The IRC channel is #sakai on the freenode network, you can join using a web 
client at http://webchat.freenode.net/?channels=sakai and you can view the 
complete logs at http://sakai.iriscouch.com/irc/_design/viewer/index.html

13:00 <SakaiGithub> [3akai-ux] bp323 pushed 1 new commit to 1.5.0: 
http://git.io/IPIi-g
13:03 <SakaiGithub> [3akai-ux] bp323 pushed 1 new commit to 1.5.0: 
http://git.io/rGTpvg
13:03 <SakaiGithub> [3akai-ux/1.5.0] Merge remote-tracking branch 
'origin/master' into 1.5.0 - Bert Pareyn
13:26 <SakaiGithub> [3akai-ux/1.5.0] SAKIII-5948 - add support for TED videos - 
Christian Vuerings
14:02 <SakaiGithub> [3akai-ux] bp323 pushed 1 new commit to 1.5.0: 
http://git.io/pc56dw
14:37 <SakaiGithub> [3akai-ux] bp323 pushed 1 new commit to 1.5.0: 
http://git.io/hMregg
14:37 <SakaiGithub> [3akai-ux/1.5.0] SAKIII-5739 turned clickable div into a 
button - jsloane
14:49 <SakaiGithub> [3akai-ux] bp323 pushed 1 new commit to 1.5.0: 
http://git.io/xSX1oA
14:49 <SakaiGithub> [3akai-ux/1.5.0] SAKIII-5777 added screen reader text to 
See All links - jsloane
15:39 <zacht> ping mrvisser
15:39 <MrVisser> hey zacht
15:40 <zacht> mrvisser Do you have a cloudbees account? I want to see if I can 
have you added to sonar administrators.
15:40 <zacht> mrvisser cool, I'm on it.
15:40 <MrVisser> great, thanks!
15:42 <zacht> mrvisser They have some kind of sso going on. If you're logged in 
at cloudbees.com, I _think_ you'll now be able to see this: 
https://sakai-oae.sonar.cloudbees.com
15:43 <MrVisser> zach: I wasn't authenticated, but it asked me to log in. when 
I did and tried to visit that page, I got the 500 error:
15:43 <MrVisser> java.lang.IllegalStateException: Fail to validate openId token
15:43 <MrVisser>        
org.sonar.plugins.openid.OpenIdValidationFilter.doFilter(OpenIdValidationFilter.java:77)
15:43 <MrVisser>        
org.sonar.server.platform.ServletFilters$GodFilterChain.doFilter(ServletFilters.java:102)
15:43 <MrVisser>        
org.sonar.server.platform.ServletFilters.doFilter(ServletFilters.java:73)
15:43 <MrVisser>        
org.sonar.server.ui.DatabaseSessionFilter.doFilter(DatabaseSessionFilter.java:34)
15:43 <MrVisser>        
org.mortbay.servlet.UserAgentFilter.doFilter(UserAgentFilter.java:78)
15:43 <MrVisser>        
org.mortbay.servlet.GzipFilter.doFilter(GzipFilter.java:131)
15:44 <zacht> grrr. I created a sonar user using your email address. I thought 
that would do it.
15:44 <MrVisser> seems I can load grandcentral.cloudbees.com and it recognizes 
me fine
15:44 <MrVisser> zacht: try using just 'mrvisser' instead
16:24 <raydavis> I'm going to try wrangling a Solr upgrade pull request by end 
of day -- it's a tough window to fit through because there's so much chance of 
merge conflicts.
16:25 <zacht> raydavis Can we help by giving you an exclusive lock?
16:26 <raydavis> zacht, thanks, but I'll give it a go without one today. If 
that doesn't work, though, I'll definitely take you up on that!
16:27 <raydavis> I really want to get this massive change in early so we can 
start finding the places that aren't covered in integration tests.
16:32 <zacht> stuart++
16:32 <zacht> gah!
