fluid-work IRC Logs-2010-12-14

Owned by Fluid IRC Log Bot
Last updated: Dec 14, 2010Version comment
35 min read

[08:50:43 CST(-0600)] <Justin_o> mlam: hello.. what are you up to today
[08:51:10 CST(-0600)] <mlam> hey Justin_o, getting the image gallery back up right now so that i can finish testing up the uploader
[08:51:20 CST(-0600)] <mlam> you need me to do anything?
[08:54:45 CST(-0600)] <Justin_o> mlam: i think you can do that first.. please let me know when you're all finished
[08:55:00 CST(-0600)] <mlam> i'll be finished in about 15 mins
[09:02:33 CST(-0600)] <mlam> Justin_o: would you happen to have a 10-20MB file lying around?
[09:03:04 CST(-0600)] <Justin_o> mlam: of anytype?
[09:03:09 CST(-0600)] <Justin_o> any type
[09:03:14 CST(-0600)] <mlam> yah
[09:04:12 CST(-0600)] <jessm> Justin_o: how's everything?
[09:04:28 CST(-0600)] <Justin_o> mlam: i have a 50mb one i think
[09:04:47 CST(-0600)] <Justin_o> jessm: it's okay... heidi posted the patch
[09:04:57 CST(-0600)] <Justin_o> just need to get that reviewed and committed now
[09:05:05 CST(-0600)] <Justin_o> antranig's patches aren't quite working
[09:05:19 CST(-0600)] <jessm> Justin_o: any available reviewers for heidi's patch?
[09:06:17 CST(-0600)] <Justin_o> jessm: she was hoping that jhung would take a look at it, and I think one other set of eyes as well.. since I think he worked on it too..
[09:06:30 CST(-0600)] <Justin_o> not yet sure who that other set of eyes would be
[09:06:32 CST(-0600)] <jessm> jhung: what are you up to?
[09:07:33 CST(-0600)] <jhung> jessm: Was just checking up on some Decapod stuff. I can take a look at Heidi's patch.
[09:08:04 CST(-0600)] <jessm> jhung: Decapod?!
[09:09:17 CST(-0600)] <jhung> jessm: armin sent me a link with some nifty DIY scanners similar to Book Liberator. I was taking a look at how some of them were designed.
[09:09:18 CST(-0600)] <jessm> jhung: please have a look at heidi's patch
[09:09:23 CST(-0600)] <jhung> jessm: yup
[09:09:32 CST(-0600)] <jessm> jhung: thanks
[09:11:46 CST(-0600)] <heidi> thanks jhung !
[09:15:56 CST(-0600)] <Justin_o> jhung: when you are done and if you are happy about it can you let me know... i think you probably did some work on that as well so it might be good to get a quick second look and then commit
[09:16:16 CST(-0600)] <jhung> justin_o: will do.
[09:17:04 CST(-0600)] <Justin_o> jhung: thanks
[09:20:57 CST(-0600)] <anastasiac> Justin_o, can I help with heidi's patch? which JIRA was that?
[09:21:22 CST(-0600)] <anastasiac> 3283?
[09:21:24 CST(-0600)] <heidi> anastasiac 3283
[09:21:26 CST(-0600)] <heidi> yep
[09:21:45 CST(-0600)] <heidi> jess and i wnt through all the demos last night and cleaned up the descriptions/instructions
[09:22:46 CST(-0600)] <anastasiac> Justin_o, do you want me to be the second set of eyes on 3283?
[09:23:20 CST(-0600)] <jessm> anastasiac: do you know which one is correct? Press the Stop button. or Press the 'Stop' button. or Press the "Stop" button.
[09:23:29 CST(-0600)] <jhung> anastasiac: want to pair on this one?
[09:23:40 CST(-0600)] <Justin_o> anastasiac: yes please
[09:24:00 CST(-0600)] <anastasiac> jessm, to me that's a style issue - consistency probably rules over "correctness"
[09:24:35 CST(-0600)] <jessm> hrm /me 's mom would think it a grammar issue
[09:24:45 CST(-0600)] * jessm digs for the logic rule on this one
[09:24:55 CST(-0600)] <jessm> anastasiac: but yes, our approach was consistency
[09:25:28 CST(-0600)] <jessm> my hunch is this: Press the 'Stop' button. Press "Stop" when you are done.
[09:29:04 CST(-0600)] <mlam> Justin_o: what else needs testing?
[09:30:05 CST(-0600)] <heidi> justin_o i'm avail to test too
[09:30:45 CST(-0600)] <Justin_o> colinclark, mlam: file queue view tests seem to be failing
[09:31:04 CST(-0600)] <colinclark> Justin_o: I'm on it
[09:31:22 CST(-0600)] <Justin_o> also noticing that FluidIoC, ModelTransformation and fileQueueView tests won't even run on IE 6
[09:33:57 CST(-0600)] <Justin_o> also the reorderer tests seem to be failing in IE 6 (sad)
[09:34:15 CST(-0600)] <Justin_o> heidi: do you have a copy of IE 6?
[09:35:01 CST(-0600)] <colinclark> Justin_o: I'll take a look at the Uploader tests in IE 6
[09:35:13 CST(-0600)] <Justin_o> colinclark: thanks
[09:35:19 CST(-0600)] <heidi> justin_o no
[09:35:41 CST(-0600)] <Justin_o> colinclark: just to note the fileQueueView one was also failing on safari 5.. probably other places too
[09:35:58 CST(-0600)] <Justin_o> heidi: which version of IE do you have?
[09:36:15 CST(-0600)] <heidi> justin_o ie8, which i can switch to ie7 w dev tools
[09:36:59 CST(-0600)] <Justin_o> heidi: can you test out antranig's patch from fluid-3487
[09:37:13 CST(-0600)] <heidi> justin_o yeah fo' shizzle
[09:37:16 CST(-0600)] <Justin_o> i was noticing here that the dropdown manual test wouldn't open in IE 6 an IE 7
[09:37:31 CST(-0600)] <Justin_o> if it won't open in the IE 7 mode, can you see if you can figure out why
[09:37:44 CST(-0600)] <heidi> sure
[09:37:46 CST(-0600)] <heidi> ill check it out
[09:39:17 CST(-0600)] <heidi> justin_o what does dead man's blur do? how to test?
[09:44:06 CST(-0600)] <Justin_o> heidi: so on the dropdown manual tests.. you would click on the dropdown and select another item with the mouse... basically dead mans blur will prevent the blur event from firing on certain items in a given condition...
[09:44:29 CST(-0600)] <Justin_o> if the dead mans blur isn't working.. when you go to click on a different item in the drop down.. it will just close...and not change the value
[09:45:02 CST(-0600)] <heidi> justin_o okay cool, thanks
[09:48:42 CST(-0600)] <jessm> anastasiac: are you on the FLUID-3799 changes that need to happen?
[09:49:04 CST(-0600)] <heidi> justin_ it works on my IE8 (compat & normal) and IE7
[09:49:04 CST(-0600)] <colinclark> Justin_o: The failing tests in fileQueueView relate to Antranig's change to the ARIA labeller
[09:49:24 CST(-0600)] <colinclark> It implies in fact that I'm testing behaviour for another component, which is arguably unnecessary
[09:49:29 CST(-0600)] <colinclark> I'll fix them right away
[09:49:35 CST(-0600)] <anastasiac> jessm, yes - the issue needs a bit of discussion, and I've commented on the JIRA and posted to the list about it, as well as working on the second comment.
[09:50:26 CST(-0600)] <jessm> anastasiac: i'm guessing the larger convo will have to happen after 1.3 – for 1.3 have you sorted out what code needs to be fixed and patched? i see the jira that mentions a particular line of code
[09:51:19 CST(-0600)] <Justin_o> colinclark: are there tests for the ARIA labeller itself
[09:51:28 CST(-0600)] <colinclark> Justin_o: I believe there are, yes
[09:51:30 CST(-0600)] <anastasiac> jessm, yes: i'll fix the issue mentioned for line 56
[09:54:15 CST(-0600)] <anastasiac> jhung, I've attached a revised patch to FLUID-3283 - have at it
[09:54:17 CST(-0600)] <jessm> heidi: ping me when you're done testing what yer doing
[09:54:33 CST(-0600)] <heidi> jessm done
[09:54:55 CST(-0600)] <jessm> heidi: can we chat for a moment about the keyboard a11y demo in chrome?
[09:54:58 CST(-0600)] <jhung> anastasiac: thanks! Will take a look.
[09:55:03 CST(-0600)] <jessm> heidi: i miss the '.' at the end entirely
[09:55:18 CST(-0600)] <heidi> jessm sure
[09:56:37 CST(-0600)] <colinclark> mlam: I'm going to get those fileQueueView tests running again
[09:56:40 CST(-0600)] <jessm> heidi: maybe it isn't an issue on chrome on win.
[09:56:43 CST(-0600)] <colinclark> Do you have time to take a look at IE 6?
