500)] <michelled> thx Justin_o - the new JIRA for UI Enhancer is 4262
[11:52:14 CDT(-0500)] <Justin_o> heidi: was that just for overriding the images
[11:52:17 CDT(-0500)] <Justin_o> michelled: thanks
[11:52:38 CDT(-0500)] <michelled> np
[11:53:21 CDT(-0500)] <heidi> Justin_o i think maybe they were created pre-using theme roller themes? they have things like .fl-theme-hc .ui-tabs .ui-tabs-nav for jquery ui tabs?
[11:56:30 CDT(-0500)] <michelled> cindyli: you have a pull request in for 4210 - this is on top of what was already merged, right?
[11:56:45 CDT(-0500)] <cindyli> yes, michelled
[11:56:53 CDT(-0500)] <michelled> ok, I'll reopen the JIRA
[12:00:49 CDT(-0500)] <heidi> Justin_o ignore what i just said. these styles ARE from theme roller.... duhh
[12:08:56 CDT(-0500)] <Justin_o> heidi: (smile) okay.. it's been forgotten
[12:09:05 CDT(-0500)] <heidi> hehe
[13:46:58 CDT(-0500)] <heidi> jameswy tomorrow we should talk about what happens to the layouts when text size/line-spacing/etc is changed
[13:47:49 CDT(-0500)] <heidi> the fat panel gets messy when things get big... so might have to add overflow scrollbars, or figure out clean wrapping, etc
[13:48:35 CDT(-0500)] <jameswy> heidi: Sure thing. In a nutshell though, for vertical issues, the panel should resize.
[13:48:56 CDT(-0500)] <jameswy> And for horizontal, segments should wrap.
[13:49:13 CDT(-0500)] <heidi> jameswy not sure what you mean
[13:49:52 CDT(-0500)] <jameswy> heidi: An increase in line spacing should force the panel to resize to be longer.
[13:51:42 CDT(-0500)] <heidi> jameswy and the tabs?
[13:54:54 CDT(-0500)] <jameswy> What about the tabs?
[13:56:50 CDT(-0500)] <heidi> jameswy do they get longer?
[13:57:13 CDT(-0500)] <jameswy> heidi Do they get longer vertically, you mean?
[13:57:21 CDT(-0500)] <heidi> jameswy right
[13:58:14 CDT(-0500)] <jameswy> heidi No--they're sectioned off from the rest of the "content"
[13:58:35 CDT(-0500)] <heidi> jameswy the "content" being the tab panel content?
[13:58:49 CDT(-0500)] <jameswy> heidi: Correct.
[13:59:16 CDT(-0500)] <heidi> okay... sectioned off in that.... what besides no line-height?
[13:59:33 CDT(-0500)] <colinclark> mlam: got a sec?
[13:59:36 CDT(-0500)] <jameswy> Sorry, what do you mean?
[13:59:47 CDT(-0500)] <mlam> colinclark: ya, for sure
[13:59:50 CDT(-0500)] <colinclark> I'm trying to think through your FLUID-4256 pull request
[13:59:57 CDT(-0500)] <colinclark> it works
[14:00:08 CDT(-0500)] <colinclark> had a bit of commented-out code in it that I cleaned up
[14:00:15 CDT(-0500)] <colinclark> but otherwise, it's functional
[14:00:25 CDT(-0500)] <colinclark> it has a bit of a curious design quality to it
[14:00:39 CDT(-0500)] <athena> curiouser and curiouser?
[14:00:51 CDT(-0500)] <colinclark> I imagine you were hanging out with yura_ while working on it (wink)
[14:00:53 CDT(-0500)] <colinclark> athena: (smile)
[14:00:55 CDT(-0500)] <colinclark> How's the conference?
[14:00:57 CDT(-0500)] <mlam> hahaha yah, i was (smile)
[14:00:59 CDT(-0500)] <athena> terrific
[14:01:00 CDT(-0500)] <colinclark> I really wish I was there
[14:01:09 CDT(-0500)] <colinclark> I wore my Jasig t-shirt this weekend
[14:01:09 CDT(-0500)] <athena> i'm about to present my 8th hour of material
[14:01:13 CDT(-0500)] <athena> lol nice
[14:01:18 CDT(-0500)] <colinclark> athena: You're always the star of the show
[14:01:19 CDT(-0500)] <athena> we wish you were here too
[14:01:27 CDT(-0500)] <colinclark> uMobile is taking over the world, according to Twitter
[14:01:30 CDT(-0500)] <athena> matt was complaining about the lack of UX representation here this year
[14:01:39 CDT(-0500)] <athena> you download the conference app? (smile)
[14:01:45 CDT(-0500)] <athena> appears to still be running
[14:01:50 CDT(-0500)] <athena> really glad that server is happy
[14:01:51 CDT(-0500)] <colinclark> I haven't yet, where do I get it?
[14:01:51 CDT(-0500)] <heidi> jameswy ill skype you and show you
[14:01:57 CDT(-0500)] <athena> itunes or the android marketplace
[14:02:04 CDT(-0500)] <athena> just search for jasig - it's called jasig '11
[14:02:20 CDT(-0500)] <colinclark> I'll install it on my brand new iPad when I get home tonight, athena (smile)
