Skip to end of metadata
Go to start of metadata

[09:22:44 CST(-0600)] <michelled> Justin_o, yzen1, jvass: you'll probably find this interesting:

[09:24:13 CST(-0600)] <yzen> kasper_: ayt?

[09:25:17 CST(-0600)] <kasper_> iam

[09:25:19 CST(-0600)] <kasper_> yu

[09:25:23 CST(-0600)] <kasper_> yzen:

[09:26:55 CST(-0600)] <yzen> kasper_: I was wondering if there's a way to make our issue tracker and wiki a little but more open, anastasiac it trying to help us out with some of the documentation related stuff but has to go through a lot of steps to get access to them

[09:29:28 CST(-0600)] <kasper_> yzen: yeah, it IS a bit of a pain right now

[09:29:56 CST(-0600)] <kasper_> it's my sense that mediawiki is pretty prone to spam

[09:30:58 CST(-0600)] <kasper_> but probably not the issue tracker though plus that one is easier to see bots/ppl messing with

[09:31:23 CST(-0600)] <Justin_o> michelled: that's pretty cool..

[10:03:17 CST(-0600)] <machines> Sorry to bother you but what is Bert's name

[10:03:43 CST(-0600)] <yzen> kasper: which node version on win7 do we need to use?

[10:05:05 CST(-0600)] <kasper> yzen: let me just fire up my VM and check

[10:08:28 CST(-0600)] <kasper> yzen: 0.6.14 is what I've got

[10:08:54 CST(-0600)] <anastasiac> kasper, thanks, I'll try that version

[10:10:25 CST(-0600)] <kasper> anastasiac: did it explode with a more recent version?

[10:10:37 CST(-0600)] <anastasiac> I tried 0.6.9

[10:10:56 CST(-0600)] <anastasiac> yay! 0.6.14 works much better, thanks kasper

[10:11:03 CST(-0600)] <kasper> cool!

[10:36:30 CST(-0600)] <avtar> fluid-everyone: still on a call so here's my update: working with boris@nebula to find a workaround for the ubuntu 10.04 image not consuming more than 1.4 gigs of free space on any sized instance. and hoping to resolve the oercommons issue with alexn1 before resuming amanda & drbd work.

[11:10:24 CST(-0600)] <colinclark> yzen1: You joining us at the architecture meeting?

[11:35:06 CST(-0600)] <anastasiac> Bosmon, are you there?

[11:35:39 CST(-0600)] <Bosmon> Hi anastasiac

[11:35:48 CST(-0600)] <anastasiac> welcome back, and happy new year, Bosmon

[11:35:51 CST(-0600)] <Bosmon> I added some comments on the video player PUL request as I saw it going by

[11:35:58 CST(-0600)] <anastasiac> I hope your break was good

[11:36:03 CST(-0600)] <Bosmon> So michelled, cindyli might be interested since I think they are involved with this branch?

[11:36:08 CST(-0600)] <Bosmon> Who else is the team?

[11:36:14 CST(-0600)] <Bosmon> Thanks, it was relatively splendid : P

[11:36:17 CST(-0600)] <Bosmon> I hope yours was too

[11:36:30 CST(-0600)] <anastasiac> thanks for the comments on that request.

[11:36:49 CST(-0600)] <anastasiac> I have another question for you, Bosmon, about injecting an aggregate event into another component

[11:36:54 CST(-0600)] <Bosmon> Ah good

[11:37:14 CST(-0600)] <Bosmon> Hopefully nothing special should happen there

[11:37:20 CST(-0600)] <Bosmon> A boiled event is still just an event

[11:37:20 CST(-0600)] <anastasiac> I have tried to do this in the VP and it didn't work; the aggregate event never fired.

[11:37:32 CST(-0600)] <anastasiac> I created a test case that replicates the setup:

[11:37:38 CST(-0600)] <anastasiac>

[11:37:42 CST(-0600)] <Bosmon> Thanks, that sounds great

[11:37:44 CST(-0600)] <Bosmon> Let me have a look at it

[11:38:15 CST(-0600)] <anastasiac> it's at the bottom of the file, Bosmon. I haven't filed a JIRA yet since I wasn't sure if I was possibly expecting something that shouldn't work

[11:38:21 CST(-0600)] <anastasiac> sounds like it should, so I'll file something

[11:38:28 CST(-0600)] <Bosmon> I think it's always worthy of a JIRA

[11:38:49 CST(-0600)] <Bosmon> Since any properly organised instance of believing there to be a framework issue is worthy of being tracked (smile)

[11:39:00 CST(-0600)] <Bosmon> Especially if you have gone to the extent of producing a test case

[11:39:01 CST(-0600)] <anastasiac> (smile)

[11:42:00 CST(-0600)] <Bosmon> Hi anastasiac - by eye, it looks like you have tried to express the same event in two different ways, rather than transferring the desired event in the right direction

[11:42:25 CST(-0600)] <Bosmon> In the subcomponent record for "comp2" you say that comp2's onReady should be injected down from comp1