[09:56:54 CST(-0600)] <mlam> ok, cool
[09:56:57 CST(-0600)] <heidi> justin_o not sure if you got my last msg. the dropdown.html works fine on my IE8 & IE7
[09:56:59 CST(-0600)] <jessm> heidi: looking here it doesn't seem chrome on mac is supported http://wiki.fluidproject.org/display/fluid/Browser+Support+Chart
[09:57:16 CST(-0600)] <mlam> yes, i can take a look.
[09:57:22 CST(-0600)] <colinclark> thanks, mlam
[09:57:25 CST(-0600)] <mlam> np
[09:57:37 CST(-0600)] <heidi> jessm look ok on chrome win?
[09:57:46 CST(-0600)] <jessm> heidi: i don't have chrome win
[09:57:54 CST(-0600)] <heidi> jessm ok ill look
[09:59:16 CST(-0600)] <Justin_o> heidi: oh yes.. thanks for checking...
[09:59:30 CST(-0600)] <Justin_o> unfortunately that will mean that i'll have to check it out on an actual system
[09:59:43 CST(-0600)] <Justin_o> jhung: do you still have IE 6 or IE 7 running
[10:00:03 CST(-0600)] <jhung> justin_o: I have IE6 in WinXP
[10:00:17 CST(-0600)] <jessm> anastasiac: when you're done patching 3799 can you ping me?
[10:00:29 CST(-0600)] <anastasiac> jessm, sure
[10:00:59 CST(-0600)] <Justin_o> jhung: great.. can you help me track some stuff down?
[10:01:25 CST(-0600)] <jhung> justin_o: I can help in a few mins. Just need to get this patch reviewed.
[10:01:32 CST(-0600)] <jhung> I can ping you when done.
[10:01:34 CST(-0600)] <mlam> colinclark: fileQueueView tests are broken in IE6. All other tests are passing
[10:01:38 CST(-0600)] <colinclark> michelled: I've just committed a minor change to the FileQueueView tests in Uploader to make them sparklingly green again
[10:01:45 CST(-0600)] <colinclark> Would you mind taking a moment to review?
[10:01:50 CST(-0600)] <colinclark> mlam: Update and try again
[10:01:54 CST(-0600)] <mlam> ok
[10:01:55 CST(-0600)] <colinclark> I just committed a fix for those tests
[10:01:59 CST(-0600)] <mlam> cool
[10:02:21 CST(-0600)] <heidi> jessm the keyboard html file works fine, however none of the demo.html pages load in chrome
[10:02:41 CST(-0600)] <heidi> jhung can you verify that when you have a sec? open a demo.html file in win chrome
[10:02:54 CST(-0600)] <Justin_o> jhung: sure... when you're ready there are a few things to choose from 1) reorderer tests are failing, 2) antranig's patch for FLUID-3487
[10:03:34 CST(-0600)] <michelled> colinclark: looking now
[10:03:45 CST(-0600)] <heidi> jhung however the build.fluid site demos load fine
[10:03:45 CST(-0600)] <colinclark> So, Justin_o, I have to make a huge apology
[10:03:54 CST(-0600)] <colinclark> I forgot to lint my unit tests in Uploader
[10:04:01 CST(-0600)] <colinclark> which is likely causing the IE 6 breakage
[10:04:04 CST(-0600)] <colinclark> stray commas
[10:04:16 CST(-0600)] <colinclark> I suck so bad, and I promise never to make this mistake again
[10:04:16 CST(-0600)] <Justin_o> colinclark: ah okay... glad you spotted that...
[10:04:18 CST(-0600)] <jhung> heidi: b-b-b-Borked in Chrome. (smile)
[10:04:19 CST(-0600)] <colinclark> for my own sanity (wink)
[10:04:28 CST(-0600)] <heidi> jhung doh! but build.fluid ok?
[10:04:57 CST(-0600)] <jhung> Which?
[10:05:07 CST(-0600)] <jessm> heidi: jhung: huh? demos don't work in chrome on win?
[10:05:10 CST(-0600)] <heidi> jhung the demos on build.fluidproject.org
[10:05:10 CST(-0600)] <Justin_o> colinclark: i'm tempted to just lint the entire codebase
[10:05:23 CST(-0600)] <jhung> jessm: the portal doesn't work. The demos are okay.
[10:05:51 CST(-0600)] <heidi> jhung wait, the demos are okay on yours? when i open any demo.html the frame is empty
[10:05:53 CST(-0600)] <jhung> heidi: yeah, work when served from Build.
[10:06:05 CST(-0600)] <heidi> oh n/m i get it
[10:06:17 CST(-0600)] <jhung> (smile)
[10:06:23 CST(-0600)] <heidi> so chrome isn't loading the frames. hmm
[10:06:28 CST(-0600)] <heidi> why would it work on build?
[10:06:34 CST(-0600)] <jessm> jhung: translate – what does "portal doesn't work" mean?
[10:07:00 CST(-0600)] <heidi> jessm if you load a demo.html (portal) vs like html/inlineedit.html (actual demo)
[10:07:11 CST(-0600)] <jhung> jessm: there are empty containers where the code snippets and demo instance would appear.
[10:07:36 CST(-0600)] <Justin_o> heidi: do you have some time to help me with linting
[10:07:36 CST(-0600)] <jhung> only happens on local copy loading demo.html.
[10:07:53 CST(-0600)] <Justin_o> jhung: that's probably because of local ajax calls... not sure though
[10:08:18 CST(-0600)] <heidi> yeah i think so too, cos pager doesn't work
[10:08:27 CST(-0600)] <heidi> justin_o i can help lint
[10:09:08 CST(-0600)] <Justin_o> heidi: okay great... how do you want to break it up...
[10:09:14 CST(-0600)] <Justin_o> so we don't duplicate work
[10:09:21 CST(-0600)] <heidi> justin_o what are we linting?
[10:09:56 CST(-0600)] <heidi> oh, everything
[10:10:04 CST(-0600)] <Justin_o> heidi: all of our js files... so we won't lint the 3rd party stuff but everything else
[10:10:25 CST(-0600)] <heidi> justin_o do you want to start in components and i can start in demos ?
[10:10:44 CST(-0600)] <Justin_o> heidi: great.. i was just going to suggest that (smile)
[10:15:36 CST(-0600)] <mlam> Bosmon: Just did another functional review of the 2 pager JIRAs (FLUID-3822, FLUID-2244) . Everything's looking great
[10:16:46 CST(-0600)] <mlam> michelled: ^^
[10:23:01 CST(-0600)] <colinclark> Justin_o, heidi: I've just linted all of Uploader
[10:23:10 CST(-0600)] <colinclark> and I'll do the framework tests now, too
[10:23:18 CST(-0600)] <heidi> cool
[10:23:39 CST(-0600)] <heidi> my sweater needs linting too
[10:23:40 CST(-0600)] <jhung> heidi, jessm, anastasiac: FLUID-3283 is ready to go. Should I commit? http://issues.fluidproject.org/browse/FLUID-3283
[10:23:52 CST(-0600)] <Justin_o> colinclark: thanks.. did you commit that
[10:23:54 CST(-0600)] <anastasiac> jhung, fine with me
[10:24:44 CST(-0600)] <colinclark> Justin_o: Not yet
[10:24:46 CST(-0600)] <colinclark> almost done
[10:24:53 CST(-0600)] <colinclark> nothing like whitespace corrections on Bosmon's files (wink)
[10:24:58 CST(-0600)] <Justin_o> (smile)
[10:25:34 CST(-0600)] <michelled> thanks mlam
[10:46:43 CST(-0600)] <Justin_o> jhung: when you get to the IE 6 can you look at Bosmon's patch for FLUID-3487
[10:46:52 CST(-0600)] <jhung> yep.
[10:47:00 CST(-0600)] <Justin_o> jhung: thanks
[10:47:24 CST(-0600)] <Justin_o> anastasiac, jhung: is FLUID-3283 all done?
[10:47:47 CST(-0600)] <jhung> justin_o: yep. Just committed before standup.
[10:47:51 CST(-0600)] <jhung> I can close it now.
[10:47:54 CST(-0600)] <Justin_o> jhung: thanks
[10:58:39 CST(-0600)] <jessm> heidi: i'm just realizing that in 3283 we tossed the letter key commands for image reorderer
[11:00:34 CST(-0600)] <heidi> ah
[11:00:43 CST(-0600)] <heidi> should add those back in - all reorderers or just that one?
[11:01:30 CST(-0600)] <heidi> jessm do oyu want to make a patch ?
[11:02:15 CST(-0600)] <jessm> heidi: i don't know the answer to that – do anastasiac or jhung have anything to add?
[11:03:16 CST(-0600)] <mlam> Justin_o: anything else on tap that needs help?
[11:03:42 CST(-0600)] <anastasiac> jessm, heidi: first, good catch jessm - I missed that. Personally, I think we should try to add the letter commands back in. It's a feature of the reorderer that it can be configured with different keystrokes, so we should show that off by letting people know about it
[11:05:15 CST(-0600)] <jessm> anastasiac: just for the layout reorderer or the whole family?
[11:05:48 CST(-0600)] <Justin_o> mlam: do you have ie 6?
