Camelia, the Perl 6 bug

IRC log for #mojo, 2012-11-22

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

All times shown according to UTC.

Time Nick Message
00:04 mattastrophe joined #mojo
00:30 good_news_everyone joined #mojo
00:30 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/eSFNsg
00:30 good_news_everyone mojo/master bfe7584 Sebastian Riedel: added continue event to Mojo::Transaction::HTTP
00:30 good_news_everyone left #mojo
00:33 sri it's a bit on the crazy side, but makes testing easier and you can redecorate the new response object after receiving the 100 Continue
00:34 sri we have to support 100 continue client side anyway... might as well make it somewhat useful
00:39 sri hmmm
00:40 * sri wonders if we could just make Mojo::UserAgent remove Expect headers if they contain 100-Continue and rip out all 100 Continue code
00:42 sri it didn't work client side and is not required server side, so we don't break anything
00:50 egopro joined #mojo
01:02 xaka joined #mojo
01:04 good_news_everyone joined #mojo
01:04 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/ZPB0TA
01:04 good_news_everyone mojo/master 842814f Sebastian Riedel: improved Mojo::UserAgent to remove Expect headers if they are set to "100-Continue"
01:04 good_news_everyone left #mojo
01:04 sri marcus, tempire, crab: i'd like to hear your opinions
01:05 laouji joined #mojo
01:05 sri no browser implements 100 Continue
01:05 sri a correct client side implementation waits a moment for a 100 Continue response, and then sends the request body anyway
01:06 sri so it shouldn't matter for the few clients that actually get it right
01:06 SmokeMac_ joined #mojo
01:11 hlin joined #mojo
01:17 Mike-PerlRecruiter_ joined #mojo
01:20 d4rkie joined #mojo
01:27 tl joined #mojo
01:41 egopro joined #mojo
01:43 egopro_ joined #mojo
02:38 good_news_everyone joined #mojo
02:38 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/PeVHUg
02:38 good_news_everyone mojo/master 9a977da Sebastian Riedel: fixed Mojo::UserAgent to ignore unexpected 1xx responses
02:38 good_news_everyone left #mojo
02:38 sri ok, that should be the most correct solution :)
02:45 dog_chocolate joined #mojo
02:52 Foxcool joined #mojo
03:15 good_news_everyone joined #mojo
03:15 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/SVN06A
03:15 good_news_everyone mojo/master 9436eef Sebastian Riedel: modernized application tests
03:15 good_news_everyone left #mojo
03:19 noganex joined #mojo
03:38 rem_lex| joined #mojo
03:40 egopro joined #mojo
03:41 egopro_ joined #mojo
03:52 good_news_everyone joined #mojo
03:52 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/TIQCRA
03:52 good_news_everyone mojo/master 50329f3 Sebastian Riedel: small optimizations
03:52 good_news_everyone left #mojo
03:57 egopro joined #mojo
04:03 egopro joined #mojo
04:54 good_news_everyone joined #mojo
04:54 good_news_everyone [mojo] kraih pushed 2 new commits to master: http://git.io/25lkbA
04:54 good_news_everyone mojo/master af62a06 Sebastian Riedel: test tweaks
04:54 good_news_everyone mojo/master efee8c9 Sebastian Riedel: documentation tweaks
04:54 good_news_everyone left #mojo
05:03 davido joined #mojo
05:07 good_news_everyone joined #mojo
05:07 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/Ls3rJg
05:07 good_news_everyone mojo/master 64cd88d Sebastian Riedel: modernized a few more tests
05:07 good_news_everyone left #mojo
05:08 * davido looks around for signs of life.
05:08 * sri hides
05:45 aptituz joined #mojo
05:46 ovnimancer joined #mojo
06:31 kitt_vl joined #mojo
06:34 spleenjack joined #mojo
06:40 dpetrov_ joined #mojo
07:00 avkhozov joined #mojo
07:16 Vandal joined #mojo
07:20 ladnaV joined #mojo
07:22 HtbaaPi joined #mojo
07:39 Vandal joined #mojo
07:48 yakudza joined #mojo
08:10 dod joined #mojo
08:19 Vandal joined #mojo
08:26 ladnaV joined #mojo
08:27 dod joined #mojo
08:39 dod joined #mojo
08:42 ver joined #mojo
08:48 batman joined #mojo
08:55 alexeiras joined #mojo
09:07 D4RK-PH0ENiX joined #mojo
09:17 memowe joined #mojo
09:18 fhelmber_ joined #mojo
09:38 lukep joined #mojo
09:41 dotan joined #mojo
09:42 cosimo joined #mojo
09:45 Britzel joined #mojo
09:49 egopro joined #mojo
09:50 dotan joined #mojo
10:01 d4rkie joined #mojo
10:05 Averna joined #mojo
10:20 egopro joined #mojo
10:23 nic I'm slightly unnerved that our resident evil genius seems to never sleep
10:24 nic Maybe that's why your mother told you not to befriend any evil geniuses
