fluid-work IRC Logs-2012-09-25

Owned by Fluid IRC Log Bot
Last updated: Sept 25, 2012Version comment
7 min read

[08:56:35 CDT(-0500)] <kasper> good morning mr. JURA

[08:56:38 CDT(-0500)] <kasper> (yura)

[09:00:16 CDT(-0500)] <yura> kasper:

[09:05:28 CDT(-0500)] <kasper> so I finally go around to looking at my old pull request you commented

[09:05:35 CDT(-0500)] <kasper> https://github.com/GPII/universal/pull/48

[09:05:52 CDT(-0500)] <kasper> yura ^

[09:06:28 CDT(-0500)] <kasper> problem is if I do not include the test-framework in universal, then I'm not able to include it in the Linux/Windows branch

[09:06:32 CDT(-0500)] <kasper> as far as I can tell

[09:07:17 CDT(-0500)] <yura> ya i think as far as i remember you could load necessary files explicitly

[09:08:25 CDT(-0500)] <kasper> yes, but unless I use the full path, I dont believe node's module search will find the testFramework module

[09:09:11 CDT(-0500)] <kasper> (this is for linux and windows – in universal it finds the testFramework just fine)

[09:11:19 CDT(-0500)] <kasper> yura: I might be doing something stupid / missing something obvious

[09:59:24 CDT(-0500)] <alexn> colinclark: ayt?

[09:59:32 CDT(-0500)] <colinclark> alexn: marginally

[09:59:35 CDT(-0500)] <colinclark> what's on your mind?

[10:00:02 CDT(-0500)] <alexn> did you have any chance to review those 2 pull requests (smile) ? https://github.com/fluid-project/videoPlayer/pull/62 and https://github.com/fluid-project/videoPlayer/pull/58

[10:00:08 CDT(-0500)] <alexn> just asking

[10:00:11 CDT(-0500)] <colinclark> If you did...

[10:00:14 CDT(-0500)] <colinclark> Sorry

[10:00:16 CDT(-0500)] <colinclark> If I did...

[10:00:18 CDT(-0500)] <colinclark> you'd know (wink)

[10:00:26 CDT(-0500)] <alexn> (smile)

[10:00:38 CDT(-0500)] <colinclark> I'll see what I can do this afternoon

[10:00:50 CDT(-0500)] <colinclark> I've had constant meetings for the past few days

[10:00:58 CDT(-0500)] <alexn> yeah no worries. I just wanted to verify that I did not miss anything

[10:01:16 CDT(-0500)] <alexn> you always have a lot on your plate

[10:04:33 CDT(-0500)] <yura> kasper: so you mean the comment in XMLSettingsHandlerTests ?

[10:04:47 CDT(-0500)] <colinclark> alexn: Did you use some code to generate the new timestamps in these files? https://github.com/fluid-project/videoPlayer/pull/62/files

[10:05:49 CDT(-0500)] <alexn> colinclark: yes I did (smile)

[10:05:55 CDT(-0500)] <colinclark> where is that code?

[10:06:03 CDT(-0500)] <alexn> I was reusing the same function from captionator. Let me find it.

[10:08:33 CDT(-0500)] <alexn> colinclark: I just used the same one as Captionator so that times would be aligned with the ones which exist in Amara https://github.com/fluid-project/videoPlayer/blob/demo/lib/captionator/js/captionator.js#L110-L117

[10:11:27 CDT(-0500)] <colinclark> so how did you use this function, specifically?

[10:12:29 CDT(-0500)] <colinclark> did you write a little program that read the universal subtitles JSON and called your unisubJSONtoVTT() function?

[10:13:12 CDT(-0500)] <alexn> well it was a quick script written and it was more like my copy-paste the reply from Amara into JSON object and parsing it (smile)

[10:13:46 CDT(-0500)] <kasper> yura: yes and no (smile) .. your suggestion does work for the XML SH as it is in universal repo .. but trying to do the same trick with one of the platform specific tests doesn't work, as it is unable to find the testframework module from there

[10:13:48 CDT(-0500)] <alexn> since we do not want to support our JSON-WebVTT format I decided not to spend time on creating something more complex which would pull JSON and convert

[10:14:08 CDT(-0500)] <alexn> but yes I basically reused captionator code

[10:14:46 CDT(-0500)] <colinclark> ok

[10:14:54 CDT(-0500)] <colinclark> this one is going in

[10:15:28 CDT(-0500)] <alexn> I found my code which I used. I can copy paste it somewhere and show it to you if you want

[10:17:04 CDT(-0500)] <alexn> oh wait my bad it is not it

[10:17:52 CDT(-0500)] <alexn> I have this one left only http://pastie.org/4797923

[10:18:36 CDT(-0500)] <alexn> I remember know I was creating our JSON using returned Amara's JSON. And then I converted our JSON to WebVTT

[10:19:14 CDT(-0500)] <alexn> I do not have the code example for time conversion but I can quickly sketch it for you, if you like colinclark

[10:19:32 CDT(-0500)] <colinclark> It doesn't really matter, no

[10:20:00 CDT(-0500)] <yura> kasper: and this is all for the node tests not browser tests

[10:20:01 CDT(-0500)] <yura> ?

