Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Comment: Migrated to Confluence 4.0

[07:54:03 CST(-0600)] <colinclark> Justin_o: Morning, King
[07:54:11 CST(-0600)] <Justin_o> colinclark: good morning
[07:54:20 CST(-0600)] <colinclark> A quick update...
[07:54:37 CST(-0600)] <colinclark> mlam and I made it surprisingly far on Friday and over the weekend with Uploader
[07:55:13 CST(-0600)] <colinclark> There are a few quirks still to iron out
[07:55:29 CST(-0600)] <Justin_o> colinclark: okay... so almost ready to go?
[07:55:42 CST(-0600)] <colinclark> but the biggest issue now is really styling the HTML 5 version of the browse button
[07:55:46 CST(-0600)] <colinclark> Out of the box, it's quite ugly
[07:55:57 CST(-0600)] <Justin_o> ah
[07:56:11 CST(-0600)] <colinclark> I have no idea how styleable it even is, so it'll be a fun task for someone who wants to learn a bit about the HTML 5 elements
[07:56:37 CST(-0600)] <colinclark> I suspect there are probably some substantial AT issues that have historically been masked by Flash
[07:56:55 CST(-0600)] <colinclark> so I guess we need to get on some screen reader testing of it as soon as we can, too
[08:01:20 CST(-0600)] <colinclark> That's a good question
[08:01:28 CST(-0600)] <colinclark> I think we've probably got a couple of hours to iron out the last few tweaks
[08:01:47 CST(-0600)] <colinclark> We should do at least a partial round of QA on it to see how it's working
[08:02:03 CST(-0600)] <colinclark> Styling might take a little longer, or it might turn out to be really easy
[08:02:10 CST(-0600)] <colinclark> I honestly don't know--super new territory
[08:03:05 CST(-0600)] <Justin_o> colinclark: okay.. so wondering where we stand for today
[08:03:17 CST(-0600)] <Justin_o> i guess we'll need to push off code freeze for one more day?
[08:03:34 CST(-0600)] <colinclark> Justin_o: Well, you and I probably need to decide what it even means to freeze today
[08:03:44 CST(-0600)] <Justin_o> and do you think we'll have time to get all the i dotting and t crossing done today?
[08:03:49 CST(-0600)] <colinclark> So, if we look at this as an "AEGIS demo freeze," then I think we're quite close
[08:03:55 CST(-0600)] <colinclark> From a code freeze perspective, there's still more
[08:04:05 CST(-0600)] <colinclark> As in, the 1.3 release...
[08:04:20 CST(-0600)] <colinclark> The Pager demo is pretty unpolished for a new release (but might be sufficient for the testing)
[08:04:34 CST(-0600)] <colinclark> Uploader still needs to be made to work with an actual server
[08:04:41 CST(-0600)] <colinclark> along with options backward compatiblity
[08:04:52 CST(-0600)] <Justin_o> colinclark: you mean in it's styling i guess
[08:05:04 CST(-0600)] <colinclark> Justin_o: Pager? Yes, styling and visual design
[08:05:33 CST(-0600)] <Justin_o> colinclark: those are two big things left for uploader.. and yes the previous comment was about pager
[08:05:39 CST(-0600)] <colinclark> yep
[08:05:58 CST(-0600)] <colinclark> When everyone left on Friday I wasn't entirely clear what was left to be done for Pager
[08:06:00 CST(-0600)] <Justin_o> colinclark: did you hear from Bosmon over the weekend at all?
[08:06:05 CST(-0600)] <colinclark> for the "demo freeze"
[08:06:07 CST(-0600)] <colinclark> Justin_o: Nope
[08:06:18 CST(-0600)] <colinclark> He fixed a minor bug in IoC, that's all I've seen
[08:06:25 CST(-0600)] <Justin_o> colinclark: (sad)
[08:06:35 CST(-0600)] <colinclark> We'll survive (smile)
[08:06:57 CST(-0600)] <Justin_o> okay.. i'll try to talk to Golam about the pager bug fixes this morning when i see him and with others about the pager demo
[08:07:00 CST(-0600)] <colinclark> So, when jhung, heidi, and golam arrive we need to get a plan for the last steps on Pager
[08:07:06 CST(-0600)] <Justin_o> colinclark: yep
[08:07:11 CST(-0600)] <colinclark> And for Uploader, we need two things:
[08:07:19 CST(-0600)] <colinclark> 1. Someone to help with screen reader testing now
[08:07:23 CST(-0600)] <colinclark> 2. A buddy to help with styling
[08:07:49 CST(-0600)] <colinclark> jessm: This has been an interesting conversation ^
[08:08:09 CST(-0600)] <Justin_o> colinclark: okay... did you want to decide on who does those things now or wait till after everyone checks in?
[08:08:30 CST(-0600)] <colinclark> Well, we might be able to decide on Uploader #1
[08:08:44 CST(-0600)] <colinclark> anastasiac did some screen reader testing of Inline Edit last week... seemed like it was helpful
[08:08:45 CST(-0600)] <jessm> colinclark: lemme catch up then
[08:09:03 CST(-0600)] <colinclark> Any chance we could do the same with Uploader, Justin_o and anastasiac?
[08:09:31 CST(-0600)] <Justin_o> colinclark, anastasiac that would be good
[08:09:49 CST(-0600)] <anastasiac> colinclark, Justin_o, I'd be happy to help with testing the uploader. Is it in trunk, and on the nightly build right now?
[08:10:17 CST(-0600)] <colinclark> anastasiac: It's in an incubator branch. We've had problems getting access to the incubator on the daily build server, but perhaps it's just forgetfulness
[08:10:23 CST(-0600)] <colinclark> Let me get you the SVN URL, at least
[08:10:44 CST(-0600)] <colinclark> http://source.fluidproject.org/svn/incubator/uploader/FLUID-3722/
[08:10:47 CST(-0600)] <anastasiac> thanks colinclark - I can check it out onto the test machine
[08:11:02 CST(-0600)] <colinclark> anastasiac: Can you spend a few minutes seeing if you can get to it on the daily build?
[08:11:14 CST(-0600)] <colinclark> We'll need it publicly visible for the actual testing next week, anyway
[08:11:27 CST(-0600)] <anastasiac> colinclark, do you mean by a new continuum task?
[08:11:44 CST(-0600)] <colinclark> No, I mean, the incubator is already ostensibly built by Continuum
[08:11:50 CST(-0600)] <colinclark> it's just, "where is it?"
[08:12:01 CST(-0600)] <anastasiac> ah, ok, gotcha
[08:12:13 CST(-0600)] <colinclark> Justin_o: In terms of styling the Browse button
[08:12:28 CST(-0600)] <colinclark> My list might include one of jameswy, heidi, or jhung
[08:13:03 CST(-0600)] <colinclark> I'm finding this input type=file multiple="" control in HTML 5 quite odd in a few ways
[08:13:09 CST(-0600)] <colinclark> understandably so
[08:13:24 CST(-0600)] <colinclark> You can't click it programmatically, so it can't be hidden
[08:13:54 CST(-0600)] <Justin_o> colinclark: really
[08:14:03 CST(-0600)] <colinclark> I guess this makes sense because you want the user to actively choose to select files, not have some sketchy script do it automatically in a way that might confuse the user into sharing access to files they wouldn't otherwise share.
[08:14:05 CST(-0600)] <Justin_o> is that just because of something to do with jquery not supporting it yet?
[08:14:13 CST(-0600)] <colinclark> I don't know for certain
[08:14:16 CST(-0600)] <Justin_o> ah okay
[08:14:18 CST(-0600)] <colinclark> But I'm assuming it's like the Flash 10 thing
[08:14:23 CST(-0600)] <Justin_o> yep
[08:14:38 CST(-0600)] <colinclark> So far, I also haven't figured out how to relabel the Browse button
[08:14:45 CST(-0600)] <colinclark> which by default in Firefox is labelled "Browse..."
[08:14:57 CST(-0600)] <colinclark> but we'd love it if it changed to "Add More," too
[08:15:03 CST(-0600)] <colinclark> that could just be laziness on my part
[08:15:34 CST(-0600)] <Justin_o> colinclark: interesting...
[08:16:16 CST(-0600)] <Justin_o> i suppose in the end it wouldn't be a deal breaker for adding html5 upload if we can't change the button label
[08:17:15 CST(-0600)] <colinclark> Justin_o: It wouldn't be a deal breaker, but it's awfully ugly as-is
[08:17:51 CST(-0600)] <Justin_o> colinclark: okay
[08:18:10 CST(-0600)] <colinclark> hey jameswy
[08:18:50 CST(-0600)] <jameswy> Howdy colinclark.
[08:19:11 CST(-0600)] <colinclark> Feel like a fun-ish HTML 5 styling mission?
[08:19:31 CST(-0600)] <anastasiac> colinclark, a quick double-check: should the uploader in the 3722 branch actually work? in FF on a Mac?
[08:19:35 CST(-0600)] <jameswy> colinclark: Haha. What do you have in mind?
[08:19:40 CST(-0600)] <colinclark> anastasiac: lol
[08:19:47 CST(-0600)] <colinclark> "Please test my completely broken component"
[08:19:51 CST(-0600)] <colinclark> Yes, it should work
[08:19:56 CST(-0600)] <anastasiac> hm
[08:20:12 CST(-0600)] <anastasiac> I just checked it out on my machine to double-check what I saw on the server, and it doesn't work fo rme
[08:20:23 CST(-0600)] <anastasiac> I get a dialog, but I can't click the upload button
[08:20:29 CST(-0600)] <colinclark> anastasiac: Let's start with the basics. What version of Firefox?