10:25 nic I have a few modules almost ready to release
10:25 nic they have almost nothing to do with web apps (directly) but they rely on having Mojo installed
10:26 nic and IMO they would benefit greatly from having a Lite app included as an example illustrating the functionality
10:26 nic one example: a syntax untangler for MySQL SQL
10:27 nic I'd like to include a Lite app that lets people try it out before coding against it
10:27 nic the question is, I really don't know where I would put such add-on examples
10:27 nic it's kind of documentation, but that you execute
10:27 nic or maybe people feel it's bad practice to bundle add-ons like that?
10:29 nic Maybe the pod should just link to a wiki page that let's them download it if they want it
10:42 crab do you have a resident (evil genius) or a (resident evil) genius?
10:45 spleenjack joined #mojo
10:46 dabudabu joined #mojo
10:55 nic or is it just a badly worded biology text, and we only have a resident weevil genus?
11:11 cosimo joined #mojo
11:24 alexeiras joined #mojo
11:26 egopro joined #mojo
12:09 mjy joined #mojo
12:09 mjy hi, am I doing something wrong or does Mojo::UserAgent actually connect to port 80 when given an FTP URL? :-/
12:13 yakudza mjy, yes. U use HTTP client on a FTP server
12:13 yakudza Mojo::UserAgent - Non-blocking I/O HTTP and WebSocket user agent
12:13 mjy well mojo uses URLs and not hostnames, so it should parse them and spit out an error message if it can't handle the protocol?
12:17 mjy i.e. my $tx = $ua->get('ftp://foo.com') should not do an HTTP request imo
12:19 trone question: is it possibile increase the flash size limit? By default it is 4096 byte.
12:23 yakudza mjy, imo U r right
12:23 trone I thought to use a different way to storage sessions (MojoX::Session::Store::Libmemcached) but flash size isn't increased (maybe it use another mechanism)
12:52 ObseLeTe joined #mojo
12:57 d4rkie joined #mojo
13:11 jontaylor joined #mojo
13:17 Mike-PerlRecruiter_ joined #mojo
13:23 jontaylor My app does a lot of webservice calls so I made it so that the app object has the webservice object I use
13:23 jontaylor but then later I decided I wanted to boost performance by adding some caching, but I only wanted the caching per request
13:23 jontaylor so I need to change the scope of my webservice object from being on the $app, to being per request somehow
13:24 jontaylor I know I can do a before_dispatch, to set it up, before each request, but I don't know where I should put the object
13:24 jontaylor does the stash seems like a sensible choice ?
13:27 egopro joined #mojo
13:53 sri o/
13:54 trone \o/
13:55 Foxcool joined #mojo
13:57 sh4 joined #mojo
14:21 mire__ joined #mojo
14:28 egopro joined #mojo
14:28 ryozi joined #mojo
14:30 trone I would manage session in memcached. Some hint/plugin?
14:42 jontaylor trone: http://search.cpan.org/~jontaylor/Moj​olicious-Plugin-ServerSession-0.01/li​b/Mojolicious/Plugin/ServerSession.pm
14:50 trone jontaylor: thanks. It seems really close I need. :)
14:50 jontaylor you could probably make it a ton better
14:50 jontaylor fork it on github :)
15:15 trone I'm missing something. Session keep arbitrary data (they are in process memory), but flash is managed in cookie so it's limited to 4096B. Reading http://mojolicio.us/perldoc/​Mojolicious/Controller#flash seem flash goes in session … I'm confused
15:15 jontaylor In mojolicious the session is stored in an encrypted cookie
15:16 jontaylor so the flash is in the session, which is stored in the cookie, which has a 4096 limit
15:17 jontaylor Its exactly this reason why I made that plugin, but I didn't want to mess with the internals of mojo too much, so I just made a plugin that gives you an additional session method
15:18 trone jontaylor: ok. Then I don't understand why if I write $self->flash( key => 'a' x 14000 ) mojolicious log (rightly) «[error] Cookie "mojolicious" is bigger than 4096 bytes.» but if I write $self->session( key => 'a' x 14000 ) this don't give me [error] line...
15:19 trone ok, I lie.
15:19 trone Also $self->session spam [error] line. sorry.
15:19 jontaylor no problem
15:19 trone So, all seem right! :D
15:20 trone Actually, I have to speak with my coworkers because I'm not sure that we want "a parallel session" system.
15:22 jontaylor and they'd be right, the better solution would to replace the existing session instead of adding an extra one
15:22 jontaylor If you write a plugin that does it well, i'll adopt your solution over mine
