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

IRC log for #mojo, 2014-12-18

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

All times shown according to UTC.

Time Nick Message
00:06 akhasanov joined #mojo
00:11 good_news_everyon joined #mojo
00:11 good_news_everyon [mojo] kraih tagged v5.70 at 7591b7e: http://git.io/2iaEiw
00:11 good_news_everyon left #mojo
00:12 good_news_everyon joined #mojo
00:12 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/1J7dUw
00:12 good_news_everyon mojo/master ce72a3e Sebastian Riedel: bump version
00:12 good_news_everyon left #mojo
00:46 marty joined #mojo
01:06 denny joined #mojo
01:19 t4nk710 joined #mojo
01:31 bobkare joined #mojo
01:45 denny joined #mojo
01:47 davido_lt joined #mojo
01:50 davido_lt joined #mojo
02:07 klapperl joined #mojo
02:47 akhasanov joined #mojo
02:56 perlang joined #mojo
03:03 noganex_ joined #mojo
03:21 doby joined #mojo
03:51 woz joined #mojo
03:59 noganex joined #mojo
04:36 sh4 joined #mojo
04:38 sh4 joined #mojo
04:38 hesperaux_ joined #mojo
04:39 hshong joined #mojo
04:47 irq joined #mojo
05:02 woz joined #mojo
05:06 KCL_ joined #mojo
05:37 zivester joined #mojo
06:02 Grinnz joined #mojo
06:04 lb_ joined #mojo
06:17 rem_lex|pivo joined #mojo
06:35 Eke- joined #mojo
06:51 woz joined #mojo
06:54 batman Grinnz_: thanks! i'll look at it later today :)
06:54 batman and yes. it was supposed to be "ioloop"... guess i should deprecate it in favor of "loop" though, to make a more similar API to other mojo modules
06:56 Eke- joined #mojo
07:00 dod joined #mojo
07:13 bobkare joined #mojo
07:40 sugar joined #mojo
08:12 Vandal joined #mojo
08:15 trone joined #mojo
08:17 eseyman joined #mojo
08:23 amon joined #mojo
08:25 hotsnow joined #mojo
08:40 woz joined #mojo
08:43 sugar_ joined #mojo
08:44 nathanael joined #mojo
08:49 Insane joined #mojo
08:53 Insane hi all, in my app I have helper, which communicates with external API via Mojo::UserAgent, that helper has check if $ua returned an error and peroforms some debug actions. One of actions is creating a hash, describing what has hapened: $ua requests and API responses. Here is the code: http://pastebin.com/LF4JmAtQ
08:53 Insane Sometimes line with "my $WFDialog = {" fails with an error "No such file"
08:54 Insane but there are not any files at all, how to debug it?
08:54 Insane I suppose there might be something inside Mojo which causes this error
09:04 firnsy Insane: it's your asset lines that are ...
09:04 firnsy ... he left :(
09:07 coolo joined #mojo
09:15 sugar__ joined #mojo
09:18 Eke- joined #mojo
09:19 Eddy joined #mojo
09:19 coolo I'm having problems with websockets in Mojo::UserAgent when the HTTP server redirects me to https - at that point the upgrade stops
09:19 coolo does anyone have experience with that ?
09:27 denis_boyun joined #mojo
09:31 fhelmberger joined #mojo
09:32 mattastrophe joined #mojo
09:38 nathanael joined #mojo
09:40 nathanael joined #mojo
09:50 hshong joined #mojo
09:51 irq joined #mojo
09:54 nathanael joined #mojo
10:28 coolo solution: use wss:// directly
10:29 woz joined #mojo
10:29 nathanael joined #mojo
10:29 neyasov____ joined #mojo
11:29 kwa joined #mojo
11:44 denis_boyun joined #mojo
11:56 irq joined #mojo
12:00 Eke|| joined #mojo
12:18 woz joined #mojo
12:20 neilhwatson joined #mojo
12:29 absolut_todd joined #mojo
13:09 Kripton joined #mojo
13:31 ignacio_ joined #mojo
13:34 akhasanov joined #mojo
13:46 jberger hahahahahahah
13:46 purl LOLCON 6.5 reached.
