Camelia, the Perl 6 bug

IRC log for #mojo, 2012-02-25

| Channels | #mojo index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:03 sri vervain: they are not covered by the deprecation policy, that's the point
00:03 sri templates, javascript and images just change
00:04 sri even more so than command line flags
00:05 sri how would a deprecation policy on design even work? :)
00:08 vervain I just stripped this from the main source http://pastie.org/3451169... with a little bit of sanitation (class config and what not) it would make a good helper (IMO)... Maybe the best way for me to present my idea is to present the plugin. :-)
00:10 mercutioviz joined #mojo
00:11 black joined #mojo
00:28 jamesw joined #mojo
00:56 sri oh noes
00:57 sri it was actually a bug in hypnotoad
00:57 sri yay for more log messages i guess
00:58 diegok I've updated and now my app refuses to set helper 'config' with my own object. It says "Helper "config" already exists, replacing." but then it's undef... :-(
01:08 diegok ^ just this fails: https://gist.github.com/1905030
01:08 Alias_ joined #mojo
01:08 diegok any idea? :)
01:16 GitHub133 joined #mojo
01:16 GitHub133 [mojo] kraih pushed 1 new commit to master: http://git.io/Sp9igA
01:16 GitHub133 [mojo/master] fixed Hypnotoad upgrade timeout bug - Sebastian Riedel
01:16 GitHub133 left #mojo
01:32 kitt_vl1 joined #mojo
01:38 GitHub87 joined #mojo
01:38 GitHub87 [mojo] kraih pushed 1 new commit to master: http://git.io/kaoyZA
01:38 GitHub87 [mojo/master] tweaked Hypnotoad log messages a little - Sebastian Riedel
01:38 GitHub87 left #mojo
01:40 sri i think we should consider another bugfix release, that hypnotoad bug can be nasty
01:40 sri you should test everything first though
01:41 vervain Is there an example of testing for an element that exists after a login redirect via flash?
01:43 jb360 joined #mojo
01:49 vervain Ah... the error was earlier in the test... works as expected now.
01:54 crab a login redirect... via flash?
01:54 vervain Bad word order... I thought that right after I typed it...
01:54 vervain May ignore the flash all together.
01:54 vervain :-)
01:55 * crab imagines vervain as flash gordon
01:55 vervain What I saw made me think it was looking for the element in the 'redirect response' rather than the final target... But I was mistaken.
01:56 vervain My years of going out to raves covered in gold paint are long over. :)
01:57 vervain Only kidding... it was Danial that liked gold paint... I pretty much stuck to purple. :-)
02:07 predgtr1 joined #mojo
02:52 kitt_vl joined #mojo
02:56 GitHub189 joined #mojo
02:56 GitHub189 [mojo] kraih pushed 1 new commit to master: http://git.io/KJboqA
02:56 GitHub189 [mojo/master] improved resilience of Mojo::IOLoop::Stream - Sebastian Riedel
02:56 GitHub189 left #mojo
03:01 kitt_vl joined #mojo
03:39 athenot joined #mojo
03:51 Elraraniel joined #mojo
03:58 jnap joined #mojo
04:05 arpadszasz joined #mojo
04:31 Kulag joined #mojo
04:33 tempire ./script/hypnotoad -f examples/hello.pl works fine for me
04:33 tempire what else should I do to test?