[14:02:26 CDT(-0500)] <athena> we haven't publicized it widely just to keep from accidentally overburdening the server and making our lives uh-oh
[14:02:28 CDT(-0500)] <athena> ooh fun
[14:02:41 CDT(-0500)] <athena> we actually will have some updates to fix some of the ipad issues in a week or two
[14:02:56 CDT(-0500)] <athena> but i know ian's been using it on his ipad and said it was usable
[14:03:12 CDT(-0500)] <athena> lots of other good stuff here too - hoping to catch up on bedework this afternoon
[14:03:14 CDT(-0500)] <colinclark> yeah, he seemed impressed, again on Twitter
[14:03:22 CDT(-0500)] <colinclark> I gather Bosmon is there again today, too
[14:03:24 CDT(-0500)] <athena> yes
[14:03:25 CDT(-0500)] <colinclark> Is he giving a talk today?
[14:03:31 CDT(-0500)] <athena> i finally just saw him for the first time 10 min ago
[14:03:36 CDT(-0500)] <colinclark> wow
[14:03:40 CDT(-0500)] <colinclark> Can you say hi to everyone for me?
[14:03:42 CDT(-0500)] <colinclark> To Matt
[14:03:44 CDT(-0500)] <colinclark> and to Ian Dolphin
[14:03:45 CDT(-0500)] <colinclark> and Jens
[14:03:48 CDT(-0500)] <colinclark> and to yourself (tongue)
[14:03:55 CDT(-0500)] <athena> yeah, he presented this morning but was scheduled at the same time as me so i couldn't go (sad) annoyed about that
[14:03:58 CDT(-0500)] <athena> yes, will do!
[14:04:01 CDT(-0500)] <athena> and hi to you too
[14:04:01 CDT(-0500)] <colinclark> thanks!
[14:04:03 CDT(-0500)] <athena> really miss you guys
[14:04:20 CDT(-0500)] <athena> and matt's right that there's a hole where all the UX awesomeness should go
[14:05:30 CDT(-0500)] <athena> trying to convince bosmon to come hang out at dev days so i can make him explain all the cool new IoC stuff to me
[14:05:53 CDT(-0500)] <colinclark> yes!
[14:06:00 CDT(-0500)] <colinclark> he should
[14:06:29 CDT(-0500)] <colinclark> mlam: Sorry, back to my curiosity...
[14:06:37 CDT(-0500)] <colinclark> So, I can spot yura_ a mile away
[14:06:43 CDT(-0500)] <colinclark> with his invokers and his expanders
[14:06:44 CDT(-0500)] <colinclark> (tongue)
[14:07:07 CDT(-0500)] <colinclark> Let me grab some lines of code for us to look at
[14:07:10 CDT(-0500)] <colinclark> and see if we can think it through
[14:07:17 CDT(-0500)] <mlam> Ok, cool
[14:08:27 CDT(-0500)] <colinclark> So this is the code: https://github.com/mlam/infusion/blob/FLUID-4256/src/webapp/components/uploader/js/HTML5UploaderSupport.js#L254-265
[14:08:51 CDT(-0500)] <colinclark> So, doFormDataUpload() used to create its own FormData object
[14:08:57 CDT(-0500)] <colinclark> but that made it impossible for us to inject a mock
[14:09:07 CDT(-0500)] <mlam> Right
[14:09:12 CDT(-0500)] <colinclark> so you refactored it to take an instance of a FormData object
[14:09:19 CDT(-0500)] <colinclark> so that it doesn't care where it came from, so long as it gets a form data
[14:09:21 CDT(-0500)] <mlam> Yes
[14:09:46 CDT(-0500)] <colinclark> And then, this being an invoker, the problem was "how do we get a FormData instance to it?"
[14:09:58 CDT(-0500)] <colinclark> So you have a new function that creates a plain old FormData instance
[14:10:14 CDT(-0500)] <mlam> Yes
[14:10:25 CDT(-0500)] <colinclark> and then you're using an deferredInvokeCall expander quite trickily
[14:10:48 CDT(-0500)] <colinclark> So you have a demands block that specifies this expander as an argument to our doFormDataUpload() invoker
[14:10:54 CDT(-0500)] <mlam> Right
[14:11:22 CDT(-0500)] <colinclark> and IoC-when resolving the arguments each time doFormDataUpload() is called-will actually call createFormData() for us
[14:11:26 CDT(-0500)] <colinclark> which works perfectly
[14:11:41 CDT(-0500)] <colinclark> I tested in Firefox 3.6, 4, and IE 6 without any trouble
[14:11:51 CDT(-0500)] <colinclark> I made one minor change to the demands block...
[14:12:08 CDT(-0500)] <colinclark> We were using an older, ugly syntax for specifying arguments to the invoker
[14:12:15 CDT(-0500)] <colinclark> ["@0", "@1", "@2" ...
[14:12:25 CDT(-0500)] <colinclark> That can now be expressed in a somewhat more readable form as:
[14:12:58 CDT(-0500)]