13:46 jberger http://pastie.org/9788163
13:54 sri rofl
13:56 dotan joined #mojo
13:56 jberger that gives me a lot of faith in the packt publishing process :-)
13:56 jberger I think I will consider publishing with them in the future
13:56 jberger hehe
13:57 jberger they would probably git that twitter girl who keeps using the #mojolicious tag to review my Mojo book xD
13:58 sri hahaha
14:02 Dandre joined #mojo
14:06 woz joined #mojo
14:17 da5id sri: Would it be an idea to use something like Hash::Merge when loading the config files with M::P::Config ?
14:17 da5id As it is now .conf and $mode.conf is kinda merged, but only on the top-level
14:18 zivester joined #mojo
14:19 jberger da5id: I usually hand-roll something like that when I need ti
14:19 jberger I would listen to arguments
14:20 da5id "It would be neat" ;-)
14:21 jberger https://github.com/TechEmpower/FrameworkBenchmarks/blob/master/frameworks/Perl/mojolicious/app.pl#L9-L24
14:21 da5id Right now it kinda merges the default config with the mode specific one, you could have a different number of hypnotoad childs in prod vs in testing but only have that one key be defined in the mode specific config
14:24 jberger hmmmm, on the counter side, its a very wordy module
14:24 da5id jberger: That "merge" does the same as Mojo, but for the keys in hypnotoad
14:25 da5id I was just annoyed that I had to replicate a full config block in each config file, when only one or two values needed to be changed.
14:25 jberger da5id: right, mine is used to merge a value which is one deeper than the root
14:26 jberger I don't usually use mode-specific config files (yet)
14:26 da5id To see the difference one would have to read and compare the block in each mode plus the default config - instead of just the actual keys
14:27 jberger da5id: I get the upside, I'm now trying to find the downside
14:27 jberger and seeing if there is a useful pattern that could be implemented to avoid the non-core dependency
14:28 da5id yeah, the only downside I see is the added dependency
14:29 jberger da5id: I think the easiest solution would be to make a config plugin on cpan that uses Hash::Merge
14:29 da5id The Hash::Merge module is quite small, and we could easily replicate just the part we need here
14:29 da5id Yeah, that would work too
14:30 jberger at $work I actually just overload the config method in my app class to use Config::JSON (not really by choice)
14:31 jberger da5id: https://metacpan.org/pod/Mojolicious::Plugin::ConfigHashMerge
14:31 jberger dotan has already got something for you ^^
14:31 * jberger reads
14:31 * da5id slaps forehead
14:36 ignacio_ joined #mojo
14:38 da5id Well, I still think ConfigHashMerge should be the default behaviour
14:39 Ptolemarch joined #mojo
14:39 genio ahhhh, crap.  How do I convince my $boss that someone else should take on this old shitty SOAP app other than me?  I don't ever want to touch SOAP again.
14:41 da5id genio: http://shadow.cat/blog/matt-s-trout/mstpan-13/
14:44 jberger the post linked from that one might help too
14:46 genio I think I'm going to need lots of bourbon
14:48 * jberger fixes genio a mint julep
14:48 oetiker joined #mojo
14:51 oetiker batman: still struggeling with Mojo::SNMP can I run several prepare statements on the same Mojo::SNMP object, or may only one prepare be active at a time ?
14:59 marty joined #mojo
15:00 marty joined #mojo
15:06 batman oetiker: you should be able to run as many as you like
15:08 batman but i think your question is far more complex than what you're telling me :/
15:09 batman got any code?
15:16 * genio wonders if batman will work on his SOAPy code for him
15:18 * sri takes genio's soap and throws it out of the channel
15:18 punter joined #mojo
15:21 * genio mutters and contemplates typing 'atom .'
