Camelia, the Perl 6 bug

IRC log for #mojo, 2011-06-16

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

All times shown according to UTC.

Time Nick Message
00:38 mattastrophe joined #mojo
01:18 crab tonyc: why? because it's likely to be changed?
01:38 j3nnn1 joined #mojo
01:45 gshank joined #mojo
01:49 gshank_ left #mojo
01:53 Foxcool left #mojo
01:56 gshank_ joined #mojo
02:01 gshank left #mojo
02:16 Foxcool joined #mojo
02:33 bangbang joined #mojo
03:06 diegok left #mojo
03:06 diegok joined #mojo
03:08 MisterHatt left #mojo
03:08 perlrocks Twitter: "Tying agreement. [Mojolicious to deploy a web design] http://is.gd/DLqZ7Q" (ja) --onagatani http://twitter.com/onagatan​i/status/81196533248294912
03:09 diegok left #mojo
03:14 diegok joined #mojo
03:33 MisterHatt joined #mojo
03:42 jwang left #mojo
04:04 arthas left #mojo
04:28 AigleNoi1 left #mojo
04:34 AigleNoir joined #mojo
04:37 mmcleric joined #mojo
04:44 AigleNoir left #mojo
05:24 Kulag left #mojo
05:26 Kulag joined #mojo
05:42 crab sri: did you look at the -08 websockets draft? any code changes? (sorry, have been too busy to keep up)
05:45 sri crab: do you mean -09?
05:45 crab yes
05:45 sri yes, no changes at all
05:45 sri actual changes are very very unlikely
05:45 sri they are just changing the wording
05:46 sri for the rfc process
05:46 crab *nod*
05:47 crab i know how it goes, i have two RFCs ;-)
05:47 sri :)
06:01 GitHub37 joined #mojo
06:01 GitHub37 [mojo] kraih pushed 1 new commit to master: http://bit.ly/jCX6cP
06:01 GitHub37 [mojo/master] updated application embedding recipe - Sebastian Riedel
06:01 GitHub37 left #mojo
06:11 koban` joined #mojo
06:21 janus left #mojo
06:29 fhelmber_ joined #mojo
06:34 AmeliePoulain joined #mojo
06:52 GitHub103 joined #mojo
06:52 GitHub103 [mojo] kraih pushed 1 new commit to master: http://bit.ly/iOjCHy
06:52 GitHub103 [mojo/master] fixed typo - Sebastian Riedel
06:52 GitHub103 left #mojo
07:02 Alias left #mojo
07:08 d4rkie joined #mojo
07:14 perlrocks Twitter: "Drawing on the ZIP file into memory / Mojolicious Reference - Getting Started with Perl code samples http://bit.ly/l1j6Ec" (ja) --wPerlnews http://twitter.com/wPerlnew​s/status/81258217786720256
07:23 Foxcool left #mojo
07:40 Foxcool joined #mojo
07:57 Sugar joined #mojo
08:03 d4rkie left #mojo
08:15 spleenjack joined #mojo
08:18 marcus hmpf
08:18 marcus kinda sucks that the tt renderer is hard coded to compile templates to $home/tmp/ctpl
08:19 marcus anyone sent a patch to ask to default to tmpdir?
08:26 yko you?
08:26 purl you is behind the times bro
08:27 marcus yko: I will if there isn't one yet.
08:27 marcus yko: Just don't like to duplicate other people's efforts
08:29 yko at least there's no related issues
08:31 spleenjack left #mojo
08:34 cosmincx joined #mojo
08:36 spleenjack joined #mojo
08:53 tholen42 left #mojo
08:56 tholen joined #mojo
09:24 yakudzo left #mojo
09:33 Drossel joined #mojo
09:34 Kulag left #mojo
09:41 Drossel left #mojo
09:41 Kulag joined #mojo
09:47 Kulag left #mojo
09:52 Drossel joined #mojo
09:58 Sugar left #mojo
10:04 Sugar joined #mojo
10:31 j3nnn1 left #mojo
10:40 crab if i ->render_static in my action, then the after_static_dispatch hook will see a tx without a res->code, right? i.e. it'll be as if it weren't a static dispatch, even though it's the static dispatcher that gets invoked behind the scenes to render the file?