[08:20:32 CST(-0600)] <colinclark> And what errors are you seeing?
[08:20:42 CST(-0600)] <colinclark> jameswy: The Uploader now has support for HTML 5
[08:20:43 CST(-0600)] <anastasiac> FF 3.6.8
[08:20:54 CST(-0600)] <colinclark> anastasiac: Okay, that's good. Errors in Firebug?
[08:21:04 CST(-0600)] <anastasiac> ah, weird
[08:21:04 CST(-0600)] <colinclark> jameswy: It
[08:21:20 CST(-0600)] <anastasiac> I opened the console to double-check any messages, and now it's working just fine...
[08:21:23 CST(-0600)] <colinclark> jameswy: It's accomplished using an < input type="file" multiple="" > control
[08:21:42 CST(-0600)] <colinclark> By default, this is rendered in a rather ugly way
[08:21:58 CST(-0600)] <colinclark> With a text field that shows the paths to each file and a button beside it labelled "Browse"
[08:22:25 CST(-0600)] <colinclark> As I was saying to Justin_o, my worst fear is that the browse won't let you style these very much for "security reasons"
[08:22:41 CST(-0600)] <colinclark> but I don't see any reason why it shouldn't be possible to make it look like our old Browse button
[08:22:45 CST(-0600)] <colinclark> or like whatever you want (smile)
[08:23:08 CST(-0600)] <jameswy> colinclark: I'm up for it (smile)
[08:23:16 CST(-0600)] <colinclark> Okay, here's the URL to the branch...
[08:23:21 CST(-0600)] <anastasiac> ok, colinclark, if I open the firebug console, the uploader is fine, but if I close it again, it's not. The problems are:
[08:23:28 CST(-0600)] <anastasiac> no 'switch to standard' link
[08:23:29 CST(-0600)] <colinclark> jameswy: http://source.fluidproject.org/svn/incubator/uploader/FLUID-3722/
[08:23:42 CST(-0600)] <anastasiac> after you browse, you don't get a file queue listing
[08:23:48 CST(-0600)] <anastasiac> the upload button is inactive
[08:23:58 CST(-0600)] <colinclark> anastasiac: You know the weird thing...
[08:24:00 CST(-0600)] <colinclark> I can totally reproduce that
[08:24:05 CST(-0600)] <colinclark> wowza
[08:24:13 CST(-0600)] <anastasiac> I can't imagine what would happen in IE :-P
[08:24:21 CST(-0600)] <colinclark> Well, nothing will happen in IE
[08:24:24 CST(-0600)] <colinclark> It's HTML 5!
[08:24:33 CST(-0600)] <colinclark> I guess maybe the IE 9 beta might work
[08:24:38 CST(-0600)] <anastasiac> FYI, http://build.fluidproject.org/incubator/uploader/FLUID-3722/src/webapp/components/uploader/html/Uploader.html
[08:24:46 CST(-0600)] <colinclark> sweet, thanks
[08:24:48 CST(-0600)] <anastasiac> I'll test it on the test Windows machine now
[08:25:13 CST(-0600)] <colinclark> Justin_o: Okay, that's a big one to add to our list
[08:25:21 CST(-0600)] <Justin_o> colinclark: (smile)
[08:25:30 CST(-0600)] <colinclark> A component that doesn't work when the Firebug window is closed?!?
[08:26:17 CST(-0600)] <colinclark> It might be simple... console logs
[08:26:19 CST(-0600)] <colinclark> let me check
[08:27:01 CST(-0600)] <heidi__> hi everyone - what can i help with?
[08:27:06 CST(-0600)] <anastasiac> colinclark, there does seem to be one console log on load: "afterReady"
[08:27:16 CST(-0600)] <anastasiac> but just because the window is closed
[08:27:19 CST(-0600)] <anastasiac> ?
[08:27:21 CST(-0600)] <anastasiac> that's just weird
[08:27:25 CST(-0600)] <Justin_o> heidi__: do you know where the pager demo was left off at?
[08:27:28 CST(-0600)] <colinclark> anastasiac: There are console logs on every event in the Uploader right now
[08:27:46 CST(-0600)] <colinclark> Yep, that'll do it
[08:27:51 CST(-0600)] <heidi__> justin_o golam would know best - he was working on screen reader issues fri aft
[08:28:06 CST(-0600)] <colinclark> It looks like when Firebug is closed, all of its API is also removed from the window
[08:28:21 CST(-0600)] <Justin_o> heidi__: so other than a11y fixes the pager demo is essentially done?
[08:28:40 CST(-0600)] <heidi__> justin_o i think so. the styling is done etc yes
[08:29:58 CST(-0600)] <colinclark> anastasiac: I've just committed a fix
[08:30:23 CST(-0600)] <Justin_o> heidi__: thanks
[08:30:38 CST(-0600)] <heidi__> justin_o anything else i can help with? i could test etc
[08:31:35 CST(-0600)] <Justin_o> heidi__: the pager demo isn't working in the build site
[08:31:58 CST(-0600)] <heidi__> jhung committed last - jon any updates ?
[08:32:48 CST(-0600)] <jhung> justin_o, heidi: I haven't put in any changes since last Friday. Let me check out the build site...
[08:33:51 CST(-0600)] <jhung> heidi_, justin_o: works fine here on http://build.fluidproject.org/infusion/demos/pager/html/pager.html#. Heidi_, confirm?
[08:34:11 CST(-0600)] <Justin_o> jhung: but not here http://build.fluidproject.org/infusion/demos/pager/demo.html
[08:34:23 CST(-0600)] <anastasiac> colinclark, excellent - I'll redeploy (unless you already did)
[08:34:24 CST(-0600)] <jhung> justin_o: yes, I see that now.
[08:34:31 CST(-0600)] <colinclark> anastasiac: nope
[08:34:33 CST(-0600)] <jhung> Probably because the demo.html code is out of date.
[08:34:34 CST(-0600)] <colinclark> go for it
[08:34:49 CST(-0600)] <anastasiac> hm
[08:34:54 CST(-0600)] <jhung> justin_o: we need to update the reference to the demo's html file. I'll do that now.
[08:35:27 CST(-0600)] <anastasiac> colinclark, I just updated from svn, and it's still broken for me
[08:36:02 CST(-0600)] <colinclark> anastasiac: Okay, why?
[08:36:05 CST(-0600)] <colinclark> Errors, anything?
[08:36:36 CST(-0600)] <anastasiac> no - same behaviour: with console, it works, without, nothing
[08:37:02 CST(-0600)] <anastasiac> hm, I'm still getting console logs - is that what you removed?
[08:37:07 CST(-0600)] <colinclark> yes
[08:37:09 CST(-0600)] <colinclark> cache?
[08:37:14 CST(-0600)] <anastasiac> disabled
[08:37:17 CST(-0600)] * anastasiac check file paths
[08:38:33 CST(-0600)] <colinclark> anastasiac: You're not using the demo page, by chance?
[08:38:47 CST(-0600)] <anastasiac> colinclark, it's fine now - I don't know what the problem was... weird
[08:41:40 CST(-0600)] <heidi__> justin_o would just going through and testing teh demos be helpful?
[08:42:17 CST(-0600)] <jhung> anastasiac: I'm trying to get the pager demo working in its demo portal page. The demo isn't showing up in the demo pane, but the HTML and CSS are showing up in tabs, but not the JS file. Any ideas?
[08:42:58 CST(-0600)] <anastasiac> jhung, that's odd - let me refresh my memory of how the portal layout works
[08:43:22 CST(-0600)] <colinclark> heidi__: Can you run through the QA plan for the demo portion of the Uploader using the HTML 5 version that mlam and I have been working on?
[08:43:40 CST(-0600)] <Justin_o> jhung: if you check in firebug i think it shows you the paths it is trying to ajax the files in from.. you will probably have to go into the "net" tab
[08:43:57 CST(-0600)] <jhung> anastasiac: k I'll take a look there
[08:44:12 CST(-0600)] <heidi__> colinclark i can test it. what specific qa plan are you thinking?
[08:44:25 CST(-0600)] <colinclark> The specific QA plan (smile)
[08:44:29 CST(-0600)] <colinclark> Let me grab you a link
[08:44:34 CST(-0600)] <heidi__> thanks
[08:44:44 CST(-0600)] <anastasiac> jhung, the html, css and js file all have to have the same root string, and that string has to match the "demo.name" in the main demo.html file - check that
[08:44:57 CST(-0600)] <colinclark> heidi__: http://wiki.fluidproject.org/display/fluid/Uploader+QA+Test+Plan
[08:45:02 CST(-0600)] <jhung> anastasiac: yes. I see that now.
[08:45:14 CST(-0600)] <colinclark> Follow the steps for the local version of the Uploader--the server side is not yet hooked up.
[08:45:25 CST(-0600)] <colinclark> Justin_o can explain (smile)
[08:45:26 CST(-0600)] <jhung> anastasiac: I change the filenames in the demo to match.
[08:45:35 CST(-0600)] <anastasiac> jhung, did that fix it?
[08:45:52 CST(-0600)] <Justin_o> heidi__: let me know if you have any questions about the test plan and I can try to help
[08:46:02 CST(-0600)] <jhung> anastasiac: I think it should. The path is looking for pager.js and it's named demo-pager.js in the repo.
