The web in a box - a next generation web framework for the Perl programming language

IRC log for #mojo, 2014-12-12

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

All times shown according to UTC.

Time Nick Message
00:06 sri i think AnyEvent::DNS and IO::Async::Resolver need to fit into the api as well
00:07 sri both should be valid options with the EV backend
00:07 sri which means no globals for switching resolvers
00:08 Grinnz_ POE::Component::Client::DNS! /runs
00:08 sri also a valid choice
00:08 Grinnz_ (the only one of these i've actually used before)
00:09 sri POE should work through the EV backend too
00:12 sri mst: i guess you could have mentioned cpamin.us one-liners today too :)
00:12 Grinnz_ POE::Component::Resolver may be more appropriate, havent seen it before
00:13 sri UV::getaddrinfo is also an option
00:14 sri starting to wonder if we should just have Mojo::Reactor::getaddrinfo
00:15 sri downside is of course that you have to subclass a reactor
00:16 sri Mojo::IOLoop->singleton->reactor(MyReactor->new)
00:47 mattastrophe joined #mojo
00:56 neyasov_____ joined #mojo
01:01 tempire batman, marcusr: I'm using Mojo::Redis2 in production
01:01 tempire I quite like it.
01:07 disputin joined #mojo
01:20 human39 joined #mojo
01:42 damaya Hey, what do you all think of this: http://pastebin.com/JNHEkc0u
01:42 damaya I.e., doing a $res->data returns a hashref
01:43 damaya Is that intuitive? Stupid? Do it some other way?
01:51 KCL_ joined #mojo
02:09 woz joined #mojo
02:15 klapperl joined #mojo
02:34 akhasanov joined #mojo
02:37 good_news_everyon joined #mojo
02:37 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/-R-e-A
02:37 good_news_everyon mojo/master bf3dec3 Sebastian Riedel: more portable path handling
02:37 good_news_everyon left #mojo
02:52 franzkafka Is there an easy way to remove the trailing slash from a Mojo::URL ?
03:10 noganex_ joined #mojo
03:20 oetiker joined #mojo
03:47 kua joined #mojo
03:58 franzkafka The more I work with all the pieces of Mojolicious, the more I love it. Now that I have a grasp on the documentation, I'm finding everything a lot easier to figure out too.
03:58 woz joined #mojo
03:59 franzkafka <3 Mojo::UA, <3 Mojo::Base, <3 Mojo::IOLoop, <3 Mojo::EventEmitter
04:16 Eke- joined #mojo
04:27 franzkafka sri, where's your perltidy rcfile?
04:28 franzkafka found it
04:53 irq joined #mojo
05:07 good_news_everyon joined #mojo
05:07 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/cEuZsQ
05:07 good_news_everyon mojo/master e468e77 Sebastian Riedel: localhost does not need to be a special case
05:07 good_news_everyon left #mojo
05:13 good_news_everyon joined #mojo
05:13 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/jBclRQ
05:13 good_news_everyon mojo/master 5f1aedd Sebastian Riedel: the function is not needed anymore
05:13 good_news_everyon left #mojo
05:33 good_news_everyon joined #mojo
05:33 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/hAqm0A
05:33 good_news_everyon mojo/master 3f5bbfc Sebastian Riedel: be more specific in description
05:33 good_news_everyon left #mojo
05:47 woz joined #mojo
05:59 Eke- joined #mojo
06:11 rem_lex|pivo joined #mojo
07:17 damaya joined #mojo
07:18 damaya Say I have Package, and then Package::Stuff. I create a new Package::Stuff object in Package. Needing an Package accessor in Package::Stuff, I have has => package; and in Package I do Package::Stuff->new(package => $self); Is it considered bad practice to pass the object like that?
