/
fluid-work IRC Logs-2010-06-17

fluid-work IRC Logs-2010-06-17

Owned by Fluid IRC Log Bot
Last updated: Jun 17, 2010Version comment

[03:15:08 CDT(-0500)] * thomas____ (~thomasain@213.246.129.247) has joined #fluid-work
[07:03:05 CDT(-0500)] * kasper (~kasper@189.188.54.2) has joined #fluid-work
[07:13:17 CDT(-0500)] * michelled (~michelled@142.150.154.101) has joined #fluid-work
[07:19:03 CDT(-0500)] * Justin_o (~Justin@142.150.154.171) has joined #fluid-work
[07:34:30 CDT(-0500)] * jessm (~Jess@c-71-232-3-151.hsd1.ma.comcast.net) has joined #fluid-work
[07:49:32 CDT(-0500)] * anastasiac (~team@142.150.154.193) has joined #fluid-work
[08:36:15 CDT(-0500)] * michelled (~michelled@user201-32.wireless.utoronto.ca) has joined #fluid-work
[08:58:08 CDT(-0500)] * EricDalquist (~dalquist@2607:f388:e:0:221:9bff:fe37:e768) has joined #fluid-work
[09:54:45 CDT(-0500)] * yura_ (~yura@142.150.154.114) has joined #fluid-work
[09:56:37 CDT(-0500)] * michelled (~michelled@142.150.154.101) has joined #fluid-work
[09:57:27 CDT(-0500)] * Justin_o (~Justin@142.150.154.171) has joined #fluid-work
[09:58:50 CDT(-0500)] * Bosmon2 (~a@c-67-176-80-187.hsd1.co.comcast.net) has joined #fluid-work
[10:14:15 CDT(-0500)] <Justin_o> jamon: can you remind me about our testswarm setup... does the perl script still do the checkout of the source code and the cron job updates it?
[10:19:25 CDT(-0500)] * athena (~athena@c-76-121-97-221.hsd1.wa.comcast.net) has joined #fluid-work
[10:22:15 CDT(-0500)] * clown (~clown@142.150.154.202) has joined #fluid-work
[10:36:09 CDT(-0500)] <kasper> justin_o: one variable that you have to set is the $SWARM (# The location of the TestSwarm that you're going to run against.)
[10:36:43 CDT(-0500)] <kasper> unsure what this means..
[10:38:51 CDT(-0500)] <jamon> Justin_o: ah let me see, svn hook runs the perl script, the perl script does a checkout and submits the job to testswarm, the cronjob on swarm.fluidproject.org does a checkout of svn to serve for tests
[10:38:57 CDT(-0500)] <jamon> Justin_o: pretty sure that's how it works
[10:39:28 CDT(-0500)] <jamon> kasper: set that $SWARM variable to swarm.fluidproject.org i think, it's the url of where the jobs live
[10:40:44 CDT(-0500)] <kasper> jamon: ok, thanks
[10:42:57 CDT(-0500)] <kasper> jamon, reading you answer to justin_o: So do I understand correctly, that your testswarm install will keep a local pool of checkouts to tested against?
[10:43:18 CDT(-0500)] <jamon> kasper: yeah that's the unresolved issue we haven't figured out
[10:43:33 CDT(-0500)] <jamon> kasper: it seems like testswarm expects a full checkout of the code for every revision
[10:43:58 CDT(-0500)] <jamon> kasper: as it stands now, i think every test gets run against the most recent checkout e.g. head
[10:44:06 CDT(-0500)] <kasper> ok
[10:44:22 CDT(-0500)] <jamon> kasper: is that ok with you for the time being?
[10:44:34 CDT(-0500)] <kasper> yeah, that's fine
[10:44:36 CDT(-0500)] <jamon> Justin_o and i should have time next week to sort that out somehow
[10:44:48 CDT(-0500)] <jamon> ok great
[10:45:32 CDT(-0500)] <Justin_o> jamon: yes... i wasn't sure if testswarm was somehow managing to request revisions... based on the checked out code... but i forgot that the tests are run against something else
[10:46:42 CDT(-0500)] <jamon> Justin_o: yeah i'm sure there's a way to do that, what i'm guessing is that the intent is to not store too many copies of code around, so once a test is run to satisfaction e.g. browser coverage and bug free, that checkout can be removed from the server
[10:47:10 CDT(-0500)] <jamon> otherwise, there must be a way to serve svn revisions as not plain text via mod_dav, just not sure how at the moment
[10:49:59 CDT(-0500)] <Justin_o> jamon: yep the trick will be that all of the specified browsers will have to have run... or at least a defined subset... yep... I think it would be better if we could serve revisions as not plain text, but I have no idea how to do this either..
[10:51:18 CDT(-0500)] <kasper> justin_o, jamon: how long time did it take you to get an auth code?
[10:53:47 CDT(-0500)] <kasper> sorry if that question made absolutely no sense without context: When signing up for test-swarm I requested an auth key/code for collectionspace. I figured this will be sent once the test-sw
[10:54:01 CDT(-0500)] <kasper> arm folks have checked out cspace
[10:54:19 CDT(-0500)] <kasper> do you remember how long it took to get that code, for you?
