fluid-work IRC Logs-2011-05-30

Owned by Fluid IRC Log Bot
Last updated: May 30, 2011Version comment
16 min read

[06:44:10 CDT(-0500)] <tom_i> when I run a clean install of Ringo using ringo main.js I receive the following…
[06:44:13 CDT(-0500)] <tom_i> Cannot find module 'config' (ringo/jsgi/connector.js#27)
[06:46:01 CDT(-0500)] <tom_i> I just pulled from git
[06:48:11 CDT(-0500)] <tom_i> ignore ^
[06:48:14 CDT(-0500)] <tom_i> coffee!
[08:16:36 CDT(-0500)] <heidi> Justin_o is the mergey UIO of michelles on the build site anywhere?
[08:17:09 CDT(-0500)] <Justin_o> heidi: not yet sorry... i am having some problems getting it to build a branch other than master
[08:17:59 CDT(-0500)] <heidi> Justin_o ok cool, just wonderin
[08:18:32 CDT(-0500)] <Justin_o> heidi: okay.. if i have time today i'll try to look at it again.. although not sure I will
[08:18:39 CDT(-0500)] <heidi> no worries
[09:17:44 CDT(-0500)] <mlam> michelled: I'm getting something really strange with your FLUID-4267 branch. If I do a clean clone of your entire fork, checkout this FLUID-4267 branch, a git diff, there's a big change in src/webapp/lib/jquery/ui/js/ui.ariaTabs.js . I'm wondering if you're getting the same
[09:18:17 CDT(-0500)] <michelled> that sounds very odd mlam
[09:19:47 CDT(-0500)] <michelled> mlam: not happening for me
[09:19:53 CDT(-0500)] <michelled> do you have uncommitted changes?
[09:19:56 CDT(-0500)] <mlam> ok, thanks
[09:20:08 CDT(-0500)] <mlam> Not in the cloned fork
[09:20:20 CDT(-0500)] <michelled> mlam: you can use 'stash' to set aside uncommitted changes if you want to
[09:20:56 CDT(-0500)] <michelled> and the clone is completely separate from your other clones mlam?
[09:21:04 CDT(-0500)] <mlam> yes michelled
[09:37:53 CDT(-0500)] <mlam> Justin_o: I have this weird message after making a fresh clone of michelled's fork: "CRLF will be replaced by LF" Would you know off the top of your head which git setting I need to set?
[09:41:11 CDT(-0500)] <mlam> Justin_o: nvm, i have it
[09:41:39 CDT(-0500)] <Justin_o> mlam: okay..
[09:51:04 CDT(-0500)] <cindyli> colinclark: michelled, i'm stuck at calculating "line spacing" by "times" due to an IE bug. in detail, if "line-height" is set to 1em in css, container.css("line-height") returns "1px" in IE. the same code works well with other browsers. This is a known bug recorded in jquery (http://bugs.jquery.com/ticket/2671) but rooted to IE and jquery says it is not going to tackle. any idea?
[09:52:31 CDT(-0500)] <michelled> how about we switch to cspace supported browsers (wink)
[09:53:04 CDT(-0500)] <jhung> justin_o: I'm ready to help out with your design task today.
[09:53:17 CDT(-0500)] <michelled> cindyli: it looks like you have no choice but to put in some browser sniffing and adjust for the bug in IE
[09:55:33 CDT(-0500)] <michelled> cindyli: the conversation on the bug report seems to say that if you are explicit with and say 'em' it will work
[09:55:37 CDT(-0500)] <michelled> cindyli: can you try that?
[09:56:05 CDT(-0500)] <jhung> colinclark: a little bird told me there's design work to be done on the Image Uploader?
[10:03:07 CDT(-0500)] <cindyli> michelled: yes, if the unit (em or px etc) is specified, it would work, but can we force users to specify it since if the unit is missing, "em" is the default.
[10:03:30 CDT(-0500)] <mlam> cindyli: the auto-save work has been committed to michelled's branch
[10:03:32 CDT(-0500)] <michelled> cindyli: we could add 'em' if it's not specified
[10:04:30 CDT(-0500)] <cindyli> michelled: means we need to scan thru css and compute the value?