[11:42:44 CST(-0600)] <Bosmon> Whereas in comp2's own implementation you SEPARATELY say that this is a boiled event, composed from comp3 and comp4's onready events

[11:42:49 CST(-0600)] <Bosmon> You can't have both of these at the same time

[11:42:57 CST(-0600)] <Bosmon> So one event definition just overwrite the other

[11:43:35 CST(-0600)] <anastasiac> ok, so maybe I should describe what I'm trying to accomplish in the VideoPlayer, Bosmon

[11:44:14 CST(-0600)] <anastasiac> the controllers component has 5 subcomponent. I want the controller's onReady to be an aggregate of the 5 subcomponents' onReady events, so that controllers.onReady only fires after all 5 subcomponents are ready

[11:44:36 CST(-0600)] <Bosmon> Yes - is this related to the issue you had before the New Year?

[11:44:42 CST(-0600)] <anastasiac> I want the controllers' parent, the VideoPlayer proper, to share the controllers' onReady event so it can know when the controllers are ready

[11:44:52 CST(-0600)] <Bosmon> We had been thinking at that point that simply using the "onCreate" event for the parent component would be sufficient

[11:45:08 CST(-0600)] <Bosmon> Has something happened since then to suggest to you that that isn't good appropriate?

[11:45:11 CST(-0600)] <anastasiac> can you remind me when, in the lifecycle, onCreate fires?

[11:45:31 CST(-0600)] <Bosmon> It is definitely fired AFTER all non-delayed subcomponents have been created

[11:45:42 CST(-0600)] <Bosmon> But firstly - can we confirm that this is the same situation we were talking about before?

[11:46:30 CST(-0600)] <anastasiac> I suspect it does follow from that issue; my memory of exactly what we were talking about is blurred by turkey and stuffing

[11:46:44 CST(-0600)] <anastasiac> what do you mean by non-delayed subcomponents?

[11:47:27 CST(-0600)] <anastasiac> I'm looking for a way of knowing when a subcomponent is actually ready, as in ready to be used, as in it's fully set up, templates loaded, all rendered, etc. One of the subcomponents is dependent on a template

[11:49:31 CST(-0600)] <Bosmon> Ok

[11:49:36 CST(-0600)] <Bosmon> So there is some genuine asynchrony here

[11:50:13 CST(-0600)] <Bosmon> I meant that those subcomponents that are not marked as "createOnEvent"

[11:50:23 CST(-0600)] <anastasiac> ok

[11:57:25 CST(-0600)] <anastasiac> Bosmon, you can see the actual VideoPlayer branch where I was trying to get this working here:

[11:59:07 CST(-0600)] <Bosmon> So I think the issue with your test is that you are trying to inject the same event DOWN from the top as the same time you are boiling it UP from the bottom

[11:59:26 CST(-0600)] <Bosmon> There must only be exactly one definition of each event firer ... or rather, only one definition can be acted on

[12:01:34 CST(-0600)] <anastasiac> so what would be the right way to accomplish this, Bosmon?

[12:01:49 CST(-0600)] <Bosmon> Well, that's an excellent question (smile)

[12:02:07 CST(-0600)] <Bosmon> Once the GPII meeting ends, I should be able to think about this a bit more clearly

[12:02:35 CST(-0600)] <anastasiac> oh! I didn't realize you were on a call - sorry about that!

[12:02:37 CST(-0600)] <Bosmon> This is sadly one of those cases which really does depend on having the GINGER WORLD working properly

[12:09:55 CST(-0600)] <Bosmon> What you WANT to express in comp 1's definition is that its event is injected upwards from comp2

[12:10:24 CST(-0600)] <Bosmon> Actually I think this actually is possible, given our existing pitiful implementation of gingerness"

[12:10:32 CST(-0600)] <Bosmon> It is the one case it is implemented for

[12:10:39 CST(-0600)] <Bosmon> "Early dependence on option from subcomponent"....

[12:11:02 CST(-0600)]

<Bosmon> If you say that comp1's event is "

Unknown macro: {comp1}" it should work

[12:11:08 CST(-0600)] <Bosmon> anastasiac ^

[12:11:25 CST(-0600)] <Bosmon> It will recognise that the path segment "comp2" refers to an uninstantiated component and immediately move on to instantiating it

[12:11:35 CST(-0600)] <Bosmon> "possibly"

[12:11:53 CST(-0600)] <anastasiac> "possibly" - "hopefully"? (smile)

[12:12:00 CST(-0600)] <anastasiac> I'll give it a try

[12:12:53 CST(-0600)] <Bosmon> Based on my memory of the implementation, I think the chances are high

[12:13:09 CST(-0600)] <Bosmon> At least, that is what it was designed to do, I don't believe anyone really relied on this feature before

[12:13:18 CST(-0600)] <Bosmon> But it was the best I could do about "gingerness" given the existing model

[12:13:48 CST(-0600)] <anastasiac> Bosmon, would this replace the comp2 onReady definition in the comp2 subcomponent declaration?