07:19 damaya Man, don't ever watch Mind of the Chef when hungry and no food in the house
07:36 woz joined #mojo
07:39 kua left #mojo
07:41 kua joined #mojo
07:43 kua left #mojo
08:06 eseyman joined #mojo
08:14 basiliscos joined #mojo
08:16 amon joined #mojo
08:26 trone joined #mojo
08:33 sunil left #mojo
08:33 Vandal joined #mojo
08:50 dod joined #mojo
09:03 sugar joined #mojo
09:07 kua joined #mojo
09:08 punter joined #mojo
09:11 sugar_ joined #mojo
09:11 fhelmberger joined #mojo
09:12 irq joined #mojo
09:16 Foxcool joined #mojo
09:25 woz joined #mojo
10:05 arthas joined #mojo
10:18 batman tempire: cool :)
10:18 batman let me know if something is odd about it
10:33 irq I read in https://metacpan.org/pod/Mojolicious::Controller (sorry only two-line paste):
10:33 irq # "/myapp/perldoc?foo=bar" if application is deployed under "/myapp"
10:33 irq $c->url_for('/perldoc')->query(foo => 'bar');
10:33 irq how the application or controller knows about /myapp deployment point?
10:53 Eke- joined #mojo
10:56 disco joined #mojo
10:58 Eke- joined #mojo
11:00 batman irq: are you using a reverse proxy?
11:00 irq yes
11:02 batman have a look at this: https://metacpan.org/pod/Mojolicious::Plugin::RequestBase
11:03 batman you might also need to rewrite the request path before sending it to the backend
11:03 batman which server do you have in front?
11:05 irq I am testing, so far, so I run with ./myapp daemon.  I want the application to be reloacatable to any prefix under  http://example.com/prefix/
11:06 damaya Who is this amazing man Marcus Ramberg?
11:06 irq ... so I can place under example.com/app1, example.com/app2, ... more applications.
11:07 damaya irq, have you seen this: http://mojolicio.us/perldoc/Mojolicious/Plugin/Mount
11:08 damaya I think that may be what you're looking for.
11:08 damaya Daniel Amaya finds Marcus Ramberg amazing.
11:09 damaya batman, are you using convos?
11:10 batman damaya: i am the creator of convos (together with marcus)
11:10 batman and yes, i am using convos
11:11 damaya it does not respond to CTCP version.
11:11 damaya I would like a spiffy version string.
11:12 damaya I will have to talk to the incredible Marcus Ramberg
11:12 batman that is strange indeed
11:12 batman which version of Mojo::IRC do you use?
11:13 damaya batman, you're not responding to CTCP version
11:13 damaya franzkakfa (also me) is using convos
11:13 damaya I am using xchat on my laptop :/
11:13 batman https://metacpan.org/pod/Mojo::IRC#ctcp_version # it should work...
11:13 damaya I need to install convos on here
11:13 batman please file a bug report
11:13 damaya ok, will do
11:13 damaya but only if you promise that it will be spiffy
11:14 woz joined #mojo
11:14 batman that would be wrong.
11:14 damaya batman, are you using latest everything?
11:15 batman i believe so
11:15 damaya Just to verify, let me go install latest version everything. One second
11:16 damaya I am always running the install script from some stupid location
11:19 damaya It goes without saying, every pacman -Syu will replace at least 10 things.
11:23 marcusr joined #mojo
11:24 marcusr tempire: sweet.
11:24 marcusr tempire: I'm very much in doubt if I want to port convos to Mojo::Redis2 or to Mojo::Pg
11:24 rofl__ Marcus Ramberg!
11:24 marcusr hi rofl__
11:25 * marcusr has his own uio.no utv server
11:25 rofl__ utv? dev?
11:25 marcusr ya
11:25 rofl__ holy moly
11:25 marcusr we have utv/test/prod
11:25 rofl__ nice
11:25 marcusr would prefer dev/stage/prod
11:25 rofl__ agree
11:26 rofl__ since thats what we are using
11:28 neyasov_____ joined #mojo
11:33 damaya Batman, another possible bug. I just started convos, connected to irc.perl.org. It says "You are not connected to magnet. Doesn't support SSL/TLS..." (something along those lines). Where I'd type it says "Connecting to backend." I see though that I am actually connected in that when I click the little talk bubble in the corner the last thing in the window is "End of /MOTD command"
11:33 damaya [Fri Dec 12 03:29:58 2014] [warn] [imatesting:magnet] This IRC network (irc.perl.org) does not support SSL/TLS.
11:33 damaya A whois on imatesting show's I am connected
11:33 purl somebody said connected was better...
11:34 damaya purl, you had me at hello.
11:34 purl damaya: what?
11:34 damaya purl, hello.