[10:04:47 CDT(-0500)] <cindyli> because ie would return "1px" if "em" is missing, cannot rely on
[10:05:19 CDT(-0500)] <michelled> cindyli: which css would this exist in? I think I don't know enough about the issue yet
[10:06:56 CDT(-0500)] <cindyli> michelled: the setting is eventually applied to the user's page, the css can be in any css files that are included in the the user's page
[10:07:50 CDT(-0500)] <cindyli> thanks, mlam!
[10:07:57 CDT(-0500)] <michelled> and are you seeing this in any of our examples?
[10:08:17 CDT(-0500)] <cindyli> yes, with full page with preview
[10:08:44 CDT(-0500)] <cindyli> default element style is line-height: 1
[10:10:28 CDT(-0500)] <michelled> I think we should do 3 things: 1. change our example to specify 'em'. 2. put in a minimum line height that we don't go smaller then (ask jameswy what this should be. 3. open a JIRA that described the problem and mention that style sheets specifying line height should include units
[10:10:31 CDT(-0500)] <michelled> what do you think?
[10:10:34 CDT(-0500)] <michelled> is that reasonable?
[10:11:38 CDT(-0500)] <colinclark> jhung: Yep, I sent an email about it to the mailing list
[10:11:46 CDT(-0500)] <jameswy> michelled, cindyli: line-height, like text size, is a multiplier on the current line heights. The minimum/default should be 1.0x
[10:11:55 CDT(-0500)] <colinclark> We may need Justin_o to get it installed on the daily build server
[10:12:14 CDT(-0500)] <colinclark> but then I think it could use some love, jhung
[10:12:39 CDT(-0500)] <cindyli> sounds good, michelled. thx, jameswy
[10:12:40 CDT(-0500)] <michelled> jameswy: the issue is that people can specify line height in a style sheet without units and in all browsers except IE it is assumed to be em
[10:12:49 CDT(-0500)] <michelled> jameswy: in IE it's assumed to be px
[10:13:01 CDT(-0500)] <jhung> colinclark: okay. I'll take a look at the content in github now, but it'll help to have a functioning demo to work from.
[10:13:11 CDT(-0500)] <michelled> so a style sheet specifying line height '1' ends up as 1px in IE
[10:13:26 CDT(-0500)] <michelled> I don't think we want to let users do that to themselves (smile)
[10:13:46 CDT(-0500)] <michelled> cindyli: can you tell in IE that it's being set to pixels?
[10:13:58 CDT(-0500)] <michelled> other then by looking at it I mean (smile)
[10:14:07 CDT(-0500)] <michelled> programatically
[10:18:29 CDT(-0500)] <Justin_o> jhung: did you have a minute to look at the work i've done so far for the conference demo.. if you could let me know what you think
[10:19:36 CDT(-0500)] <jameswy> michelled: Is there any way around the IE issue?
[10:19:43 CDT(-0500)] <jhung> justin_o: is that available on the build server?
[10:19:59 CDT(-0500)] <Justin_o> jhung: saddly not.. but i can send you a link to my github repo
[10:20:05 CDT(-0500)] <jhung> k
[10:21:11 CDT(-0500)] <Justin_o> jhung: you can download this branch https://github.com/jobara/workshops/tree/FLUID-4269
[10:22:09 CDT(-0500)] <Justin_o> jhung: i'm trying to demostrate some of the new input types in html5, in particular here will be date, search, url, tel, number and datalist (although this doesn't work in safari so may be dropped)
[10:22:32 CDT(-0500)] <Justin_o> i've tried to make the demo look like one of those travel search widgets, and styled it something like a boarding pass
[10:22:33 CDT(-0500)] <michelled> jameswy: cindyli is looking into it but she doesn't think so
[10:23:00 CDT(-0500)] <jameswy> michelled: So, in short, it's impossible to do a line-height multiplier in IE?
[10:23:20 CDT(-0500)] <jhung> Justin_o: It looks pretty good! I'll take a closer look. You need design and design implementation work done, right?
