/
fluid-work IRC Logs-2009-10-13

fluid-work IRC Logs-2009-10-13

Oct 13, 2009

[07:09:39 EDT(-0400)] * heidi (n=thesumme@bas5-oshawa95-1176469158.dsl.bell.ca) has joined #fluid-work
[08:01:25 EDT(-0400)] * Justin_o (n=Justin@142.150.154.171) has joined #fluid-work
[08:44:42 EDT(-0400)] * athena (n=athena@adsl-75-58-127-15.dsl.wlfrct.sbcglobal.net) has joined #fluid-work
[09:44:50 EDT(-0400)] <Justin_o> fj4000: http://wiki.fluidproject.org/display/fluid/Testing+using+multiple+browsers
[09:49:51 EDT(-0400)] * anastasiac (n=team@142.150.154.189) has joined #fluid-work
[09:55:31 EDT(-0400)] * clown (n=clown@142.150.154.101) has joined #fluid-work
[09:57:42 EDT(-0400)] * yura (n=yura@142.150.154.163) has joined #fluid-work
[10:13:29 EDT(-0400)] * colinclark (n=colin@142.150.154.101) has joined #fluid-work
[10:19:42 EDT(-0400)] * jgarciavila (n=joan@62.97.110.250) has joined #fluid-work
[10:36:15 EDT(-0400)] <jgarciavila> michelle?
[10:36:38 EDT(-0400)] <anastasiac> jgarciavilla, michelled is here, but in a meeting right now
[10:37:45 EDT(-0400)] <jgarciavila> ok
[10:44:54 EDT(-0400)] * michelled (n=michelle@142.150.154.193) has joined #fluid-work
[10:49:31 EDT(-0400)] <michelled> jgarciavila, colinclark did you want to meet now to talk about the bug parade email?
[10:49:43 EDT(-0400)] <colinclark> I'm free for a few minutes
[10:49:48 EDT(-0400)] <colinclark> Sounds good
[10:49:53 EDT(-0400)] <jgarciavila> ok
[10:50:15 EDT(-0400)] * Justin_o (n=Justin@142.150.154.171) has joined #fluid-work
[10:50:24 EDT(-0400)] <michelled> let's try breeze
[10:50:27 EDT(-0400)] <jgarciavila> ok
[10:59:06 EDT(-0400)] <anastasiac> Justin_o: I was hoping to help with testing so I took on the Builder testing FF, OS X. Should the Builder be working today?
[10:59:43 EDT(-0400)] <Justin_o> anastasiac: it should, colinclark is just fixing up something that isn't quite working on the test site at the moment
[11:00:06 EDT(-0400)] <anastasiac> hm... The url on the QA test plan page leads to a site that looks absolutely horrible: http://142.150.154.58/builder/
[11:00:24 EDT(-0400)] <anastasiac> is that what he's fixing?
[11:01:33 EDT(-0400)] <colinclark> anastasiac: It looks like laurel has been maintaining the Builder for awhile by hand-modifying code on the server.
[11:01:54 EDT(-0400)] <colinclark> So there's a different version of Infusion on the server, causing problems with the Builder.
[11:02:04 EDT(-0400)] <anastasiac> ah
[11:02:31 EDT(-0400)] <colinclark> So I'm tidying things up on the dev website so that the code there is a direct copy of a clean checkout of the Builder, rather than some kind of mangled hybrid thingy. (smile)
[11:02:48 EDT(-0400)] <anastasiac> so I shouldn't try to test yet...
[11:03:17 EDT(-0400)] <colinclark> anastasiac: It depends if you like testing nothingness
[11:03:32 EDT(-0400)] <colinclark> If you are testing for complete failure, we're doing great!
[11:03:39 EDT(-0400)] <anastasiac> (smile)
[11:18:47 EDT(-0400)] * mackrauss (n=Armin@142.150.154.101) has joined #fluid-work
[11:30:21 EDT(-0400)] <Justin_o> colinclark, fj4000, and others: As James was looking over some of the engage work, it came up that if there are no images provided for an artifact on the browse page, it should renderer a default image
[11:30:40 EDT(-0400)] <Justin_o> I've worked on this but haven't committed it yet because there was no jira on bug parade
[11:31:00 EDT(-0400)] <Justin_o> what do you think of me adding http://issues.fluidproject.org/browse/ENGAGE-118 to bug parade
[11:31:20 EDT(-0400)] <colinclark> Justin_o: Perhaps it's worth filling the issue out a little bit
[11:43:04 EDT(-0400)] * elicochran (n=elicochr@169.229.212.36) has joined #fluid-work
[11:53:38 EDT(-0400)] <Justin_o> colinclark: i've updated it a bit
[12:03:03 EDT(-0400)] <fj4000> Justin_o, colinclark: I have another option to think about for this image thing
[12:03:15 EDT(-0400)] <colinclark> fj4000: ?
[12:03:21 EDT(-0400)] <fj4000> we could add a class name on the container to use CSS to draw a "Not Found" image
[12:03:39 EDT(-0400)] <fj4000> colinclark: 11:30, Justin's idea
[12:03:44 EDT(-0400)] <colinclark> ok
[12:10:03 EDT(-0400)] <Justin_o> colinclark, fj4000: so the downside is that it would be a bit more complicated to implement, the benefit would be that there would be one less call to bring in an image
[12:10:56 EDT(-0400)] <Justin_o> colinclark: so the current solution would be to renderer the image if a URL is provided, or to renderer the URL from the template if one isn't
[12:11:39 EDT(-0400)] <Justin_o> colinclark: with fj4000's suggestion, if the URL isn't provided, the img tag won't be rendered and a class toggled instead
[12:12:14 EDT(-0400)] <fj4000> it would save a round trip to the server to fetch an image
[12:17:38 EDT(-0400)] <colinclark> I'm sorry, I can't respond to this stuff right now... I'll check in later
[12:37:45 EDT(-0400)] <jgarciavila> colin. michelled: i've downloaded the "progress-indicator" as a kind-of-template and also i've read the tutorial.
[12:38:19 EDT(-0400)] <jgarciavila> colin. michelled: Executing "progress-indicator.html" "progress-indicator2.html" I get errors.