11:34 purl "Hello, I'm damaya and I'm a Perl hacker." "Hi, damaya!"
11:35 damaya Yeah, no other debug output other than that.
11:35 damaya And then clicking the gear I get: sidebar://form.sidebar (Firefox doesn't know how to open this address, because one of the following protocols (sidebar) isn't associated with any program or is not allowed in this context.)
11:36 damaya Hrm, the only thing I can click is the talk bubble in the left corner :/
11:37 damaya OK, so killed the daemon and restarted it... it automatically connects and I am stuck with 'Connecting to backend' again and no links working other than the talk bubble.
11:38 damaya batman, Sorry to bombard you with all that.
11:51 denis_boyun joined #mojo
12:00 damaya batman, been sitting here for 30 minutes: i.imgur.com/UYDD8RI.png
12:03 mishantil damaya: which wm are you using?
12:04 damaya awesome
12:04 damaya mishantil, it's called awesome :P
12:04 mishantil looks good
12:05 damaya mishantil, I use it on my laptops since I don't have dualscreens and I like tiling
12:05 mishantil damaya: Yeah, I have heard of Awesome previously. I considered it when I moved to xmonad myself.
12:05 mishantil damaya: I use xmonad because it works so well with tiling and dualscreen (but also on my laptop).
12:05 damaya mishantil, I've been using it for a few years now and I really like it. But when I have dualscreens I tend to use XFCE.
12:06 damaya awesome works pretty well with dualscreen, but I don't tend to use the features of a tiling wm with dualscreen
12:06 damaya though actually, it's not really a bad idea
12:06 mishantil I wee.
12:06 denis_boyun_ joined #mojo
12:06 mishantil Fuck.
12:06 mishantil s/wee/see
12:06 damaya I would look like a hacker at work with this set up
12:07 damaya though I'm no longer allowed to use Linux at work, so that kinda craps on that idea
12:07 mishantil What? "not allowed"?
12:07 damaya Funny that I work for a company who makes a distro of Linux and I have to run Windows.
12:07 damaya hahaha
12:07 mishantil Please do elaborate.
12:08 damaya Yeah, my work basically pulled support for Linux for a ton of tools we use then said, "You have a choice: Apple laptop or Windows."
12:08 damaya :/
12:09 damaya So, for example, we use a tool for web conferences, but that tool will no longer work in Linux (it's web-based, java app).
12:09 damaya And by "makes a distro of Linux," I basically mean they take Redhat, repackage it with some bullshit and sell support.
12:09 damaya And by bullshit, I mean (seriously) McAffee AV.
12:10 mishantil Oh my.
12:11 damaya wtf, people are naming their kids damaya??
12:11 batman sorry. my computer broke down :(
12:11 batman what is "pacman -Syu" ?
12:11 damaya oh, I updated my system :/
12:12 damaya Arch Linux. At least they notify you now though.
12:12 damaya I remember updating and they moved to systemd without so much as a notification.
12:12 damaya Kind of like gentoo news, but a bit more discrete I guess.
12:12 batman damaya: which version are you running?
12:13 damaya batman, version of Convos?
12:13 marcusr damaya: looks from your screenshot like you're connected to IRC. What is the problem? Can't join channels?
12:13 batman damaya: something is terrible wrong. check your javascript console.
12:13 mishantil damaya: Well, people are calling their kids Fifi Trixibelle, Hitler and all kinds of interesting things.
12:15 batman damaya: yes. version of convos
12:15 batman marcusr: javascript gone crazy. damaya can't even reveal the sidebars
12:16 damaya batmat, imgur.com/PF6SpaI
12:16 damaya s/batmat/batman
12:16 neilhwatson joined #mojo
12:17 damaya convos I just pulled from github
12:17 damaya Actually, I did the curl install
12:17 batman i wonder why your links doesn't have http infront...
12:17 damaya I know it installed Mojo::IRC 0.13
12:18 damaya Convos is 0.8602
12:18 marcusr seems like asset management is failing
12:18 marcusr no jquery...
12:18 batman damaya: which version of Mojolicious::Plugin::AssetPack ?
12:18 marcusr damaya: Are you using master?
12:19 marcusr possibly try the release branch which has assets precompiled?