15:22 mattastrophe joined #mojo
15:24 sri yay, postgresql 9.4 is out \o/
15:24 Grinnz o:
15:24 Grinnz just in time to not get included with fedora 21
15:25 jberger http://imgur.com/woMxv5U
15:25 Grinnz haha
15:26 oetiker batman: I am trying to come up with a test ... must be something within my code that is causing the second (concurrent) request to take 5 seconds extra to complete
15:26 oetiker since in the test two invocations are running nicely in paralell
15:27 genio ooooh.  lots of shiny things in 9.4
15:28 sri http://www.postgresql.org/about/news/1557/
15:29 * Grinnz is so fucking jealous of people that get to use JSONB
15:33 batman genio: i don't do soap. i just do xml over http, with a "SOAP-Method" header ;)
15:34 batman oetiker: ok. i'm sorry you had to find so many weird things in the code :(
15:34 batman but i'm thankful that you report them back
15:34 batman :)
15:34 oetiker batman: got it :)
15:35 oetiker batman: the problem is that if I create multiple instances of Mojo::SNMP it starts going 'slow'
15:35 oetiker it seems that there must be only one instance
15:35 batman i was afraid of that :(
15:35 batman i don't think that can be fixed.
15:35 batman or at least i have no idea how to fix it...
15:36 batman at the moment
15:36 purl rumour has it at the moment is just goes BZZZZZZZT if you do anything slightly wrong or learning how to use it
15:36 oetiker not a problem
15:37 oetiker not really anyway :)
15:37 oetiker I can work around it ...
15:37 oetiker though
15:37 batman oetiker: i think this is the problem: https://github.com/jhthorsen/mojo-snmp/blob/master/lib/Mojo/SNMP.pm#L436
15:37 oetiker oops yes
15:37 oetiker assigning to globals
15:38 oetiker so how about makeing your magic dispatcher a singleton ?
15:38 batman i think that would work... after all, it already is... kind of :/
15:38 oetiker hihi exactly
15:38 batman need to think a bit about it, but could you try it out locally?
15:39 oetiker I can also make Mojo::SNMP a class variable
15:39 oetiker then I don't see the problem anymore :)
15:40 oetiker I also have a test to demonstrate if that helps
15:40 batman that would be awesome. could you make a new issue?
15:42 oetiker will  do
15:46 jnbek joined #mojo
15:48 sh4 joined #mojo
15:55 woz joined #mojo
16:05 oetiker batman you got your issue
16:05 * batman got issues
16:05 batman :P
16:12 sh4 joined #mojo
16:14 dod joined #mojo
16:17 batman oetiker: new release on it's way to cpan :)
16:20 batman its
16:20 batman :(
16:21 genio heh.  we think mastering apostrophes is difficult now... just imagine if we were French!  :)
16:23 oetiker batman 0.901 is an option
16:23 oetiker batman ... or use semantic versioning from the getgo :)
16:30 sri oetiker: how do you use semantiv cersioning in perl?
16:30 sri 5.001001?
16:32 sri s/c/v/
16:34 oetiker hmm I just did ... at least pause did not get angry with me it seems http://search.cpan.org/~oetiker/Mojolicious-Command-generate-qx_mojo_app-0.2.2/
16:34 sri yea, that doesn't work
16:35 oetiker http://search.cpan.org/~dwheeler/SemVer-v0.5.0/
16:36 sri well, at least it used to not work a few years ago
16:36 oetiker hihi
16:36 howitdo joined #mojo
16:38 sri pretty sure you're just asking for trouble though
16:38 sri any numeric comparison of VERSIONS will fail
16:39 sri after all
16:39 purl i guess after all is said and done.
16:39 sri perl -E 'say 0+"5.6.7"'
16:40 sri i'd be surprised if "use Foo 5.6.7" worked
16:42 jberger check using version.pm
16:42 jberger sri: that should work
16:43 batman oetiker: did I release 0.0
16:43 jberger or does it need v
16:43 jberger hmmmm
16:43 batman Oops! I meant 0.901?