[12:39:11 EDT(-0400)] <jgarciavila> colin, michelledd: ReferenceError: fluid_1_2 is not defined
[12:39:45 EDT(-0400)] <jgarciavila> colin, michelled: at [Top-level script]() (progress-indicator/js/progress-indicator2.js:4)
[12:42:52 EDT(-0400)] * justin_o (n=jmo@142.150.154.101) has joined #fluid-work
[12:45:06 EDT(-0400)] <jgarciavila> hi justin.
[12:49:48 EDT(-0400)] <jgarciavila> justin: i was looking at the progress bar wiki links that you gave me. And I see that all javascript on the component is inside one div.
[12:50:44 EDT(-0400)] <jgarciavila> justin: some js comes from wiki download urls. How can I create something similar to start the component and the page where the components has to be renderer?
[12:53:26 EDT(-0400)] <anastasiac> jgarciavilla: colinclark and michelled are out of the office right now
[12:53:35 EDT(-0400)] <anastasiac> Justin_o is pair-programming with jura
[12:53:47 EDT(-0400)] <anastasiac> sorry, yura, not jura
[12:54:08 EDT(-0400)] <anastasiac> justin_o: ^
[12:59:16 EDT(-0400)] * clown (n=clown@142.150.154.101) has left #fluid-work
[13:00:47 EDT(-0400)] <justin_o> jgarciavila: I'm not sure that is the best approach that i took... but it was what I needed to do at the time to get it up on the wiki. Basically i just uploaded the necessary files as attachments on to the wiki
[13:02:03 EDT(-0400)] <jgarciavila> justin_o: thanks.
[13:02:19 EDT(-0400)] <justin_o> jgarciavila: np
[13:17:42 EDT(-0400)] <elicochran> anyone there played around with any of the jquery autocomplete plugins and if so do you have a preference?
[13:45:08 EDT(-0400)] * colinclark (n=colin@142.150.154.101) has joined #fluid-work
[13:49:13 EDT(-0400)] * michelled_ (n=team@142.150.154.193) has joined #fluid-work
[13:51:52 EDT(-0400)] * laurel1 (n=laurel@207.112.104.122) has joined #fluid-work
[13:52:29 EDT(-0400)] <laurel1> hey fj4000: I just saw your email...problems?
[13:53:14 EDT(-0400)] <colinclark> laurel1: I'm having quite a challenge with the Builder on the website today.
[13:53:26 EDT(-0400)] <laurel1> hi colinclark
[13:53:36 EDT(-0400)] <laurel1> can i help you?
[13:53:55 EDT(-0400)] <colinclark> laurel1: Well, I'm not sure.
[13:54:07 EDT(-0400)] <colinclark> It seems like I'm best to just teach myself how this thing is deployed
[13:54:19 EDT(-0400)] <colinclark> But it's quite a chunk of spaghetti at the moment
[13:54:44 EDT(-0400)] <colinclark> laurel1: Were you hand-modifying code to change the Fluid version number?
[13:54:44 EDT(-0400)] <colinclark> On the server?
[13:55:19 EDT(-0400)] <laurel1> colinclark: there is one place where it is hand modified - in the config file
[13:55:39 EDT(-0400)] <colinclark> laurel1: Well, it looks like there are several places where the PHP code is hand modified...
[13:55:40 EDT(-0400)] * clown (n=clown@142.150.154.101) has joined #fluid-work
[13:55:53 EDT(-0400)] <colinclark> including the "DISTANT_PATH" point to the config file, etc
[13:56:00 EDT(-0400)] <colinclark> But I was referring to JavaScript code.
[13:56:16 EDT(-0400)] <colinclark> Your latest patched version of customBuild.js had been updated to link against fluid_1_1
[13:56:24 EDT(-0400)] <colinclark> But there's an older version of Infusion installed here
[13:56:32 EDT(-0400)] <laurel1> colinclark, no then, in the js code, the version is obtained from infusion itself
[13:56:44 EDT(-0400)] <laurel1> but I had to upload infusion myself
[13:57:08 EDT(-0400)] <laurel1> as I noted to you I think last friday or thursday - a level of concern because it wasn't completely up to date
[13:57:18 EDT(-0400)] <colinclark> Yeah, so I wonder how this ever worked
[13:57:48 EDT(-0400)] <colinclark> So we had either been running an old version of all the code, or perhaps you hand-modified the fluid_1_x version in the JS code to get it all working.
[13:58:03 EDT(-0400)] <colinclark> Either way, it seems apparent that we haven't really been testing this code in a realistic environment
[13:58:15 EDT(-0400)] <colinclark> We've been testing it against and old version of Infusion, etc.
[13:58:20 EDT(-0400)] <laurel1> i'm sorry you are frustrated by this colinclark
[13:58:42 EDT(-0400)] <colinclark> I'm a tiny bit frustrated, but mostly focused on getting it all working so we can carry on with QA testing.
[13:59:03 EDT(-0400)] <laurel1> ok...well, can i help?
[13:59:17 EDT(-0400)] <laurel1> or are you confident that you can figure it out
[13:59:23 EDT(-0400)] <colinclark> I'm not at all confident
[13:59:30 EDT(-0400)] <colinclark> But things seem to be going okay.
[13:59:36 EDT(-0400)] <colinclark> You can answer a quick dumb question
[13:59:45 EDT(-0400)] <laurel1> never dumb questions, remember?
[13:59:48 EDT(-0400)] <colinclark> Do the PHP pages get cached by the web server at all?
[14:00:08 EDT(-0400)] <colinclark> In other words, if I update a PHP file, do I need to do anything else to make it work?
[14:00:22 EDT(-0400)] <laurel1> no you don't have to do anything else to make it work
[14:00:32 EDT(-0400)] <laurel1> BUT
[14:00:50 EDT(-0400)] <laurel1> we did use a cache to save the result of the json parsing
[14:01:15 EDT(-0400)] <laurel1> so if you change the json data then you do need to remove the cached file in order for it to write again
[14:01:38 EDT(-0400)] <colinclark> laurel1: Which JSON data?