10:46 d4rkie joined #mojo
11:13 koban` left #mojo
11:14 koban joined #mojo
11:37 spleenjack left #mojo
11:38 spleenjack joined #mojo
11:52 j3nnn1 joined #mojo
12:00 koban left #mojo
12:11 d4rkie left #mojo
12:19 koban joined #mojo
12:33 Akron joined #mojo
13:16 spleenjack left #mojo
13:17 spleenjack joined #mojo
13:39 elb0w sri see anything wrong here? This is my app and my javascript for the web socket, I tried aurora but couldnt get onconnect to fire http://pastie.org/2077498
13:39 elb0w onopen sorry
13:42 spleenjack left #mojo
13:43 spleenjack joined #mojo
13:50 metaperl joined #mojo
14:11 AmeliePoulain left #mojo
14:12 sh9 joined #mojo
14:16 kaare joined #mojo
14:41 koban left #mojo
14:55 perlrocks Twitter: "@ Antonfin You can read a short report about web development in Perl (Mojolicious + Rose:: DB:: Object + PSGI / Plack + deploy on dotcloud)" (ru) --koorchik http://twitter.com/koorchi​k/status/81374230880595969
15:07 marty Mojotopia Lives!!!!   http://www.youtube.com/watch?v=_DVeRz5VuBI
15:09 amoore joined #mojo
15:09 perlrocks Twitter: "@ Koorchik Leave only Mojolicious. This is not a specialized conference on perl" (ru) --antonfin http://twitter.com/antonfi​n/status/81377790540251136
15:12 perlrocks Twitter: "@ Antonfin Well if there would be specialized in pearls, I would have talked only about Mojolicious. And then about the web development should be:)" (ru) --koorchik http://twitter.com/koorchi​k/status/81378551189880832
15:18 perlrocks Twitter: "@ Antonfin better surface and about everything, but with an emphasis on Mojolicious: take the framework + ORM, quickly write the draft and lay out two clicks to internet" (ru) --koorchik http://twitter.com/koorchi​k/status/81380146002661376
15:19 perlrocks Twitter: "@ Koorchik Mojolicious + PSGI / Plack - such as limits opportunities Mojolicious, as well as CGI, FastCGI, WebSocket" (ru) --antonfin http://twitter.com/antonfi​n/status/81380318250156032
15:22 elb0w take a shot of vodka
15:22 elb0w whoops
15:23 perlrocks Twitter: "@ Antonfin About Disabilities - did not understand. Mojolicious works fine in PSGI mode, and many use the Starman, and not the Hypnotoad." (ru) --koorchik http://twitter.com/koorchi​k/status/81381300254162944
15:24 baton8 left #mojo
15:26 perlrocks Twitter: "@ Antonfin "PSGI / Plack - such as limits opportunities Mojolicious" -, about a loss of functionality are you talking about?" (ru) --koorchik http://twitter.com/koorchi​k/status/81382059519651842
15:28 baton8 joined #mojo
15:33 perlrocks Twitter: "@ Antonfin Damn, found the problem. You just need to say that Mojolicious can operate in any mode:). But I still think about the topic." (ru) --koorchik http://twitter.com/koorchi​k/status/81384010139435008
15:35 trone_ joined #mojo
15:39 trone left #mojo
15:43 bangbang left #mojo
15:46 Akron left #mojo
15:46 Akron joined #mojo
16:12 sri elb0w: i told you how to test if websockets work... i'm not gonna look at your broken app...