04:34 sri zero downtime upgrades
04:35 sri also if you like the new log messages
04:35 sri kill a worker too
04:36 sri i've made special info messages for when a worker dies without reason and when it doesn't have a heartbeat
04:38 * sri just activated filevault2 for his ssd
04:38 sri took 30 minutes to encrypt everything, you can work normally during the whole time
04:38 sri no performance difference noticeable
04:39 sri now activating it for the time machine disk... that will take hours to encrypt as far as i can see :S
04:40 sri 3% after 15 minutes
04:45 Kulag joined #mojo
04:48 sri oh damn, i made a mistake
04:48 tempire for failvault or hypnotoad
04:48 sri hypno
04:49 sri there is an expected unexpected case where a worker dies
04:49 tempire I like the log messages.  it was disconcerting before, not seeing immediate feedback on what was going on
04:49 tempire now I feel connected to the hypnotoad
04:52 tempire hot reload works
04:52 tempire logging works
04:52 GitHub164 joined #mojo
04:52 GitHub164 [mojo] kraih pushed 1 new commit to master: http://git.io/nHXVoA
04:52 GitHub164 [mojo/master] tweaked Hypnotoad log messages some more - Sebastian Riedel
04:52 GitHub164 left #mojo
04:52 sri i suppose this should be debug if it's not special
04:55 sri i kinda want it to be info or even warn, but dead workers are expected and kinda spammy
05:13 sri the upgrade timeout fix is kinda important, so i vote for a quick release
05:15 metys joined #mojo
05:15 cosimo joined #mojo
05:25 vel joined #mojo
05:37 crab ok
05:39 inokenty joined #mojo
05:47 crab cpanify once more gives me lots of timeouts.
05:49 crab and indeed, the timeout appears absurdly short. investigating.
05:51 crab oh. the default connect timeout is 3s(!)
05:52 Elraraniel left #mojo
05:53 GitHub190 joined #mojo
05:53 GitHub190 [mojo] amenonsen pushed 1 new commit to master: http://git.io/CbcCfQ
05:53 GitHub190 [mojo/master] Bump VERSION after release - Abhijit Menon-Sen
05:53 GitHub190 left #mojo
06:05 poundyz joined #mojo
06:27 Foxcool joined #mojo
07:13 preflex_ joined #mojo
07:18 kaare joined #mojo
08:08 noganex joined #mojo
08:35 tempire are there any other json modules supporting pointers?
09:18 zpmorgan joined #mojo
09:24 jamesw joined #mojo
09:25 tempire huh.
09:25 tempire I've never attempted to test a lite app.
09:31 tempire ah.
09:31 tempire not so hard.
09:37 Alias_ joined #mojo
09:38 tempire sri: I'd like to set HARNESS_OPTIONS=c in the test command
09:40 tempire also, I'd like to add a generate test command
09:41 tempire would streamline the testing explanation in the mojocast.
09:41 tempire and the code impact is minimal.
09:42 inokenty joined #mojo
09:50 crab hi tempire.
09:50 tempire hi crab
10:09 black joined #mojo
10:13 black joined #mojo
10:30 Kulag joined #mojo
10:39 Kulag joined #mojo
11:05 diegok morning!
11:05 purl Mornings are great. Every time you experience morning, you're not dead yet!
11:06 diegok purl: I love you ;-)
11:06 purl diegok: sorry...
11:10 diegok Any idea how can I fix this: https://gist.github.com/1905030 ?
11:10 diegok I need to overwrite the config helper...
11:33 vervain You have defined a helper... but I don't think thats the same as $self->config, does $self->config still look like a hash?
11:34 vervain If you want $self->config to be your config then maybe you need to do something like $self->config = My::Config->new
11:35 Kulag joined #mojo
11:37 vervain caveat... I'm still learning too... maybe helpers do get defined as $self->foo?
11:39 diegok vervain: $self->config = xxx will never work
11:41 diegok vervain: problem is that config helper is now "official" and it warns about being overwritten
11:41 vervain Gotcha.
11:41 diegok vervain: but still, it's undef then :(
11:41 diegok vervain: It was working just like this before upgrade.
11:42 vervain Ah... I have a working example... let me see if an upgrade breaks it.
11:54 vervain diegok: Is this example missing anything that you need? http://pastie.org/3454662
11:54 vervain ^ That works on 2.53
11:57 diegok vervain: hm... strange. I should isolate my problem. It looks almost the same (I'm not using lite but the idea is the same)
11:57 vervain I'm just sketching up a non-lite version of same.
11:58 diegok vervain: I should probably move away as hypnotoad now reads it's config from ->config hash...
11:58 diegok vervain: but still I would like to see what I'm doing wrong here...