[14:02:03 EDT(-0400)] <laurel1> There are 2 parts to the php
[14:02:44 EDT(-0400)] <laurel1> 1) the jsonParser, which goes through all the json files in infusion and creates a data structure to display the module information
[14:03:06 EDT(-0400)] <colinclark> btw laurel, here's a nascent wiki page I'm maintaining with the steps I think are required to deploy a fresh, up to date copy of the Builder: http://wiki.fluidproject.org/display/fluid/Deploying+the+Infusion+Builder
[14:03:48 EDT(-0400)] <laurel1> 2) the postProcessor, which actually does the build
[14:04:49 EDT(-0400)] <laurel1> so what I was talking about was 1)
[14:05:16 EDT(-0400)] <colinclark> laurel1: Ok, makes sense
[14:05:25 EDT(-0400)] <laurel1> so for instance, if you upload a new copy of infusion you would want to remove the cache
[14:05:44 EDT(-0400)] <laurel1> dictated by cache_file
[14:06:18 EDT(-0400)] <laurel1> anything else I can help with?
[14:06:48 EDT(-0400)] <laurel1> your instructions seem to be getting there
[14:07:00 EDT(-0400)] <colinclark> Not sure, hang on one second
[14:07:01 EDT(-0400)] <colinclark> Can you read that wiki page and see if it seems sane
[14:07:37 EDT(-0400)] <colinclark> We need to get the organization of Builder in SVN to better reflect its deployed structure
[14:07:45 EDT(-0400)] <colinclark> Having to do this all by hand is really awkward
[14:07:50 EDT(-0400)] <laurel1> it does seem sane
[14:07:58 EDT(-0400)] <laurel1> I agree, that is a good idea
[14:08:09 EDT(-0400)] <colinclark> I'd like to also have config "profiles," so we'll actually have a server version of config.php, minus only the DB username and password
[14:08:14 EDT(-0400)] <colinclark> That'll have to come after 1.1.2
[14:08:24 EDT(-0400)] <laurel1> i guess review should have extended to "deployment" in the case of the infusion builder
[14:09:05 EDT(-0400)] <laurel1> sounds good.
[14:10:18 EDT(-0400)] <laurel1> So the other thing that isn't actually in your page is the implementation of the page in cmsms
[14:10:30 EDT(-0400)] <colinclark> laurel1: Yep, that's exactly what I was thinking. Anyway, this is review now, even if it is a little late
[14:10:36 EDT(-0400)] <laurel1> which I know you'd prefer not to have to do
[14:10:49 EDT(-0400)] <colinclark> laurel1: So to confirm, do the PHP pages need to have specific permissions to be served correctly?
[14:11:09 EDT(-0400)] <laurel1> btw...are you a newly uncled guy today?
[14:11:11 EDT(-0400)] <laurel1> yes they do
[14:11:14 EDT(-0400)] <colinclark> not yet
[14:11:16 EDT(-0400)] <colinclark> still on baby watch
[14:11:23 EDT(-0400)] <colinclark> laurel1: The files you had there before have ugo+x on them
[14:11:30 EDT(-0400)] <laurel1> yup
[14:11:34 EDT(-0400)] <colinclark> ok
[14:12:58 EDT(-0400)] <colinclark> So, laurel1, in short we have a distinction here between code that is "public" (builder.php + Infusion + Builder CSS & JS) and internal code (all the stuff used by builder.php)
[14:13:02 EDT(-0400)] <colinclark> Does that sound right?
[14:13:04 EDT(-0400)] <laurel1> I'm pretty sure I had the permissions right, but can honestly say now I'm second guessing myself...let's chedk with fj4000
[14:13:18 EDT(-0400)] <laurel1> yes that sounds right.
[14:13:43 EDT(-0400)] <laurel1> fj4000 suggested (and I agreed) that we didn't want people to be able to hit the other php files directly
[14:13:47 EDT(-0400)] <laurel1> make sense?
[14:14:39 EDT(-0400)] <colinclark> laurel1: I'm sure you had the permissions at least sorta right
[14:14:46 EDT(-0400)] <colinclark> I just need to get everything here in order
[14:14:58 EDT(-0400)] <colinclark> This separation between private and public is a good thing
[14:15:08 EDT(-0400)] <colinclark> Unfortunately, I've still got this stuff looking for its dependencies at the wrong path
[14:15:18 EDT(-0400)] <laurel1> oh that's not good
[14:15:35 EDT(-0400)] <laurel1> that's what the "distant-path" is for.
[14:15:43 EDT(-0400)] <laurel1> I guess my instructions weren't very good
[14:15:45 EDT(-0400)] <colinclark> So I thought
[14:15:54 EDT(-0400)] <colinclark> It wasn't the instructions
[14:16:08 EDT(-0400)] <laurel1> did you trash the config that was there before?
[14:16:23 EDT(-0400)] <laurel1> because I have a copy of it on the machine I use at the office
[14:16:30 EDT(-0400)] <laurel1> which you can use as a model
[14:16:53 EDT(-0400)] <colinclark> laurel1: No, I've got backups of everything that was there
[14:17:03 EDT(-0400)] <colinclark> I wouldn't have been able to get the config file right without an example of the old one
[14:17:09 EDT(-0400)] <colinclark> Which is why I suggest we have config profiles instead
[14:17:14 EDT(-0400)] <colinclark> So no one has to do this again
[14:17:15 EDT(-0400)] <colinclark> (wink)
[14:17:50 EDT(-0400)] <colinclark> So this is what the PHP code is returning when Builder tries to GET the module data: http://142.150.154.58/custom-build/php/builder.php?var=fluid.customBuild.dependencies
[14:17:57 EDT(-0400)] <colinclark> and those paths just aren't right
[14:18:23 EDT(-0400)] <laurel1> ok.