[08:46:10 CST(-0600)] <heidi__> justin_o okay thanks ... ill go through this
[08:46:21 CST(-0600)] <anastasiac> right, jhung, the string and all the file names have to match
[08:46:38 CST(-0600)] <colinclark> heidi__: So the new Uploader is in a branch of the incubator
[08:46:52 CST(-0600)] <jhung> anastasiac: incidentally, I think our demo files should have the "demo-" prefix to reduce confusion between the similarly named files belonging to the component.
[08:46:54 CST(-0600)] <heidi__> colinclark, ah okay. cool
[08:47:00 CST(-0600)] <jamon> Justin_o: what's up with that url you were asking about on friday?
[08:47:01 CST(-0600)] <colinclark> anastasiac can fire you a link to the daily build of it
[08:47:05 CST(-0600)] <colinclark> anastasiac: You rebuilt it, right?
[08:47:07 CST(-0600)] <anastasiac> jhung, good idea
[08:47:19 CST(-0600)] <anastasiac> colinclark, yes
[08:47:22 CST(-0600)] <anastasiac> everyone: http://build.fluidproject.org/incubator/uploader/FLUID-3722/src/webapp/components/uploader/html/Uploader.html
[08:47:32 CST(-0600)] <Justin_o> jhung: i would agree.. it gets really confusing at time which file you're looking at
[08:48:21 CST(-0600)] <Justin_o> jamon: it's okay.. anastasia figured it out
[08:48:31 CST(-0600)] * anastasiac is heading over to another machine to test the uploader with screen readers. if you need me, you'll have to shout out loud
[08:48:36 CST(-0600)] <mlam> colinclark: so last steps in prep for the demo is to style the browse button?
[08:48:40 CST(-0600)] <Justin_o> jamon: were you able to get the testswarms check out updated
[08:49:03 CST(-0600)] <colinclark> mlam: Yep. Read the channel logs for the whole story, but I'll give a quick update on where everyone is at
[08:49:19 CST(-0600)] <colinclark> heidi__ is working on running through the Uploader QA plan against our HTML 5 version
[08:49:29 CST(-0600)] <jamon> Justin_o: oh i hadn't touched it, but i will try a checkout for you if you like
[08:49:31 CST(-0600)] <colinclark> anastasiac is doing screen reader testing of HTML 5 Uploader
[08:49:50 CST(-0600)] <mlam> cool
[08:49:57 CST(-0600)] <colinclark> jhung is working on getting the Pager demo working again
[08:50:19 CST(-0600)] <colinclark> We're waiting on Golam for an update on Pager a11y fixes and Antranig on Reorderer a11y fixes
[08:50:30 CST(-0600)] <Justin_o> jamon: thanks
[08:50:37 CST(-0600)] <Justin_o> hopefully that will work
[08:50:57 CST(-0600)] <mlam> I'll get golam into the channel right when he gets in
[08:51:09 CST(-0600)] <colinclark> no worries
[08:51:19 CST(-0600)] <jamon> Justin_o: it will take some fiddling i think, to get all the commits that have been missed into the queue though
[08:51:22 CST(-0600)] <colinclark> So, mlam things are looking pretty decent with the demo version of HTML 5 Uploader
[08:51:37 CST(-0600)] <colinclark> thanks for catching that addFiles() bug last night... I was totally confused (smile)
[08:51:55 CST(-0600)] <colinclark> mlam: you and I are on the hook for any more fixes and polish that need to be done
[08:52:09 CST(-0600)] <colinclark> One thing I noticed is that deleting items from the queue seems to not quite work how you'd expect
[08:52:21 CST(-0600)] <mlam> no problem (smile) that's the least i can do. thanks for all your help with the uploader
[08:52:40 CST(-0600)] <colinclark> the little "Choose Browse files to add files to the queue" reappears before all the files have actually been removed from the queue
[08:52:54 CST(-0600)] <colinclark> Start in on that while I hop on my bike
[08:52:59 CST(-0600)] <mlam> ok
[08:54:22 CST(-0600)] <Justin_o> jamon: okay.. thanks for your help. I suspect that the testswarm just runs all the tests on the latest checkout though.. so it may not be an issue
[08:54:40 CST(-0600)] <jamon> Justin_o: it runs the tests on every commit
[08:55:06 CST(-0600)] <jamon> Justin_o: if it is fine to skip revisions though, and just move to head, that makes things a lot easier
[08:55:27 CST(-0600)] <Justin_o> jamon: yep, but if there were no browsers to run it on a commit x before commit y occurs.. i think it runs the x tests on the y checkout.
[08:56:56 CST(-0600)] <jamon> Justin_o: should be all set then
[08:57:34 CST(-0600)] <Justin_o> jamon: okay
[08:57:37 CST(-0600)] <Justin_o> thanks
[08:59:38 CST(-0600)] <jhung> justin_o: I've committed changes to the pager demo. Try rebuilding.
[09:00:56 CST(-0600)] <Justin_o> jhung: thanks
[09:01:30 CST(-0600)] <heidi__> uploader just killed my safari :|
[09:01:39 CST(-0600)] <heidi__> "quit unexpectedly"
[09:03:35 CST(-0600)] <heidi__> twice...
[09:04:25 CST(-0600)] <jessm> heidi__: what version of safari?
[09:04:53 CST(-0600)] <heidi__> safari 5
[09:07:20 CST(-0600)] <mlam> heidi__: that's not good. does it happen every time?
[09:07:31 CST(-0600)] <heidi__> mlam yes
[09:07:34 CST(-0600)] <heidi__> with voiceover on
[09:07:37 CST(-0600)] <heidi__> it kills it every time
[09:08:04 CST(-0600)] <jhung> justin_o: what can I help with now?
[09:13:07 CST(-0600)] <Justin_o> jhung: this still doesn't seem to be working
[09:13:08 CST(-0600)] <Justin_o> http://build.fluidproject.org/infusion/demos/pager/demo.html
[09:13:32 CST(-0600)] <Justin_o> i rebuilt the daily build site
[09:13:38 CST(-0600)] <heidi__> mlam when it says "Choose Browse files to add files to the queue" i'm guessing that means "choose file" button. why are they diff names?
[09:14:06 CST(-0600)] <jhung> justin_o: yes. I see that it's still looking for an old pager file that's been deprecated.
[09:14:33 CST(-0600)] <jhung> justin_o: going to look into that now...
[09:14:57 CST(-0600)] <mlam> heidi__: we haven't styled the Browse button yet
[09:15:56 CST(-0600)] <Justin_o> mlam, heidi__ : jameswy is looking into that now
[09:15:59 CST(-0600)] <heidi__> mlam i think it just says something diff on safari
[09:16:18 CST(-0600)] <heidi__> so the instructions are for ff
[09:16:38 CST(-0600)] <jameswy> heidi__, mlam: correct. input file is different across browsers.
[09:16:40 CST(-0600)] <heidi__> it's not a big deal, just my eye scanned for "browse"
[09:20:07 CST(-0600)] <jameswy> mlam: heads-up on an uploader bug: I can't remove the last file
[09:20:18 CST(-0600)] <jameswy> (on a multiple file list)
[09:20:27 CST(-0600)] <jameswy> (FF 3.6, OS X 10.5)
[09:20:49 CST(-0600)] <mlam> jameswy: thanks. adding it to my list
[09:21:49 CST(-0600)] <jameswy> mlam: actually, it seems to be the last few, depending on how many files there are (maybe your index calculations are off a bit?)
[09:22:25 CST(-0600)] <heidi__> mlam another one: on safari5 , os 10.6 it says "choose file, no file selected" below the 3 files i just added. should say "add more" or something
[09:30:32 CST(-0600)] <heidi__> mlam... is it useful to just post bugs here or should i send the full list at the end?
[09:30:53 CST(-0600)] <mlam> i think a full list at the end would be really helpful
[09:31:08 CST(-0600)] <heidi__> i deleted the last file out of 3, and the "to upload" changed from 3 to 1 file, instead of the 2 listed still
[09:31:09 CST(-0600)] <mlam> i'm just working on a couple of the bugs now, so i may miss something once i get through them
[09:31:10 CST(-0600)] <heidi__> okay
[09:31:13 CST(-0600)] <heidi__> will do
[09:31:19 CST(-0600)] <mlam> thanks
[09:31:45 CST(-0600)] <justin_o> golam: hello... could you let me know where you're at with the pager fixes
[09:33:49 CST(-0600)] <jhung> justin_o: I've reverted the filenaming back to the way it was with the old demo. However, I am now getting an error: "fluid.setScopedData is not a function
[09:33:49 CST(-0600)] <jhung> /webapp/framework/core/js/jquery.keyboard-a11y.js
[09:33:49 CST(-0600)] <jhung> Line 474"
[09:34:25 CST(-0600)] <golam> justin_o: the problem with IE8 still exist with the new demo when trying to announce the message "Current page"and I am looking into the bug right now
[09:35:00 CST(-0600)] <jhung> justin_o: seems related to something that Bosmon checked in on Friday night? Bosmon, any tips?
[09:35:01 CST(-0600)] <mlam> heidi__, jameswy: fixed the bug to allow us to delete all files from the file queue
[09:35:22 CST(-0600)] <heidi__> mlam okay updating and will try again
[09:35:30 CST(-0600)] <mlam> thanks
[09:35:54 CST(-0600)] <heidi__> mlam still troubles
[09:36:15 CST(-0600)] <heidi__> it deletes fine, but the remaining file count is wrong.