[10:23:39 CDT(-0500)] <michelled> jameswy: no, it should be fine if the user specifies units when setting line height
[10:23:50 CDT(-0500)] <michelled> the issue is if they don't specify units
[10:24:07 CDT(-0500)] <Justin_o> jhung: yes.. specifically the first one, but sure if you want to help out with the second (smile)
[10:24:26 CDT(-0500)] <jhung> justin_o: I'll see what I can do. (smile)
[10:24:38 CDT(-0500)] <Justin_o> jhung: the other thing i was planning on looking at was to be able to transform the layout for mobile.. possibly using mobile-fss and media queries
[10:24:54 CDT(-0500)] <jhung> justin_o: so text input fields is what is demonstrated - no spinners, radios, etc?
[10:25:10 CDT(-0500)] <jhung> justin_o: interesting...
[10:25:14 CDT(-0500)] <Justin_o> jhung: yes... let me send you a link to some info on the new stuff
[10:25:17 CDT(-0500)] <jhung> justin_o: how high of a priority is that?
[10:25:24 CDT(-0500)] <jhung> (re: mobile transformation)
[10:25:50 CDT(-0500)] <Justin_o> jhung: it would be nice if possible.. just so you know it would need to be done by tomorrow since the conference is wednesday morning
[10:26:04 CDT(-0500)] <Justin_o> so if we don't have enough time then i guess it will have to go
[10:26:18 CDT(-0500)] <jhung> justin_o: okay
[10:26:27 CDT(-0500)] <colinclark> Justin_o: yay media queries
[10:27:20 CDT(-0500)] <Justin_o> colinclark: they are pretty cool.. i wonder if we'll have to reevaluate our implementation of mobile-fss or provide another bridge between it and the standard fss
[10:28:02 CDT(-0500)] <colinclark> I guess maybe the simplest thing to do for now might be to do an @import based on a media query
[10:28:03 CDT(-0500)] <colinclark> if it's possible
[10:28:19 CDT(-0500)] <colinclark> meaning, if we see that, say, the screen size is smaller than x, @import the mobile-fss
[10:28:52 CDT(-0500)] <colinclark> michelled: Does this article change your mind about your suggested approach to handling line height? http://meyerweb.com/eric/thoughts/2006/02/08/unitless-line-heights/
[10:28:52 CDT(-0500)] <Justin_o> jhung: here's some more info about the new inputs http://html5accessibility.com/ http://diveintohtml5.org/forms.html
[10:29:22 CDT(-0500)] <Justin_o> colinclark: i guess you wouldn't be able to go back though
[10:29:30 CDT(-0500)] <colinclark> go back?
[10:29:54 CDT(-0500)] <jhung> justin_o: ok got it. I'll get started on that now.
[10:30:09 CDT(-0500)] <Justin_o> colinclark: i think you can't use the regular layout and mobile layout at the same time
[10:30:20 CDT(-0500)] <Justin_o> jhung: thanks a lot
[10:31:46 CDT(-0500)] <jhung> justin_o: any chance on getting an image uploader running on a server so I can see what I does currently? I don't have a server.
[10:41:17 CDT(-0500)] <jhung> whoops. Totally missed stand-up.
[10:41:24 CDT(-0500)] <Justin_o> jhung: hmm... there's an old version that is being served
[10:41:28 CDT(-0500)] <Justin_o> jhung: (smile) oops
[10:41:51 CDT(-0500)] <Justin_o> jhung: i'm not sure if any of the visual stylings of it have changed, cindy and colin might know
[10:42:16 CDT(-0500)] <Justin_o> jhung: actually i might be able to just switch to latest version in colin's repo
[10:42:18 CDT(-0500)] <Justin_o> one second
[10:49:03 CDT(-0500)] <Justin_o> jhung: sorry.. there's some sort of permissions error.. i'll have to see if i can get colinclark to delete the temp files... then i should be able to update the image gallery and build it for you
[10:49:16 CDT(-0500)] <jhung> Thanks!
[10:58:36 CDT(-0500)] <michelled> mlam: can you also update the 4267 with the branches you have and are merging in?