[14:18:57 EDT(-0400)] <laurel1> i'm not sure how to help...let me think
[14:19:54 EDT(-0400)] <colinclark> I wonder if paths are somehow relative to builder.php
[14:20:01 EDT(-0400)] <laurel1> ok...the config.php isn't being found
[14:20:03 EDT(-0400)] * elicochran_ (n=elicochr@dhcp-169-229-212-36.LIPS.Berkeley.EDU) has joined #fluid-work
[14:20:35 EDT(-0400)] <colinclark> DISTANT_PATH is definitely set correctly, and so builder.php is finding its dependencies
[14:20:54 EDT(-0400)] <colinclark> It's those dependencies that aren't finding their dependencies
[14:20:56 EDT(-0400)] <colinclark> if that makes sense
[14:21:00 EDT(-0400)] <laurel1> yes, but the error message is pretty clear
[14:21:29 EDT(-0400)] <laurel1> so let me check more
[14:26:55 EDT(-0400)] <laurel1> nope you are right...it doesn't make sense. but I'm still looking
[14:27:59 EDT(-0400)] <laurel1> it's looking here /usr/webserver/content/fluidpro/docs/custom-build/php/jsonProcessor.php
[14:28:28 EDT(-0400)] <colinclark> yes
[14:28:32 EDT(-0400)] <laurel1> wierdness
[14:30:41 EDT(-0400)] <colinclark> Yeah, it's pretty clearly decided to use paths relative to the builder.php file
[14:31:31 EDT(-0400)] <colinclark> So all the calls to include_once are being resolved against that directory
[14:31:48 EDT(-0400)] <colinclark> Whereas if I prefixed them all, everywhere, with DISTANT_PATH, I think it would work
[14:31:54 EDT(-0400)] <colinclark> Which strikes me as terribly problematic
[14:33:29 EDT(-0400)] <laurel1> ok if I just do a quick edit in jsonProcessor.php?
[14:33:57 EDT(-0400)] <colinclark> laurel1: Tell me first
[14:34:01 EDT(-0400)] <colinclark> I'm editing several of these files
[14:34:24 EDT(-0400)] <laurel1> ok. I was thinking of trying ./config.php
[14:34:39 EDT(-0400)] <laurel1> i see you've tried DISTANT_PATH
[14:34:51 EDT(-0400)] <colinclark> laurel1: And it seems to be working better
[14:35:04 EDT(-0400)] <laurel1> but really truly, it was working. seriously
[14:35:06 EDT(-0400)] <laurel1> sigh
[14:35:09 EDT(-0400)] <laurel1> sheesh
[14:35:30 EDT(-0400)] <colinclark> I don't doubt that it was working
[14:35:35 EDT(-0400)] <laurel1> i do (smile)
[14:36:08 EDT(-0400)] <colinclark> ./ definitely doesn't work
[14:36:18 EDT(-0400)] <laurel1> ok...well it was worth a try
[14:37:32 EDT(-0400)] <laurel1> what an interesting mystery
[14:38:52 EDT(-0400)] <colinclark> laurel1: I'm quite close
[14:39:02 EDT(-0400)] <colinclark> So tell me again about this jsonCache
[14:39:08 EDT(-0400)] <colinclark> Do I need to create it if it doesn't already exist?
[14:39:24 EDT(-0400)] <laurel1> no you do not
[14:39:36 EDT(-0400)] <colinclark> It will create it for us?
[14:39:40 EDT(-0400)] <laurel1> yes it will
[14:40:32 EDT(-0400)] <colinclark> Here's the error I'm getting now...
[14:40:32 EDT(-0400)] <colinclark> http://142.150.154.58/custom-build/php/builder.php?var=fluid.customBuild.dependencies
[14:40:34 EDT(-0400)] <laurel1> she says with trepidation
[14:40:54 EDT(-0400)] <laurel1> i see that. so it's trying to open the build.properties file and can't find it now either
[14:41:39 EDT(-0400)] <laurel1> or the file isnt readable
[14:41:49 EDT(-0400)] <colinclark> Ah!
[14:41:50 EDT(-0400)] <colinclark> of course
[14:41:52 EDT(-0400)] <colinclark> that's my bad
[14:41:58 EDT(-0400)] <laurel1> I believe I had to fiddle with the permissions there
[14:41:59 EDT(-0400)] <laurel1> sigh
[14:42:09 EDT(-0400)] <colinclark> there are just too many directories called customBuild or custom-build or something else
[14:42:14 EDT(-0400)] <laurel1> i agree
[14:42:16 EDT(-0400)] <colinclark> we need to get the code all nice and organized
[14:42:38 EDT(-0400)] <laurel1> sounds good to me. I tried to hide it all in the same place, but that doesn't work
[14:43:03 EDT(-0400)] <colinclark> Intriguing: http://142.150.154.58/custom-build/php/builder.php?var=fluid.customBuild.dependencies
[14:43:16 EDT(-0400)] <colinclark> No errors, no response
[14:43:33 EDT(-0400)] <colinclark> Victory!! http://142.150.154.58/custom-build/php/builder.php?var=fluid.customBuild.dependencies
[14:44:16 EDT(-0400)] <colinclark> I had, in desperation, tried to give it an empty jsonCache. It was kindly using it for me.
[14:44:18 EDT(-0400)] <laurel1> ok...so here's where we start to add little echo statements within the jsonProcessor code
[14:44:39 EDT(-0400)] <laurel1> see if we can figure out where it is dying
[14:44:49 EDT(-0400)] <colinclark> It's all working
[14:44:55 EDT(-0400)] <laurel1> oh, is it?
[14:44:58 EDT(-0400)] <colinclark> http://142.150.154.58/builder/
[14:45:18 EDT(-0400)] <colinclark> nearly
[14:45:19 EDT(-0400)] <colinclark> there must be a hardcoded path here somewhere
[14:45:26 EDT(-0400)] <colinclark> try downloading a build
[14:45:53 EDT(-0400)] <laurel1> ah
[14:46:07 EDT(-0400)] <laurel1> so now we're into postProcessor.php
[14:46:33 EDT(-0400)] <laurel1> and obviously an error checking that I missed (wink)
[14:49:21 EDT(-0400)] <laurel1> my guess. it isn't finding the zip file
[14:50:09 EDT(-0400)] <laurel1> ah...i know
[14:50:16 EDT(-0400)] <laurel1> here you need to do 2 things
[14:50:28 EDT(-0400)] <laurel1> 1) you need to clear out the cache
[14:50:40 EDT(-0400)] <laurel1> and 2) you need to clean out the cache table in the mysql
[14:51:04 EDT(-0400)] <laurel1> bet the mysql isn't clear, so is pointing to a location that doesn't exist
[14:52:31 EDT(-0400)] <laurel1> also, I'm not seeing any new build files in /tmp where I'm expecting to see them
[14:52:44 EDT(-0400)] <laurel1> so I'm guessing that the build isn't actually happening
[14:53:14 EDT(-0400)] <laurel1> I only have a few more mins...I'm not feeling very helpful
[14:53:26 EDT(-0400)] * elicochran_ (n=elicochr@dhcp-169-229-212-36.LIPS.Berkeley.EDU) has joined #fluid-work
[14:57:26 EDT(-0400)] <laurel1> colinclark: I think if you've had to change all the paths in jsonProcessor, you probably have to change all the paths in postProcessor too
[14:57:46 EDT(-0400)] <colinclark> i've changed all the paths in all the php files
[14:58:01 EDT(-0400)] <colinclark> ok, justin_o and michelled and I have been talking
[14:58:08 EDT(-0400)] <colinclark> I think we're going to have to ship Infusion 1.1.2 without the Builder.