[11:05:57 CST(-0600)] <mlam> no, i don't
[11:06:03 CST(-0600)] <mlam> but i can use the test machine
[11:06:03 CST(-0600)] <anastasiac> jessm, I'd say the whole family (customized, of course: list would likely only mention m and i)
[11:06:26 CST(-0600)] <jessm> anastasiac: can you work on a patch? sorry i didn't catch this sooner (sad)
[11:06:38 CST(-0600)] <anastasiac> sure - I'll get to it after 3799
[11:06:51 CST(-0600)] <jessm> anastasiac: great
[11:06:51 CST(-0600)] <Justin_o> mlam: thanks.. can you start looking into why the reorderer tests are failing in IE 6
[11:07:23 CST(-0600)] <mlam> ok
[11:09:59 CST(-0600)] <jhung> justin_o: reorderer tests failing on IE7 in WinXP.
[11:10:33 CST(-0600)] <Justin_o> jhung: thanks.. mlam is starting to look into that now
[11:11:18 CST(-0600)] <jhung> k. going to take a look at FLUID-3487 patch now
[11:13:51 CST(-0600)] <Justin_o> jhung: thaqnks
[11:37:15 CST(-0600)] <anastasiac> Justin_o: I've attached a patch to FLUID-3799 and assigned it to Bosmon - it's in his hands for now
[11:45:50 CST(-0600)] <anastasiac> Justin_o, earlier jessm caught the fact that the reorderer demo instructions don't mention that letter keys can be used in addition to arrow keys. I have a patch for that, that I can attach to the JIRA - is that something that can be committed?
[11:46:11 CST(-0600)] <anastasiac> I can't attach the patch to the JIRA without re-opening
[11:46:15 CST(-0600)] <anastasiac> not sure how important it is?
[11:46:46 CST(-0600)] <Justin_o> anastasiac: i think you should probably re-open comment about the problem... commit, get someone to review and close
[11:47:00 CST(-0600)] <Justin_o> unless it was jess who made the patch... then i guess you did the review
[11:47:01 CST(-0600)] <anastasiac> Justin_o, thanks - will do
[11:47:05 CST(-0600)] <Justin_o> anastasiac: thanks
[11:47:27 CST(-0600)] <jhung> jessm, anastasiac: are the letter combinations required?
[11:48:03 CST(-0600)] <anastasiac> jhung, not required, but multiple keysets is one of the features of the Reorderer. It was my opinion that we might want to highlight that fact
[11:48:54 CST(-0600)] <anastasiac> jhung, do you disagree?
[11:49:16 CST(-0600)] * anastasiac has mouse hovered over the 'reopen' button
[11:50:13 CST(-0600)] <jhung> anastasiac: I don't disagree, but I'm afraid it will make the instructions harder to follow. I think if we want to demonstrate multiple keysets, I think there may be a better way to demonstrate this.
[11:51:06 CST(-0600)] <anastasiac> jhung, here's what I'm proposing as the change: "Arrow keys (or 'i,' 'k,' 'm' or 'k') move the focus. CTRL + Arrow moves the selection." What do you think?
[11:51:38 CST(-0600)] <anastasiac> except that should be 'j' of course
[11:52:35 CST(-0600)] <jhung> anastasiac: we'll need to add CTRL + Arrow (or 'i' 'j' 'k' 'm') as well?
[11:53:20 CST(-0600)] <anastasiac> I thought about that, but for brevity I left it out, on the assumption that users would likely be able to figure that out. Am I giving users too much credit?
[11:53:42 CST(-0600)] <anastasiac> I toyed with "CTRL + same keys" but didn't like it
[11:53:45 CST(-0600)] <jhung> anastasiac: maybe "Arrow or 'i' 'j' 'k' 'm' are direction keys. Direction keys move focus, and CTRL + Direction Key moves the selection. Or "moves the item"? sounds better?
[11:54:13 CST(-0600)] <anastasiac> oh, I like that.
[11:54:17 CST(-0600)] <anastasiac> thanks
[11:54:27 CST(-0600)] <jhung> cool
[11:55:41 CST(-0600)] <jhung> justin_o: I just realized I can not test the patch for FLUID-3487 because reorderer is broken in IE7.
[11:57:50 CST(-0600)] <anastasiac> jhung: slight refinement? better? "Arrow keys, or 'i,' 'k,' 'm' and 'j,' are direction keys."
[11:59:06 CST(-0600)] <jhung> how about "Arrow or i'j'
[11:59:17 CST(-0600)] <jhung> hmmm
[11:59:33 CST(-0600)] <anastasiac> trying to balance brevity with grammatical correctness...
[11:59:35 CST(-0600)] <jhung> it sounds a bit funny because Arrow keys are direction keys too.
[12:00:21 CST(-0600)] <anastasiac> well, yes, that's what it says.
[12:00:47 CST(-0600)] <jhung> How about flipping it around and saying: "Direction keys are: arrow keys, or i-j-k-m."?
[12:01:44 CST(-0600)] <anastasiac> how about "Direction keys (arrow keys or i-j-k-m) move focus. CTRL..."
[12:02:48 CST(-0600)] <jhung> yeah, that works. But replace brackets with commas. My momma always tol' me that stuff in dem brack'ts can be ignored.
[12:04:02 CST(-0600)] <anastasiac> not sure about commas - I think it makes the sentence harder to parse
[12:04:22 CST(-0600)] <jhung> k. I'm okay with the brackets.
[12:04:57 CST(-0600)] <anastasiac> michelled, hallway usability test: Does this make sense as keyboard instructions for the reorderer? "Direction keys (arrow keys or i-j-k-m) move focus. CTRL + direction keys move selection"
[12:06:16 CST(-0600)] <michelled> I wonder if people will understand the difference between focus and selection
[12:06:19 CST(-0600)] <michelled> I don't
[12:06:32 CST(-0600)] <jhung> michelled: I'm thinking the same thing.
[12:06:40 CST(-0600)] <michelled> but I'd probably just try both and figure out what it does
[12:07:03 CST(-0600)] <anastasiac> that wasn't even the point of the question (smile) it was the combination of arrow keys plus the letter keys
[12:07:10 CST(-0600)] <anastasiac> it used to only mention arrow keys
[12:07:11 CST(-0600)] <jhung> anastasiac: maybe "move selection" -> "moves item"?
[12:07:23 CST(-0600)] <anastasiac> bettter
[12:09:04 CST(-0600)] <Justin_o> jhung: sorry just saw your message
[12:09:15 CST(-0600)] <Justin_o> the reorderer is broken in IE 7?
[12:09:37 CST(-0600)] <jhung> yeah. the same reason the unit tests for the reorderer in IE7 fail.
[12:12:18 CST(-0600)] <jhung> justin_o: however FF3.6 + NVDA + WinXP works nicely with the patch.
[12:12:24 CST(-0600)] <Justin_o> jhung: right mike's looking into that.. i think FLUID-3487 is something else though.. you would be testing the dropdown inline edit manual test.. and the FluidView unit tests
[12:13:12 CST(-0600)] <jhung> oh geez.
[12:13:13 CST(-0600)] <jhung> lol
[12:13:41 CST(-0600)] <jhung> justin_o: helps if I read the bugs. (smile) I thought I was testing bosmon's patch for reorderer postioning feedback.
[12:14:01 CST(-0600)] <anastasiac> Justin_o, FYI: FLUID-3283 reopened, committed, resolved and assigned to jessm for review
[12:14:01 CST(-0600)] <jhung> justin_o: going to test FLUID-3487 now.
[12:14:07 CST(-0600)] <Justin_o> jhung: oh (smile) well good to know that that is working fine in NVDA and FF
[12:14:12 CST(-0600)] <Justin_o> jhung: thanks
[12:17:33 CST(-0600)] <jhung> bosmon: for FLUID-3487, aside from running the FluidView-test.html and dropdown.html test, how else should I test the patch?
[12:19:57 CST(-0600)] <Justin_o> mlam: just committed the reorderer linting
[12:20:02 CST(-0600)] <mlam> ok, cool thanks
[12:20:22 CST(-0600)] <anastasiac> Justin_o, I've resolved FLUID-3881, commenting unsupported functions. I updated any new framework and renderer functions, so I think we're good to go
[12:20:43 CST(-0600)] <Justin_o> anastasiac: great.. did antranig get back to you about reorderer changes?
[12:25:41 CST(-0600)] <jhung> I can't seem to run manual-tests\html\dropdown.html in IE7. The page loads and then IE7 gives me an "operation aborted" dialog.
[12:25:48 CST(-0600)] <jhung> Checking IE 8 now.
[12:34:26 CST(-0600)] <Justin_o> heidi: just committed your demo linting
[12:34:35 CST(-0600)] <heidi> sweet
[12:40:29 CST(-0600)] <jessm> anastasiac: FLUID-3283 looks good. Thanks for that quick fix. I'll close now < Justin_o
[12:41:23 CST(-0600)] <Justin_o> jhung: yes... that is the problem
[12:41:36 CST(-0600)] <Justin_o> jhung: would you be able to track down what is causing the issue
[12:41:54 CST(-0600)] <jhung> justin_o: maybe. I'll do some digging.