[09:40:07 CST(-0600)] <jhung> anastasiac, colinclark: can someone talk me through how to trace this error properly? I'm getting "fluid.setScopedData is not a function" on line 474 in jquery.keyboard-a11y.js.
[09:40:54 CST(-0600)] <justin_o> jhung: i think the jquery.keyboard-a11y.js is dependent on fluid.documents.js
[09:40:59 CST(-0600)] <justin_o> is that there?
[09:41:30 CST(-0600)] <justin_o> golam: so what is left to do to get all of the pagers accessbility fixes done?
[09:41:30 CST(-0600)] <jhung> justin_o: it is not. I'll put that in now.
[09:42:48 CST(-0600)] <jhung> justin_o: correction. FluidDocument.js is there.
[09:43:19 CST(-0600)] <justin_o> jhung: yes.. that's the correct nam
[09:43:21 CST(-0600)] <justin_o> name
[09:44:06 CST(-0600)] <golam> justin_o: when hover over page link it's not saying anything in IE8 about the visiting link
[09:44:35 CST(-0600)] <justin_o> golam: so you were able to get the tooltip stuff taken care of on Friday with Michelle?
[09:44:56 CST(-0600)] <golam> justin_o: no that wasn't done
[09:45:15 CST(-0600)] <justin_o> golam: okay.. so there are just those two issues left then?
[09:46:06 CST(-0600)] <golam> justin_o: well latest JAWS and FireFox gives the message about the visiting link
[09:46:29 CST(-0600)] <golam> justin_o: yes I need to make it work on IE8
[09:46:49 CST(-0600)] <justin_o> golam: ah okay..
[09:47:08 CST(-0600)] <justin_o> golam: where did you and michelle leave off at for the tooltip?
[09:48:02 CST(-0600)] <golam> justin_o: I guess at the meeting we all had in the morning
[09:48:12 CST(-0600)] <justin_o> golam: okay
[09:48:13 CST(-0600)] <justin_o> thanks
[09:53:18 CST(-0600)] <justin_o> jhung: the path to FluidDocument.js is wrong
[09:53:29 CST(-0600)] <justin_o> it should only have ../../ not ../../../
[09:53:41 CST(-0600)] <jhung> yes
[09:53:48 CST(-0600)] <jhung> I just found that. Committing it now.
[09:54:09 CST(-0600)] <jhung> thanks justin_o.
[09:54:25 CST(-0600)] <harriswong> colinclark: should i work on the uploader tests on the trunk or incubator 3722 branch?
[09:54:49 CST(-0600)] <justin_o> jhung: thanks
[09:55:07 CST(-0600)] <jhung> justin_o: should be working now...
[09:56:13 CST(-0600)] <justin_o> jhung: thanks i'll rebuild the daily build site
[09:56:24 CST(-0600)] <justin_o> jhung: can you take a look at the changes i made to inline edit over the weekend
[09:56:49 CST(-0600)] <colinclark> heidi__: I'm just catching up on the channel logs
[09:56:57 CST(-0600)] <jhung> justin_o: sure. Anything in particular I should keep an eye on?
[09:57:05 CST(-0600)] <colinclark> I can't reproduce your crash of Uploader with Safari 5 and VoiceOver. It works pretty nicely for me.
[09:57:27 CST(-0600)] <justin_o> jhung: do you get the commit e-mails?
[09:57:31 CST(-0600)] <colinclark> As for the button issues-the fact that the button label doesn't match the instructions, and the fact that it doesn't change to "Add More"-that's the issue that jameswy is working on at the moment.
[09:57:45 CST(-0600)] <colinclark> It seems like this particular HTML 5 element is actually pretty hard to style correctly
[09:57:46 CST(-0600)] <heidi__> i just restarted my comp
[09:57:53 CST(-0600)] <heidi__> to make sure there's no weirdness going on
[09:58:42 CST(-0600)] <colinclark> ok
[10:00:00 CST(-0600)] <heidi__> colinclark should i be testing a local svn copy or the build site? i just updated and the local copy "browse" link doesn't do anything.
[10:00:08 CST(-0600)] <jhung> yes, but they don't go into my inbox. I guess the commit on Saturday afternoon?
[10:00:12 CST(-0600)] <justin_o> jhung: these were the changes that i made
[10:00:14 CST(-0600)] <justin_o> http://fluidproject.org/pipermail/commits/2010-November/005817.html
[10:00:20 CST(-0600)] <justin_o> jhung: yep on saturday
[10:00:23 CST(-0600)] <jhung> k
[10:00:30 CST(-0600)] <jhung> justin_o: I'll take a look.
[10:00:35 CST(-0600)] <colinclark> heidi__: Either one is fine, but testing off the daily build is probably easier for us all to coordinate on. mlam, did you rebuild after your latest commit?
[10:00:52 CST(-0600)] <justin_o> jhung: thanks
[10:01:05 CST(-0600)] <mlam> no, i have not
[10:02:46 CST(-0600)] <colinclark> mlam: Do you want me to do it, if you're busy?
[10:02:57 CST(-0600)] <mlam> yes please
[10:04:16 CST(-0600)] <colinclark> k
[10:04:38 CST(-0600)] <heidi__> okay so will continue testing http://build.fluidproject.org/incubator/uploader/FLUID-3722/src/webapp/components/uploader/html/Uploader.html
[10:04:48 CST(-0600)] <colinclark> heidi__: I'm going to rebuild it now, so wait one sec
[10:04:52 CST(-0600)] <heidi__> k
[10:04:54 CST(-0600)] <colinclark> Then you'll be able to test off the latest and greatest
[10:05:28 CST(-0600)] <heidi__> do i have to run my local copy through localhost? not sure why it's not working
[10:07:20 CST(-0600)] <colinclark> heidi__: Nope, you don't
[10:07:25 CST(-0600)] <colinclark> Anything in Firebug?
[10:07:40 CST(-0600)] <colinclark> Need more details
[10:08:36 CST(-0600)] <colinclark> anastasiac: I'm just trying to parse all the Uploader issues you've found
[10:08:42 CST(-0600)] <colinclark> but the first one in JAWS is interesting
[10:08:59 CST(-0600)] <anastasiac> you mean the fact that it doesn't say anything in browse mode?
[10:08:59 CST(-0600)] <colinclark> do you think an application role would improve the experience in JAWS?
[10:09:06 CST(-0600)] <colinclark> and in NVDA?
[10:09:46 CST(-0600)] <anastasiac> good question - I don't really know what effect that has on screen readers, but it's quite possible
[10:10:06 CST(-0600)] <heidi__> i'm running FLUID-3722/src/webapp/demos/uploader/html/uploader.html locally and i don't have the "switch to" links at the top and the browse button doesn't work. the errors i get are "fluid.demands is not a function" and "fluid.uploader.fileQueue is not a function"
[10:10:17 CST(-0600)] <colinclark> anastasiac: The application role should cause a very modern screen reader to switch modes automatically.
[10:10:24 CST(-0600)] <colinclark> heidi__: Seems like a file pathing issue
[10:10:34 CST(-0600)] <colinclark> Did you check out the root of the FLUID-3722 branch?
[10:10:41 CST(-0600)] <anastasiac> colinclark, that would help, I would think
[10:11:01 CST(-0600)] <colinclark> anastasiac: I'm going to try it and commit it--will you test it out?
[10:11:09 CST(-0600)] <anastasiac> certainly!
[10:11:11 CST(-0600)] <colinclark> btw should rebuilding the incubator take as long as it takes?
[10:11:35 CST(-0600)] <colinclark> 4'10"
[10:11:36 CST(-0600)] <colinclark> pretty slow
[10:11:41 CST(-0600)] <colinclark> heidi__: it's rebuilt
[10:11:42 CST(-0600)] <heidi__> i checked out http://source.fluidproject.org/svn/incubator/uploader
[10:11:48 CST(-0600)] <heidi__> okay
[10:11:59 CST(-0600)] <anastasiac> yeah, that is slow - but there's a lot in there
[10:12:30 CST(-0600)] <anastasiac> lots of check-outs of externals
[10:12:38 CST(-0600)] <colinclark> heidi__: Based on the errors you've got, you're clearly not getting the right file dependencies
[10:12:47 CST(-0600)] <heidi__> okay
[10:12:52 CST(-0600)] <colinclark> "fluid.demands is not a function" means that the FluidIOC.js file isn't being found
[10:13:02 CST(-0600)] <heidi__> did i check out the wrong path
[10:13:04 CST(-0600)] <colinclark> "fluid.uploader.fileQueue" means that FileQueue.js isn't being found
[10:13:17 CST(-0600)] <colinclark> heidi__: Well, you could have checked it out one level deeper, but it shouldn't make any difference at all
[10:13:53 CST(-0600)] <heidi__> okay... i'm not sure why then. ill just test build for now
[10:17:15 CST(-0600)] <colinclark> anastasiac: If I commit this application role, will you have to rebuild to test?
[10:17:39 CST(-0600)] <jhung> justin_o: going over the changes now. It looks good! However, I do notice that the Undo Edit link in Rich text editor gets a visited state when clicked. I think if Undo is available, it should never appear as visited.