[10:59:04 CDT(-0500)] <mlam> Ok, sure
[10:59:25 CDT(-0500)] <michelled> thx
[10:59:42 CDT(-0500)] <mlam> np
[11:11:48 CDT(-0500)] <harriswong> mlam: i am having problems with master uploader test, could it be just my machine, can you verify please..
[11:12:02 CDT(-0500)] <mlam> yup, i'll take a look
[11:12:11 CDT(-0500)] <mlam> what browser harriswong?
[11:12:19 CDT(-0500)] <harriswong> FF4 on ubuntu
[11:12:26 CDT(-0500)] <mlam> ok
[11:12:32 CDT(-0500)] <harriswong> Uploader-test.html mlam, thanks
[11:13:33 CDT(-0500)] <mlam> harriswong: working for me
[11:13:36 CDT(-0500)] <mlam> which test is failing?
[11:14:00 CDT(-0500)] <harriswong> mine halts on #4
[11:14:06 CDT(-0500)] <harriswong> i wonder if it's cause I have console open
[11:18:31 CDT(-0500)] <harriswong> mlam: try this: http://172.16.2.83/fluid/infusion/src/webapp/tests/component-tests/uploader/html/Uploader-test.html
[11:20:47 CDT(-0500)] <mlam> yah, it's broken on test #4 for me.
[11:21:09 CDT(-0500)] <mlam> i'm not sure why it's happening, but it's definitely running fine off the most up-to-date master
[11:23:26 CDT(-0500)] <harriswong> mlam: thanks!
[11:26:29 CDT(-0500)] <harriswong> Justin_o: so for http://issues.fluidproject.org/browse/FLUID-4169, to ensure all views are linked to DataBinding, is it sufficient enough to ensure all unit tests passed?
[11:26:53 CDT(-0500)] <Justin_o> harriswong: it would be for the unit tests
[11:27:02 CDT(-0500)] <Justin_o> harriswong: but you would have to make sure that all the html files that need it, import it
[11:28:06 CDT(-0500)] <harriswong> Justin_o: Right, I meant on top, demos are working and throwing no errors as well.
[11:31:12 CDT(-0500)] <Justin_o> harriswong: okay.. great.. did you happen to look through the integration-demos, stand alone demos and manual tests
[11:31:17 CDT(-0500)] <Justin_o> if that's all done then i guess we're good
[11:31:54 CDT(-0500)] <harriswong> Justin_o: Will process with integration-demo, manual tests, etc.
[11:32:36 CDT(-0500)] <Justin_o> harriswong: thanks
[11:57:59 CDT(-0500)] <cindyli> michelled: there's a solution for line-height issue with ie. i poked directly into ie of currentStyle.lineHeight, if given "1.875", ie actually returns "1.875", it was jquery that was trying to convert all units into px had no idea what the missing unit means so it added in "px".
[11:58:35 CDT(-0500)] <michelled> cindyli: colinclark is also looking at the issue and found the same thing
[11:58:37 CDT(-0500)] <cindyli> so the solution would be, if (ie && unit is missing) we assume "em". is this ok?
[11:59:24 CDT(-0500)] <michelled> he also pointed out that my earlier suggestion would do strange things to child elements and that we should actually not use units when setting line height
[11:59:33 CDT(-0500)] <colinclark> cindyli: Hang on one sec
[11:59:39 CDT(-0500)] <colinclark> I need to do a little bit more testing
[11:59:55 CDT(-0500)] <cindyli> thanks, colinclark
[12:00:12 CDT(-0500)] <colinclark> this is a really interesting problem (smile)
[12:00:21 CDT(-0500)] <cindyli> agree
[12:02:10 CDT(-0500)] <colinclark> cindyli: Okay, so yes
[12:02:13 CDT(-0500)] <colinclark> You're right
[12:02:17 CDT(-0500)] <colinclark> I ran a few extra tests
[12:02:33 CDT(-0500)] <colinclark> my biggest concern was ensuring that a parent's lineHeight style would be correctly set on a child element
[12:02:53 CDT(-0500)] <colinclark> So, let's say that the UIE's container isn't the whole page
[12:03:07 CDT(-0500)] <colinclark> let's say there was something like this:
[12:03:59 CDT(-0500)] <colinclark> <div id="parent" style="line-height: 2.5;"> <div id="ui-enhanceable"> page content is here </div> </div>
[12:04:10 CDT(-0500)] <colinclark> We really don't even know that "parent" exists
[12:04:13 CDT(-0500)] <colinclark> and shouldn't
[12:04:54 CDT(-0500)] <colinclark> but will a call to document.getElementById("ui-enhanceable").currentStyle.lineHeight provide a real value?