[12:41:56 CST(-0600)] <Justin_o> jessm, anastasiac: sure
[12:42:04 CST(-0600)] <Justin_o> jhung: thanks
[13:03:58 CST(-0600)] <Bosmon> jhung - those two things are pretty much what you can do for the patch
[13:04:45 CST(-0600)] <jhung> Bosmon: okay. There seems to be a problem with the dropdown manual test in IE7. It doesn't load. Gives an "Operation aborted" error.
[13:09:00 CST(-0600)] <Bosmon> jhung: Any idea why?
[13:09:07 CST(-0600)] <Bosmon> Unfortunately I don't have IE7 available to test
[13:09:12 CST(-0600)] <jhung> bosmon: I'm looking into it now.
[13:17:01 CST(-0600)] <jhung> bosmon: changing to jquery.selectbox-0.6.1.js gives a mildly better error message.
[13:17:19 CST(-0600)] <Bosmon> That's intriguing
[13:17:21 CST(-0600)] <Bosmon> Had we upgraded?
[13:17:36 CST(-0600)] <Bosmon> Also, did this demo used to work in the 1.2 release in IE7?
[13:17:38 CST(-0600)] <jhung> no. it is using selectbox-0.5.js
[13:17:40 CST(-0600)] <Bosmon> And what is the error message
[13:18:19 CST(-0600)] <jhung> Using selectbox-0.6.1, the error is: "Line 756, Char 9, Error 'nodeName' is null or not an object.
[13:18:27 CST(-0600)] <Bosmon> wow
[13:19:41 CST(-0600)] <Bosmon> Does that line refer to selectbox itself?
[13:19:46 CST(-0600)] <Bosmon> I can't match it up yet to any of our common files
[13:20:22 CST(-0600)] <jhung> Bosmon: it may be. Looking at it now...
[13:21:06 CST(-0600)] <jhung> unfortunately debuggin in IE7 is painful and the line number is not useful.
[13:22:11 CST(-0600)] <jhung> The error only appears when I select the editable field.
[13:22:51 CST(-0600)] <jhung> I'm going to revert to see if this test ever worked in IE7.
[13:23:08 CST(-0600)] <Bosmon> That will be useful to know
[13:23:35 CST(-0600)] <Bosmon> Since it was not any particularly kind of official test, for a component we never formally released, I think it has only had spotty QA
[13:24:04 CST(-0600)] <jhung> indeed.
[13:24:12 CST(-0600)] <jhung> Looking for an older revision now.
[13:29:31 CST(-0600)] <jhung> bosmon: works for Infusion 1.2.
[13:30:24 CST(-0600)] <Bosmon> gah
[13:33:17 CST(-0600)] <Bosmon> Well, JURA has proposed a simplified implementation of DMB that should be functionally equivalent to the one we had in 1.2
[13:33:28 CST(-0600)] <Bosmon> I will make a revised patch and see if this fixes our IE issues too
[13:38:25 CST(-0600)] <jhung> bosmon: okay. I'll tinker a little longer on this. Let me know when you want me to test out a patch.
[13:45:23 CST(-0600)] <Bosmon> jhung - I've attached a revised patch FLUID-3487-b to the issue
[13:45:30 CST(-0600)] <Bosmon> See if it works any better for you on IE7
[13:45:48 CST(-0600)] <Bosmon> If it doesn't I'll be REALLY surprised since its effect should now be identical to the implementation we shipped in 1.2
[13:49:10 CST(-0600)] <Bosmon> fluid-everyone - did anyone take a look at FLUID-3899?
[13:50:30 CST(-0600)] <Justin_o> Bosmon: yura did
[13:53:25 CST(-0600)] <jhung> bosmon: still no go with IE7 with the revised patch.
[13:59:02 CST(-0600)] <yura_> Bosmon: I cant apply patch b into trunk
[13:59:15 CST(-0600)] <yura_> FluidViewTests dont apply
[14:00:42 CST(-0600)] <colinclark> fluid-everyone: I've got a list of our last to-dos based on conversations with Justin_o, Bosmon, michelled, and others.
[14:01:03 CST(-0600)] <colinclark> 1. fluid.deadMansBlur() is broken. Bosmon, jhung, and yura_ are heads down with this one.
[14:01:20 CST(-0600)] <colinclark> 2. Uploader is freaking in IE 7 (and possibly 6). mlam and cindyli are looking at this one, and Justin_o will join them
[14:01:42 CST(-0600)] <colinclark> We wonder if it might be related to the ariaLabeller and IE hating modern tags and attributes, but that might just be wishful thinking.
[14:01:54 CST(-0600)] <colinclark> 3. Reorderer tests are failing in IE 6 and 7.
[14:02:02 CST(-0600)] <colinclark> 4. Linting everything. heidi is working on this one.
[14:02:12 CST(-0600)] <colinclark> 5. Framework code review. This one is on my plate
[14:02:28 CST(-0600)] <Bosmon> fluid-everyone - OK, I have a fix for these IE issues
[14:02:34 CST(-0600)] <colinclark> 6. Uploader code review. michelled will review this, and if Bosmon manages to fix #1 we'll have him involved, too
[14:02:36 CST(-0600)] <Bosmon> The answer is unfortunately very disagreeable
[14:02:49 CST(-0600)] <colinclark> fluid-everyone: If you're not working on one of these six tasks, let me know.
[14:03:01 CST(-0600)] <Bosmon> http://www.clientcide.com/code-snippets/manipulating-the-dom/ie-and-operation-aborted/
[14:03:04 CST(-0600)] <anastasiac> colinclark, I'm available to help where needed
[14:03:14 CST(-0600)] <Bosmon> Everyone interested in issue #1 please take a look at this link
[14:03:22 CST(-0600)] <colinclark> anastasiac: Can you help heidi with #4, now that Justin_o is moving over to help with Uploader?
[14:03:30 CST(-0600)] <anastasiac> sure
[14:03:34 CST(-0600)] <Bosmon> Unfortunately this suggests that something we have recommended as a "best practice" is not something that can be made to work on IE
[14:03:56 CST(-0600)] <anastasiac> heidi, where are you at? which chunk of files are still lint-ful?
[14:04:02 CST(-0600)] <Bosmon> Basically a page which begins to perform some kind of DOM manipulation too early during page loading will fail on IE6 and IE7 with the "operation aborted" message
[14:04:29 CST(-0600)] <Justin_o> heidi: can you send anastasiac all your patches... except for the first one that i committed
[14:04:34 CST(-0600)] <heidi> anastasiac i'm almost done actually. just /tests to do. justin_o i'm guessing we don't do /lib
[14:04:42 CST(-0600)] <heidi> justin_o sure
[14:04:43 CST(-0600)] <Justin_o> heidi: nope
[14:04:55 CST(-0600)] <anastasiac> heidi, sure - send me the patches, I'll get them committed
[14:05:13 CST(-0600)] <colinclark> Bosmon: I'm reading it now
[14:05:13 CST(-0600)] <heidi> anastasiac ok ill send one and the other's coming in a bit
[14:06:48 CST(-0600)] <mlam> colinclark: the error i'm getting in IE7 for the uploader is pointing to a line of code in the fileQueueView that calls fluid.updateAriaLabel
[14:08:05 CST(-0600)] <colinclark> mlam: Maybe my theory is right, then
[14:08:09 CST(-0600)] <mlam> yeah
[14:08:12 CST(-0600)] <colinclark> keep digging
[14:08:23 CST(-0600)] <colinclark> Unfortunately one of the only techniques you can do is to strip out chunks of code
[14:08:27 CST(-0600)] <colinclark> finding what works and doesn't
[14:08:32 CST(-0600)] <colinclark> have you tried IE 6 yet, mlam?
[14:08:58 CST(-0600)] <mlam> no, not yet
[14:09:00 CST(-0600)] <Justin_o> anastasiac: so all of the component, and demo files have been linted...
[14:09:29 CST(-0600)] <Justin_o> i sent you a patch for the start of the framework linting.. although one of the unit tests is now failing.. so i must have done something wrong... likely in the fluid.js file
[14:09:52 CST(-0600)] <Justin_o> anastasiac: heidi is working on the other areas
[14:10:08 CST(-0600)] <anastasiac> thanks, Justin_
[14:10:15 CST(-0600)] <anastasiac> Justin_0
[14:10:27 CST(-0600)] <Justin_o> anastasiac: thank you
[14:10:35 CST(-0600)] <jhung> bosmon: any idea why the error would go away if using selectbox 0.6.1? Or is it the same thing just manifest differently?
[14:10:59 CST(-0600)] <Bosmon> jhung - please read the link - also, I have posted up a "c" version of the patch which works for me under IE6
[14:11:04 CST(-0600)] <Bosmon> Please could you test it under IE7
[14:19:07 CST(-0600)] <justin_o> fluid-everyone: so member:mlam was just saying that fluid.updateArialabel is what is breaking in IE 7 and IE 6... we can take a look at trying to get it to work... i'm also thinking that we could just not use it in IE 7 and IE 6 since they don't support ARIA anyways
[14:19:28 CST(-0600)] <colinclark> That's certainly an option
[14:19:36 CST(-0600)] <colinclark> Let's look at it once we know what exactly is causing the problem
[14:19:52 CST(-0600)] <colinclark> since it might be revealing a problem in our implementation
[14:19:56 CST(-0600)] <Bosmon> Ok, so I have linted the FluidIoCTest file which has restored it to running under IE6
[14:20:04 CST(-0600)] <Bosmon> As usual, the problem was of the "stray comma" variety
[14:20:04 CST(-0600)] <colinclark> or it might just be revealing a problem in Microsoft's implementation (wink)
[14:20:09 CST(-0600)] <colinclark> Bosmon: I did that earlier today
[14:20:11 CST(-0600)] <colinclark> did that not go in?