16:13 sri crab: yes
16:14 xantus morn
16:14 xantus is now known as xantus[]
16:14 xantus[] ;)
16:15 * sri waves
16:16 xantus[] maybe websocket support could be a plugin
16:16 xantus[] but I think its just too integrated
16:17 elb0w sri, im using the example and 1 js function
16:17 xantus[] its probably not possbile to mount() a previous version of mojo
16:19 sri xantus[]: no
16:20 xantus[] elb0w: Your browser version probably uses an old websocket impl.  You would need to switch your mojo to 1.16
16:20 elb0w Im using Aurora
16:20 xantus[] I think thats the right version
16:20 sri xantus[]: i told him all that earlier
16:20 elb0w Also I have the most recent mojo install from cpan
16:20 sri and how to ensure it all works...
16:20 elb0w sri: but I did what you said
16:21 sri elb0w: then the test i gave you was successful?
16:21 xantus[] sri: yep, and I'
16:21 xantus[] sri: yep, and I'm repeating it :)
16:22 elb0w I see the GET request come
16:22 xantus[] FF6 uses websocket v7
16:22 elb0w hmm
16:22 sri FF6 uses ietf-08/09
16:22 sri the same as us
16:23 elb0w ff6 is the only one that gets this version?
16:23 sri yes
16:23 trone_ is now known as trone
16:23 xantus[] ahhhh, I'm looking at the bugzilla for ff6. # 640003
16:23 sri the test agan.... run "mojo daemon" and connect with aurora to "http://127.0.0.1:3000/diag/websocket"
16:23 elb0w yeah I did that, I see the response
16:24 sri then websockets work
16:24 elb0w when I tried just going to websocket '/echo' ... I didnt
16:24 xantus[] lol
16:24 * xantus[] * sleep mode activated
16:26 * sri updates aurora
16:27 sri elb0w: did you really get the message that websockets work?
16:27 elb0w yes
16:28 elb0w when I went to the diag
16:28 sri it seems broken in the latest aurora version, like not working at all
16:28 elb0w the diag is the only time ive gotten any feedback
16:28 elb0w when I try with just the example websocket I dont get anything, unless I use chrome
16:28 xantus[] downloading ff6
16:28 elb0w sri, http://pastie.org/2078209
16:29 sri elb0w: i don't know what you mean with 'just the example websocket'
16:29 trone_ joined #mojo
16:30 elb0w if I do,  websocket '/echo' => ... then try from js to connect to it using new WebSocket('ws://localhost:3000/echo')... I dont get any response in FF
16:30 elb0w however when I went to diag/websocket I saw that it made a request
16:31 elb0w But if I try in chrome, the javascript shows me the request
16:31 elb0w (In the myapp.pl output that is)
16:34 trone left #mojo
16:35 elb0w sri: is this just me not understanding websockets well enough? Sorry to be such a bother, this is all new to me
16:36 xantus[] whats the url for the code you'
16:36 xantus[] are running
16:36 sri interesting, in the latest aurora version you need to prefix "new WebSocket"
16:36 sri "new MozWebSocket"
16:36 xantus[] are you effing kidding?
16:36 sri sadly not
16:37 sri they must have changed it a few days ago
16:37 sri i think they will release FF6 like that :S
16:37 elb0w http://pastie.org/2078239
16:38 xantus[] sri: https://bugzilla.mozilla.o​rg/show_bug.cgi?id=659324
16:40 Akron left #mojo
16:41 elb0w http://replay.isit.nu/item?id=2600711
16:45 xantus[] it'll be changed back after release
16:47 kaare left #mojo
16:50 sri yea, until then we have a little more boilerplate
16:50 GitHub62 joined #mojo
16:50 GitHub62 [mojo] kraih pushed 1 new commit to master: http://bit.ly/mA7phJ
16:50 GitHub62 [mojo/master] updated WebSocket diagnostics test in Mojo::HelloWorld for latest Firefox Aurora - Sebastian Riedel
16:50 GitHub62 left #mojo
16:51 sri ok, that test should pass with latest aurora
17:01 janus joined #mojo
17:07 alnewkirk joined #mojo
17:37 spleenjack left #mojo
17:41 Sugar left #mojo
17:46 perlrocks Twitter: "[Mojolicious] Some people I'm in China I Tsukatsu" (ja) --trinityCRZ http://twitter.com/trinityC​RZ/status/81417342193631232
18:01 sri ALL GLORY TO THE HYPNOTOAD!