16:44 batman I don't get what you mean oetiker :/
16:44 sri mojolicious originally used x.x.x versions, but andk made me switch, so i'm pretty sure there are serious problems
16:44 sri https://github.com/kraih/mojo/blob/master/Changes#L4073
16:44 jberger that's true, I wouldn't recommend them
16:44 batman Who cares, as long as it gets higher..? :)
16:45 batman oetiker: did you mean I should have called it 0.091 instead of 0.10?
16:45 Grinnz_ if you do use version; our $VERSION = version->declare('vX.X.X'); that should work
16:45 sri that also reminds me that i have no clue how v-string versions actually work... perl -E 'say v102.111.111'
16:45 batman If so... I don't care. At all.
16:45 * jberger goes to go get sonar-ed
16:45 Grinnz_ i wouldnt attempt it with v-strings
16:46 sri DBD::Pg actually uses a v-string version
16:46 Grinnz_ as long as it numerically compares, i guess it's fine
16:46 sri run the one-liner ;p
16:47 Grinnz_ yeah...
16:47 Grinnz_ vstrings are weird :P
16:51 sri jberger: if you're using version, it all makes sense
16:51 sri but $VERSION = v1.2.3 or $VERSION = "1.2.3" have me confused
16:52 Grinnz_ yeah, without the version module, it really should be 0.000000 format
16:52 meredith it's a bit amusing that the SemVer package is a v0.x.x so it's flagging itself as having a potentially unstable interface, under its own system
16:52 meredith of course we have a long tradition of "stable" 0. stuff
16:54 sri aaah
16:54 sri for the record... DBD::Pg uses "use version; our $VERSION = qv('3.4.2');"
16:54 sri that seems quite sensible
16:55 sri although... it does appear to make "use DBD::Pg VERSION" complicated
16:55 Grinnz_ yeah i like the "use version" way when i'm working with version numbers that aren't entirely perl specific
16:55 sri you'd have to know to do "use DBD::Pg 3.004002"
16:55 sri (or actually, i have no clue if that's correct ;p)
16:56 Grinnz_ what about "use DBD::Pg v3.4.2"
16:56 Grinnz_ :P
16:56 sri i doubt "use DBD::Pg 3.4.2" works
16:56 Grinnz_ i suppose it just does a numeric comparison
16:56 sri v-strings don't work that way
16:57 sri v3.4.2 is equivalent to chr(3) . chr(4) . chr(2)
16:57 sri at least unless there is some deeper magic involved
16:57 denny joined #mojo
16:57 sri please explain if anyone here knows the details :)
16:58 Grinnz_ maybe use version; use DBD::Pg version->parse('3.4.2')
16:58 Grinnz_ (i hope that doesn't work)
16:58 Ptolemarch joined #mojo
16:59 sri actually... "use DBD::Pg 3.4.2" actually works
16:59 Grinnz_ o rly
16:59 sri s/actually//
17:08 oetiker batman: should the walk method be able to handle multiple OIDs ? I think it does not at least not when running multiple requests concurrently
17:09 * oetiker goes away for a bit
17:11 sri btw. this was the ticket in question https://rt.cpan.org/Public/Bug/Display.html?id=40672
17:12 sri it was mostly about switching the versioning scheme
17:13 sri people who already do "use Mojolicious 5.70" will be in trouble with version 5.70.1
17:14 sri "0.8.4 is less than 0.8. 0.800.0 is equal to 0.8. 0.800.1 is larger than 0.8."
17:14 sri not that i wanted to switch to semver anyway ;p
17:15 sri versioning in perl is messed up though... and i will certainly have forgotten everything we talked about today in a few days
17:18 good_news_everyon joined #mojo
17:18 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/ufqMYQ
17:18 good_news_everyon mojo/master 25e121d Sebastian Riedel: updated jQuery to version 2.1.3
17:18 good_news_everyon left #mojo
17:19 jberger I always go back to xdg's blog post, weep, then do exactly as he recommends
17:20 Grinnz_ sri: yeah, you just have remember that versions with only one decimal point compare as 3 digits, so its actually Mojolicious version 5.700 :P
17:21 sri i always knew that, still didn't remember the actual problem
17:21 Grinnz_ so when you do 5.70.1 it compares to 5.700.0
17:21 batman oetiker: whenever you have a question... i'm pretty sure the answer will be "write a test - file a bug report"
17:21 jberger http://www.dagolden.com/index.php/369/version-numbers-should-be-boring/
17:22 sri Grinnz_: and just ignore the problems with switching versioning schemes and bare v-strings?