[12:05:01 CDT(-0500)] <colinclark> reflecting the line height inherited by its parent
[12:05:05 CDT(-0500)] <colinclark> and the answer appears to be yes
[12:05:21 CDT(-0500)] <colinclark> So, this seems like a viable option to you, cindyli?
[12:10:55 CDT(-0500)] <cindyli> colinclark: here u go about line-height "normal", which is actually abnormal: http://meyerweb.com/eric/thoughts/2008/05/06/line-height-abnormal/
[12:22:24 CDT(-0500)] <jhung1> justin_o: I have a design ready for you. Where do you want it?
[12:25:03 CDT(-0500)] <Justin_o> jhung: thanks.. i guess you can throw it up on the jira
[12:25:09 CDT(-0500)] <Justin_o> it's um.. let find it
[12:26:59 CDT(-0500)] <Justin_o> jhung: here it is http://issues.fluidproject.org/browse/FLUID-4269
[12:28:47 CDT(-0500)] <jhung> Justin_o: Done.
[12:29:00 CDT(-0500)] <jhung> justin_o: after you've looked at it, maybe we can talk it over?
[12:31:58 CDT(-0500)] <Justin_o> cindyli: do you know if the image gallery that you worked will have a problem if the temp directory is already existing
[12:32:18 CDT(-0500)] <cindyli> Justin_o: no
[12:32:22 CDT(-0500)] <Justin_o> jhung: sure taking a look now
[12:32:24 CDT(-0500)] <Justin_o> cindyli: thanks
[12:32:44 CDT(-0500)] <Justin_o> jhung: actually give me a few minutes.. will try to get the image gallery served first
[12:32:44 CDT(-0500)] <cindyli> np, Justin_o, u may get an unix error at running install.sh though
[12:32:52 CDT(-0500)] <Justin_o> cindyli: oh really
[12:32:54 CDT(-0500)] <jhung> thanks justin_o
[12:33:05 CDT(-0500)] <cindyli> think so, that's where temp dir is created
[12:33:17 CDT(-0500)] <Justin_o> cindyli: anyway to check for it's presence first?
[12:33:38 CDT(-0500)] <cindyli> Justin_o: ah, u mean in install.sh? ya, we could
[12:34:20 CDT(-0500)] <Justin_o> cindyli: yes.. i think so... we had something in the build script on the server that would remove it first, but it doesn't have permissions to delete the files from within the temp directory so was failing
[12:35:04 CDT(-0500)] <cindyli> icic. Justin_o, let me improve install.sh to check the dir existence
[12:37:45 CDT(-0500)] <Justin_o> cindyli: thanks
[12:38:04 CDT(-0500)] <cindyli> np
[12:42:25 CDT(-0500)] <Justin_o> jhung: okay.. did you want to chat now about the design?
[12:42:42 CDT(-0500)] <jhung> Sure
[12:42:52 CDT(-0500)] <Justin_o> jhung: skype?
[12:42:56 CDT(-0500)] <jhung> sure
[12:43:00 CDT(-0500)] <Justin_o> okay.. logging in
[12:55:03 CDT(-0500)] <harriswong> Justin_o: for FLUID-4169, i have gone through the files listed in the comment; am I missing anything? If not, what would be the next step? thanks.
[12:58:14 CDT(-0500)] <Justin_o> harriswong: if they were all working then i guess all the issues were caught already... so if you didn't need to make any changes you can just close the jira off
[13:01:52 CDT(-0500)] <harriswong> Justin_o: okay. which other issue would you like me to work on next?