11:58 vervain Ya... I was wondering if there would be less pain to rename your helper my_config or something. :-)
11:59 vervain I hear that... I love figuring out subtitles... it's where the most learning is. :-)
11:59 diegok vervain: I've already renamed. But will think about it. I prefere to have only one big config for all my stuff
12:00 diegok vervain: this app is part of some more tools and some more apps all reading same config
12:00 diegok vervain: config came from db at deploy time
12:01 diegok and could be different en every box
12:02 diegok it reads a json and adds some methods over what it reads.
12:06 vervain diegok: Do you know what version it last worked on for you?
12:07 diegok vervain: 2.47 for sure, but I've jumped directly to 2.52
12:09 diegok vervain: where you able to reproduce?
12:09 vervain Yes... on 2.48 as well... I can go back further but I want to double check something first.
12:18 vervain I made this example: http://pastie.org/3454792
12:19 sri crab++
12:20 vervain diegok: doesn't work in 2.48... should I go back further or can you spot a reason why the example isn't valid?
12:20 sri crab: 3s is absurdly short?
12:21 vervain diegok: I just discovered that it's not convenient for me to go back further... can you test that example on your 2.47?
12:22 diegok vervain: thank you. I just also created the example app and I'm looking at this. :)
12:22 diegok ^sure
12:39 diegok vervain: it works on .46 ~:-o
12:40 vervain diegok: It may create pieces to clean up... but when I said somthing like $self->config = it would seem that I meant $self->{config} = $config;  This works: http://pastie.org/3454896
12:41 diegok vervain: I'll se why this got broken but then I'll move away of doing this because don't want to break further things depending on config being config :)
12:51 diegok vervain: 2.47 added experimental config and 2.52 "Removed experimental status from config method in Mojo" so better to do it different on my side :-)
12:52 GitHub144 joined #mojo
12:52 GitHub144 [mojo] kraih pushed 1 new commit to master: http://git.io/qCuvcg
12:52 GitHub144 [mojo/master] small documentation tweaks - Sebastian Riedel
12:52 GitHub144 left #mojo
12:54 black joined #mojo
13:02 GitHub2 joined #mojo
13:02 GitHub2 [mojo] kraih pushed 1 new commit to master: http://git.io/QIQgAg
13:02 GitHub2 [mojo/master] fixed small rendering bug in Mojolicious::Plugin::PODRenderer - Sebastian Riedel
13:02 GitHub2 left #mojo
13:17 tempire hmm.  maybe the generate test doesn't make much sense.  but the HARNESS_OPTIONS=c definitely does.
13:20 tempire maybe generate lite_test
13:20 sri tempire: i'm ok with setting c
13:21 tempire and lite_test?
13:22 sri not with that
13:22 tempire making noobs figure out and type in http://search.cpan.org/~ams/Mojoliciou​s-2.53/lib/Mojolicious/Lite.pm#Testing seems like a lot
13:23 crab sri: on my connection it seems to timeout without connecting to something about 50% on average (but much more often than that for pause).
13:23 GitHub147 joined #mojo
13:23 GitHub147 [mojo] kraih pushed 1 new commit to master: http://git.io/ffCiVQ
13:23 GitHub147 [mojo/master] improved test command to colorize by default - Sebastian Riedel
13:23 GitHub147 left #mojo
13:23 tempire then again, creating a t directory doesn't seem good either.
13:24 sri tempire: after our recent failure with the gitignore generator i'd rather be more conservative and only add new ones when not having them becomes an actual problem :S
13:24 tempire I know.  I'm just talking at this point.
13:25 sri nobody stops you from making a plugin though :)
13:25 tempire yeah, was thinking about that too.
13:26 sri crab: hmm, what do you propose?
13:27 sri increase it to 10 seconds?
13:31 vel joined #mojo
13:34 GitHub62 joined #mojo
13:34 GitHub62 [mojo] kraih pushed 1 new commit to master: http://git.io/vhPjOA
13:34 GitHub62 [mojo/master] increased default connect timeout from 3 to 10 seconds in Mojo::UserAgent and Mojo::IOLoop::Client - Sebastian Riedel
13:34 GitHub62 left #mojo
13:34 sri lets see how that works
13:38 sri hmm
13:38 sri or was i too quick with that one?