12:20 damaya Unknown command "upgrade", maybe you need to install it?
12:20 damaya :/
12:20 damaya ok, let me pull latest versions of Mojolicious*
12:20 batman i don't get why it says "syntax error"
12:20 batman can you look at public/packed/*.js ?
12:21 damaya yeah, any particular file in there, or do you just want a directory listing?
12:22 batman one thing first.. how do you start convos?
12:23 damaya ok, so latest version of mojolicious::plugin::assetpack and mojolicious, still same thing
12:23 damaya ./vendor/bin/carton exec script/convos daemon --listen http://*:3000
12:24 damaya then I just connected on http://localhost:3000
12:25 batman can you start it with MOJO_MODE=development ?
12:25 batman maybe it's easier to track down the js error then...
12:25 Eke|| joined #mojo
12:28 damaya http://imgur.com/lZQ5Wag
12:28 damaya that's all the info I could fit on the screen
12:28 damaya Mojolicious::Plugin::AssetPack version, convos version output, and the error
12:28 damaya yeah, will do
12:29 sugar__ joined #mojo
12:33 denis_boyun joined #mojo
12:34 damaya http://imgur.com/fnXvRgc
12:37 damaya yeah, it's just not going to run on my system for some reason
12:37 damaya setting the MOJO_MODE to development it does connect, but it looks terrible
12:37 damaya removing it I get errors in firebug console (the errors you saw)
12:37 imatesting joined #mojo
12:37 imatesting :)
12:39 damaya redis v2.8.18-1
12:45 imatesting joined #mojo
12:48 imatesting joined #mojo
12:49 imatesting joined #mojo
12:49 damaya in development mode it does work (kinda), but the interface is ugly (no js/css loaded).
12:49 damaya In production mode I get the js errors that you saw
12:49 marcusr the error above indicates that you need sass.
12:49 damaya holy crap, I already said all that...
12:50 damaya marcusr, that only happens in debug mode. everything looks fine in production mode.
12:51 marcusr except the javascript assets are fscked.
12:52 damaya Is sass something you even install?
12:52 marcusr yes, it's a rubygem
12:53 damaya Alright, just installed sass
12:53 jberger damaya: you just updated Mojolicious right?
12:53 imatesting joined #mojo
12:53 damaya same errors in production
12:54 jberger there was a nasty little bug that affected carton exec which was recently fixed
12:54 imatesting joined #mojo
12:55 damaya arg, same damn errors about sass in development mode
12:55 damaya I did a gem install sass
12:55 damaya Successfully installed sass-3.4.9
12:55 damaya Installing carton 1.0.12
12:56 jberger sometimes ruby gems don't make it into the PATH
12:56 imatesting joined #mojo
12:57 damaya http://i.imgur.com/XrupN7j.png
12:57 damaya jberger, how do I check that?
12:58 damaya ok
12:58 damaya one second
12:58 purl rumour has it one second is second is the duration of 9,192,631,770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the caesium-133 atom
13:00 damaya ok! well sass is no longer giving me errors in development mode (thank you jberger)... but now I'm just getting the firebug console errors like I was in production.
13:01 imatesting joined #mojo
13:01 imatesting lies!
13:01 amon joined #mojo
13:01 imatesting it works perfectly in development mode, just not in production
13:01 imatesting yeah, sidebar links work too
13:02 damaya marcusr, do you want me to open an issue to the fact that convos does not respond to CTCP version requests?
13:03 woz joined #mojo
13:04 imatesting joined #mojo
13:04 imatesting ok, kick ass, production mode is working now too
13:04 imatesting I guess it was because SASS was not installed?
13:04 damaya well, that's what seemed to fix it
13:05 neilhwatson joined #mojo
13:05 damaya goodnight everyone. marcusr, let me know if you want me to open a bug w/ regards to CTCP version. I'll check backlog when I wake up.
13:07 marcusr damaya: Please do.