[12:16:28 CST(-0600)] <Bosmon> anastasiac - you would remove comp2's definition completely in the subcomponent declaration

[12:16:42 CST(-0600)] <Bosmon> And this would instead be written as the definition for comp1's onReady event

[12:16:52 CST(-0600)] <Bosmon> By this scheme, the injection proceeds in steps all the way from the bottom to the top

[12:17:04 CST(-0600)] <anastasiac> ok, that's what I thought

[12:17:11 CST(-0600)] <Bosmon> And you will only have ONE definition for comp2's event in the system

[12:17:16 CST(-0600)] <Bosmon> That is, the correctly boiled one

[12:21:14 CST(-0600)] <anastasiac> hm. comp2's onReady fires, but not comp1's….

[12:21:23 CST(-0600)] <Bosmon> Ok

[12:21:32 CST(-0600)] <Bosmon> If you set a test case with THAT version in, I can try to make it work (smile)

[12:21:43 CST(-0600)] <Bosmon> And that would definitely be worth a JIRA

[12:21:51 CST(-0600)] <anastasiac> ok (smile)

[12:21:52 CST(-0600)] <Bosmon> "improper gingerness during event injection"....

[12:29:28 CST(-0600)] <Bosmon> Although it would seem impossible, that if comp1 got any event at all, that it wouldn't fire

[12:29:36 CST(-0600)] <Bosmon> Since either there is no event, or the right event

[12:29:41 CST(-0600)] <Bosmon> I don't see any other possibility

[12:30:03 CST(-0600)] <anastasiac> well, maybe I've done the injections wrong...

[12:30:17 CST(-0600)] <anastasiac> fluid.defaults("fluid.tests.comp1", {

[12:30:17 CST(-0600)] <anastasiac> gradeNames: ["fluid.eventedComponent", "autoInit"],

[12:30:19 CST(-0600)] <anastasiac> events: {

[12:30:20 CST(-0600)]

<anastasiac> onReady: "

Unknown macro: {comp1}"

[12:30:22 CST(-0600)] <anastasiac> },

[12:30:23 CST(-0600)] <anastasiac> components: {

[12:30:25 CST(-0600)] <anastasiac> comp2: {

[12:30:26 CST(-0600)] <anastasiac> type: "fluid.tests.comp2"

[12:30:28 CST(-0600)] <anastasiac> }

[12:30:28 CST(-0600)] <anastasiac> }

[12:30:29 CST(-0600)] <anastasiac> });

[12:30:32 CST(-0600)] <Bosmon> That seems correct to me

[12:30:44 CST(-0600)] <Bosmon> If comp1 gets any event, I can't see how it could avoid being the correct event

[12:31:03 CST(-0600)] <Bosmon> There could be a bugg

[12:31:37 CST(-0600)] <Bosmon> Do you know if there is a community meeting today?

[12:33:52 CST(-0600)] <anastasiac> Bosmon:

[12:34:14 CST(-0600)] <anastasiac> with a new branch with the new test case

[12:34:25 CST(-0600)] <anastasiac> with a new branch with the new test case

[12:34:37 CST(-0600)] <Bosmon> The JIRA could do with a slightly better description (smile)

[12:34:49 CST(-0600)] <Bosmon> Maybe it's time for another of those "how to write a good JIRA" meetings again....

[12:35:12 CST(-0600)] <anastasiac> I was hoping the test case would suffice

[12:35:21 CST(-0600)] <Bosmon> Of course it can't (smile)

[12:35:29 CST(-0600)] <Bosmon> There should be words explaining what about the test case is significant

[12:35:54 CST(-0600)] <Bosmon> Also, about the situation in which you encountered the problem, what you were trying to do, and our characterisation of the issue so far

[12:36:08 CST(-0600)] <Bosmon> The JIRA should contain EVERYTHING that a person from the future needs, in order to completely characterise what the problem is

[12:36:16 CST(-0600)] <Bosmon> In case we forget, or someone else needs to understand

[12:36:55 CST(-0600)] <anastasiac> ok, I'll update it with more information

[12:37:02 CST(-0600)] <Bosmon> thanks so much, anastasiac!

[12:43:59 CST(-0600)] <anastasiac> Bosmon, yes: there is a community meeting today, to discuss "Where should components have their issues tracked" and, I believe, whether or not we should version components separately from the framework

[12:55:12 CST(-0600)] <Bosmon> oh dear : P

[13:48:13 CST(-0600)] <Justin_o_> fluid-everyone: would anyone like to join the community meeting remotely this week

[13:48:21 CST(-0600)] <Justin_o_> the topic is about how to organize our components in Jira

[14:16:05 CST(-0600)] <Bosmon> Hi Justin_o - sorry to be so late - is the meeting still going on?

[14:16:09 CST(-0600)] <Bosmon> Is it happening over Skype?

[14:16:49 CST(-0600)] <Justin_o> Bosmon: yes.. i'll Skype you in

[14:16:58 CST(-0600)] <Bosmon> great!

[14:16:59 CST(-0600)] <Bosmon> Thanks

  • No labels