<colinclark> args: ["

Unknown macro: {arguments}

.0", "

.1", "

Unknown macro: {arguments}

.2",...


[14:13:08 CDT(-0500)] <colinclark> anyway, that's a separate issue altogether
[14:13:53 CDT(-0500)] <mlam> ahh ok, looks much cleaner, i guess the same should be done for line 246 in the other doUpload demands block
[14:17:35 CDT(-0500)] <colinclark> yep, I did that one too
[14:17:40 CDT(-0500)] <colinclark> So, I'm wondering aloud, mlam
[14:17:51 CDT(-0500)] <colinclark> whether all of this isn't a sign that we outgrew these invokers
[14:18:38 CDT(-0500)] <mlam> ^^ I'm not following. What do you mean by outgrowing the invokers?
[14:18:58 CDT(-0500)] <colinclark> Well, what we've got is sort of complex
[14:19:06 CDT(-0500)] <colinclark> And I'm thinking we're missing a component
[14:19:21 CDT(-0500)] <colinclark> it seems to me that there are now a few different pieces of behaviour that are sort of interrelated
[14:19:49 CDT(-0500)] <colinclark> 1. Create an XHR
[14:19:53 CDT(-0500)] <colinclark> 2. Create a FormData
[14:19:59 CDT(-0500)] <colinclark> 3. Do something with #1 and #2
[14:20:07 CDT(-0500)] <colinclark> In the past #1 and #2 were just implicit
[14:20:20 CDT(-0500)] <colinclark> they aren't a lot of lines of code, so they were subsumed within #3
[14:20:31 CDT(-0500)] <colinclark> but then when you wanted to start writing tests with mocks of these, you had to split them out
[14:20:37 CDT(-0500)] <colinclark> in the case of #1 you created a new invoker
[14:20:51 CDT(-0500)] <colinclark> I was intrigued by the fact that you didn't choose to do #2 the same way
[14:21:16 CDT(-0500)] <colinclark> and I guess, in part, that was due to the fact that an invoker can't have an invoker
[14:21:22 CDT(-0500)] <colinclark> which would be insane (smile)
[14:21:40 CDT(-0500)] <colinclark> I guess the question I have is, "is this a thing?"
[14:21:55 CDT(-0500)] <colinclark> and as part of answering that question, I guess another question is
[14:22:13 CDT(-0500)] <colinclark> when do we need to change the implementation of #1 and #2?
[14:22:39 CDT(-0500)] <colinclark> So we have two different "strategies" for the HTML5 remote...
[14:23:01 CDT(-0500)] <mlam> I initially had #2 as an invoker, but then we came across an issue in FF3.6 where FormData wasn't supported. The invoker to create the FormData object would run every time the HTML5 strategy was resolved. So I had to ask yura_ for help on a different approach
[14:23:20 CDT(-0500)] <colinclark> hmm
[14:23:27 CDT(-0500)] <colinclark> I'm not sure what that code looked like
[14:23:39 CDT(-0500)] <colinclark> but I can't see how an invoker would be run every time the strategy was resolved
[14:24:06 CDT(-0500)] <colinclark> Do you remember what the code looked like?
[14:24:26 CDT(-0500)] <mlam> the invoker was originally called before the doUpload() function
[14:25:08 CDT(-0500)] <mlam> and only when the we used the FormData to send the file, did I pass along the formData parameter
[14:25:11 CDT(-0500)] <colinclark> ahhh
[14:25:27 CDT(-0500)] <colinclark> So, you were trying to create a FormData object, regardless of the doUpload implementation we were using
[14:25:34 CDT(-0500)] <mlam> and I think because I called it far too soon, every time the HTML5 strategy was used, we would have a formData not supported error in FF 3.6
[14:25:36 CDT(-0500)] <colinclark> in other words, we have two different implementations of doUpload
[14:25:48 CDT(-0500)] <colinclark> a. manual multipart upload, used by FF 3.6
[14:25:54 CDT(-0500)] <colinclark> b. form data upload, used by everyone else
[14:26:11 CDT(-0500)] <colinclark> and you were trying to create a FormData regardless of whether you were using a. or b.
[14:26:13 CDT(-0500)] <colinclark> is that right?
[14:26:24 CDT(-0500)] <mlam> yes