16:32 <zacht> stuartf++
16:34 <stuartf> sakai-bot: karma empty stuart
16:39 <zacht> stuartf Woah, I didn't know karma was so fragile. :-)
16:40 <stuartf> yeah, that doesn't check for auth or anything, honor system
16:44 <SakaiGithub> [3akai-ux] nicolaasmatthijs pushed 3 new commits to 1.4.0: 
http://git.io/ce7REw
16:44 <SakaiGithub> [3akai-ux/1.4.0] SAKIII-6068 fix IE content upload - jsloane
16:44 <SakaiGithub> [3akai-ux/1.4.0] SAKIII-6068 fix IE content upload - jsloane
16:44 <SakaiGithub> [3akai-ux/1.4.0] SAKIII-6068 Fix version upload for IE - 
Bert Pareyn
16:45 <SakaiGithub> [3akai-ux] nicolaasmatthijs pushed 1 new commit to 
1.4.0-release: http://git.io/09mYyA
16:45 <SakaiGithub> [3akai-ux/1.4.0-release] Merge branch '1.4.0-release' of 
github.com:sakaiproject/3akai-ux into 1.4.0-release - Nicolaas Matthijs
17:08 <SakaiGithub> [3akai-ux] christianv pushed 1 new commit to 1.4.1: 
http://git.io/57pMXw
17:08 <SakaiGithub> [3akai-ux/1.4.1] Merge bp323/SAKIII-6068 into 1.4.1 - 
Christian Vuerings
17:14 <SakaiGithub> [3akai-ux] christianv pushed 2 new commits to 1.5.0: 
http://git.io/tjT4IA
17:14 <SakaiGithub> [3akai-ux/1.5.0] Merge remote-tracking branch 
'bp323/SAKIII-6068' into 1.5.0 - Christian Vuerings
17:14 <SakaiGithub> [3akai-ux/1.5.0] Merge bp323/SAKIII-6068 into 1.5.0 - 
Christian Vuerings
17:17 <SakaiGithub> [3akai-ux] christianv pushed 3 new commits to master: 
http://git.io/dfXvjg
17:17 <SakaiGithub> [3akai-ux/master] Merge pull request #2304 from 
jsloane/SAKIII-5838 - Bert Pareyn
17:17 <SakaiGithub> [3akai-ux/master] Merge remote-tracking branch 
'bp323/SAKIII-6068' - Christian Vuerings
17:17 <SakaiGithub> [3akai-ux/master] NOJIRA - Update snapshot versions - 
Christian Vuerings
17:18 <MrVisser> zacht: do you know much about the cropit processor?
17:18 <zacht> mrvisser I've been in there a time or two. It's knowable. :-)
17:20 <MrVisser> zacht: good to know :) I think I'm experience some sort of 
delay-related issue. When I create a user and upload their profile picture and 
crop it through the UI, it works fine. When I do it using the model loader, I 
get a NPE, which suggests that the image it wants to crop does not exist. Does 
this ring any kind of bells to you?
17:21 <zacht> mrvisser Does the model wait for the image upload before making 
the request to crop?
17:25 <MrVisser> zacht: the request to crop the picture is done in the callback 
of the request that uploads it.
17:33 <zacht> raydavis whew, that solr artifact is BIG.
17:34 <PhysX> denbuzze ping
17:34 <raydavis> zacht, yes! You can see why I don't want to download it every 
time. :)
17:34 <raydavis> Plus Apache doesn't like when people get distribution binaries 
from the central URL.
17:34 <denbuzze> PhysX: pong
17:34 <PhysX> denbuzze morning! would you mind taking a look at the 3 PR's that 
are assigned to you for review some time today?
17:34 <denbuzze> PhysX: already working on the first one (build on Windows)
17:35 <PhysX> denbuzze awesome
17:36 <denbuzze> PhysX: btw I noticed we need to be very careful with cross 
patching PR's - it sometimes updates some of the .xml files with different 
snapshot versions
17:37 <zacht> mrvisser I'm not aware of any delay introduced by the image 
upload (I'm pretty sure indexing is not involved).
17:37 <PhysX> denbuzze: oh right
17:37 <PhysX> denbuzze: we should
17:38 <zacht> mrvisser But of course that doesn't mean there isn't something 
that takes some time to finish.
17:38 <denbuzze> PhysX: They all should be ok now - but maybe it's good to 
double check (1.4.0, 1.4.0-release, 1.4.1, 1.5.0 and master ^^)
17:38 <PhysX> yeah
17:41 <MrVisser> zacht: thanks for sniffing it out for me :)
17:55 <ctweney> hey zacht, check it out — cuke tests for the files bundle : 
https://github.com/ctweney/nakamura/commit/595e79321588cc3f774155bb2907c224bd522693
17:56 <zacht> ctweney You are making my day!
17:57 <zacht> ctweney How are you finding it?