[14:58:19 EDT(-0400)] <laurel1> crap...sorry
[14:58:27 EDT(-0400)] * justin_o (n=jmo@142.150.154.101) has joined #fluid-work
[14:58:37 EDT(-0400)] <laurel1> i don't see the changes in postProcessor though...fyi
[14:58:48 EDT(-0400)] <colinclark> ok
[14:58:51 EDT(-0400)] <colinclark> i'll take a look
[14:59:00 EDT(-0400)] <colinclark> you're totally right, laurel1, we should have extended review to include the server
[14:59:22 EDT(-0400)] <laurel1> well, i'm much sorry for the challenges you've had today
[14:59:38 EDT(-0400)] <laurel1> must have given you all heart attacks
[14:59:54 EDT(-0400)] <laurel1> I'm having a small one here.
[15:01:10 EDT(-0400)] <laurel1> i need to go - but i guess making the decision not to include this in the release will relieve the pressure for today and we'll get it sorted before 1.2
[15:01:23 EDT(-0400)] * athena (n=athena@adsl-75-58-127-15.dsl.wlfrct.sbcglobal.net) has joined #fluid-work
[15:02:36 EDT(-0400)] <laurel1> ok...see everyone bright and early tomorrow.
[15:03:07 EDT(-0400)] * laurel1 (n=laurel@207.112.104.122) has left #fluid-work
[15:05:47 EDT(-0400)] <Bosmon> Am I here?
[15:06:52 EDT(-0400)] * mackrauss (n=Armin@142.150.154.101) has joined #fluid-work
[15:07:49 EDT(-0400)] <athena> as far as i can tell, Bosmon
[15:09:12 EDT(-0400)] <colinclark> Bosmon: Hi
[15:09:17 EDT(-0400)] <colinclark> Justin_o was looking for you earlier
[15:10:09 EDT(-0400)] <justin_o> Bosmon: seems the query string is returning the query twice.. so instead of "mccord&virtual" we get "mccord&virtual&mccord&virtual"
[15:11:04 EDT(-0400)] <anastasiac> athena, I just visited our uPortal deployment using Safari on Mac OS X. It seems to be mistakenly assuming that it's running on a mobile browser, and displaying quite differently than it does in FireFox. Might you have any idea why this is happening? Is it intentional? http://build.fluidproject.org/uPortal/render.userLayoutRootNode.uP
[15:11:28 EDT(-0400)] <athena> not entirely intentional, but known
[15:11:41 EDT(-0400)] <athena> do you have a good list of webkit mobile device user agents somewhere?
[15:11:56 EDT(-0400)] <athena> i don't happen to have one, so for the time being all webkit browsers are getting routed to the mobile theme
[15:11:56 EDT(-0400)] <anastasiac> I can't answer that, but fj4000 might be able to
[15:12:01 EDT(-0400)] <athena> which needs to get fixed (smile)
[15:12:50 EDT(-0400)] <Bosmon> Ah, that is odd
[15:13:00 EDT(-0400)] <Bosmon> Although really "mccord&virtual" is outside the spec...
[15:13:00 EDT(-0400)] <athena> how's it looking other than that, anastasiac?
[15:13:04 EDT(-0400)] <fj4000> anastasiac: i dont have a list, no
[15:13:16 EDT(-0400)] <athena> we need to do some other stuff, like address the login form
[15:13:17 EDT(-0400)] <fj4000> but i would think feature detection might work best
[15:13:30 EDT(-0400)] <athena> feature detection?
[15:13:32 EDT(-0400)] <fj4000> since no two webkits are the same
[15:13:47 EDT(-0400)] <anastasiac> athena, it seems fine. I'm just doing some testing of the links on our build site, and I came across this unexpected appearance
[15:13:49 EDT(-0400)] <fj4000> athena: yes, like looking for capabilities instead of browser names
[15:13:52 EDT(-0400)] <athena> and i'd like to make it not require a page reload when you go back to the main navigation
[15:13:52 EDT(-0400)] <justin_o> Bosmon: what should the spec be?
[15:13:56 EDT(-0400)] <athena> ah
[15:14:02 EDT(-0400)] <athena> yeah unfortunately we can't do that fj4000 (smile)
[15:14:05 EDT(-0400)] <Bosmon> We may only really support key-value pairs
[15:14:10 EDT(-0400)] <fj4000> no?
[15:14:18 EDT(-0400)] <athena> uportal actually supports sending completely different "themes" for different browsers
[15:14:42 EDT(-0400)] <Bosmon> I mean, "blunt URLs" like that are sort of within the RFC, but can cause problems in various environments
[15:14:45 EDT(-0400)] <fj4000> im not sure I understand....
[15:14:48 EDT(-0400)] <Bosmon> And also, it is just not clear what they mean (tongue)
[15:15:04 EDT(-0400)] <athena> it's much more than just css/javascript - the pipeline actually renders content by passing in through several different XSLTs
[15:15:10 EDT(-0400)] <fj4000> oh
[15:15:29 EDT(-0400)] <athena> so what actually happens in the backend is that before the page is ever rendered, the user's XML layout is put through a structure transform, then a theme transform
[15:15:32 EDT(-0400)] <fj4000> but how do you tell, now, which browser your dealing with?
[15:15:34 EDT(-0400)] <athena> that last step builds the HTML itself
[15:15:45 EDT(-0400)] <athena> so we do matching based on the user agent in the java code
[15:16:01 EDT(-0400)] <fj4000> oh....server side agent sniffing?