[10:20:10 CDT(-0500)] <alexn> colinclark: well I can assure you that I used captionator's functions for time conversions.

[10:20:47 CDT(-0500)] <kasper> yura: yes

[10:21:01 CDT(-0500)] <colinclark> thank you for the assurance, alexn (tongue)

[10:21:37 CDT(-0500)] <alexn> colinclark: (smile)

[10:22:00 CDT(-0500)] <yura> kasper: and you assume that the universal is checked out into a sibbling node_modules dir right ?

[10:23:36 CDT(-0500)] <colinclark> alexn: It's interesting. Your comment about having to cut and paste the code from Captionator into your little quick script tells us a lot about code factoring...

[10:23:42 CDT(-0500)] <kasper> yura: yes - ie: gpii/linux, gpii/node_modules/universal, gpii/node_modules/universal/gpii/node_modules/testFramework

[10:24:00 CDT(-0500)] <colinclark> You were, yourself, a client of that JSON to WebVTT API

[10:24:07 CDT(-0500)] <colinclark> and because of its factoring, couldn't use it directly

[10:24:37 CDT(-0500)] <colinclark> since it was just an instance method you added to Captionator, rather a function available to all

[10:25:00 CDT(-0500)] <colinclark> michelled's new branch has this code factored in a way that would have allowed you to use it directly, with no cutting or pasting required: https://github.com/michelled/videoPlayer/blob/FLUID-4701/js/VideoPlayer_html5Captionator.js#L156-166

[10:25:15 CDT(-0500)] <alexn> Also it is hard to parse WebVTT to create a JSON. So I decided to patse Amara's JSON to create our JSON. Then parse our JSON to create WebVTT

[10:33:20 CDT(-0500)] <colinclark> alexn: FLUID-4793 is pushed and resolved

[10:37:41 CDT(-0500)] <colinclark> alexn: It doesn't look like your FLUID-4787 pull request is quite there

[10:37:56 CDT(-0500)] <colinclark> I actually don't fully understand it--it seems half-finished?

[10:38:41 CDT(-0500)] <avtar> fluid-everyone: i'm still on a call so here's my standup/connections update. yesterday didn't turn out as planned – had to look into atutor/mediawiki issues with cindy, set up a wiki for idi, and troubleshoot issues with the nebula cluster. the latter has been escalated and their support team is looking into it. today i'm going to prepare for the 'backup curve' xen server work before meeting yong, and then work with bert in the data

[10:38:41 CDT(-0500)] <avtar> centre after that.

[10:38:44 CDT(-0500)] <colinclark> For example, you introduced a new test utility function called "setStaticEnvironment" which appears to hard code the components that can be set: https://github.com/fluid-project/videoPlayer/pull/58/files#L6R66

[10:39:18 CDT(-0500)] <colinclark> And then below you have a similar function: https://github.com/fluid-project/videoPlayer/pull/58/files#L6L55

[10:39:33 CDT(-0500)] <colinclark> sorry, this line: https://github.com/fluid-project/videoPlayer/pull/58/files#L6R76

[10:40:22 CDT(-0500)] <colinclark> The use of "integrations" is odd, further down… https://github.com/fluid-project/videoPlayer/pull/58/files#L8R158

[10:40:55 CDT(-0500)] <colinclark> Where you seem to be making a step towards allowing test developers to specify which components they want to add to the static environment, but then it also seems to be hard-coded

[10:41:26 CDT(-0500)] <jessm> avtar: when are we meeting next?