[14:20:13 CST(-0600)] <Bosmon> However this reveals that the final test in the file is actually running
[14:20:18 CST(-0600)] <Bosmon> er failing
[14:20:31 CST(-0600)] <Bosmon> Sorry, I guess I just didn't update
[14:20:37 CST(-0600)] <colinclark> Okay, so the failing test
[14:20:45 CST(-0600)] <colinclark> seems to be due to its reliance on local Ajax
[14:20:45 CST(-0600)] <justin_o> Bosmon: lots of linting coming then
[14:20:54 CST(-0600)] <colinclark> which I always disallow in FF, just so I notice these
[14:21:01 CST(-0600)] <colinclark> it was passing in michelled
[14:21:04 CST(-0600)] <colinclark> 's browser
[14:21:04 CST(-0600)] <Bosmon> ok
[14:21:07 CST(-0600)] <colinclark> as well as in Safari
[14:21:12 CST(-0600)] <Bosmon> So local AJAX doesn't work in IE6?
[14:21:17 CST(-0600)] <colinclark> erm
[14:21:20 CST(-0600)] <colinclark> hrrm
[14:21:24 CST(-0600)] <colinclark> No, in FF
[14:21:28 CST(-0600)] <colinclark> It's failing in IE6?
[14:21:29 CST(-0600)] <colinclark> Which test?
[14:21:30 CST(-0600)] <Bosmon> Yes
[14:21:34 CST(-0600)] <Bosmon> Deferred expander test
[14:21:46 CST(-0600)] <colinclark> Yeah, that's the one that was failing on my Firefox by not michelled's
[14:21:54 CST(-0600)] <colinclark> And which passed on my Safari
[14:22:01 CST(-0600)] <colinclark> But as far as I know, local ajax works on IE 6
[14:22:18 CST(-0600)] <Bosmon> It also fails in Opera 9.5, for what it's worth (smile)
[14:22:23 CST(-0600)] <colinclark> bizarre
[14:31:18 CST(-0600)] <mlam> colinclark: i commented out the defaults block for deadMansBlur in FluidView.js as well as the updateAriaLabel call in fileQueueView and the uploader is working in IE7. I'm going to look more deeply into the updateAriaLabel function
[14:32:09 CST(-0600)] <colinclark> one at a time, yes
[14:32:21 CST(-0600)] <colinclark> I expect that DMB has no impact on Uploader
[14:32:25 CST(-0600)] <colinclark> we don't use it
[14:32:51 CST(-0600)] <colinclark> So you can exclude that by simply commenting out all of DMB, ensuring that Uploader still fails, put it back in, and then be confident that the labeller is our culprit
[14:32:55 CST(-0600)] <colinclark> make sense, mlam?
[14:34:03 CST(-0600)] <heidi> anastasiac lots of js in /tests if you have time to help after looking at patches. just ping me (smile)
[14:34:34 CST(-0600)] <anastasiac> will do, heidi - I'm linting the framework right now, which is going to take a bit of time
[14:34:49 CST(-0600)] <heidi> okee
[14:35:45 CST(-0600)] <colinclark> mind-numbing work
[14:36:21 CST(-0600)] <jhung> bosmon: tested patch-c under IE7 and it works.
[14:36:21 CST(-0600)] <Bosmon> Ok, it looks like the failure in IE6 on the deferredFetcher is that the JSON is being considered invalid
[14:36:26 CST(-0600)] <Bosmon> jhung, great
[14:36:36 CST(-0600)] <Bosmon> justin_o, colinclark - does this mean we can commit it?
[14:36:37 CST(-0600)] <mlam> yes, makes sense. there was an unnecessary comma at the end of the DMB defaults block that caused one of the errors
[14:36:51 CST(-0600)] <Bosmon> mlam - that is gone in the recent 2 patches
[14:37:05 CST(-0600)] <colinclark> Bosmon: Can I review it first? Send me a link, I'll review, you commit
[14:37:08 CST(-0600)] <colinclark> Let's go slow here
[14:37:11 CST(-0600)] <colinclark> Fix, review, then commit
[14:37:14 CST(-0600)] <Bosmon> (smile)
[14:37:24 CST(-0600)] <colinclark> Here's our mantra today, fluid-everyone: slow and steady
[14:37:32 CST(-0600)] <jessm> +1
[14:37:39 CST(-0600)] <Bosmon> http://issues.fluidproject.org/secure/attachment/11466/FLUID-3487-c.patch
[14:37:47 CST(-0600)] <Bosmon> colinclark: this is the patch which jhung just tested
[14:37:52 CST(-0600)] <colinclark> ok
[14:38:23 CST(-0600)] <Bosmon> It might be worth retesting it with focusin() put back in, but yura_ insists that this causes some problems
[14:38:29 CST(-0600)] <Bosmon> I have asked him to make up a test case demonstrating that
[14:39:53 CST(-0600)] <Bosmon> So.... any ideas why the string "[1, 2, 3]" might be considered invalid JSON under IE6?
[14:40:00 CST(-0600)] <colinclark> hmm
[14:40:05 CST(-0600)] <colinclark> what does JSLint think?
[14:40:16 CST(-0600)] <Bosmon> I didn't know that JSLint could lint JSON (tongue)
[14:40:21 CST(-0600)] <colinclark> it can indeed
[14:40:29 CST(-0600)] <colinclark> although I think yura_ has some tool he likes better
[14:40:55 CST(-0600)] <Bosmon> How do I tell it that I want it to lint JSON?
[14:40:57 CST(-0600)] <justin_o> yura_: uses jsonlint which also formats it for you
[14:41:29 CST(-0600)] <Bosmon> jsonlint claims it is valud
[14:41:31 CST(-0600)] <Bosmon> valid
[14:42:25 CST(-0600)] <colinclark> Bosmon: A minor note: after you've committed your patch (in a separate commit) can you lint the file? I see new additions not matching Crockfordism
[14:42:35 CST(-0600)] <colinclark> you'll need to coordinate with anastasiac, who may have already touched that file
[14:42:45 CST(-0600)] <colinclark> anastasiac: Please avoid committing against FluidView.js for a moment
[14:43:08 CST(-0600)] <Bosmon> Oh... this is embarrassing (smile)
[14:43:11 CST(-0600)] <colinclark> Bosmon: What does the ^ operator do?
[14:43:17 CST(-0600)] <Bosmon> I think I know why this is invalid JSON....
[14:43:23 CST(-0600)] <Bosmon> It is because JSON.js is not included in the test file (smile)
[14:43:29 CST(-0600)] <Bosmon> colinclark: This is the exclusive OR operator
[14:43:29 CST(-0600)] <colinclark> oops (tongue)
[14:43:30 CST(-0600)] <colinclark> + jqUnit.assertTrue(message + " - Blur handler has not executed", shouldBlur ^ !blurReceived);
[14:43:32 CST(-0600)] <colinclark> ah
[14:43:33 CST(-0600)] <colinclark> wow
[14:43:42 CST(-0600)] <colinclark> who knew?
[14:45:52 CST(-0600)] <colinclark> Bosmon: I'm just going to apply and test your patch
[14:45:55 CST(-0600)] <colinclark> but it looks just fine
[14:46:10 CST(-0600)] <colinclark> And apparently a stray comma was causing the Uploader to freak, which I imagine you noticed
[14:46:12 CST(-0600)] <colinclark> or will when you lint
[14:46:25 CST(-0600)] <colinclark> I think we're going to have to get an automated lint post-commit hook installed
[14:46:33 CST(-0600)] <colinclark> Not linting is absolutely killing us
[14:48:50 CST(-0600)] <Bosmon> Ok, so readding json2.js as a dependency restores the FluidIoC test to running under IE6 again
[14:49:11 CST(-0600)] <colinclark> Bosmon: Your patch is failing to apply
[14:49:20 CST(-0600)] <Bosmon> No surprise, everything has been linted (tongue)
[14:49:30 CST(-0600)] <Bosmon> Shall I regenerate it?
[14:49:47 CST(-0600)] <Bosmon> It even runs under Opera 9.5 (tongue)
[14:49:57 CST(-0600)] <colinclark> Bosmon: Could you please?
[14:55:02 CST(-0600)] <colinclark> Bosmon: A code review question when you have a chance. Ignore me until you do...
[14:55:25 CST(-0600)] <colinclark> For the ariaLabeller, you have an invoker called generateLiveElement(), whose signature looks like this:
[14:55:27 CST(-0600)] <colinclark> fluid.ariaLabeller.generateLiveElement = function(that) {
[14:55:43 CST(-0600)] <colinclark> Why did you choose to pass the full ariaLabeller instance as its only argument?