[15:16:05 EDT(-0400)] <athena> there's support for matching profiles to user agents based on regexes
[15:16:06 EDT(-0400)] <athena> yep
[15:16:24 EDT(-0400)] <Bosmon> I'm not sure if you gave it a "valid" URL the duplication would go away?
[15:16:42 EDT(-0400)] <athena> so we'll probably need to build regexes for the iphone, android, etc.
[15:16:48 EDT(-0400)] <justin_o> okay...
[15:16:51 EDT(-0400)] <justin_o> Bosmon: ^
[15:16:52 EDT(-0400)] <athena> and then we'll need to do the same thing in cas
[15:17:04 EDT(-0400)] <fj4000> well, all I know is the iphone passes in the string "iPhone" into it
[15:17:11 EDT(-0400)] <fj4000> in the agent
[15:17:17 EDT(-0400)] <athena> yeah
[15:17:27 EDT(-0400)] <athena> i'm guessing we just need to collect a bunch of statements like that (smile)
[15:17:58 EDT(-0400)] <athena> do you happen to know what mobile devices are likely to render mFSS with lots of javascript well?
[15:18:10 EDT(-0400)] <athena> i'm guessing that list is something like recent webkit-based devices
[15:20:00 EDT(-0400)] <justin_o> Bosmon: i'll try seeing if the output ends of being different if i use the key/value pairs for the url, at the moment i'm trying to track down another problem, but will let you know what I find, when I get a chance to check it
[15:23:54 EDT(-0400)] <Bosmon> If it is a bug, can we still fix it?
[15:24:46 EDT(-0400)] <anastasiac> justin_o: I'm testing the links on the build site, and one of the 'escalated tests' doesn't look right at all: http://build.fluidproject.org/infusion/tests/escalated-tests/html/jquery-tabs.html
[15:25:54 EDT(-0400)] <justin_o> anastasiac: i guess the styling is broken
[15:26:19 EDT(-0400)] <anastasiac> should I file a JIRA? what is the proper process, at this stage of testing?
[15:27:15 EDT(-0400)] <colinclark> anastasiac: that test seems to still show what it is intended for
[15:27:20 EDT(-0400)] <colinclark> So what's the problem?
[15:27:41 EDT(-0400)] <anastasiac> I guess I wasn't/am not clear on what it's supposed to be showing (smile)
[15:29:29 EDT(-0400)] <colinclark> anastasiac: "In Opera, the <li> tags above form a closed loop in tab order, preventing focus from escaping elsewhere in the browser."
[15:29:50 EDT(-0400)] <colinclark> If you try it in Opera, you do get stuck within the list
[15:30:15 EDT(-0400)] <anastasiac> ok, I wasn't clear if the fact that it's billed as a "tabs" example and it doesn't show tabs was a problem
[15:30:19 EDT(-0400)] <colinclark> "escalated" tests are supposed to show bugs in code that isn't ours.
[15:30:19 EDT(-0400)] <anastasiac> apparently, not
[15:31:41 EDT(-0400)] * mackrauss_ (n=Armin@142.150.154.101) has joined #fluid-work
[15:34:02 EDT(-0400)] <colinclark> jamon: This day is not working out quite as expected
[15:34:05 EDT(-0400)] <colinclark> I'm in another meeting
[15:34:12 EDT(-0400)] <colinclark> We can wait until tomorrow to talk website
[15:34:17 EDT(-0400)] <colinclark> Laurel will be around, too
[15:34:25 EDT(-0400)] <colinclark> jamon: Sorry to keep you waiting for me for so long. (sad)
[15:45:44 EDT(-0400)] <athena> fj4000: do you have any suggestions for what we should do w/ the jQuery UI tabs on mobile devices?
[15:46:01 EDT(-0400)] <athena> i don't know if maybe we can just skin them so they'll be more likely to display well on the page
[15:46:08 EDT(-0400)] <fj4000> um
[15:46:19 EDT(-0400)] <fj4000> well, how far can you go?
[15:46:29 EDT(-0400)] <fj4000> would you like them to look the same, just smaller?
[15:46:42 EDT(-0400)] <fj4000> or more like the mobile fss tabs for the iphone?
[15:46:51 EDT(-0400)] <athena> potentially both
[15:47:06 EDT(-0400)] <athena> we might eventually make some portlets mobile-aware such that they could send different markup
[15:47:18 EDT(-0400)] <athena> or markup that gets ignored in a non-mobile device
[15:47:33 EDT(-0400)] <athena> but it would be nice to make things sort of display ok if the markup is old, too
[15:47:38 EDT(-0400)] <athena> are there mobile fss tabs styles?
[15:47:43 EDT(-0400)] <fj4000> without fiddling with markup, im not too sure
[15:47:44 EDT(-0400)] <fj4000> yes
[15:47:50 EDT(-0400)] <fj4000> mobile fss has a tabs demo
[15:47:53 EDT(-0400)] <athena> i guess that sounds like a place to start then
[15:47:57 EDT(-0400)] <fj4000> (tongue)
[15:48:17 EDT(-0400)] * fj4000 goes link hunting
[15:49:02 EDT(-0400)] <athena> ah, i see something in the tabs/1.html
[15:49:07 EDT(-0400)] <fj4000> yup
[15:49:08 EDT(-0400)] <athena> i wasn't sure what the intended usage of those was
[15:49:10 EDT(-0400)] <fj4000> beat me to it
[15:49:21 EDT(-0400)] <fj4000> they are the mobile flavour of fl-tabs
[15:49:30 EDT(-0400)] <athena> is that supposed to be in the header, or does it make sense to use those in the page below the header?
[15:49:41 EDT(-0400)] <fj4000> both are fine
[15:49:51 EDT(-0400)] <fj4000> demo should include both usages
[15:50:11 EDT(-0400)] <athena> ok
[15:50:15 EDT(-0400)] <fj4000> ach!