13:09 Averna joined #mojo
13:10 denis_boyun joined #mojo
13:30 neilhwatson joined #mojo
13:33 PotatoGim^Home joined #mojo
13:36 neilhwatson joined #mojo
13:39 neilhwatson joined #mojo
13:41 neyasov______ joined #mojo
13:41 neilhwatson joined #mojo
13:47 neilhwatson joined #mojo
13:49 neilhwatson joined #mojo
13:50 sugar joined #mojo
13:57 neilhwatson joined #mojo
14:01 akhasanov joined #mojo
14:06 neilhwatson joined #mojo
14:09 neilhwatson joined #mojo
14:32 ignacio_ joined #mojo
14:34 neilhwatson joined #mojo
14:39 neilhwatson joined #mojo
14:42 neilhwatson joined #mojo
14:52 woz joined #mojo
14:53 Grinnz looks like oleg found a way to keep the threads issue from segfaulting NDN https://github.com/kraih/mojo/issues/716#issuecomment-66774108
15:00 eseyman joined #mojo
15:10 zivester joined #mojo
15:16 sri hmm
15:17 sri does that change anything?
15:17 sri like, do we want to just bring back NDN support as it was?
15:18 sri looks like it's unreleased
15:20 jberger have we released since it was removed?
15:20 * jberger checks, doesn't think so
15:20 jberger no
15:21 sri not yet, but the deadline runs out in a few hours
15:21 jberger what else drives the deadline?
15:21 jberger if there is a fix in on his end, and if he will release soon, I say wait
15:21 sri is it a fix?
15:22 jberger it dies on load, so our eval will catch it
15:23 jberger https://github.com/olegwtf/p5-Net-DNS-Native/commit/40086874797dca25b3547375219ef169d89d2453#diff-f3399ebd755ffce2be4c5d3bce7f329cR32
15:29 sri i'm still sceptical
15:30 jberger I can't help that
15:30 jberger perhaps Oleg can chime in?
15:31 sri giving NDN some time to even out the kinks might not be the worst idea
15:32 jberger you're not wrong
15:32 sri at this point i'm rather scared of segfault bugs, we've had more reports recently where we didn't know the cause
15:32 jberger being in mojo is (was?) a good stress test for it, but we also can't be open to segfaults
15:39 sh4 joined #mojo
15:51 doby joined #mojo
16:03 Ptolemarch joined #mojo
16:10 sugar_ joined #mojo
16:13 Oleg joined #mojo
16:14 Ptolemarch joined #mojo
16:16 Oleg I am here. So, decision is to leave all as is and try to implement pluggable API?
16:22 sri we have no proposal for a pluggable api yet
16:24 sri right now i think we only have two options, 1) you release a new NDN, and we put support back as it was, or 2) we release without NDN support, and wait until a pluggable api proposal emerges, or NDN had more time to evolve
16:24 Oleg I think not so much people interested in DNS stuff at all, so we'll not get it in nearest time
16:27 batman sri: i'm also skeptical on having ndn support in core. +1 on pluggable api.
16:27 Oleg 1 looks reasonable, let's give it a second chance :)
16:28 sri batman: nobody is working on a pluggable api though
16:28 sri there are some problems that make it rather hard
16:28 batman no. just outing my opinions
16:28 batman sorry for not being around more :(
16:29 batman no idea why my life suddenly got so busy...
16:29 sri marcusr, tempire, crab: opinions?
16:29 * batman will go and celebrate a birthday now
16:29 jberger is it batman's birthday?
16:29 batman no.
16:29 batman gf's
16:29 Oleg happy birthday, batman
16:30 batman haha. _not_ my birthday :)
16:30 batman thanks anyway :)
16:30 Oleg :)
16:30 jberger then again, batman wouldn't say if it was his birthday, it would give away his secret identity
16:30 sri happy birthday, batman
16:30 mst happy birthday, batgirl? :D
16:30 jberger happy birthday batman!! :)(
16:31 jberger haha, batgirl, I like it
16:31 akhasanov joined #mojo
16:31 * batman 's birthday is in february. don't wanna get older until then
16:31 batman mst: yeah. i'll forward it. i like to call her "supergirl" though ;)
16:32 mst batman: that's awesome and I'm totally going to forget it before next time it's relevant ;)
16:32 batman "older" is pretty awesome. the alternative is so much worse...
16:33 jberger mst: you can't remember tons of one-off comments about random Perlers whom you've never actually met?