14:04 * sri shrugs
14:12 batman joined #mojo
14:22 fibo joined #mojo
14:45 d4rkie joined #mojo
14:55 Foxcool joined #mojo
15:22 crab sri->patch_timeout(3600)
15:22 * sri waits
15:22 memowe o/
15:22 sri \o
15:24 inokenty joined #mojo
15:33 sri damn
15:33 sri encrypting my backup drive took 10 hours
15:36 ZadYree wow
15:37 sri but it's pretty neat once it's all done
15:37 sri entirely transparent and not noticeable
15:37 memowe sounds great
15:38 sri you enter a password right after pushing the power button (before os x boots) and that's it
15:40 sri and of course every time you wake your laptop up from standby
15:40 crab before os x boots?
15:40 sri yes
15:41 sri it's the normal password widget though
15:41 sri think it works with the recovery partition
15:41 sri which is unencrypted
15:46 sri also love the way they did the conversion process
15:47 sri everything just happens in the background, and you can normally work while it converts one block after another
15:48 sri 20 minutes for the 120mb ssd and 10 hours for the 1tb backup usb hd
15:51 sri hope marcus won't get mad if i add 4 new env vars :o
15:51 * sri is considering MOJO_CONNECT_TIMEOUT, MOJO_INACTIVITY_TIMEOUT, MOJO_WEBSOCKET_TIMEOUT and MOJO_REQUEST_TIMEOUT
16:28 * sri pokes crab, tempire and marcus
16:33 vel joined #mojo
16:34 GitHub190 joined #mojo
16:34 GitHub190 [mojo] kraih pushed 1 new commit to master: http://git.io/qKpYGQ
16:34 GitHub190 [mojo/master] added support for MOJO_CONNECT_TIMEOUT, MOJO_INACTIVITY_TIMEOUT, MOJO_REQUEST_TIMEOUT and MOJO_WEBSOCKET_TIMEOUT environment variables - Sebastian Riedel
16:34 GitHub190 left #mojo
16:34 sri i think they fit in and are quite useful
17:03 crab sri: no objections, seems good
17:11 fibo hi marcus, are you aka oleber ?
17:12 crab haha
17:12 crab no. oleber is aka oleber.
17:12 crab but he's not been on #mojo for a very long time.
17:12 crab marcus is innocent.
17:13 GitHub163 joined #mojo
17:13 GitHub163 [mojo] kraih pushed 1 new commit to master: http://git.io/es7bIQ
17:13 GitHub163 [mojo/master] fixed timeout test - Sebastian Riedel
17:13 GitHub163 left #mojo
17:14 * sri blames marcus
17:17 crab marcus is only innocent of being oleber. he's steeped in guilt otherwise.
17:17 sri :)
17:19 marcus crab: guilty as charged.
17:47 vervain tempire: MojoExample has Text::Database which exists in the wild.
17:48 vervain Test::Database
17:48 purl somebody said Test::Database was configure once, test everything
18:13 inokenty joined #mojo
19:05 xaka joined #mojo
20:05 marcus http://annevankesteren.nl/​2012/02/standards-red-pill
20:05 marcus Anne++
20:21 preflex_ joined #mojo
20:46 jnap joined #mojo
20:52 abra joined #mojo
21:26 tempire vervain: yes, I'm not worried about it, though.  Since it's just an example, I don't see a reason for it to interfere.
21:28 mire joined #mojo
21:53 bpmedley joined #mojo
21:54 xaka joined #mojo
21:59 perlite joined #mojo
22:01 Mad_Dud joined #mojo
22:26 vel joined #mojo
23:18 noganex_ joined #mojo
23:26 vel joined #mojo
23:36 noganex joined #mojo

| Channels | #mojo index | Today | | Search | Google Search | Plain-Text | summary