[14:57:47 CST(-0600)] <Bosmon> Sheer laziness? (tongue)
[14:58:05 CST(-0600)] <Bosmon> Ok, version "d" of FLUID-3487 is now up which applies to current trunk
[14:59:09 CST(-0600)] <colinclark> Bosmon: Thanks. I'll apply it and test now
[15:09:26 CST(-0600)] <justin_o> jhung: would you mind helping heidi and anastasiac with linting
[15:09:31 CST(-0600)] <justin_o> there's a lot of work with that stuff
[15:10:01 CST(-0600)] <jhung> justin_o: okay.
[15:10:06 CST(-0600)] <justin_o> jhung: thanks
[15:10:07 CST(-0600)] <colinclark> justin_o: We just hit a big fat Progressive Enhancement blocker
[15:10:14 CST(-0600)] <colinclark> due to my stupidly avoiding testing on IE 6
[15:10:35 CST(-0600)] <colinclark> We're looking for XMLHttpRequest directly, which will cause an error on non-standard browsers
[15:10:38 CST(-0600)] <colinclark> this is probably the correct feature check
[15:11:00 CST(-0600)] <colinclark> var canSendBinary = window.FormData || (window.XMLHttpRequest && window.XMLHttpRequest.prototype.sendAsBinary);
[15:11:27 CST(-0600)] <justin_o> colinclark: okay... i can file a jira for that...
[15:11:35 CST(-0600)] <colinclark> Thanks, King
[15:11:37 CST(-0600)] <justin_o> sounds like a blocker i guess
[15:11:56 CST(-0600)] <colinclark> Yepper
[15:12:25 CST(-0600)] <colinclark> mlam: Coincidentally, I get the semi-enhanced (i.e. Flash) version of Uploader when I fixed this
[15:13:28 CST(-0600)] <mlam> yah makes sense.
[15:15:35 CST(-0600)] <colinclark> Okay, I can even get the demo to work
[15:15:38 CST(-0600)] <colinclark> We're in business
[15:15:53 CST(-0600)] <Bosmon> colinclark: So we did confirm then that "window.x" is a safe way to test for things which may be "doubly undefined" as we call it? (tongue)
[15:15:55 CST(-0600)] <colinclark> Bosmon: Your patch looks good, tests pass, Uploader is partially restored as a result
[15:16:08 CST(-0600)] <colinclark> Bosmon: I believe it's one way to safely test, yes
[15:16:24 CST(-0600)] <colinclark> alternatively typeof(globallyThing) === "undefined" is appropriate
[15:16:31 CST(-0600)] <Bosmon> But messy
[15:16:37 CST(-0600)] <colinclark> I don't have any evidence to think that they're any different
[15:16:40 CST(-0600)] <colinclark> and yes, as you say, messier
[15:16:45 CST(-0600)] <colinclark> which is why I chose the former
[15:17:05 CST(-0600)] <colinclark> Bosmon: Please go ahead and commit your FLUID-3487-d.patch
[15:17:09 CST(-0600)] <Bosmon> Well, per our private chat last week, this really does suggest that for things to be "doubly undefined" is really a hole in the language impl/spec
[15:17:18 CST(-0600)] <colinclark> I agree
[15:17:55 CST(-0600)] <justin_o> colinclark, mlam: here's the jira for the uploader blocker http://issues.fluidproject.org/browse/FLUID-3901
[15:18:28 CST(-0600)] <colinclark> mlam: You feel comfortable making up a patch with the fix?
[15:18:45 CST(-0600)] <mlam> sure , i can do that
[15:20:07 CST(-0600)] <colinclark> Thanks, mlam
[15:20:16 CST(-0600)] <colinclark> So justin_o, you had mentioned Reorderer failures in IE 6?
[15:20:20 CST(-0600)] <colinclark> Can you elaborate?
[15:20:27 CST(-0600)] <mlam> np
[15:20:52 CST(-0600)] <mlam> colinclark: all tests but one were failing due to the stray comma
[15:21:00 CST(-0600)] <mlam> for the reorderer
[15:21:07 CST(-0600)] <colinclark> ok
[15:21:23 CST(-0600)] <colinclark> Bosmon's patch kindly addresses that issue, too
[15:21:39 CST(-0600)] <justin_o> colinclark: that's good.. still trying to trackdown this last one...
[15:21:57 CST(-0600)] <justin_o> seems that the aria-label isn't getting changed properly
[15:22:09 CST(-0600)] <colinclark> Okay, tell me more
[15:23:18 CST(-0600)] <justin_o> i believe a move operation is performed and then a new item is focused.. so when the new item is focused the label of the old one should be changed to not have "moved from" in it. but this is not happenening
[15:23:31 CST(-0600)] <colinclark> ok
[15:23:33 CST(-0600)] <colinclark> Interesting
[15:23:45 CST(-0600)] <colinclark> I'm seeing somewhat odd behaviour in the layout reorderer demo in IE 6
[15:23:54 CST(-0600)] <colinclark> I sort of feel like it's styling related
[15:24:02 CST(-0600)] <colinclark> and I can't quite pin down what is happening
[15:24:28 CST(-0600)] <justin_o> really
[15:24:29 CST(-0600)] <justin_o> hmm
[15:24:53 CST(-0600)] <colinclark> Portlets move into the bottom of the other column
[15:25:53 CST(-0600)] <colinclark> I'm going to leave this for now, but I expect it's probably a styling issue
[15:26:07 CST(-0600)] <colinclark> Bosmon: Did you managed to get your patch in?
[15:26:30 CST(-0600)] <justin_o> colinclark: not sure i'm seeing what you're seeing
[15:26:47 CST(-0600)] <colinclark> 1. In IE 6, move everything into the right-hand column
[15:27:15 CST(-0600)] <colinclark> 2. move one back to the left column, and it'll move to the bottom of that column, with a huge whitespace gap between the first and the last
[15:27:41 CST(-0600)] <Bosmon> colinclark: not yet... yura is still trying to explore whether "focusin" is usable after all
[15:27:59 CST(-0600)] <Bosmon> Since although this patch works, it would be quite awkward to use in many cases without focus delegation
[15:28:15 CST(-0600)] <Bosmon> So far all versions with "focusin" work for me too
[15:31:32 CST(-0600)] <justin_o> colinclark: ah.. i'm not seeing that... seems like in ie6 the right hand column wraps under the other ones
[15:31:43 CST(-0600)] <justin_o> jhung, heidi: ^ have you seen that
[15:31:55 CST(-0600)] <colinclark> My smaller window is causing it, justin_o?
[15:33:33 CST(-0600)] <heidi> i havent
[15:33:37 CST(-0600)] <heidi> but i don't have ie6
[15:35:28 CST(-0600)] <justin_o> colinclark: i think so
[15:35:51 CST(-0600)] <colinclark> heidi: We have to get you this hot IE 6 VM when you're here on Thursday
[15:35:54 CST(-0600)] <colinclark> it even has a debugger!
[15:36:02 CST(-0600)] <heidi> ooo sweet
[15:36:05 CST(-0600)] <heidi> (smile)
[15:37:53 CST(-0600)] <colinclark> Bosmon: Last review question for your ariaLabeller...
[15:38:02 CST(-0600)] <colinclark> I don't see any unit tests for your new live region functionality
[15:38:28 CST(-0600)] <colinclark> Any reason not to write them?
[15:38:34 CST(-0600)] <colinclark> other than time?
[15:38:56 CST(-0600)] <Bosmon> No reason... other than I'm not entirely sure what they could usefully test
[15:39:01 CST(-0600)] <Bosmon> That some text appears in a live region?
[15:39:18 CST(-0600)] <colinclark> Yeah, it's analogous to the test you actually did write (wink)
[15:46:23 CST(-0600)] <colinclark> Bosmon: I wrote you a test for this functionality as part of my review
[15:46:41 CST(-0600)] <colinclark> points out the oddness that you need to constantly tell the labeller that you want an update to be dynamic, but it passes
[15:46:45 CST(-0600)] <colinclark> I'll make a patch
[15:47:05 CST(-0600)] <Bosmon> Well, is that odd? It is conceivable you might want a non-dynamic update
[15:47:12 CST(-0600)] <Bosmon> Actually, it is essential
[15:47:52 CST(-0600)] <Bosmon> If you look at the reorderer, ALL of the updates are non-dynamic except for the one for the element which just moved
[15:49:48 CST(-0600)] <Bosmon> colinclark: I think we have concluded that focusin probably does work after all.... sorry to create extra bother, but could I get you or someone to review an "i" version of the FLUID-3487 patch?
[15:49:59 CST(-0600)] <Bosmon> YURA has written even more taxing test cases for it
[15:51:05 CST(-0600)] <colinclark> Bosmon: I'd argue then that we should be consistent in how we express the dynamicLabel flag
[15:51:29 CST(-0600)] <colinclark> In the case of fluid.updateAriaLabel, we have the label itself, then an optional options block
[15:51:50 CST(-0600)] <colinclark> whereas in the case of update(), we take both the label and the dynamicLabel as options
[15:52:01 CST(-0600)] <colinclark> anyway, it's a minor distinction
[15:52:07 CST(-0600)] <colinclark> and yes, I'll review an i version of the patch
[15:52:08 CST(-0600)] <jhung> justin_o: the jquery-tabs.html escalated test is broken. So I have no way of verifying if linting has had any negative effect. Recommendation?