16:33 * batman is out of here
16:33 jberger batman: o/ have fun
16:34 mst jberger: I normally have a better memory for personal details of perlers' partners than of the perlers themselves; IME most of us seem to have partners who're more organised than we are
16:34 sri na na na na na na na na na na na na na na
16:34 mst so it's worthwhile currying favour to get them to catherd the perler half of the relationship for me
16:34 jberger mst: "partners who're more organised than we are" so true
16:35 mst e.g. never, ever ask larry about turning up to something; instead, you email gloria, who figures it out and then tells him what he's doing
16:36 genio Yea, my better half is always telling me what I'm doing.  I never know until a few hours before.  "get dressed, we're about to ..."
16:37 mst programmerdom tends to correlate with poor executive function
16:37 mst though in my case, work-wise, it's usually mdk who organises me
16:37 crab how about function execution?
16:37 mst my partners tend to mostly organise my social life ;)
16:38 btyler d
16:39 genio I mostly lack a social life these days.  Go home, play with dogs, do yard work, do mechanic work, rinse, repeat.
16:39 btyler whoops, sorry. ssh disconnect snafu
16:41 woz joined #mojo
16:51 Oleg released NDN 0.14
16:52 sri Oleg++
16:52 sri jberger, batman, marcusr, tempire, crab: allright, can i get a vote? do we try again with NDN support or release without and wait?
16:53 tempire It's released, might as well try
16:55 jberger sri: +1 on try again
16:56 mst if I had a vote I think it would be for "try again"
16:57 akhasanov joined #mojo
16:57 Grinnz_ do you (can you) have a test for what was segfaulting before?
16:58 Grinnz_ i suppose its not generalizable since it only happens when perl isn't built with pthreads
16:58 jberger Grinnz_: he can check to see if pthreads was compiled in and bail out of the tests
16:59 jberger I did link to the commit earlier, I admit I didn't look for tests
16:59 Oleg It is not easy to test. I was able to reproduce it with 500 simultaneous resolving of localhost
16:59 Oleg but this may be too much for test
17:00 mst that sounds like the sort of test that'd be worth having under AUTOMATED_TESTING so cpan testers runs it
17:00 mst DBIC has a bunch of stress tests that run that way
17:01 Oleg yes, I also used AUTOMATED_TESTING for long running tests
17:02 Oleg the bigger problem is opened files limit for 500 simultaneous resolving this will be at least 1000 files
17:02 mst so, doing that, along with having the test diag() or whatever as much information as possible about the system, so that you can confirm the characteristics of systems upon which things go wrong and refuse to install on such systems later, seems like a reasonable plan
17:03 mst note that I'm aware I may be stating the obvious here
17:11 irq joined #mojo
17:12 good_news_everyon joined #mojo
17:12 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/FbK9Cg
17:12 good_news_everyon mojo/master ce0b908 Sebastian Riedel: try again with Net::DNS::Native 0.14
17:12 good_news_everyon left #mojo
17:18 Ptolemarch joined #mojo
17:19 Oleg really good news :), sri++
17:54 elsanchez joined #mojo
18:02 skiehn joined #mojo
18:04 mattastrophe joined #mojo
18:06 sugar__ joined #mojo
18:22 irq joined #mojo
18:22 rem_lex joined #mojo
18:29 tencendur joined #mojo
18:30 woz joined #mojo
18:44 mattastrophe joined #mojo
18:53 basiliscos joined #mojo
19:15 bwf joined #mojo
19:22 Ptolemarch joined #mojo
19:31 irq joined #mojo
19:39 denis_boyun joined #mojo
19:45 Eke- joined #mojo
19:50 denis_boyun joined #mojo
19:53 basiliscos joined #mojo
19:56 good_news_everyon joined #mojo
19:56 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/hZnoZg
19:56 good_news_everyon mojo/master 5f6e035 Sebastian Riedel: mention that * is a shortcut for 0.0.0.0
19:56 good_news_everyon left #mojo
19:57 dod joined #mojo
20:03 denis_boyun_ joined #mojo
20:13 denis_boyun joined #mojo
20:14 denis_boyun__ joined #mojo
20:16 denis_boyun___ joined #mojo
20:19 woz joined #mojo
20:20 sugar joined #mojo
20:28 mattastrophe joined #mojo
20:28 denis_boyun joined #mojo
20:32 denis_boyun_ joined #mojo
20:34 denis_boyun__ joined #mojo
20:36 disputin joined #mojo
20:37 denis_boyun joined #mojo
20:42 akhasanov joined #mojo
20:42 denis_boyun_ joined #mojo
20:43 meredith hm, there's now a dist called "Minions"
20:44 jberger for when your Minion needs Minions?