[10:41:45 CDT(-0500)] <alexn> colinclark: I tried to use this code as a good example of how to do it -> https://github.com/fluid-project/infusion/blob/master/src/webapp/tests/component-tests/uploader/js/UploaderTests.js#L525-578

[10:41:56 CDT(-0500)] <colinclark> alexn: I think you could also make your "testFunction" in the test configuration more declarative: https://github.com/fluid-project/videoPlayer/pull/58/files#L8R186

[10:42:10 CDT(-0500)] <colinclark> I like the steps you are taking--they're definitely in the right direction

[10:42:19 CDT(-0500)] <colinclark> and show a really nice push towards declarative tests

[10:42:34 CDT(-0500)] <colinclark> but it just doesn't look to me like you've quite finished the job yet

[10:43:53 CDT(-0500)] <alexn> colinclark: what about setStaticEnvironment. I used the code which we have in Uploader inside of Infusion as my guide on how to do it

[10:44:26 CDT(-0500)] <colinclark> Right. That isn't a sufficiently good example, as Bosmon and I both mentioned last week

[10:44:38 CDT(-0500)] <colinclark> It hopefully points you in the right direction

[10:44:42 CDT(-0500)] <colinclark> but it's probably not quite there

[10:45:01 CDT(-0500)] <colinclark> you have to go a little further in the design process with this

[10:45:02 CDT(-0500)] <alexn> colinclark: Ok. I will take a loot to make it more dynamic

[10:45:08 CDT(-0500)] <colinclark> nice, thanks

[10:45:08 CDT(-0500)] <avtar> jessm: are you free on thursday morning?

[10:45:30 CDT(-0500)] <jessm> avtar: that's bert-o-clock for me, not available

[10:51:45 CDT(-0500)] <avtar> jessm: i adium'd you

[10:51:50 CDT(-0500)] <avtar> anastasiac: i'm looking into that error

[10:58:13 CDT(-0500)] <anastasiac> avtar, thanks

[11:27:32 CDT(-0500)] <avtar> anastasiac: this was going on http://www.mediawiki.org/wiki/Extension_talk:ConfirmAccount#Accept_Requests_doesn.27t_work_after_update

[11:28:05 CDT(-0500)] <avtar> which is weird because the plugin's snapshot is from yesterday

[11:28:28 CDT(-0500)] <anastasiac> odd. David did eventually get a password - is that because you fixed something, avtar?

[11:29:47 CDT(-0500)] <avtar> yes, i gave it another shot with bits retrieved from here https://gerrit.wikimedia.org/r/gitweb?p=mediawiki/extensions/ConfirmAccount.git

[11:31:52 CDT(-0500)] <avtar> it works but the plugin's config options have changed so i'll need to take another look at this once i'm back from the data centre later today

[11:50:39 CDT(-0500)] <avtar> anastasiac: nm, it's good to go. just had to move some of the options into an array.

[11:51:46 CDT(-0500)] <Justin_o> cindyli: i'm looking at the test_convetionatl.py file

[11:52:03 CDT(-0500)] <Justin_o> do we expect the file name to be "capture-0_0.jpg" or "capture_0_0.jpg)

[11:53:02 CDT(-0500)] <cindyli> yes, Justin_o

[11:53:14 CDT(-0500)] <Justin_o> cindyli: the first one right?

[11:53:26 CDT(-0500)] <cindyli> ya, my preference is the first one.

[11:54:14 CDT(-0500)] <Justin_o> cindyli: okay… no problem.. just wanted to double check before i renamed the image files we are using for mockCapture

[11:54:17 CDT(-0500)] <Justin_o> and testing

[11:55:57 CDT(-0500)] <cindyli> thanks, Justin_o

[12:05:40 CDT(-0500)] <cindyli> Justin_o: we have a multiCapture var that either equals to simCapture or sequentialCapture. however, the simCapture and sequentialCapture take in different set of parameters. sim has 3 more. In order to call one multiCapture for both, these 2 funcs should accept same set of parameters. Do you think if it's proper to fake in 3 more parameters for sequentialCapture?

[12:08:11 CDT(-0500)] <Justin_o> cindyli: (sad)

[12:08:35 CDT(-0500)] <Justin_o> we could change it to take in **kwargs instead

[12:11:15 CDT(-0500)] <cindyli> ah, good point. thanks, Justin_o

[12:11:52 CDT(-0500)] <Justin_o> cindyli: then we can access the pieces through kwargs['property']

[12:12:25 CDT(-0500)] <cindyli> great

[14:35:29 CDT(-0500)] <Justin_o> cindyli: okay.. i pushed up my changes.. still need to implement DELETE though..

[14:35:45 CDT(-0500)] <cindyli> ok, thanks, Justin_o

[14:36:05 CDT(-0500)] <Justin_o> cindyli: talk to you tomorrow

[14:36:12 CDT(-0500)] <cindyli> ok

[14:57:10 CDT(-0500)] <avtar> anastasiac: the logo is up

[14:57:31 CDT(-0500)] <anastasiac> avtar, thanks!!