[15:52:36 CST(-0600)] <justin_o> jhung: how is it broken?
[15:52:39 CST(-0600)] <colinclark> Especially if you'll allow me to add my test to it, Bosmon
[15:52:43 CST(-0600)] <colinclark> http://issues.fluidproject.org/browse/FLUID-3730
[15:52:59 CST(-0600)] <Bosmon> colinclark: For sure
[15:53:02 CST(-0600)] <jhung> justin_o: see here http://build.fluidproject.org/infusion/tests/escalated-tests/html/jquery-tabs.html
[15:53:11 CST(-0600)] <Bosmon> I'm still puzzling about the inconsistency you see in the API
[15:53:27 CST(-0600)] <Bosmon> It's true that "text" is broken out in the "component-free" API
[15:53:28 CST(-0600)] <colinclark> it's actually an issue of how you specify the label itself
[15:53:33 CST(-0600)] <colinclark> yes
[15:53:37 CST(-0600)] <mlam> colinclark: tested my patch for FLUID-3901 in both IE6 and IE8. working well
[15:53:39 CST(-0600)] <colinclark> it's confusing
[15:53:42 CST(-0600)] <colinclark> mlam: super awesome
[15:53:58 CST(-0600)] <justin_o> jhung: hmm... it's probably missing some dependency
[15:54:02 CST(-0600)] <Bosmon> colinclark: Can you think of an API you would prefer? Do you think both APIs should just take one argument, the way "update" does?
[15:54:22 CST(-0600)] <colinclark> Bosmon: Perhaps it's a side-effect of the fact that I know the implementation, but when I invoked fluid.updateAriaLabel(), I expected my options to "stick"
[15:54:36 CST(-0600)] <Bosmon> ah, I see
[15:54:40 CST(-0600)] <colinclark> whereas the dynamicLabel option didn't
[15:54:52 CST(-0600)] <colinclark> So these aren't really "options" in the traditional sense
[15:54:55 CST(-0600)] <jhung> justin_o: probably. Looking now...
[15:54:59 CST(-0600)] <Bosmon> They're not, no
[15:55:09 CST(-0600)] <colinclark> Bosmon: Anyway, I don't think this is a blocker for 1.3
[15:55:10 CST(-0600)] <justin_o> jhung: thanks
[15:55:16 CST(-0600)] <colinclark> I just want to be thorough in my review (smile)
[15:55:17 CST(-0600)] <Bosmon> Although this is an interesting issue of the kind we are facing with the CSpace Confirmation dialog
[15:55:23 CST(-0600)] <Bosmon> Which I think points to a somewhat new kind of API
[15:55:35 CST(-0600)] <Bosmon> If we were JAVER-ites, we would call it a "component factory method"
[16:01:05 CST(-0600)] <colinclark> Bosmon: Well, I think the two issues are separate
[16:01:25 CST(-0600)] <colinclark> I'm quite comfortable with these sorts of things
[16:01:37 CST(-0600)] <colinclark> "component factory method" aka. "just another function"
[16:01:47 CST(-0600)] <colinclark> It's the problem of an option not behaving like an option
[16:02:02 CST(-0600)] <colinclark> dynamicLabel is not "state" for the component
[16:02:28 CST(-0600)] <colinclark> it's relevant on a per-update basis
[16:04:46 CST(-0600)] <colinclark> Bosmon, yura_: To confirm...
[16:05:10 CST(-0600)] <justin_o> Bosmon: i'm trying to figure out why the layoutReorderer has a failing test in IE 6
[16:05:12 CST(-0600)] <heidi> justin_o there are still some dirs in /tests that need linting. i won't be able to stay late to finish them. i'm half-way done component-tests, jhung is doing escalated-tests, and the rest remain
[16:05:14 CST(-0600)] <colinclark> I'm going to commit FLUID-3487-i.patch, with my additional ARIA live region test.
[16:05:17 CST(-0600)] <colinclark> Bosmon, yura_: yes?
[16:05:26 CST(-0600)] <justin_o> Bosmon: but i'm not sure i understand how the labeller works...
[16:05:27 CST(-0600)] <colinclark> I have reviewed it and am sufficiently happy
[16:05:32 CST(-0600)] <justin_o> Bosmon: can you give me a summary
[16:06:10 CST(-0600)] <colinclark> justin_o: Do you just want to come over here and I can walk you through the code?
[16:06:13 CST(-0600)] <colinclark> Since i've just reviewed it?
[16:06:22 CST(-0600)] <justin_o> heidi: okay.. can you please send anastasiac what you have
[16:06:35 CST(-0600)] <heidi> will do
[16:06:41 CST(-0600)] <justin_o> colinclark: thanks
[16:07:37 CST(-0600)] <anastasiac> 426
[16:18:37 CST(-0600)] <Bosmon> anastasiac: I have some problems with your delinting of the renderer files
[16:18:55 CST(-0600)] <Bosmon> Especially since they undo some changes that I repeatedly keep having to make again when they are "delinted"
[16:18:59 CST(-0600)] <anastasiac> oh, good - I'm glad you looked at it. what's up?
[16:19:03 CST(-0600)] <anastasiac> did I break somethign?
[16:19:08 CST(-0600)] <anastasiac> all the tests still pass...
[16:19:10 CST(-0600)] <Bosmon> I think we should be clear that the check for "hasOwnProperty" in particular is not something we do in any of our code
[16:19:18 CST(-0600)] <anastasiac> why not?
[16:19:24 CST(-0600)] <Bosmon> Whenever I find these, I just remove them again (tongue)
[16:19:33 CST(-0600)] <Bosmon> because we code to standards whereby they are not necessary
[16:19:45 CST(-0600)] <Bosmon> I thought we had established this months, if not years ago (tongue)
[16:20:54 CST(-0600)] <anastasiac> sorry, I guess I must have been at the dentist that day. I was going by the the guidelines on the wiki
[16:21:28 CST(-0600)] <anastasiac> I will revert those changes. was there anything else, Bosmon?
[16:23:07 CST(-0600)] <Bosmon> Yes, also you have converted all the "else" calls into the single-line fall
[16:23:11 CST(-0600)] <Bosmon> single-line form
[16:23:48 CST(-0600)] <Bosmon> I can't believe that there is anywhere that says that is a better practice than using a block
[16:23:56 CST(-0600)] <Bosmon> Can you point me to the wiki guidelines pages?
[16:24:04 CST(-0600)] <Bosmon> Does it actually still recommend we use "hasOwnProperty"?
[16:24:45 CST(-0600)] <anastasiac> no, it says to go by "the good parts" recommendations
[16:24:50 CST(-0600)] <anastasiac> http://wiki.fluidproject.org/display/fluid/Coding+and+Commit+Standards
[16:25:41 CST(-0600)] <Bosmon> Well, if you look at the screenshot, it has a tick there on the setting "tolerate unfiltered for in"
[16:26:25 CST(-0600)] <Bosmon> If possible, I would like to revert this commit since I wonder if there is a way of delinting the file such that the whole file doesn't change
[16:26:30 CST(-0600)] <Bosmon> At the moment the diff covers the entire file
[16:26:57 CST(-0600)] <michelled> the problem with using a screen shot here is that what is checked for 'good parts' does change over time
[16:27:05 CST(-0600)] <Bosmon> I see
[16:27:29 CST(-0600)] <michelled> the screen shot itself doesn't conform to the settings that are described below - notice that 'one var per function' is checked
[16:27:36 CST(-0600)] <Bosmon> How about the "else" issue
[16:27:39 CST(-0600)] <Bosmon> I don't see a setting for this
[16:27:51 CST(-0600)] <Bosmon> But if there was one, I would assume it would be for the REVERSE of what happened here (tongue)
[16:27:59 CST(-0600)] <anastasiac> michelled, so what settings should we be using?
[16:27:59 CST(-0600)] <Bosmon> That is, to convert "dangling else" into the block form
[16:28:28 CST(-0600)] <anastasiac> should I go by the text description only?
[16:28:49 CST(-0600)] <Bosmon> function renderDebugMessage(message){
[16:28:53 CST(-0600)] <michelled> I'm really surprised that lint wanted the else to be single line - what did it say?
[16:28:58 CST(-0600)] <Bosmon> Also, spaces have been removed between argument ends and open blocks
[16:29:16 CST(-0600)] <Bosmon> Could we revert this commit, and continue with this issue using patches as we are for all our other current issues?
[16:29:53 CST(-0600)] <Bosmon> I see you're continuing to comit delinting directly to trunk (sad)
[16:30:06 CST(-0600)] <anastasiac> I'm following the king's instructions on this one
[16:30:13 CST(-0600)] <Bosmon> justin_o !
[16:30:24 CST(-0600)] <michelled> Bosmon: if it is reverted could you lint the renderer?