20:44 meredith nah, not related to Minion at all, it's a class building kit
20:44 jberger MINION ALL THE THINGS!!!!
20:44 jberger yay, another class building kit
20:44 * jberger decides not even to look
20:45 * jberger just wrote a $url_like ephemeral test method, for testing after redirects
20:47 * jberger kinda likes it
20:48 cfedde TimTowtdi
20:48 purl There Is More Than One Way To Do It or a crappy way to write a tutorial
20:49 denis_boyun joined #mojo
20:56 denis_boyun_ joined #mojo
20:58 punter joined #mojo
21:01 denis_boyun joined #mojo
21:03 denis_boyun__ joined #mojo
21:07 denis_boyun_ joined #mojo
21:07 tempire What am I missing
21:08 tempire https://github.com/kraih/mojo/blob/master/lib/Mojolicious/Plugin/DefaultHelpers.pm#L79
21:08 tempire I don't understand how the 'my $tx' helps anything if it just goes out of scope once _delay is finished
21:08 jberger tempire: inside catch
21:08 jberger it is closed over
21:09 jberger that holds a strong reference
21:11 denis_boyun joined #mojo
21:20 denis_boyun_ joined #mojo
21:29 dod joined #mojo
21:31 Averna joined #mojo
22:07 denis_boyun joined #mojo
22:07 woz joined #mojo
22:37 absolut__ joined #mojo
22:41 jberger reference counting is hard
22:41 jberger lets go .... ALL GLORY TO THE HYPNOTOAD!!!!
22:53 cfedde bbbbbbbbb.b.bb.b.b.b.b.bb.b..b.b.b.b.b..bb.b.b.b.b..b.b..b.b.b
22:58 sugar joined #mojo
23:01 neyasov______ joined #mojo
23:03 Ptolemarch joined #mojo
23:05 Grinnz_ one reference... two reference... one reference... ah ah ah
23:08 * sri wonders if we should even try to load IO::Socket::IP after Net::DNS::Native
23:09 sri anyone may load it before us anyway
23:10 Grinnz_ if they load it before, and it causes NDN to die, then they can fix that in their code
23:10 sri might be less surprising to just fail consistently there
23:10 sri they have to figure it out, and i'm not keen on documenting that shit
23:11 Grinnz_ yeah :/ but better than having no way for it to work IMO
23:11 sri they could just build a better perl
23:11 Grinnz_ hehe
23:12 sri if you don't control the perl, i doubt you'd be able to properly use it anyway
23:13 Grinnz_ what if you had a centralized discussion of the optional dependencies in the docs somewhere, then you could put a note about NDN there
23:13 sri as in, you'd want to use kqueue/epoll, tune descriptor limits and so on
23:13 sri Grinnz_: how do you even explain the problem?
23:14 jberger sri: people who don't know the problem won't care, those who do will find the appropriate docs
23:14 sri the problem is not IO::Socket::IP specific
23:14 Grinnz_ you don't have to explain it, so much as say that loading IO::Socket::IP or <other stuff that die message says> before Mojolicious could cause NDN to not be used
23:14 jberger non-blocking resolution only matters for certain high performance cases, and those people read the docs
23:14 jberger (or should anyway
23:14 jberger )
23:15 Grinnz_ if they need more info... the NDN docs should be consulted
23:15 sri IO::Socket::IP just happens to call some Socket functions that trigger the problem
23:17 sri i don't think the NDN docs explain it yet
23:17 sri there's just the die error, which we catch
23:17 sri https://metacpan.org/source/OLEG/Net-DNS-Native-0.14/lib/Net/DNS/Native.pm#L31
23:17 Grinnz_ would be a good thing for them to document
23:18 Grinnz_ they just have a note about installation, but it's not just an installation issue anymore
23:19 sri i guess NDN should explain this
23:20 sri still can be quite surprising
23:20 sri "mojo version" may show NDN being loaded, and your app fails because you load some module using Socket functions at the wrong time
23:21 Grinnz_ right
23:21 sri well... not actually fails... just performs badly because lookups are suddenly blocking and you don't notice a thing
23:22 Grinnz_ well would that die message show up in the logs?