[10:17:45 CST(-0600)] <anastasiac> colinclark, probably. Or I could check it out onto the test machine
[10:17:57 CST(-0600)] <colinclark> I fear rebuilding again will be disruptive to heidi__
[10:18:08 CST(-0600)] <anastasiac> I can check it out, that's fine
[10:18:18 CST(-0600)] <colinclark> cool, thanks
[10:18:18 CST(-0600)] <heidi__> it's okay
[10:18:21 CST(-0600)] <colinclark> lemme commit, then
[10:20:14 CST(-0600)] <colinclark> Okay, anastasiac, it should be in
[10:20:24 CST(-0600)] <anastasiac> thanks
[10:21:48 CST(-0600)] <heidi__> mlam why would the styling be so different on build site vs checkout. i see on the local copy it has a nicely style "browse files" link
[10:21:51 CST(-0600)] <heidi__> they seem very different
[10:22:17 CST(-0600)] <mlam> heidi__: the html5 uploader isn't in trunk yet
[10:22:27 CST(-0600)] <mlam> so the trunk just has the SWF uploader
[10:22:53 CST(-0600)] <mlam> the html5 Uploader is only in the FLUID-3722 branch in the incubator
[10:22:53 CST(-0600)] <heidi__> mlam i checked out the incubator
[10:23:02 CST(-0600)] <heidi__> yeah that's the one i mean
[10:23:22 CST(-0600)] <mlam> which html file did you open? in the components directory or the demo?
[10:23:43 CST(-0600)] <heidi__> ah, that's my issue. thanks mike
[10:23:46 CST(-0600)] <heidi__> sorry to distract
[10:23:49 CST(-0600)] <mlam> np (smile)
[10:24:42 CST(-0600)] <colinclark> We need to toss the Uploader demo as soon as this AEGIS testing is done
[10:24:52 CST(-0600)] <colinclark> It's a complete cut and paste job of the actual Uploader template
[10:25:02 CST(-0600)] <colinclark> which causes exactly the sort of problems heidi__ just encountered
[10:26:30 CST(-0600)] <justin_o> jhung: okay... do you think this could wait till after the release to address that
[10:27:43 CST(-0600)] <jhung> justin_o: yes. I'll put it into a jira.
[10:27:51 CST(-0600)] <justin_o> jhung: thanks
[10:34:36 CST(-0600)] <heidi__> stand up ya'll !
[10:35:35 CST(-0600)] <jessm> anastasiac: colinclark justin_o golam mlam ^
[10:35:45 CST(-0600)] <jessm> oh, everyone!
[10:37:36 CST(-0600)] <colinclark> we're on our way,
[10:37:38 CST(-0600)] <colinclark> sorry
[10:37:42 CST(-0600)] <colinclark> most of the laptop users are in another meeting (wink)
[10:54:49 CST(-0600)] <heidi__> mlam how do you delete a file from list w/ keyboard
[10:57:33 CST(-0600)] <mlam> the uploader listens for the delete button
[10:57:48 CST(-0600)] <jameswy> colinclark, mlam, justin_o: browse button solution: http://www.shauninman.com/archive/2007/09/10/styling_file_inputs_with_css_and_the_dom
[10:57:54 CST(-0600)] <heidi__> so how do i ..use it?
[10:58:33 CST(-0600)] <mlam> if you tab through the component, once a file has gained focus, hit hte delete key, and the file should disappear
[10:58:50 CST(-0600)] <heidi__> ah, delete key.
[10:58:54 CST(-0600)] <jameswy> justin_o: http://www.shauninman.com/archive/2007/09/10/styling_file_inputs_with_css_and_the_dom
[10:59:07 CST(-0600)] <heidi__> is that standard or do we need to incl instructions for that
[10:59:24 CST(-0600)] <jameswy> colinclark, justin_o, mlam: If I understand the solution correctly, it's just making the original button invisible, and putting in an image underneath/on top.
[10:59:40 CST(-0600)] <jameswy> So when you click it, you're just clicking the invisible button.
[10:59:44 CST(-0600)] <colinclark> heidi__: There's a thread on the mailing list where I mention that
[10:59:57 CST(-0600)] <colinclark> I think we need to provide additional instructions on how to do that
[11:00:04 CST(-0600)] <heidi__> agreed!
[11:00:16 CST(-0600)] <colinclark> But the interaction is consistent with how you'd do it on the desktop (at least in Windows)
[11:00:38 CST(-0600)] <heidi__> yeah. i was expecting to be able to focus the x button
[11:00:43 CST(-0600)] <heidi__> and hit enter
[11:02:19 CST(-0600)] <colinclark> heidi__: We struggled through that interaction
[11:02:37 CST(-0600)] <colinclark> hitting tab would probably be equally weird, given that you're in the midst of an arrow-key style navigation
[11:03:00 CST(-0600)] <colinclark> I don't think we'll redesign it for this release, but it's definitely something we should consider
[11:03:14 CST(-0600)] <heidi__> i think clean instructions are enough
[11:03:23 CST(-0600)] <heidi__> the delete thing is slick i think
[11:03:35 CST(-0600)] <colinclark> I think it should at least be consistent across platforms
[11:03:42 CST(-0600)] <heidi__> ya
[11:03:44 CST(-0600)] <colinclark> which means we should also bind it to Command-Delete for the Mac
[11:03:48 CST(-0600)] <colinclark> that should be really easy to do
[11:09:47 CST(-0600)] <colinclark> heidi__: Have you found any other interesting Uploader bugs so far, aside from the button-related issues that jameswy and justin_o are working on?
[11:10:25 CST(-0600)] <heidi__> mmm, not yet. i restarted my comp and the voiceover safari dying thing isn't happening so that's good, just going through qa list
[11:10:44 CST(-0600)] <colinclark> (smile)
[11:10:46 CST(-0600)] <colinclark> cool
[11:10:53 CST(-0600)] <heidi__> j/j fix will incl "add more" right
[11:11:09 CST(-0600)] <colinclark> yep
[11:11:12 CST(-0600)] <heidi__> cool
[11:11:17 CST(-0600)] <colinclark> The solution is really nasty, but it's weirdly familiar
[11:11:18 CST(-0600)] <heidi__> yeah this thing is slick
[11:11:37 CST(-0600)] <heidi__> all the green bars zooming by i was like WOAH NICE
[11:11:40 CST(-0600)] <colinclark> lol
[11:12:45 CST(-0600)] <colinclark> jhung: Do you have time for a couple of quick Pager questions?
[11:12:52 CST(-0600)] <heidi__> so the check marks and green bar = success. will there be any actual text saying "success" ?
[11:13:27 CST(-0600)] <colinclark> heidi__: Check that thread on the list where anastasiac has been reporting her screen reader results
[11:13:33 CST(-0600)] <colinclark> But I believe it does say "uploaded"
[11:13:35 CST(-0600)] <heidi__> i will
[11:13:37 CST(-0600)] <colinclark> or something to that effect
[11:13:41 CST(-0600)] <heidi__> but for non-screen reader user?
[11:13:47 CST(-0600)] <colinclark> I'm pretty sure it's alt text on that checkmark
[11:14:01 CST(-0600)] <colinclark> For a non-screen reader user, they get several layers of information:
[11:14:10 CST(-0600)] <colinclark> (i.e. for a sighted user)
[11:14:14 CST(-0600)] <colinclark> checkmark
[11:14:15 CST(-0600)] <colinclark> green bar
[11:14:28 CST(-0600)] <colinclark> and the information in the total progress area where it tells you how many things were uploaded
[11:15:13 CST(-0600)] <heidi__> ah
[11:15:15 CST(-0600)] <heidi__> okay
[11:15:37 CST(-0600)] <colinclark> almondkey designed the interaction, but I don't see her in the channel
[11:19:43 CST(-0600)] <heidi__> colinclark stop upload doesn't work yet?
[11:19:52 CST(-0600)] <colinclark> Well, it does...
[11:19:53 CST(-0600)] <colinclark> but it's odd
[11:19:58 CST(-0600)] <colinclark> This was due to a bug in SWFUpload
[11:20:01 CST(-0600)] <heidi__> mm, how do i stop it?
[11:20:06 CST(-0600)] <colinclark> We should be able to fix it for HTML 5, but haven't
[11:20:10 CST(-0600)] <colinclark> You stop it by clicking stop
[11:20:16 CST(-0600)] <heidi__> doesn't work
[11:20:17 CST(-0600)] <colinclark> The caveat is that it waits for the current file to finish
[11:20:18 CST(-0600)] <colinclark> it does
[11:20:21 CST(-0600)] <colinclark> (tongue)
[11:20:29 CST(-0600)] <heidi__> uh.. oh
[11:20:30 CST(-0600)] <colinclark> The problem is that SWFUpload just didn't support stopping files
[11:20:34 CST(-0600)] <heidi__> so i can't stop the current file
[11:20:38 CST(-0600)] <colinclark> exactly
[11:20:41 CST(-0600)] <heidi__> ack
[11:20:42 CST(-0600)] <colinclark> you have to wait for the current one to finish
[11:20:45 CST(-0600)] <colinclark> totally sucks
[11:20:51 CST(-0600)] <heidi__> yeah
[11:21:00 CST(-0600)] <colinclark> it's already a known issue, fortunately
[11:21:03 CST(-0600)] <heidi__> k
[11:21:18 CST(-0600)] <jhung> colinclark: sorry was grabbing lunch. What's up?
[11:21:54 CST(-0600)] <heidi__> colinclark yeah it's esp weird if it's the last file you want to stop. it appears that the stop doesn't work at all.