[10:54:27 CDT(-0500)] <jamon> kasper: yeah i'll go check the database, testswarm is bleeding edge alpha code (wink)
[10:55:59 CDT(-0500)] <jamon> kasper: the username you submitted to swarm.fluidproject.org is collectionspace?
[10:56:06 CDT(-0500)] <kasper> yup
[10:56:18 CDT(-0500)] <jamon> k, pm'ed you the authorization code
[10:56:37 CDT(-0500)] <kasper> thanks a lot jamon
[10:56:52 CDT(-0500)] <jamon> np, sorry it's taking so long to get setup
[10:58:11 CDT(-0500)] <kasper> jamon: nah, its not bad at all.. the main reason for taking time, is because I'm doing 1000 other things at the same time (smile)
[10:58:43 CDT(-0500)] <jamon> hahah, we're in the same club i guess
[11:16:41 CDT(-0500)] * kasper (~kasper@189.188.54.2) has joined #fluid-work
[11:19:45 CDT(-0500)] <kasper> jamon: got another question about the script: my $SUITE (currently "http://localhost/cspace-checkout"(wink), that would point to our own checkout, right (ie. http://nightly.collectionspace.org/testswarm)?
[11:25:14 CDT(-0500)] * clown (~clown@142.150.154.202) has joined #fluid-work
[11:35:33 CDT(-0500)] <jessm> jamon: ping?
[11:37:52 CDT(-0500)] <kasper> justin_o,jamon: ok, It seems I got it properly set up now (still manually executing perl script, but besides that..) .. It is behaving a bit odd though... It seems to stall on a test, even though it is completed
[11:38:24 CDT(-0500)] <kasper> I have my chrome 5 browser open as a client, and there it says the test is complete
[11:38:51 CDT(-0500)] <kasper> but it doesnt do the cooldown and starting new test, as it did when it was running your test-suite
[11:40:44 CDT(-0500)] <kasper> hmmm... nevermind... it seems to have gone on
[11:40:55 CDT(-0500)] <kasper> just very slow compared to your tests
[11:41:18 CDT(-0500)] <kasper> but ignore me for now.. I'll probably whine again soon!
[11:46:26 CDT(-0500)] <kasper> justin_o,jamon: Ok, I'll continue my rambling now. All the tests time out. They get executed correctly, everything passes, but it seems the tests do not "return" properly. Does this sound familiar/make sense to you?
[11:52:35 CDT(-0500)] <Justin_o> kasper: i'll try to look at your tests in the swarm
[11:54:06 CDT(-0500)] <Justin_o> kasper: i think you might be using the old version of jqunit
[11:54:58 CDT(-0500)] <Justin_o> kasper: this is the latest version https://source.fluidproject.org/svn/fluid/infusion/trunk/src/webapp/tests/test-core/jqUnit/js/jqUnit.js
[11:55:19 CDT(-0500)] <Justin_o> you can see there is a function at the top for testswarm... this injects the inject.js script onto your test page
[11:56:36 CDT(-0500)] <Justin_o> anastasiac: do you know if cspace's unit test framework was updated to the latests from infusion
[11:59:43 CDT(-0500)] <anastasiac> Justin_o, yes: we did update to the version that is present in the infusion-1.2.x-collectionspace-0.7alpha tag, which was cut shortly after 1.2
[12:02:07 CDT(-0500)] <Justin_o> anastasiac: do you know if you just updated qunit and not jqunit?
[12:02:22 CDT(-0500)] <anastasiac> let me check...
[12:02:45 CDT(-0500)] <Justin_o> okay.. i just took a peak at svn and it looks like it is the old version of jqunit https://source.collectionspace.org/collection-space/src/ui/trunk/src/test/lib/jqUnit/js/jqUnit.js
[12:03:27 CDT(-0500)] <Justin_o> kasper, anastasiac: additionally the inject.js file is also missing, which is needed by testswarm
[12:03:29 CDT(-0500)] <anastasiac> Justin_o, we did update jqUnit...
[12:03:33 CDT(-0500)] <Justin_o> really... hmm
[12:05:29 CDT(-0500)] <anastasiac> but as I said, not to what's in trunk, to what's in our tag
[12:05:48 CDT(-0500)] <Justin_o> anastasiac: this is the latest in the infusion https://source.fluidproject.org/svn/fluid/infusion/trunk/src/webapp/tests/test-core/jqUnit/js/jqUnit.js
[12:05:56 CDT(-0500)] <Justin_o> i guess your tag was made before the change
[12:06:56 CDT(-0500)] <anastasiac> probably
[12:07:03 CDT(-0500)] <Justin_o> kasper: so it looks like you are missing some necessary updates to the testing code... i'm not sure what you'll want to do. You could probably just upgrade the testing framework now or wait until after your release...