23:22 sri no
23:22 sri you don't notice a thing
23:22 Grinnz_ ideally would want some notification that NDN isn't being used even though it's installed, not just for this issue
23:22 sri just falls back to the blocking resolver
23:23 Grinnz_ dunno how to tell the difference between "is installed" and "use fails" though
23:23 sri normally you'd get that notification by running "mojo version"
23:23 Grinnz_ isn't*
23:23 sri MY WHOLE POINT!!!!1
23:24 Grinnz_ a quandary indeed
23:24 sri anyway... failing more consistently seems like the bestter option
23:24 sri s/s//
23:24 jberger I think it sounds like the bestter option
23:25 * sri sets jberger on fire
23:25 * Grinnz_ still wants a centralized explanation of Mojo's optional deps somewhere <_<
23:26 good_news_everyon joined #mojo
23:26 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/vr1biQ
23:26 good_news_everyon mojo/master 0d477f2 Sebastian Riedel: fail more consistently
23:26 good_news_everyon left #mojo
23:27 * sri sets Grinnz_ on fire
23:27 * jberger gets a tan
23:27 sri like the one literally every event loop using module in the dist contains? http://mojolicio.us/perldoc/Mojo/IOLoop#DESCRIPTION
23:28 Grinnz_ yeah theres that :P
23:29 Grinnz_ IO::Socket::SSL isn't just for the event loop though is it?
23:30 punter joined #mojo
23:30 Grinnz_ thats one of the things i was confused about... i don't know exactly what features that module enables
23:30 sri TLS, that's it
23:30 Grinnz_ TLS for what, though, for Mojo::UserAgent? for the server?
23:31 sri all
23:31 Grinnz_ anything else?
23:31 purl anything else is probably going to be even worse.
23:31 jberger Grinnz_: the IOLoop is probably the most central part of the system
23:31 Grinnz_ true...
23:31 purl i think true is http://search.cpan.org/perldoc?true or http://www.youtube.com/watch?v=ikkg4NobV_w
23:32 Grinnz_ purl: you are silly
23:32 purl ...but purl is <reply> I am a (modified) flooterbuck infobot, and my owner is hachi.  Download source at http://flooterbuck.sourceforge.net/ or edenc's bitch or espertinho or bugado or stupid. or the mongodb of irc bots or Nuclear Biological Chemical or a big metal dummy....
23:32 jberger hahahaha
23:32 purl LOLCON 4 reached.
23:32 sri the mongodb of irc bots... that's brutal
23:32 Grinnz_ lol
23:32 jberger I saw that too
23:32 * sri pats purl
23:32 * purl pats sri back, harder
23:32 sri botsnack!
23:32 purl :)
23:33 * jberger pushes purl down the stairs
23:33 * purl attaches herself to jberger using suction and a ham straw
23:33 cfedde perl you are also he mongodb of irc bots
23:34 cfedde purl: you are also he mongodb of irc bots
23:34 purl okay, cfedde.
23:39 jberger cfedde: wat?
23:39 purl Watt?  Watt?  I can't hear you, there's this buzzing noise where my brain should be. or https://www.destroyallsoftware.com/talks/wat
23:40 Grinnz_ purl: you have an answer for everything, don't you
23:40 purl Grinnz_: what?
23:40 Grinnz_ almost :)
23:40 jberger purl: Mongolian cluster fuck
23:40 purl mongolian cluster fuck
23:45 cfedde purl you are silly
23:45 purl ...but purl is <reply> I am a (modified) flooterbuck infobot, and my owner is hachi.  Download source at http://flooterbuck.sourceforge.net/ or edenc's bitch or espertinho or bugado or stupid. or the mongodb of irc bots or Nuclear Biological Chemical or a big metal dummy. or he mongodb of irc bots...
23:50 akhasanov joined #mojo
23:56 woz joined #mojo
23:58 damaya joined #mojo
23:58 absolut_todd joined #mojo
23:59 imatesting joined #mojo

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