17:57 <ctweney> at first I hated the redundant verbosity
17:57 <ctweney> but now I'm finding it a nice road map to the test
17:57 <ctweney> although it's tempting to cheat and say "Then I check all my 
expected stuff" and have all the complex tests in the ruby code
17:58 <zacht> ctweney In my limited experience, I have found that you want to 
build up a kind of vocabulary of phrases that convey the most reusability.
17:59 <ctweney> yeah
18:07 <MrVisser> zacht: argh, the problem was in some modifications I made to 
the model loader to handle HTTPS
18:07 <MrVisser> there are 2 request entry points.. the upload of the image was 
flying under the radar to an HTTP URL and getting 302'd :(
18:08 <zacht> mrvisser good sleuthing.
18:09 <zacht> ctweney One of the things I really like is being able to write 
the spec and for it to very kindly show you which parts remain unimplemented.
18:19 <ctweney> zacht, I'm using RubyMine to edit cuke tests, and it is gold
18:19 <ctweney> gold, zacht, gold!
18:19 <ctweney> autocompletion of the verbose phrases and everything!
18:20 <zacht> ctweney JetBrains can have all my money.
18:20 <ctweney> but they offer open-source licenses, so you don't even have to 
pay!
18:29 <thecarlhall> zacht, what's the goal with having a minimal setup for 
testing files.impl and user.impl?  the graph reduction is pretty minimal and 
mostly seen in our own bundles
18:30 <thecarlhall> zacht, to get the full working graph, I'm going to need 
more scripts to walk the predecessor graphs for api bundles
18:32 <zacht> thecarlhall The scenario I had in mind is a Felix with only what 
you would need in it to: 1) take an HTTP request 2) invoke some service under 
test
18:32 <zacht> thecarlhall The service under test may or may not be hooked up to 
a "real" database.
18:33 <thecarlhall> zacht, right but why fewer bundles? if we have our lazy 
loading set correctly, the unused services won't be loaded
18:34 <zacht> thecarlhall if it can startup and shutdown fast, I don't mind if 
it has many bundles in it.
18:35 <thecarlhall> zacht, I ask because it seems like we're going to hit a 
marginal difference between all nakamura bundles vs. as few as needed
18:35 <zacht> thecarlhall You're using pax runner?
18:35 <thecarlhall> zacht, haven't even got that far. I'm still building the 
graph
18:35 <thecarlhall> zacht, we end up with most if not all sling bundles which 
is where the startup slowness is
18:36 <zacht> thecarlhall But what if there is not sling dependency?
18:36 <zacht> no*
18:36 <thecarlhall> zacht, we depend on sling.api and that drags in a big graph
18:37 <thecarlhall> zacht, I'm at the point of writing more script power to 
build the rest of the graph. I can go ahead with that to check my insanity
18:37 <zacht> thecarlhall It's my intention that after I have decomposed this 
bundle according to responsibilities, that the core of it won't make any 
mention of sling.
18:38 <thecarlhall> zacht, files.impl refers to sling.api. I haven't checked to 
see where yet.
18:38 <thecarlhall> zacht, but if there are servlets, they will extend 
SlingAllMethodsServlet or SlingSafeMethodsServlet
18:39 <thecarlhall> zacht, the tricky part with sling.api is that it is all of 
the sling api whereas we do smaller sets like files.api
18:40 <thecarlhall> zacht, so to find what implements the slice of sling.api we 
use, I'll have to start/stop the server to see where it breaks
18:41 <thecarlhall> here's a view of our system without the second level api -> 
impl resolution: http://ge.tt/26VbgFL/v/0
18:41 <zacht> thecarlhall We can start by spinning up the whole enchilada, with 
the goal of whittling it down over time.
18:45 <thecarlhall> zacht, we can easily cut down the number of our own bundles 
we start but lazy loaded services really handle that for us
18:45 <thecarlhall> I'll take a read through our use of immediate=true
19:05 <thecarlhall> zacht, we use immediate=true very loosely. I'm going to 
clear that up and check the startup time from there. If you want to still 
pursue a server with fewer bundles, I can finish the scripts to build up that 
bundle list.
19:11 <zacht> thecarlhall ok.
19:12 <thecarlhall> zacht, it's amazing the number of places that start 
immediately and don't need to
19:12 <zacht> thecarlhall Sounds like a good thing to look for.