15:23 trone :D Yep, I would write it, but I'm not sure we have the time
15:23 trone … :-( we'll see.
15:23 jontaylor and thats the problem I had :)
15:24 trone damnit! :D
15:24 trone jontaylor: thanks a lot for chatting. I need it.
15:25 jontaylor no worries
15:29 egopro joined #mojo
15:50 marty_ joined #mojo
16:10 mattastrophe joined #mojo
16:22 nicomen how would I test that a get redirects to the root url? $t->get_ok->status_is('302')->headers->location​_is_root_of_service_or_matches_a_named_route()
16:30 egopro joined #mojo
16:34 d4rkie joined #mojo
16:36 cosimo nicomen: $t->get_ok->status_is(302)->​header_is(Location=>'blah')
16:37 cosimo nicomen: assuming you know the 'blah' yourself
16:41 mire__ joined #mojo
16:46 nicomen the thing is, I want to not have to know the blah myself ;)
16:47 nicomen but thanks, using ugly qr{(?gmx) \A https?://[^/]+/ \z} now...
16:50 good_news_everyone joined #mojo
16:50 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/qcN3ig
16:50 good_news_everyone mojo/master 6112d18 Sebastian Riedel: test tweaks
16:50 good_news_everyone left #mojo
17:29 sinkovsky joined #mojo
17:43 mjy is there a saner way to work with Mojo::UserAgent and removing the 5MB(?) Mojo::Message limit than setting the environment variable?
17:47 alexeiras joined #mojo
17:51 mjy (I'd rather set it only for one UserAgent instance)
17:56 dod joined #mojo
17:58 dotan joined #mojo
18:00 dotan joined #mojo
18:03 inokenty joined #mojo
18:08 geira joined #mojo
18:11 davido joined #mojo
18:16 geira joined #mojo
18:20 geira joined #mojo
18:31 egopro joined #mojo
19:05 inokenty joined #mojo
19:11 ObseLeTe joined #mojo
19:17 SmokeMachine joined #mojo
19:39 d4rkie joined #mojo
19:40 D4RK-PH0ENiX joined #mojo
19:47 GabrielVieira joined #mojo
19:55 sinkovsky joined #mojo
20:03 d4rkie joined #mojo
20:04 d4rkie joined #mojo
20:17 jpn joined #mojo
20:32 egopro joined #mojo
20:42 sinkovsky joined #mojo
20:43 dpetrov_ joined #mojo
21:11 lukep joined #mojo
21:12 ki0 joined #mojo
21:16 sinkovsky joined #mojo
21:32 egopro joined #mojo
21:46 mattastrophe joined #mojo
21:57 sinkovsky joined #mojo
22:01 SmokeMac_ joined #mojo
22:05 geira joined #mojo
22:14 SmokeMac_ joined #mojo
22:14 jpn left #mojo
22:18 geira joined #mojo
22:24 trone_ joined #mojo
22:39 SmokeMac_ joined #mojo
22:49 janus hrm, not really clean, but monkey-patching Mojo::Message::Request at least does not poke into internals to get rid of content-length zero ;)
22:56 janus how about having an event emitted to do the same?
22:57 memowe joined #mojo
22:58 geira joined #mojo
23:02 SmokeMac_ joined #mojo
23:04 jzawodn joined #mojo
23:05 janus FTR: http://nopaste.info/3178467553.html
23:06 janus uhm, backslash got lost in the first line
23:10 sri janus: what's the use case for such an event?
23:10 janus broken http servers that barf on content-length: 0
23:10 mjy left #mojo
23:10 janus ... on get requests
23:10 sri that would be seriously broken
23:10 sri example?
23:11 mjy joined #mojo
23:11 sri how common is that server?
23:11 janus ah, event purpose you mean - well, just for custom fixups i guess
23:11 janus let me see...
23:12 sri if there is no good use case it's not gonna happen
23:12 janus indeed, and i can live with that monkey since its not that ugly
23:13 janus lighttpd in this case - not sure if it applies generally... checking
23:16 sri works for me
23:16 sri mojo get -v www.chefkoch.de
23:17 janus yeah, quick local setup tells the same
23:17 janus good :)
23:18 sri i'm all for adding more events... but this doesn't seem worth it
23:18 janus i agree
23:21 heytrav Question regarding hypnotoad and logging. I have an application that I'm cpanifying, deploying locally with cpanm, and running with hypnotoad. Where should the logfile be in this setup?
23:41 heytrav should this be set using: $self->home->rel_dir('...') logic like for the "public" and "templates" directories?
23:52 sri i wouldn't mind making Mojo::UserAgent a little smarter and saving a few bytes every request
23:52 sri http://pastie.org/5420795 # here's a patch
23:52 * sri pokes janus, marcus, tempire and crab
23:52 sri should that be done or not?
23:53 andrefs joined #mojo
23:53 sri it's 3 lines less code, so i would be pretty much ok with it, unless someone sees a flaw
23:55 dvinciguerra joined #mojo
23:58 Averna joined #mojo
23:59 sri and of course saves a few trees :)

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