[11:22:16 CST(-0600)] <colinclark> heidi__: Yep
[11:22:27 CST(-0600)] <colinclark> it's a fairly simple fix, but brings up an issue I want to ask jameswy
[11:22:28 CST(-0600)] <colinclark> about
[11:22:57 CST(-0600)] <colinclark> So, if SWFUpload doesn't support stopping the current file and HTML 5 does (we think)
[11:23:06 CST(-0600)] <colinclark> is it okay for the demo mode to make it look like SWFUpload does support it, or not?
[11:23:14 CST(-0600)] <colinclark> jhung: Couple of things
[11:23:33 CST(-0600)] <anastasiac> colinclark, mlam: I've filed two JIRAs describing the screen reader testing problems I found with the Uploader: FLUID-3835 and FLUID-3836. Maybe justin_o can comment on whether or not they're bug-parade worthy
[11:23:39 CST(-0600)] <colinclark> anastasiac: wicked!
[11:23:50 CST(-0600)] <colinclark> Pager demo, running locally, requires Firefox's local XHR setting to be on, right, jhung?
[11:23:54 CST(-0600)] <mlam> ok, thanks
[11:24:05 CST(-0600)] <colinclark> I never set that setting, and the demo doesn't work for me
[11:24:26 CST(-0600)] <colinclark> 2. I think the demo data is probably really confusing, with all of the numbered students, etc.
[11:24:33 CST(-0600)] <colinclark> Any chance we could make it nicer?
[11:24:43 CST(-0600)] <jhung> colinclark: I don't recall having to set that.
[11:25:29 CST(-0600)] <jhung> colinclark: checking my settings now... it could be the case.
[11:25:55 CST(-0600)] <colinclark> brb
[11:26:02 CST(-0600)] <heidi__> the changing background colour of the browse files bar makes me feel weird/confused. i want it to stay the same since it's my info/work area. i feel like a progress bar should be separate...
[11:26:29 CST(-0600)] <heidi__> but there are prob reasons i don't know as to why it's like that. just have never seen something like that before.
[11:33:18 CST(-0600)] <heidi__> mlam ... howcome the header of the uploads table and the uploads table itself are... two different tables?
[11:33:39 CST(-0600)] <jhung> colinclark: I noticed that there are some uneeded jquery libraries that are not needed in the HTML file - jquery.ui and jquery.bgiframe. Not sure if that would help the issue.
[11:34:51 CST(-0600)] <mlam> heidi__: to tell you the truth, i have no idea what the intentions were on that. i haven't looked much into the html
[11:35:52 CST(-0600)] <heidi__> mlam who wrote the html?
[11:36:07 CST(-0600)] <jhung> colinclark: incidentally, we'll probably need a code review on the pager demo since a lot has changed, and it was a modification of the Sakai integration demo.
[11:36:42 CST(-0600)] <mlam> i'm not sure...i can check the svn log for you really quickly
[11:37:28 CST(-0600)] <heidi__> mlam sure! just wondering if there's a reason its like that
[11:37:39 CST(-0600)] <heidi__> cos normally table headers are useful for.. the actual data table
[11:37:47 CST(-0600)] <mlam> heidi__: looks like Eli was the original creator
[11:38:03 CST(-0600)] <heidi__> ok
[11:38:15 CST(-0600)] <heidi__> colinclark any thoughts on why it's like that?
[11:38:31 CST(-0600)] <mlam> heidi__: you think we should just collapse the 2 tables together?
[11:39:06 CST(-0600)] <heidi__> there are 3 - the footer is a separate table as well
[11:39:16 CST(-0600)] <heidi__> it's pretty ... odd. not sure why one would do it this way
[11:39:37 CST(-0600)] <heidi__> footer shouldn't be table at all
[11:40:47 CST(-0600)] <heidi__> guessing colin's lunching
[11:40:53 CST(-0600)] <mlam> yah, i think so
[11:41:08 CST(-0600)] <justin_o> heidi__, mlam: i'm pretty sure he's out
[11:41:26 CST(-0600)] <mlam> heidi__: i think if we can simplify the html, we should
[11:41:42 CST(-0600)] <heidi__> yeah it's
[11:41:43 CST(-0600)] <heidi__> yeah
[11:44:38 CST(-0600)] <heidi__> mlam i can help with the html, i just don't want to break stuff this late in the game if the js is dependent on this structure
[11:45:04 CST(-0600)] <mlam> the feeling's mutual (smile)
[11:46:07 CST(-0600)] <colinclark> heidi__, mlam I'm back
[11:46:10 CST(-0600)] <colinclark> What was the question?
[11:46:20 CST(-0600)] <heidi__> hey colin, wb
[11:46:30 CST(-0600)] <heidi__> just wondering why there are 3 tables in the uploader html
[11:46:59 CST(-0600)] <colinclark> ah, just catching up
[11:47:03 CST(-0600)] <colinclark> I like your quote:
[11:47:03 CST(-0600)] <heidi__> i think the table header / data should = 1 table and the footer shouldn't be one at all
[11:47:06 CST(-0600)] <colinclark> "It's pretty... odd"
[11:47:20 CST(-0600)] <heidi__> mmhm
[11:47:23 CST(-0600)] <colinclark> You'll often see elicochran in the channel
[11:47:33 CST(-0600)] <colinclark> That markup is his baby, so he can probably explain the motivations
[11:47:49 CST(-0600)] <heidi__> i don't think it's the best way of doing it
[11:47:52 CST(-0600)] <colinclark> I think it's too late in the game to change it before the AEGIS testing unless they're actively complicated our AT testing
[11:48:22 CST(-0600)] <colinclark> heidi__: I agree, but if possible we should still understand the author's intentions before diving in to change it
[11:48:52 CST(-0600)] <heidi__> data tables should have table headers (smile)
[11:48:59 CST(-0600)] <heidi__> but maybe we can check in with eli
[11:49:03 CST(-0600)] <colinclark> indeed
[11:49:17 CST(-0600)] <colinclark> jacob was pretty convinced he could improve it substantially
[11:49:36 CST(-0600)] <heidi__> it's just that the first table has the caption "file upload queue"
[11:49:38 CST(-0600)] <heidi__> and it's not
[11:49:48 CST(-0600)] <heidi__> the second table is, but it doesn't have a caption
[11:50:04 CST(-0600)] <heidi__> there are semantic issues that would prob confuse a screen reader user
[11:50:10 CST(-0600)] <heidi__> i dunno. just sayin
[11:50:29 CST(-0600)] <heidi__> and the footer isn't tabular data... shouldn't be a table at all.
[11:51:07 CST(-0600)] <colinclark> I agree with you, heidi__
[11:51:07 CST(-0600)] <jessm> heidi__: file some jira and email eli (smile)
[11:51:14 CST(-0600)] <heidi__> will do!
[11:51:15 CST(-0600)] <colinclark> and cc the list!
[11:51:26 CST(-0600)] <jessm> he's on the list so that's a nice way to email
[11:51:35 CST(-0600)] <colinclark> +1
[11:51:49 CST(-0600)] <heidi__> too bad there's no time to fix
[11:52:09 CST(-0600)] <colinclark> If we find it's really affecting AT users, we can fix it earlier
[11:52:10 CST(-0600)] <jessm> heidi__: there's always time (smile) time after, time future, etc.
[11:52:15 CST(-0600)] <colinclark> otherwise it's definite candidate for 1.4
[12:00:41 CST(-0600)] <justin_o> anastasiac: i added those last two uploader jiras you filed to bug parade
[12:00:53 CST(-0600)] <anastasiac> Justin_o, cool
[12:01:11 CST(-0600)] <justin_o> cindyli: i've added FLUID-3443 to bug parade as well
[12:01:34 CST(-0600)] <cindyli> justin_o: ok
[12:02:10 CST(-0600)] <colinclark> mlam: are you in a position where you can try to fix those two jiras?
[12:02:32 CST(-0600)] <colinclark> Is FLUID-3443 the "D Button" issue?
[12:02:36 CST(-0600)] <colinclark> Sounds so funny!
[12:02:59 CST(-0600)] <mlam> I was just about to reply to email in response to anastasiac's email
[12:03:55 CST(-0600)] <heidi__> is anyone working on http://issues.fluidproject.org/browse/FLUID-3664 ?
[12:04:59 CST(-0600)] <colinclark> heidi__: This one is a mishmash of issues
[12:05:12 CST(-0600)] <heidi__> yeah lots !
[12:05:14 CST(-0600)] <colinclark> anastasiac has started to split them out based on the actual behaviour of the HTML 5 version
[12:05:24 CST(-0600)] <heidi__> cool
[12:05:27 CST(-0600)] <colinclark> These were filed against the Flash version, so many of them aren't relevant
[12:07:18 CST(-0600)] <anastasiac> heidi_, I'll test the issues in 3664 against the HTML5 version, and try to help sort things out
[12:07:48 CST(-0600)] <heidi__> yeah i just quickly went through with voiceover and there's a lot of missing feedback
[12:07:52 CST(-0600)] <heidi__> hopefully we can add that in
[12:08:36 CST(-0600)] <colinclark> Hey heidi__, do you have a VM setup with JAWS or NVDA?