18:03 perlrocks Twitter: "www/p5-Mojolicious-Plugin-Database - 1.05: - Update to 1.05 - Add LICENSE  PR:             ports/157920 Submitte... http://bit.ly/kQ4nJB" --freshports_org http://twitter.com/freshports​_org/status/81421690424397824
18:04 Foxcool left #mojo
18:07 Foxcool joined #mojo
18:13 jwang joined #mojo
18:24 marty ALL GLORY TO THE HYPNOTOAD!
18:27 Foxcool left #mojo
18:29 Foxcool joined #mojo
18:34 perlrocks Twitter: "www/p5-Mojolicious-Plugin-Mongodb - 1.07: - Update to 1.07 - Add LICENSE  PR:             ports/157921 Submitted... http://bit.ly/jSSOSn" --freshports_org http://twitter.com/freshports​_org/status/81429445419864064
18:46 Foxcool left #mojo
18:46 Foxcool joined #mojo
18:49 Foxcool left #mojo
18:50 Foxcool joined #mojo
18:51 Foxcool left #mojo
18:51 Foxcool joined #mojo
18:52 sh9 left #mojo
19:02 fhelmber_ left #mojo
19:44 SmokeMachine joined #mojo
19:53 SmokeMachine left #mojo
19:58 sri interesting
19:58 sri mount + --reload = segfault :o
20:09 sri hmm
20:10 sri even more interesting, it works fine in mounted apps
20:22 sri huh
20:22 sri now it suddenly works
20:22 * sri is very confused
20:45 amoore left #mojo
20:45 * tempire sprinkles brain glitter on sri
20:49 sri apprears to be related to DATA
20:49 tempire he's dead
20:49 sri segfaults reliably when i render something from the DATA section in a mounted app :S
20:50 dekimsey1 left #mojo
21:10 sri hmm
21:10 sri actually seems to be a template segfaulting Oo
21:11 sri oooooh
21:11 sri it's just the built in 404
21:20 alnewkirk left #mojo
21:45 sri ok, i don't get it :(
21:46 sri https://github.com/kraih/mojo/blob/master/​lib/Mojolicious/Plugin/TagHelpers.pm#L200
21:46 sri that line causes the segfault
21:47 sri i could really use some help
21:47 sri otherwise i'll prolly end up ripping out the whole reload code
21:53 sri for some reason i keep finding new ways to segfault perl :(
22:06 MisterHatt sri: segfauult golf indeed
22:06 sri i have absolutely no idea what's wrong there
22:06 sri just tried replicating the issue with a helper in a normal app, and it works flawless
22:07 sri something about that line i linked must be special
22:09 sri what the hell....
22:09 purl dot dot dot *yawn* dot
22:09 sri the stylesheet helper even works for DATA templates
22:10 sri it actually works so well it's scary
22:25 sri ok, i don't get it...
22:25 sri i've pasted the exact same template into a new mounted app and it just works
22:35 chansen any difference if you do assign the result of $old->() to a scalar and then do the concatenating?
22:36 sri no difference at all
22:36 sri but i can replicate the exact scenario now
22:37 sri it seems the segfault only triggers in the main app, not mounted ones
22:37 sri even if both are rendering the same template
22:37 chansen tried more than one version of perl?
22:38 sri yes, all the way back to 5.8.9
22:38 chansen ok, guess you need to produce a minimal case and report to p5p
22:39 sri no wait
22:39 sri 5.8.9 is not affected!
22:39 sri it's 5.10.0+
22:39 sri hah, i wish i could turn it into a minimal case :)
22:51 metaperl left #mojo
22:53 chansen perl -D might reveal some info (assuming you have an interpreter compiled with -D DEBUGGING)
23:08 j3nnn1 left #mojo
23:10 memememomo left #mojo

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