[13:02:08 CDT(-0500)] <Justin_o> harriswong: hmm.. let me take a look
[13:02:30 CDT(-0500)] <harriswong> Justin_o: Would the resolution be "Won't fix"?
[13:03:06 CDT(-0500)] <Justin_o> harriswong: i think you can make it fixed and just mention that all the issues have been caught already or something like that
[13:03:15 CDT(-0500)] <harriswong> Justin_o: k
[13:06:22 CDT(-0500)] <Justin_o> harriswong: here's the bug parade report http://issues.fluidproject.org/secure/ConfigureReport.jspa?atl_token=hLyns-VDmh&amp;filterid=10128&amp;mapper=components&amp;selectedProjectId=10001&amp;reportKey=com.atlassian.jira.plugin.system.reports%3Asinglelevelgroupby&amp;Next=Next
[13:06:35 CDT(-0500)] <Justin_o> harriswong: i think maybe start knocking off those "demos" issues
[13:07:21 CDT(-0500)] <harriswong> Justin_o: ok
[13:10:47 CDT(-0500)] <Justin_o> Bosmon: some pull request stuff to chat with you about
[13:11:39 CDT(-0500)] <harriswong> Justin_o: i think 4109 should be in "demo" instead of layoutReorderer, should I change it?
[13:12:20 CDT(-0500)] <Justin_o> harriswong: it should probably be in both.. could you add "demos" as a component for that one
[13:12:29 CDT(-0500)] <harriswong> Justin_o: k
[13:12:33 CDT(-0500)] <Justin_o> harriswong: thanks
[13:12:35 CDT(-0500)] <Justin_o> good catch
[13:12:38 CDT(-0500)] <harriswong> np
[13:13:56 CDT(-0500)] <Justin_o> Bosmon: so to start, could you please look at the FLUID-4132, and the (FLUID-4235, FLUID-4205) pull request as well
[13:15:17 CDT(-0500)] <Justin_o> Bosmon: also had some questions for you about the pull request you made
[13:15:50 CDT(-0500)] <Bosmon> Justin_o - sure
[13:16:42 CDT(-0500)] <Justin_o> Bosmon: so for the FLUID-3953 pull request you made..
[13:17:11 CDT(-0500)] <Justin_o> from the jquery 1.6.1 change log, they have a chart that shows the recommened uses of attr and prop.
[13:17:11 CDT(-0500)] <Justin_o> http://blog.jquery.com/2011/05/12/jquery-1-6-1-released/
[13:17:25 CDT(-0500)] <Justin_o> it seems like they suggest that you use attr for "id"
[13:18:44 CDT(-0500)] <Justin_o> Bosmon: also it seems like not all of the uses of .attr("id") were switched to .prop.. not sure if this was intentional or if the were just missed
[13:19:44 CDT(-0500)] <Justin_o> Bosmon: next i think we should have a proper unit test for this in the framework unit tests..
[13:20:35 CDT(-0500)] <Justin_o> Bosmon: I've tested this in IE6, IE7, IE8, safari and FF4 and it seems to have worked.
[13:31:26 CDT(-0500)] <Bosmon> Justin_o - as far as I can tell, their advice is not correct (smile)
[13:31:36 CDT(-0500)] <Bosmon> Justin_o - can you tell me which instances were not switched?
[13:31:39 CDT(-0500)] <Bosmon> I tried to catch them all
[13:31:59 CDT(-0500)] <Bosmon> I'm sure they will revise their advice in the light of evidence from IE9 (smile)
[13:33:11 CDT(-0500)] <Justin_o> Bosmon: okay.. we might want to send them a note about that
[13:33:24 CDT(-0500)] <Justin_o> Bosmon: i'm writing up the files on the pull request as we speak
[13:34:48 CDT(-0500)] <Justin_o> Bosmon: comment is on the pull request
[13:35:02 CDT(-0500)] <Bosmon> Justin_o - thanks
[13:35:07 CDT(-0500)] <Bosmon> I agree we should have a dedicated unit test
[13:35:14 CDT(-0500)] <Bosmon> It will be easy to put one in FluidJSTests.js
[13:35:28 CDT(-0500)] <Justin_o> Bosmon: great, thanks
[13:51:12 CDT(-0500)] <michelled> Bosmon, Justin_o: I think you can leave the 4235, 4205 pull request because mlam is going to merge it into the 4267 uioptions merge branch
[13:51:20 CDT(-0500)] <michelled> we can do the pull from the branch instead
[13:52:02 CDT(-0500)] <Justin_o> michelled: okay
[14:11:43 CDT(-0500)] <Justin_o> cindyli: for you the uploader server, did the url to the start page change?