[16:30:47 CST(-0600)] <michelled> I think one of the difficulties of linting things that have never passed lint is that it's difficult to see what's intended and what's not
[16:31:06 CST(-0600)] <Bosmon> Actually I have passed those files through lint several times (tongue)
[16:31:14 CST(-0600)] <Bosmon> Yes
[16:33:03 CST(-0600)] <anastasiac> michelled, so what configuration of jslint should I be following? There's lots more to go through, and I don't want to have to re-do it.
[16:34:25 CST(-0600)] <michelled> anastasiac: I think the text description is good but we should add 'tolerate unfiltered for in'
[16:34:25 CST(-0600)] <justin_o> FLUID-3901 has been reviewed, committed, tested, closed (not in that order)
[16:34:39 CST(-0600)] <michelled> Bosmon: do you think there are other settings that we should either turn on or off?
[16:35:36 CST(-0600)] <mlam> jhung_afk: anastasiac: are you guys still linting files?
[16:35:50 CST(-0600)] <anastasiac> mlam, I'm still at it
[16:36:05 CST(-0600)] <mlam> any directory you want me to pick up to make things easier?
[16:37:55 CST(-0600)] <Bosmon> Well, of the tick boxes in the picture, "Strict white space" and "tolerate sloppy line breaking" could be dodgy... I can't say what those actually do, concretely
[16:38:12 CST(-0600)] <Bosmon> Also "Allow one var statement per function" seems incorrect, but perhaps it means "one var statement per variable"
[16:38:52 CST(-0600)] <michelled> Bosmon: strict whitespace is one we all follow. it mostly tells us where to put spaces
[16:39:17 CST(-0600)] <anastasiac> mlam, I'm working my way through the framework files in alphabetical order - if you want to start at the bottom of the list, and post a patch every file or two, that would help
[16:39:22 CST(-0600)] <michelled> the 'allow one var statement per function' we never follow
[16:39:30 CST(-0600)] <heidi> hey colinclark did you lint componente-tests/uploader/js files ?
[16:39:38 CST(-0600)] <mlam> ok, what's the JIRA that i should be posting under?
[16:39:42 CST(-0600)] <Bosmon> Does it say they should be missing in lines like this? function renderDebugMessage(message){
[16:39:48 CST(-0600)] <anastasiac> FLUID-3757, mlam
[16:39:48 CST(-0600)] <Bosmon> colinclark went home
[16:39:52 CST(-0600)] <justin_o> heidi: colinclark has left
[16:39:55 CST(-0600)] <mlam> ok, thanks
[16:40:00 CST(-0600)] <heidi> ah shoot. do you know if he did justin_o
[16:40:02 CST(-0600)] <heidi> or mlam
[16:40:03 CST(-0600)] <anastasiac> Bosmon, I will revert the changes and post a patch of them
[16:40:10 CST(-0600)] <michelled> Bosmon: anastasiac is about to revert the renderer linting and create a patch
[16:40:22 CST(-0600)] <mlam> heidi: yes, colin linted the uploader test files
[16:40:28 CST(-0600)] <justin_o> heidi: he said he had linted all the uploader files.. so i'd guess so
[16:40:28 CST(-0600)] <heidi> cool thanks mike
[16:40:31 CST(-0600)] <mlam> np
[16:40:49 CST(-0600)] <Bosmon> Thanks, all
[16:41:04 CST(-0600)] <Bosmon> And also, anywhere in the framework where a "hasOwnProperty" check has been inserted
[16:41:09 CST(-0600)] <Bosmon> These shouldn't occur in any code
[16:41:53 CST(-0600)] <anastasiac> right
[16:43:40 CST(-0600)] <michelled> we can deal with looking at it and committing it tonight or tomorrow
[16:44:00 CST(-0600)] <heidi> justin_o finished tests/component-tests and sent patch to anastasiac
[16:44:53 CST(-0600)] <Bosmon> So in the linted version for example there is both this:
[16:45:02 CST(-0600)] <anastasiac> Justin_o, should I commit de-linting changes to trunk, or post a patch on the JIRA?
[16:45:05 CST(-0600)] <Bosmon> fluid.explodeSelectionToInputs = function (optionlist, opts) {
[16:45:07 CST(-0600)] <Bosmon> as well as this
[16:45:19 CST(-0600)] <Bosmon> fluid.explode = function(hash, basepath){
[16:45:27 CST(-0600)] <justin_o> anastasiac: you mean the reversal for that file
[16:45:27 CST(-0600)] <justin_o> {color}
[16:45:32 CST(-0600)] <justin_o> ?
[16:45:45 CST(-0600)] <anastasiac> no, I mean for any and all de-linting changes I make
[16:46:44 CST(-0600)] <anastasiac> michelled, Bosmon: I've reverted the renderer changes, and posted a patch of the changes on FLUID-3757
[16:47:25 CST(-0600)] <justin_o> anastasiac: maybe we'll go with patches now.. and then commit all at once
[16:48:48 CST(-0600)] <anastasiac> who should review my patches, Justin_o?
[16:50:03 CST(-0600)] <justin_o> anastasiac: good question...
[16:50:13 CST(-0600)] <justin_o> Bosmon: you want to review and commit all the linting patches
[16:51:08 CST(-0600)] <Bosmon> Yes, it will be fine
[16:51:13 CST(-0600)] <justin_o> Bosmon: thanks...
[16:51:15 CST(-0600)] <Bosmon> Thanks, all
[16:53:33 CST(-0600)] <mlam> anastasiac: posted patch for bottom 3 files.
[16:53:54 CST(-0600)] <mlam> where abouts are you in the framework list?
[16:53:57 CST(-0600)] <anastasiac> thanks, mlam - Bosmon is going to review the patches
[16:54:05 CST(-0600)] <mlam> np
[16:54:29 CST(-0600)] <anastasiac> mlam, I've committed DataBinding and Fluid, I've done FluidDebugging, and was about to start on FluidDocument
[16:54:46 CST(-0600)] <anastasiac> and will post a patch of those two, then keep going (smile)
[16:55:10 CST(-0600)] <mlam> ok, i can do the next 3 if you'd like. IoC and view and requests
[16:56:54 CST(-0600)] <anastasiac> great, thanks mlam
[16:56:58 CST(-0600)] <jessm> i'm about to head out – everyone ok in here?
[16:57:15 CST(-0600)] <mlam> np
[17:03:40 CST(-0600)] <anastasiac> mlam, do you want to modify the comment that accompanied your last de-linting patch on the JIRA to include the file name and a more explicit description of exactly which files are covered? That might make it easier for Bosom, since there are several files
[17:04:11 CST(-0600)] <mlam> ya, i will. i'm actually going to delete my patch. my linting settings were too forgiving (sad)
[17:04:55 CST(-0600)] <anastasiac> mlam, check the coding and commit standards wiki page
[17:05:11 CST(-0600)] <anastasiac> there's a description of which jslint settings to use
[17:05:13 CST(-0600)] <anastasiac> updated today
[17:05:22 CST(-0600)] <anastasiac> http://wiki.fluidproject.org/display/fluid/Coding+and+Commit+Standards
[17:05:35 CST(-0600)] <anastasiac> Justin_o, does all the delinting have to be finished today?
[17:05:49 CST(-0600)] <justin_o> anastasiac: preferably... there's a lot eh
[17:05:54 CST(-0600)] <justin_o> how much is left?
[17:06:16 CST(-0600)] <anastasiac> well, heidi accidentally didn't attach her file to the email she sent me, so I don't have her patch for the tests
[17:06:40 CST(-0600)] <anastasiac> she was working with jhung on them - I'm not sure know how far he got, either
[17:07:22 CST(-0600)] <anastasiac> with mlam's help, the framework files are in good shape - but only patches on the JIRA, not all committed
[17:07:32 CST(-0600)] <anastasiac> I think all the demos are good
[17:07:38 CST(-0600)] <anastasiac> I can double-check them
[17:12:52 CST(-0600)] <justin_o> anastasiac: okay.. i guess Bosmon can review and commit whatever is there tonight and then we'll try to knock the rest off asap tomorrow
[17:12:59 CST(-0600)] <anastasiac> so justin_o, status of de-linting: tests not completed; everything else seems to be either done or has a relevant patch attached to the JIRA (once mlam finishes the last three framework files). some of the tests have been delinted, we just don't have the patches
[17:13:17 CST(-0600)] <anastasiac> cool - finish tests tomorrow
[17:13:29 CST(-0600)] <anastasiac> I'll email heidi, in case she gets a chance to attach the patch to the jira
[17:13:31 CST(-0600)] <justin_o> anastasiac: yep.. worst case i think we can live with that if the tests all pass
[17:13:36 CST(-0600)] <justin_o> anastasiac: thanks
[17:14:08 CST(-0600)] <anastasiac> ah, email is wonky - I do have heidi's patch. I will attach it, and we'll see tomorrow where we are
[17:17:17 CST(-0600)] <anastasiac> ok Bosmon, there are several patches attached to FLUID-3757, and one or two more to come from mlam. That should cover everything except some of the test files - we'll have to finish that tomorrow. I have to take off now