[12:10:12 CDT(-0500)] <kasper> anastasiac,justin_o: I guess it depends how big a thing it is to upgrade the framework, and how error-prone it is .. I'm a bit hesitant in doing anything that could introduce new bugs
[12:11:23 CDT(-0500)] <kasper> anastasiac: would you consider this process error-prone, and also, how long do you think it would take to upgrade the testing framework?
[12:12:06 CDT(-0500)] <Justin_o> kasper:It should only be a change to the tests. Which means that it shouldn't affect any of the actual running code. In a worst case scenario you could find yourself in a position where your unit tests aren't working.
[12:14:17 CDT(-0500)] <anastasiac> kasper, it's probably a good thing to get set up for test swarm. I think the issue is just developer availability, given our timelines. I wonder if laurel might have time to help, after she's done with the HTML work she's doing for repeated fields. Or maybe even Justin_o could create a patch for us
[12:15:27 CDT(-0500)] <Justin_o> anastasiac: i could probably look into making a patch if you guys would like to do the update before the release
[12:15:31 CDT(-0500)] <Justin_o> kasper: ^
[12:16:31 CDT(-0500)] <kasper> justin_o: that would be great.. We still got 13 days to go until release, and based on you and anastasiac's comments, I think it makes sense to upgrade
[12:16:58 CDT(-0500)] <anastasiac> Justin_o - thanks so much for offering to help with the upgrade
[12:17:37 CDT(-0500)] <Justin_o> kasper, anastasiac: np, if someone could create me a jira for it i'll create a patch and put it up there when it is ready
[12:17:59 CDT(-0500)] <kasper> will do
[12:17:59 CDT(-0500)] <anastasiac> Justin_o, kasper: we have a JIRA for the upgrade we did - I'll reopen it with a comment about this
[12:18:12 CDT(-0500)] <kasper> ah ok, cool! thanks anastasiac
[12:18:39 CDT(-0500)] <anastasiac> Justin_o: http://issues.collectionspace.org/browse/CSPACE-2038
[12:18:52 CDT(-0500)] <kasper> and yes, thanks a lot justin_o for helping us out with this!
[12:26:10 CDT(-0500)] <Justin_o> anastasiac: i'm going to probably have to slightly fork the jqunit code because it has a relative path to the inject.js file which won't be the same path as it will be in collectionspace
[12:26:18 CDT(-0500)] <Justin_o> any thoughts on this
[12:29:50 CDT(-0500)] <anastasiac> Justin_o, I think that'd probably be ok.
[12:30:05 CDT(-0500)] <anastasiac> maybe include a readme explaining any differences
[12:32:50 CDT(-0500)] <Justin_o> anastasiac: good idea okay, will do that
[12:46:33 CDT(-0500)] <jamon> if that's too much work Justin_o, kasper, anastasiac, find and sed could work from the perl script
[12:47:49 CDT(-0500)] <Justin_o> jamon: i would imagine that they should update at some point, but that could be a good stop gap if they want to hold off till after the release
[12:47:50 CDT(-0500)] <jamon> that's what was being used before that jira Justin_o posted
[12:48:09 CDT(-0500)] <jamon> upgrade for sure, but as a quick and dirty hack to get it working before the release (wink)
[14:11:03 CDT(-0500)] <Justin_o> kasper, anastasiac: I've posted a patch on the jira http://issues.collectionspace.org/browse/CSPACE-2038
[14:11:21 CDT(-0500)] <kasper> wuhu! the king strikes back!
[14:11:30 CDT(-0500)] <anastasiac> Justin_o, that's great. Let me get the patch I'm currently reviewing committed, then I'll have a look
[14:11:38 CDT(-0500)] <Justin_o> anastasiac: thanks
[14:12:59 CDT(-0500)] <Justin_o> kasper: i can't quite remember what the exact code does in the perl script but i noticed that one of your html files in the tests directory isn't actually a test. So you may have to make a slight modification so that it doesn't try to run it
[14:16:59 CDT(-0500)] <kasper> ahh thanks, justin_o! just took a look at the html files and the script.. seems I just need to change a regexp to only include filenames ending in *Test.html
[14:17:09 CDT(-0500)] <kasper> thanks for noticing that
[14:37:12 CDT(-0500)] <anastasiac> kasper, Justin_o: I've committed Justin's patch upgrading the UI to support testswarm
[14:37:52 CDT(-0500)] <kasper> awesome! thanks anastasiac! and thanks justin_o for finding and fixing this!
[15:56:23 CDT(-0500)] * Justin_o (~Justin@142.150.154.171) has joined #fluid-work
[16:04:32 CDT(-0500)] * jessm (~Jess@c-71-232-3-151.hsd1.ma.comcast.net) has joined #fluid-work
[16:17:54 CDT(-0500)] * colinclark (~colin@142.150.154.101) has joined #fluid-work
[16:46:03 CDT(-0500)] * anastasiac (~team@142.150.154.193) has left #fluid-work
[17:56:52 CDT(-0500)] * Justin_o (~Justin@142.150.154.171) has left #fluid-work