[15:50:18 EDT(-0400)] <fj4000> they look broken to me
[15:50:33 EDT(-0400)] <athena> they show up for me in my checkout
[15:50:39 EDT(-0400)] <athena> though i don't know what they're supposed to look like
[15:51:08 EDT(-0400)] <athena> i'll have to play around w/ the markup
[15:51:11 EDT(-0400)] <fj4000> one sec....this could be scary
[15:51:31 EDT(-0400)] <athena> part of me wonders whether it'd make sense to just make the jquery ui theme render tabs like the mfss skin
[15:52:05 EDT(-0400)] <fj4000> using themeroller?
[15:52:20 EDT(-0400)] <athena> hm
[15:52:21 EDT(-0400)] <athena> maybe?
[15:52:29 EDT(-0400)] <athena> such a thing might be more than i'm capable of (smile)
[15:53:09 EDT(-0400)] <athena> from what i've seen your css looks more webkit-specific and less image-based than standard jQuery UI themes
[15:53:51 EDT(-0400)] <fj4000> athena: a few things
[15:54:06 EDT(-0400)] <fj4000> 1) the tabs demo is missing a critical theme class name
[15:54:28 EDT(-0400)] <fj4000> so it's broken until the fl-theme-iphone class name is added to the <body>
[15:54:33 EDT(-0400)] <athena> ah - yeah, i was looking at my older checkout which still works - i think it uses old css
[15:54:43 EDT(-0400)] <fj4000> phew
[15:54:43 EDT(-0400)] <athena> i can add that to my newer checkout though
[15:54:51 EDT(-0400)] <athena> actually, i already did that for the main page (smile)
[15:55:00 EDT(-0400)] <fj4000> awesome
[15:55:06 EDT(-0400)] <athena> it looked like some pages were maybe missing from that last time i looked?
[15:55:36 EDT(-0400)] <fj4000> we're still ripping these demos apart and putting them back together again in the demo portal
[15:55:53 EDT(-0400)] <athena> gotcha (smile)
[15:55:55 EDT(-0400)] <fj4000> for your other point, about the images, indeed these webkit based designs are for webkit browsers
[15:55:58 EDT(-0400)] <athena> i figured it was somethign like that
[15:56:07 EDT(-0400)] <fj4000> they're as image-less as possible
[15:56:10 EDT(-0400)] <athena> yeah
[15:56:16 EDT(-0400)] <athena> i imagine that's better download-wise?
[15:56:27 EDT(-0400)] <fj4000> amazingly so
[15:56:30 EDT(-0400)] <fj4000> and response time
[15:56:35 EDT(-0400)] * athena found out just how sloooooow EDGE is a couple weeks ago
[15:56:39 EDT(-0400)] <fj4000> 10x less http requests
[15:56:49 EDT(-0400)] <athena> rural wyoming does not in fact have 3G
[15:56:57 EDT(-0400)] <fj4000> (tongue)
[15:57:14 EDT(-0400)] <fj4000> so, there is a fundamental diff. between jquery ui as it currently is and mobile fss
[15:57:48 EDT(-0400)] <fj4000> although, from what i've seen, the markup pattern is similar
[15:58:38 EDT(-0400)] <fj4000> so, if you took away the jquery ui stylesheets, it might just work....
[15:58:52 EDT(-0400)] <athena> interesting
[15:58:52 EDT(-0400)] <athena> i'll have to play w/ it
[15:58:54 EDT(-0400)] <fj4000> how do they create a tab?
[15:58:54 EDT(-0400)] <athena> trying to think how much jquery css we have
[15:58:58 EDT(-0400)] <athena> it's probably mostly for datepicker and tabs
[15:59:19 EDT(-0400)] <fj4000> is it a <li> <span /> </li> ?
[15:59:25 EDT(-0400)] <athena> hm.
[15:59:28 EDT(-0400)] <athena> i'd have to double check
[15:59:32 EDT(-0400)] <fj4000> ok
[15:59:38 EDT(-0400)] <athena> definitely <li>s, but they may not have spans inside
[15:59:46 EDT(-0400)] <fj4000> if you want to pursue this, just give us a shout
[16:00:05 EDT(-0400)] <athena> will do!
[16:00:32 EDT(-0400)] <athena> one odd thing about developing for the portal environment is that you have a lot less control over the markup
[16:00:33 EDT(-0400)] <athena> so
[16:00:52 EDT(-0400)] <athena> we'd like to make things likely to render sort of reasonably even if someone hasn't updated their content
[16:00:58 EDT(-0400)] <athena> of course, we cant' guarantee perfection
[16:03:11 EDT(-0400)] * justin_o (n=jmo@142.150.154.101) has joined #fluid-work
[16:03:20 EDT(-0400)] <athena> by the way, do transitions work when you're linking to a new page entirely, or do animations only really make sense when you're staying in the same html page?
[16:03:33 EDT(-0400)] <fj4000> its like a js effect
[16:03:37 EDT(-0400)] <fj4000> so only that page
[16:03:59 EDT(-0400)] <athena> i wonder if we should leave them out then
[16:04:28 EDT(-0400)] * Justin_o_ (n=Justin@142.150.154.171) has joined #fluid-work
[16:04:31 EDT(-0400)] <athena> since we'd have animations when going back to the main navigation, but not when clicking into a navigation item
[16:04:32 EDT(-0400)] <fj4000> could you do the transition, and then load the next page?
[16:04:49 EDT(-0400)] <fj4000> maybe js could just delay the load?
[16:04:58 EDT(-0400)] <athena> i suppose maybe we could?
[16:05:06 EDT(-0400)] <athena> would that make sense from a user's perspective?
[16:05:14 EDT(-0400)] <athena> i can try it and see how it feels
[16:05:42 EDT(-0400)] <fj4000> the only issue i can think of off the top of my head is the lack of any incoming information
[16:06:10 EDT(-0400)] <fj4000> so if your doing a page to page transition (ala Screen Navigator) then you will transition your current content, but not your incoming content
[16:06:24 EDT(-0400)] <fj4000> and then it would flash into view
[16:06:32 EDT(-0400)] <fj4000> once the new page is loaded
[16:06:44 EDT(-0400)] <athena> i can try it and see (smile)
[16:06:49 EDT(-0400)] <fj4000> (smile)
[16:07:04 EDT(-0400)] <athena> i'll have to take a look at how to use the screen navigator stuff
[16:07:46 EDT(-0400)] <fj4000> screen navigator is my lame attempt at doing transitions between content
[16:08:00 EDT(-0400)] <athena> i am quite sure it is not lame
[16:08:01 EDT(-0400)] <fj4000> im not sure it will survive in it's current form though
[16:08:28 EDT(-0400)] <athena> gary and i were just talking about how nice the mobile theme looks already, for the record
[16:08:45 EDT(-0400)] <fj4000> we're thrilled your giving it a go
[16:09:15 EDT(-0400)] <athena> well, i appreciate you being willing to answer questions
[16:10:17 EDT(-0400)] <athena> and i still owe you email about the uportal FSS issues
[16:10:18 EDT(-0400)] * athena sighs
[16:10:33 EDT(-0400)] <fj4000> it's what we're here for!