[12:09:02 CST(-0600)] <heidi__> i have a VM, but no jaws. i can install nvda on it tho
[12:09:16 CST(-0600)] <colinclark> that would be cool
[12:09:25 CST(-0600)] <colinclark> Safari 5's support for ARIA still isn't really awesome
[12:09:42 CST(-0600)] <colinclark> so it's probably best to cross-check your findings with a setup that actually works (wink)
[12:09:42 CST(-0600)] <heidi__> it works okay for basic stuff
[12:09:50 CST(-0600)] <colinclark> much as it pains me to say about VoiceOver
[12:10:04 CST(-0600)] <colinclark> Sure, it's when it doesn't work that is the problem
[12:10:19 CST(-0600)] <colinclark> hard to ever know if it's the environment or the code
[12:11:46 CST(-0600)] <colinclark> jhung_afk has disappeared, but in case anyone else is working on Pager, I'm going to dive into some fixes for the demo
[12:12:02 CST(-0600)] <colinclark> 1. It doesn't work locally in Firefox because it tries to load data via Ajax against the local file system
[12:12:17 CST(-0600)] <colinclark> 2. The demo code could use some tidying up
[12:12:40 CST(-0600)] <colinclark> I think there's still maybe some styling that could be tweaked...
[12:13:25 CST(-0600)] <colinclark> focus indicator as well as the "last" page in the pager bar, which wraps to a new line and looks kind of funny
[12:23:21 CST(-0600)] <jameswy> colinclark: If SWFUpload doesn't support stopping the current file and HTML5 does, the demo mode should reflect that; we want to be transparent to the user (in short: not ok for the demo mode to make it look like SWFUpload supports stopping current file)
[12:23:46 CST(-0600)] <colinclark> jameswy: Seems a hundred percent sensible
[12:27:09 CST(-0600)] <colinclark> Like anything sensible, it's going to take a bit longer to support that
[12:27:19 CST(-0600)] <colinclark> since we'll have to parameterize that start/stop logic
[12:32:35 CST(-0600)] <colinclark> jhung: hey!
[12:32:36 CST(-0600)] <colinclark> You're back
[12:35:40 CST(-0600)] <jhung> hey colinclark. yep. reading log now.
[12:35:49 CST(-0600)] <colinclark> So, I'm diving into Pager
[12:35:53 CST(-0600)] <colinclark> but I'm wondering if you could give me a hand
[12:36:01 CST(-0600)] <colinclark> I think the data for this example is just totally out of control (wink)
[12:36:02 CST(-0600)] <jhung> colinclark: sure
[12:36:17 CST(-0600)] <colinclark> Any chance you could pare it down to only the data we actually are rendering
[12:36:32 CST(-0600)] <colinclark> and give people real names rather than the numbered students, which are pretty confusing
[12:36:56 CST(-0600)] <jhung> colinclark: you realize what you
[12:37:00 CST(-0600)] <jhung> re asking, right?
[12:37:04 CST(-0600)] <jhung> (big grin)
[12:37:10 CST(-0600)] <colinclark> I'm staring at the data file right now (tongue)
[12:37:18 CST(-0600)] <colinclark> Put it this way:
[12:37:31 CST(-0600)] <colinclark> Go to the daily build, look at the Pager demo, and click the "data" tab
[12:37:33 CST(-0600)] <colinclark> Scary, no?
[12:37:43 CST(-0600)] <jhung> colinclark: yeah.
[12:37:48 CST(-0600)] <jhung> I've seen it.
[12:38:11 CST(-0600)] <jhung> colinclark: I better start that now... maybe I'll be done by Saturday. (wink)
[12:38:36 CST(-0600)] <colinclark> lol
[12:38:39 CST(-0600)] <colinclark> Do what you can
[12:38:54 CST(-0600)] <colinclark> I'm just about to commit a change to the data file to make it work in FF without the fancy setting turned on
[12:38:56 CST(-0600)] <colinclark> so give me five minutes
[12:38:58 CST(-0600)] <heidi__> colinclark i think eli split up the table so he could make only the data scrollable?
[12:39:05 CST(-0600)] <colinclark> That's part of the reason, yes
[12:39:23 CST(-0600)] <jhung> colinclark: sure. let me know when you're done.
[12:40:53 CST(-0600)] <colinclark> jhung: I know it's drag... I found that when demoing the pager, the numbered students caused people to think that the sorting wasn't really working properly
[12:44:11 CST(-0600)] <jhung> colinclark: it's really no problem. Just wanted to give you a hard time. (wink)
[12:44:16 CST(-0600)] <colinclark> lol
[12:44:19 CST(-0600)] <colinclark> I don't blame you
[12:49:43 CST(-0600)] <anastasiac> colinclark, heidi_: I've reviewed the issues described in FLUID-3664, to see if any are present in the HTML5 version in FF with NVDA:
[12:49:48 CST(-0600)] <anastasiac> Most of the problems described there with NVDA are not present. The ones that are are already covered by JIRAs:
[12:49:52 CST(-0600)] <anastasiac> FLUID-1829: Need instructions telling Jaws users to turn cursor off
[12:49:52 CST(-0600)] <anastasiac> FLUID-2849: Single-file version switches to multi-file version automatically
[12:49:53 CST(-0600)] <anastasiac> FLUID-3835: No indication file queue is empty
[12:50:17 CST(-0600)] <colinclark> mlam: You're working on a few of these, right? ^
[12:51:01 CST(-0600)] <colinclark> jhung: A little stuck, but I'm nearly done
[12:51:20 CST(-0600)] <jhung> colinclark: no prob. I'm starting work on the data and just will copy over when done.
[12:52:08 CST(-0600)] <mlam> yes, i'm working on FLUID-3835
[12:52:15 CST(-0600)] <mlam> and FLUID-3834
[12:52:38 CST(-0600)] <mlam> sorry, not FLUID-3834, but FLUID-3836
[13:00:28 CST(-0600)] <heidi__> anastasiac is there any feedback when i file has been added or deleted from queue
[13:01:10 CST(-0600)] <anastasiac> heidi__, nothing direct. Focus moves to the upload button, which is read, so you know the addition has been completed, but there's no listing of what files were added, or anything like that
[13:01:29 CST(-0600)] <heidi__> right, you have to go back to check the list
[13:01:36 CST(-0600)] <anastasiac> for deletion, it's different
[13:01:53 CST(-0600)] <anastasiac> you're on a file when you delete it, so focus moves to the next file in the list after delete, and that next file name is ready
[13:02:16 CST(-0600)] <heidi__> but there's no like "file deleted" feedback or anything
[13:03:05 CST(-0600)] <heidi__> and after hitting upload, any feedback?
[13:03:36 CST(-0600)] <anastasiac> heidi__, you do hear progress indications: beeps for NVDA, "progress is x% complete" or Jaws
[13:03:44 CST(-0600)] <anastasiac> in NVDA, completion = silence
[13:04:00 CST(-0600)] <anastasiac> in Jaws, completion - "progress is complete" if the timing is right
[13:04:12 CST(-0600)] <colinclark> It seems like we may want to make the text in the footer a live region. That would provide users with additional feedback about addition and removal of files
[13:04:15 CST(-0600)] <heidi__> okay
[13:04:19 CST(-0600)] <colinclark> as well as when uploads start
[13:04:26 CST(-0600)] <anastasiac> that would probably help
[13:04:52 CST(-0600)] <colinclark> I can give that a try shortly
[13:05:03 CST(-0600)] <colinclark> jhung: I'm totally stuck on this demo
[13:06:16 CST(-0600)] <jhung> colinclark: anything I can help with?
[13:06:22 CST(-0600)] <colinclark> i dunno
[13:06:29 CST(-0600)] <colinclark> unless you've got some magic
[13:06:48 CST(-0600)] <colinclark> Switching this data file to load via JSONP is causing strange errors in Pager
[13:08:38 CST(-0600)] <jhung> colinclark: and it's only a problem with this particular data or any file?
[13:09:16 CST(-0600)] <colinclark> I don't know yet
[13:09:21 CST(-0600)] <colinclark> you're right to suggest I try a simpler file
[13:09:28 CST(-0600)] <colinclark> I don't actually think this file is valid JSON
[13:09:41 CST(-0600)] <colinclark> since there are apostrophes in someone's name that isn't escaped
[13:10:42 CST(-0600)] <jhung> ah. Okay. I can fix those as part of my work anyway.
[13:10:59 CST(-0600)] <colinclark> give it a try, jhung, though that isn't my problem
[13:11:16 CST(-0600)] <colinclark> I think I'm going to leave it for a bit and fix some of these Uploader a11y issues anastasiac and heidi__ are finding
[13:11:22 CST(-0600)] <jhung> sure.
[13:11:31 CST(-0600)] <colinclark> Once you've got the data looking awesome, I can come back to it
[13:11:33 CST(-0600)] <heidi__> awesome colin
[13:43:19 CST(-0600)] <mlam> colinclark: almost finished FLUID-3835 (ie. empty file queue) then i'll move to FLUID-3836
[13:43:27 CST(-0600)] <colinclark> cool
[13:44:38 CST(-0600)] <colinclark> I'm working on adding live regions to the Uploader to give better feedback when files are added and removed from the queue
[13:44:43 CST(-0600)] <colinclark> I don't know if there's a relevant JIRA yet or not
[13:44:49 CST(-0600)] <colinclark> Do you know, anastasiac or heidi__?
[13:44:59 CST(-0600)] <heidi__> colin is the 20mb upload test still used?
[13:45:07 CST(-0600)] <heidi__> i don't colin
[13:45:13 CST(-0600)] <anastasiac> let me check
[13:46:20 CST(-0600)] <anastasiac> colinclark, it doesn't seem so - do you want me to file somethign?