[14:26:32 CDT(-0500)] <colinclark> yeah, that doesn't work (smile)
[14:26:56 CDT(-0500)] <colinclark> And that was my point about an invoker not being able to have an invoker...
[14:27:13 CDT(-0500)] <colinclark> it's actually the particular doUpload() implementation that needs the FormData creator
[14:27:26 CDT(-0500)] <colinclark> Which is what brings me to the point about perhaps we've outgrown these invokers
[14:27:28 CDT(-0500)] <mlam> yah (smile)
[14:27:33 CDT(-0500)] <colinclark> I'm wondering if we actually need a component here
[14:27:40 CDT(-0500)] <colinclark> I was grasping for names
[14:27:47 CDT(-0500)] <colinclark> an "uploader" if you know what I mean
[14:27:57 CDT(-0500)] <colinclark> bad choice of name (tongue)
[14:28:04 CDT(-0500)] <colinclark> a "do-er"
[14:28:08 CDT(-0500)] <colinclark> bad choice of name (tongue)
[14:28:33 CDT(-0500)] <mlam> I guess another "strategy" component?
[14:28:37 CDT(-0500)] <colinclark> Put another way, a thing that is responsible for taking a bunch of stuff, preparing it, and sending it off
[14:28:45 CDT(-0500)] <colinclark> yeah another "strategy"
[14:29:51 CDT(-0500)] <colinclark> Any bright naming ideas?
[14:30:26 CDT(-0500)] <mlam> hmm
[14:30:53 CDT(-0500)] <colinclark> I was actually sort of thinking that it was comparable to the "engine" in the Flash strategy
[14:31:00 CDT(-0500)] <colinclark> but ultimately they're different enough that it'd be awfully confusing
[14:31:08 CDT(-0500)] <colinclark> not that that name was particularly good in the first place
[14:33:04 CDT(-0500)] <heidi> Justin_o i have a reset button, but not sure why live preview doesn't reset when i click it... thoughts? or should i just wait? (smile)
[14:35:39 CDT(-0500)] <colinclark> mlam: I think I'm going to stick a comment on the code
[14:35:42 CDT(-0500)] <colinclark> saying something like
[14:35:52 CDT(-0500)] <colinclark> "this works, but could be simplified by creating a component here"
[14:35:54 CDT(-0500)] <colinclark> and then commit it
[14:36:08 CDT(-0500)] <colinclark> We can let it sit for a bit and think about it
[14:36:48 CDT(-0500)] <mlam> Ok, sounds good colinclark
[15:01:16 CDT(-0500)] <Justin_o> heidi: sorry.. i was chatting with harriswong
[15:01:25 CDT(-0500)] <Justin_o> heidi: i'm heading out now, but we can chat about it tomorrow
[15:01:35 CDT(-0500)] <heidi> cool tty tomorrow
[15:46:51 CDT(-0500)] <colinclark> mlam: I just went ahead and did it
[15:47:02 CDT(-0500)] <colinclark> with the regrettable name of a "fileSender"
[15:47:03 CDT(-0500)] <colinclark> two types
[15:47:07 CDT(-0500)] <colinclark> rawMIMESender
[15:47:10 CDT(-0500)] <colinclark> and formDataSend
[15:47:15 CDT(-0500)] <colinclark> formDataSender, that is
[15:47:24 CDT(-0500)] <colinclark> each component has one magical method
[15:47:27 CDT(-0500)] <colinclark> send()
[15:47:35 CDT(-0500)] <colinclark> I'm going to go to code naming hell for the Uploader (tongue)
[15:48:00 CDT(-0500)] <colinclark> I'm going to push it into my Github fork, if you're willing to take a look at it
[15:48:09 CDT(-0500)] <mlam> ok, cool
[15:48:11 CDT(-0500)] <mlam> yah, i'll take a look
[15:49:04 CDT(-0500)] <colinclark> The next thing that needs to go like this is the code that listens to XHR events and fires Uploader events accordingly
[15:49:18 CDT(-0500)] <colinclark> but i'll leave that as an exercise for another day
[15:53:30 CDT(-0500)] <heidi> hey cindyli i reopened with http://issues.fluidproject.org/browse/FLUID-4216 with a few small comments re: sliders
[15:53:38 CDT(-0500)] <heidi> oh she's gone..
[15:54:19 CDT(-0500)] <mlam> colinclark: I've got to run for now, i'll take a look at this new component later tonight
[15:55:07 CDT(-0500)] <colinclark> k
[16:07:59 CDT(-0500)] <anastasiac> /leave