[14:12:00 CDT(-0500)] <cindyli> yes
[14:12:12 CDT(-0500)] <cindyli> image-gallery.php
[14:12:17 CDT(-0500)] <Justin_o> cindyli: okay.. thanks
[14:12:21 CDT(-0500)] <cindyli> np
[14:12:42 CDT(-0500)] <Justin_o> colinclark: would you be able to change the server to mount image-gallery.php instead
[14:13:11 CDT(-0500)] <Justin_o> colinclark: never mind.. the directory was mounted.. i can just change the url
[14:13:28 CDT(-0500)] <Justin_o> jhung: you can see the image gallery now http://build.fluidproject.org/uploader/hourly/image-gallery.php
[14:14:52 CDT(-0500)] <jhung> justin_o, cindyli: excellent! Thanks!
[14:15:09 CDT(-0500)] <jhung> justin_o: how soon does this design need to be done?
[14:15:46 CDT(-0500)] <Justin_o> jhung: well.. we're hoping to use this for testing which starts monday
[14:16:34 CDT(-0500)] <jhung> Justin_o: So you wanted this yesterday. (big grin)
[14:16:55 CDT(-0500)] <Justin_o> jhung: last week actually (smile)
[14:17:10 CDT(-0500)] <colinclark> So, umm, this is pretty amazing: https://developer.mozilla.org/en-US/demos/detail/doom-on-the-web
[14:17:41 CDT(-0500)] <jhung> justin_o: lol. Okay. Then perhaps I'll get started on the time machine instead.
[14:17:49 CDT(-0500)] <jhung> I'll try to get something as quickly as I can.
[14:17:57 CDT(-0500)] <Justin_o> jhung: (smile) thanks
[14:18:11 CDT(-0500)] <Justin_o> colinclark: wow.. now you can really play doom everywhere
[14:18:24 CDT(-0500)] <colinclark> pretty crazy stuff
[14:18:30 CDT(-0500)] <colinclark> I played a lot of it when I was younger
[14:18:34 CDT(-0500)] <colinclark> and now it runs in JavaScript
[14:18:46 CDT(-0500)] <Justin_o> colinclark: i don't think i ever got out of the first room
[15:04:20 CDT(-0500)] <jhung> colinclark, cindyli: Question about the behaviour of the image uploader.
[15:06:01 CDT(-0500)] <jhung> nevermind, cindyli and colinclark, I answered my own question.
[15:06:27 CDT(-0500)] <jhung> I was curious what happens if the user attempts to queue a group of files that exceed the queue space.
[15:07:58 CDT(-0500)] <cindyli> jhung: cool that u figured out. do u mean exceeding the queue setting that is set at UI? nothing happens in this case, meaning the files are not queued and no error reported, am i right?
[15:08:35 CDT(-0500)] <jhung> cindyli: yes that's exactly what happens. I hope there's the ability to report an error to the user.
[15:09:03 CDT(-0500)] <cindyli> ya, my hope as well. (tongue)
[15:14:07 CDT(-0500)] <colinclark> jhung, cindyli: That's coming
[15:14:28 CDT(-0500)] <cindyli> awesome!
[15:14:36 CDT(-0500)] <jhung> colinclark: excellent. I'm designing as though it's already there. (wink)
[15:16:08 CDT(-0500)] <colinclark> I just need to put in harriswong and mlam's error message view for the Uploader
[15:16:25 CDT(-0500)] <colinclark> Which is on my list as soon as this presentation's ready
[15:25:14 CDT(-0500)] <jhung> colinclark: is there a jira I can attach the image uploader demo design to?