[16:10:42 EDT(-0400)] <athena> (smile)
[16:12:18 EDT(-0400)] <Justin_o_> colinclark, fj4000: did you have thoughts about adding http://issues.fluidproject.org/browse/ENGAGE-118 to bug parade ?
[16:13:25 EDT(-0400)] * yura (n=yura@user152-190.wireless.utoronto.ca) has joined #fluid-work
[16:17:10 EDT(-0400)] <fj4000> I don't really have much of an opinion
[16:21:07 EDT(-0400)] <Justin_o_> fj4000, colinclark: there was also an issue with the page timing out... http://issues.fluidproject.org/browse/ENGAGE-119
[16:21:15 EDT(-0400)] <Justin_o_> would we be able to get this onto bug parade as well
[16:21:25 EDT(-0400)] <Justin_o_> it seems that the change will just be to set async to true
[16:21:56 EDT(-0400)] <fj4000> sounds good to me
[16:24:40 EDT(-0400)] <colinclark> Justin_o_: +1
[16:24:45 EDT(-0400)] <colinclark> I'll add it to the parade
[16:25:11 EDT(-0400)] <Justin_o_> colinclark: thanks
[16:26:32 EDT(-0400)] <Bosmon> Is the "live" version of kettle still the one in incubator?
[16:32:59 EDT(-0400)] <colinclark> Bosmon: Yes
[16:39:38 EDT(-0400)] <Bosmon> For how much longer, do we estimate?
[17:05:14 EDT(-0400)] <anastasiac> Justin_o: I've been doing testing, and finding bugs, and filing JIRAs. I don't think any of them are blockers, but it's not my call. What's the procedure - do you want the JIRA numbers?
[17:06:31 EDT(-0400)] <Justin_o> anastasiac: sure... if you could send them to the list and cc me
[17:06:35 EDT(-0400)] <Justin_o> thanks for the hlep
[17:08:00 EDT(-0400)] * fj40001 (n=Jacob@142.150.154.164) has joined #fluid-work
[17:11:17 EDT(-0400)] * jayshao (n=jayshao@ool-4573130c.dyn.optonline.net) has joined #fluid-work
[17:18:14 EDT(-0400)] * justin_o (n=jmo@142.150.154.101) has joined #fluid-work
[17:18:19 EDT(-0400)] <colinclark> +1 for ENGAGE-119
[17:18:31 EDT(-0400)] <colinclark> on bug parade
[17:18:36 EDT(-0400)] <justin_o> colinclark: thanks
[17:19:03 EDT(-0400)] <colinclark> justin_o: http://issues.fluidproject.org/browse/ENGAGE-119
[17:24:11 EDT(-0400)] <Bosmon> Why are we specifying an async option at all in loadData.js?
[17:24:18 EDT(-0400)] <Bosmon> It seems like it would restrict its portability...
[17:26:32 EDT(-0400)] <michelled> I don't know the detail about this but the async option 'fixed' the timeout issue. ENGAGE-119
[17:30:30 EDT(-0400)] * clown (n=clown@142.150.154.101) has left #fluid-work
[17:43:04 EDT(-0400)] <yura> colinclark: hi Colin, i updated browse and view components to work with multiple categories, but i didnt commit it, i created a patch and was wondering where I could post it
[17:43:12 EDT(-0400)] <colinclark> hey yura
[17:43:18 EDT(-0400)] <colinclark> Is there a JIRA for this issue?
[17:43:22 EDT(-0400)] <yura> no
[17:43:37 EDT(-0400)] <yura> but it fixes stuff that was breaking before
[17:43:37 EDT(-0400)] <colinclark> File a JIRA and post a post a patch there. I'm including it in our list of to-dos for the release
[17:43:41 EDT(-0400)] <colinclark> So I can add it to the bug parade
[17:43:49 EDT(-0400)] <colinclark> I think it's a good thing to include in this release
[17:43:49 EDT(-0400)] <yura> ok
[17:44:06 EDT(-0400)] <yura> are you moving the repository to trunk?
[17:44:35 EDT(-0400)] <colinclark> Tomorrow
[17:44:44 EDT(-0400)] <colinclark> I think we're in pretty good shape to get this release out soon
[17:44:51 EDT(-0400)] * anastasiac (n=team@142.150.154.189) has left #fluid-work
[17:44:57 EDT(-0400)] <colinclark> We've gone through each of the remaining tasks and estimated them
[17:44:59 EDT(-0400)] <colinclark> I'm just sending an email out to the list
[17:46:36 EDT(-0400)] <yura> this is the jira : http://issues.fluidproject.org/browse/ENGAGE-120
[17:51:32 EDT(-0400)] <colinclark> thanks, yura
[17:51:37 EDT(-0400)] <colinclark> stick a patch up there now
[17:51:42 EDT(-0400)] <colinclark> and we'll chat with the king about it tomorrow
[17:51:48 EDT(-0400)] <yura> (smile) sure
[17:51:53 EDT(-0400)] <yura> thanks
[18:52:38 EDT(-0400)] * colinclark_ (n=colin@142.150.154.130) has joined #fluid-work
[19:25:16 EDT(-0400)] * athena (n=athena@adsl-75-58-127-15.dsl.wlfrct.sbcglobal.net) has joined #fluid-work
[19:42:06 EDT(-0400)] * athena (n=athena@adsl-75-58-127-15.dsl.wlfrct.sbcglobal.net) has joined #fluid-work
[20:06:14 EDT(-0400)] * heidi (n=thesumme@bas5-oshawa95-1176469158.dsl.bell.ca) has joined #fluid-work