19:12 <thecarlhall> zacht, there's also lazy bundle activation if we want it
20:35 <ctweney> thecarlhall, zacht, do either of you have any idea what the 
"alternativeStream" bits of code in CreateContentPoolServlet are about?
20:36 <thecarlhall> ctweney, it's been a while since I read through it
20:36 <thecarlhall> ctweney, I *think* it's pages on a single piece of content
20:36 <zacht> ctweney Those are used for small, normal, and large versions of 
images.
20:36 <thecarlhall> that's the one ^^
20:36 <ctweney> aha
20:37 <ctweney> wow, it's well documented!
20:37 <raydavis>  /* Ask zacht */
20:37 <ctweney>  /* if that fails, ask raydavis */
20:38 <zacht> ctweney Exercise for the reader to see if they're used in some 
other way. :-|
20:39 <zacht> ctweney As ugly as it is, I'm _really_ glad we're drawing back 
the curtain on this stuff.
20:39 <zacht> They say sunlight is the best disinfectant.
21:06 <ctweney> zacht:  any idea if we still use ".resource" in pooled content? 
eg /p/23d8Jaw.resource ?
21:07 <zacht> ctweney I'm not familiar with that one.
21:07 <zacht> ctweney Is that one of the selectors we're not currently testing?
21:07 <ctweney> not sure if we're testing it or not. it's impossible to grep 
for ".resource" because there are > 1000 hits on that
21:08 <ctweney> aha, we do have a test for it
21:08 <ctweney> page-create-test.rb
21:09 <ctweney> denbuzze:  do you know if we ever post to 
/p/#{poolId}.resource.html paths?
21:11 <denbuzze> ctweney: afaik we don't, do you see any POST's in the server 
logs?
21:13 <ctweney> zacht:  as I go along in my files testing I notice a lot of 
unused code and half-baked features
21:13 <ctweney> and some features that are completely stoned
21:13 <ctweney> I am making notes for their eventual elimination
21:13 <ctweney> the ".resource" fork appears to be one of these
21:13 <zacht> ctweney cool.
21:17 <ctweney> zacht:  I'm also taking the approach of writing all new tests 
instead of reusing / repackaging old ones. That is a lot slower, but helps me 
understand the code better.
21:19 <zacht> ctweney That's probably a good approach. I'm writing new tests 
for things that are currently uncovered, but if I get great momentum, I'll keep 
writing new tests, even for previously covered stuff.
22:07 <SakaiGithub> [sakai-widgetlibrary/master] OAEWDGT-186 - Split into 
different pages - Christian Vuerings
22:07 <SakaiGithub> [sakai-widgetlibrary/master] OAEWDGT-186 - Apply changes 
after code review @apex13 - Christian Vuerings
22:56 <SakaiGithub> [3akai-ux/1.5.0] SAKIII-5866 fixed release build for 
windows - jsloane
22:56 <SakaiGithub> [3akai-ux/1.5.0] Merge remote-tracking branch 
'jsloane/SAKIII-5866' into SAKIII-5866 - Christian Vuerings
23:05 <SakaiGithub> [3akai-ux] zathomas tagged 1.4.0-RC2 at 3782c07: 
http://git.io/Igf_jw
23:05 <SakaiGithub> [3akai-ux/1.4.0-RC2] SAKIII-6068 fix IE content upload - 
jsloane
23:05 <SakaiGithub> [3akai-ux/1.4.0-RC2] SAKIII-6068 fix IE content upload - 
jsloane
23:05 <SakaiGithub> [3akai-ux/1.4.0-RC2] SAKIII-6068 Fix version upload for IE 
- Bert Pareyn
23:40 <zacht> thecarlhall FYI https://github.com/sakaiproject/nakamura/pull/942 
should be good to go. Thanks for your help on this.
23:41 <thecarlhall> zacht++
23:54 <GitHub159> [nakamura] thecarlhall pushed 2 new commits to master: 
http://git.io/iXMsLQ
23:54 <GitHub159> [nakamura/master] KERN-3051 Fix execute flag on kern-3038.rb 
(breaks mvn -P integration) - Ray Davis
23:54 <GitHub159> [nakamura/master] Merge pull request #946 from 
raydavis/KERN-3051 - Carl Hall

_______________________________________________
oae-dev mailing list
[email protected]
http://collab.sakaiproject.org/mailman/listinfo/oae-dev

Reply via email to