[15:25:31 CDT(-0500)] <jhung> I tried looking and didn't find one. Did I miss it?
[15:25:35 CDT(-0500)] <colinclark> jhung: I'd make a new one
[15:25:40 CDT(-0500)] <colinclark> if you don't mind
[15:25:40 CDT(-0500)] <jhung> k
[15:25:43 CDT(-0500)] <colinclark> thanks!
[15:25:45 CDT(-0500)] <jhung> np
[15:29:19 CDT(-0500)] <jhung> colinclark, cindyli, mlam, anyone else involved in the Image Uploader. Here's the design. http://issues.fluidproject.org/browse/FLUID-4274
[15:30:23 CDT(-0500)] <Bosmon> Thanks for your mail, cindyli, I'll start thinking about the issue now...
[15:31:17 CDT(-0500)] <Bosmon> The use of the urlExpander is a good idea, although it is one of these early pieces of CSpace infrastructure that are only about "three-quarters baked"....
[15:32:54 CDT(-0500)] <cindyli> cool. thank YOU!
[15:33:04 CDT(-0500)] <Bosmon> I don't think its necessarily a fatal flaw that the defaults don't work in this case - since really there could never be a "sensible default" location for the template
[15:33:24 CDT(-0500)] <Bosmon> Well, if there could be, I don't know where it would be
[15:34:07 CDT(-0500)] <cindyli> Bosmon, true. but with this solution, all templates must sit in the same dir, what if the implementer has one his own template in his own dir?
[15:34:17 CDT(-0500)] <Bosmon> The use of the renderer leads to this often awkward blurring of the boundaries between an application's code and its data.... usually HTML templates are required only by the view layer of an application
[15:34:41 CDT(-0500)] <Bosmon> OK, yes, I see the problem
[15:35:17 CDT(-0500)] <Bosmon> Perhaps one solution would be to make a little "micro-component" associated with each of the different template names
[15:35:29 CDT(-0500)] <cindyli> ic. so yura_'s idea of separating each render component to a view component attached with a sub renderer component would help
[15:35:31 CDT(-0500)] <Bosmon> That way, someone could either write a single demands block, the way you have now, which sets a common prefix
[15:35:38 CDT(-0500)] <Bosmon> Ah yes, it seems we have had the same idea
[15:36:05 CDT(-0500)] <Bosmon> So yes, in this model, you could either issue just one demands block to set the prefix, or else issue more specific demands blocks if you want to load each template from a different place
[15:36:32 CDT(-0500)] <cindyli> great. i will try
[15:37:10 CDT(-0500)] <Bosmon> Also, the user might want to specify the URL either using a prefix, or else by specifying a literal URL
[15:37:35 CDT(-0500)] <Bosmon> We might want to call "prefix" something a bit more semantic
[15:37:53 CDT(-0500)] <Bosmon> it looks like what it really is is the path to a URL where the entirety of Fluid Infusion is mounted
[15:38:14 CDT(-0500)] <Bosmon> In Kettle, we gave these particular "mount paths" meaningful names like "fluid-infusion" etc.
[15:38:26 CDT(-0500)] <Bosmon> So that it was more easy to interpret what the configuration is the configuration of
[15:39:15 CDT(-0500)] <cindyli> ic. but i'm still not convinced that the prefix would make sense to the implementers, since they will have to first be aware of the infusion location and providing a relative prefix to that location
[15:39:38 CDT(-0500)] <Bosmon> Yes... if possible they should not have to bother with the prefix specifically
[15:39:58 CDT(-0500)] <cindyli> agree
[15:40:04 CDT(-0500)] <Bosmon> They should expect all of Infusion to be mounted at some path, and then the default value of the "prefix" to just be wherever the template actually is
[15:40:34 CDT(-0500)] <cindyli> ok
[15:40:36 CDT(-0500)] <Bosmon> I think in almost every sensible case this is how the implementor will work - but we want to make it as easy as possible for them to override individual templates with their own replacement values
[15:40:54 CDT(-0500)] <cindyli> exactly