[13:46:29 CST(-0600)] <colinclark> that'd be wicked
[13:46:35 CST(-0600)] <colinclark> heidi__: ask justin_o
[13:47:54 CST(-0600)] <heidi__> justin_o is there a max file upload feature with uploader?
[13:48:35 CST(-0600)] <anastasiac> colinclark: http://issues.fluidproject.org/browse/FLUID-3838
[13:58:55 CST(-0600)] <colinclark> Okay, so it looks like Firefox 3.5 definitively won't support multiple file uploads with HTML 5
[13:59:21 CST(-0600)] <colinclark> It does include support for progress events in XHR, but it doesn't support the < input type="file" multiple="" > element at all
[13:59:49 CST(-0600)] <colinclark> Since it's not an A-Grade browser anymore, it's not the end of the world. I imagine the vast majority of 3.5 users have already upgraded to 3.6
[14:02:47 CST(-0600)] <justin_o> heidi__: there was for the flash version at least
[14:02:56 CST(-0600)] <justin_o> which i believe was set to 20mb but was configurable
[14:03:07 CST(-0600)] <heidi__> justin_o anything similar for html5 ?
[14:03:18 CST(-0600)] <justin_o> heidi__: that i'm not sure of
[14:03:48 CST(-0600)] <heidi__> mlam?
[14:04:49 CST(-0600)] <mlam> heidi__: with xhr, you can set a max file size for uploading if you use the GET request
[14:05:02 CST(-0600)] <mlam> however if you use POST, there is no file size limit
[14:20:25 CST(-0600)] <heidi__> golam are the pager a11y fixes ready?
[14:20:44 CST(-0600)] <golam> heidi__: not yet
[14:20:54 CST(-0600)] <heidi__> k
[14:21:47 CST(-0600)] <jameswy> colinclark, justin_o, mlam, heidi__: So, update on the HTML5 file upload button...
[14:21:48 CST(-0600)] <jameswy> The trick that's circulating on the 'net is to make the real upload button + field transparent, and plop your custom button image or button image + text field behind it.
[14:21:48 CST(-0600)] <jameswy> Since Safari's implementation of file upload doesn't include a text field, we'd be going with just a custom button image to maintain experience consistency across browsers.
[14:21:48 CST(-0600)] <jameswy> The problem is that the straightforward path to the solution doesn't work: Firefox forcibly puts its upload button + text field out there in full width, regardless of the width you tell it to use. It's very stubborn that way, and nothing seems to change its mind.
[14:21:48 CST(-0600)] <jameswy> There's a workaround to it, but it'd require some time to implement: it's essentially to dynamically place and position the transparent real button only when your mouse is hovering over the fake button.
[14:23:38 CST(-0600)] <jameswy> Do we want to: a) go down said route, b) continue trying to find another solution, c) use the broken solution, or d) just use the browser-specific file upload button + text field?
[14:24:28 CST(-0600)] <colinclark> sorry, i missed the earlier discussion about file size limits for the HTML 5 version of the uploader
[14:24:34 CST(-0600)] <colinclark> mlam: That's something we'll have to implement ourselves
[14:24:47 CST(-0600)] <colinclark> heidi__: And it's something that we haven't yet had an opportunity to do.
[14:24:58 CST(-0600)] <jessm> jameswy: you might need to coordinate with mlam for a minute?
[14:26:21 CST(-0600)] <jameswy> mlam: Whenever you have a minute, let's sync up on uploader
[14:26:50 CST(-0600)] <colinclark> jameswy: a) or d) strike me as the preferable options
[14:27:12 CST(-0600)] <mlam> jameswy: gimme 2 mins?
[14:27:16 CST(-0600)] <colinclark> Where a) is defined as "work within the current design wireframes, and do everything we can to make it work within that context"
[14:27:17 CST(-0600)] <colinclark> an
[14:27:52 CST(-0600)] <colinclark> and where b) is defined as "do a partial redesign that can gracefully accommodate the native (and varying) look and feel of the browser's control"
[14:28:04 CST(-0600)] <colinclark> jameswy: do you know what I mean?
[14:28:31 CST(-0600)] <heidi__> d) sounds less painful
[14:28:42 CST(-0600)] <heidi__> guess it depends on time?
[14:29:47 CST(-0600)] <jameswy> colinclark: Sounds reasonable. I think option a) might take more time than we have right now, but I'm guessing that's more up to justin_o or mlam. I'm going to spend another half-hour on b), and option d) is definitely our fall-back.
[14:32:22 CST(-0600)] <jhung> colinclark: new data put into SVN. Let me know how it goes.
[14:32:25 CST(-0600)] <heidi__> uploader killing my safari w/ VO again. hmm.
[14:35:51 CST(-0600)] <heidi__> ill assume it's a VO thing for now
[14:36:06 CST(-0600)] <colinclark> jhung: thanks, i'll take a loo
[14:36:07 CST(-0600)] <colinclark> k
[14:36:13 CST(-0600)] <colinclark> look
[14:36:14 CST(-0600)] <colinclark> not loo
[14:36:16 CST(-0600)] <colinclark> sheesh
[14:36:27 CST(-0600)] <jhung> lol
[14:36:39 CST(-0600)] <jhung> Thanks for clarifying. (smile)
[14:37:37 CST(-0600)] <heidi__> justin_o anything i can help with?
[14:41:19 CST(-0600)] <heidi__> mlam only other uploader thing was hover style on inactive delete buttons. other stuff is already on your radar (browse button stuff, stop upload on current file)
[14:41:26 CST(-0600)] <heidi__> ^super minor
[14:42:03 CST(-0600)] <heidi__> justin_o are there other QA tests i can run through that might help?
[14:42:48 CST(-0600)] <heidi__> has anyone tested uploader on IE8 ?
[14:43:18 CST(-0600)] <colinclark> heidi__: The version of Uploader we're testing is only the HTML 5 version, not the Flash version
[14:43:24 CST(-0600)] <colinclark> and IE 8 doesn't support HTML 5
[14:43:33 CST(-0600)] <heidi__> right ok
[14:43:56 CST(-0600)] <colinclark> so we're not going to sweat IE 8 until after the AEGIS testing
[14:44:21 CST(-0600)] <colinclark> So the plan with the Uploader for its release is to have a kind of tiered approach to graceful degradation
[14:44:29 CST(-0600)] <colinclark> 1. We check if the browser supports HTML 5 multi uploads
[14:44:34 CST(-0600)] <colinclark> 2. If not, we check if Flash 10 is present
[14:44:46 CST(-0600)] <colinclark> 3. If not, we deliver the old fashioned single-file uploader
[14:45:08 CST(-0600)] <heidi__> ok.
[14:55:31 CST(-0600)] <colinclark> hey, do we know that the Grid Reorderer demo is borked on the daily build?
[14:56:43 CST(-0600)] <jhung> ah. Same problem as it was in Pager. Typo I think in one of the paths.
[14:56:48 CST(-0600)] <jhung> ^colinclark
[14:56:54 CST(-0600)] <colinclark> jhung: cool
[14:57:00 CST(-0600)] <jhung> checking it out now.
[14:57:01 CST(-0600)] <colinclark> any chance you have a sec to fix it for me?
[14:57:03 CST(-0600)] <colinclark> thanks
[15:00:01 CST(-0600)] <anastasiac> jhung, the path to the new FluidDocument file is wrong in the demo
[15:00:10 CST(-0600)] <jhung> anastasiac: yeah
[15:00:22 CST(-0600)] <jhung> anastasiac: I
[15:00:30 CST(-0600)] <jhung> 'm going to check the other demos now.
[15:00:47 CST(-0600)] <anastasiac> thanks
[15:01:10 CST(-0600)] <justin_o> heidi__: sorry i missed your message
[15:01:17 CST(-0600)] <justin_o> are you still looking for stuff to test
[15:01:27 CST(-0600)] <heidi__> justin_o ya
[15:01:55 CST(-0600)] <justin_o> what version of mac os do you have?
[15:02:27 CST(-0600)] <heidi__> i tested 10.6 ff and safari and am quickly running through 10.5 safari now
[15:02:56 CST(-0600)] <heidi__> i also have win xp
[15:03:24 CST(-0600)] <justin_o> heidi__: if you have 10.5 cindyli1 might be able to use your help to do a quick builder test
[15:03:34 CST(-0600)] <heidi__> sure np - cindy?
[15:03:37 CST(-0600)] <justin_o> heidi__: if you're done with the a11y testing that is
[15:04:01 CST(-0600)] <heidi__> i'd like to test pager once golam is done, but not ready yet. anything else justin?
[15:04:15 CST(-0600)] <heidi__> cindyli url to test?
[15:04:23 CST(-0600)] <cindyli1> heidi__: that's good. thx guys. http://builder.fluidproject.org/
[15:04:48 CST(-0600)] <heidi__> is there a similar QA breakdown for build justin_o
[15:04:52 CST(-0600)] <heidi__> or pager
[15:05:16 CST(-0600)] <cindyli1> heidi__: test plan: http://wiki.fluidproject.org/display/fluid/Infusion+Builder+QA+Test+Plan
[15:05:29 CST(-0600)] <heidi__> thanks cindy!
[15:05:36 CST(-0600)] <cindyli1> heidi__: thank YOU!
[15:39:22 CST(-0600)] <jessm> justin_o: ping
[15:39:37 CST(-0600)] * jhung testing Layout Reorderer