17:22 Grinnz_ well i just avoid touching v-strings :P
17:23 jberger where are you comparing versions that you can't use version.pm?
17:24 sri perl 5.8
17:24 purl perl 5.8 is probably shipped on 10 year old centos, node.js is totally not! :)
17:25 denny joined #mojo
17:25 jberger doesn't version.pm live on cpan too?
17:34 sugar joined #mojo
17:44 woz joined #mojo
18:50 denis_boyun joined #mojo
18:54 woz joined #mojo
18:58 good_news_everyon joined #mojo
18:58 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/eJXDkA
18:58 good_news_everyon mojo/master 61b7847 Sebastian Riedel: try the new container-based stack again
18:58 good_news_everyon left #mojo
19:02 punter joined #mojo
19:02 denis_boyun joined #mojo
19:04 sri dunno, doesn't seem faster
19:08 basiliscos joined #mojo
19:13 linagee Is there a startup sub for Mojolicious::Lite?
19:17 jberger linagee: that's the script itself
19:18 linagee jberger: er sorry, I mean I want a function to execute at the beginning of my Mojolicious::Lite app, but just once.
19:18 linagee kind of like under
19:28 Grinnz_ linagee: then execute it, before the routes
19:41 tvl joined #mojo
19:46 asarch joined #mojo
19:48 Grinnz_ batman: i don't suppose you have any insight on how to get apple's key and cert BS to work for Mojo::APNS... this is getting annoying
19:48 sri that's literally what the growing guide is all about
19:49 Grinnz_ it's exported as a p12 from apples whatever, i take that and i create a key file and a cert file using openssl
19:49 Grinnz_ but it doesn't work :/
20:10 linagee Grinnz_: aha. :) thanks
20:14 tempire Grinnz: What problem are you having?
20:14 hernan604 any mac user used http://ianyh.com/amethyst/ ?
20:16 Grinnz_ tempire: think i figured it out... they are probably giving me the wrong cert
20:20 disputin joined #mojo
20:33 neyasov____ joined #mojo
21:20 tempire ugh
21:20 * tempire needs ::Delay for swift
21:20 tempire objc programmers have no design sensibility at all
21:22 preaction hence, swift :p
21:23 tempire Yeah, but they're the ones creating swift libraries
21:23 jberger tempire: I'll admit, I made a quick port of EventEmitter to python once when I needed it
21:24 sugar_ joined #mojo
21:25 woz joined #mojo
21:26 Grinnz_ Taylor::Swift?
21:43 firnsy jberger: EventEmitter in python, i done the exact same thing a couple of months ago for an experimental SDK
21:44 sugar__ joined #mojo
21:45 jberger I would have stolen delay too, but I ran into an "unfixable bug" before I could get there and had to move on
21:46 preaction by which you mean you ran into a "feature"
21:48 * jberger curses DBO
21:50 irq joined #mojo
21:53 preaction upgrade your git! https://github.com/blog/1938-git-client-vulnerability-announced
21:54 sugar joined #mojo
21:55 firnsy meh ... windows/mac only
21:55 * sri is scared
21:56 hernan604 good for companys that use svn
21:59 sri oh, the new version is already in homebrew
21:59 sri phew
21:59 preaction yeah, thank yhwh for that
21:59 jberger "We have also completed an automated scan of all existing content on github.com to look for malicious content that might have been pushed to our site before this vulnerability was discovered."
22:00 jberger I wonder what they found
22:01 jberger no updates via offical mac app store yet
22:02 jberger guess I don't go around cloning untrusted repos for a while
22:09 rem_lex|pivo joined #mojo
22:15 sri jberger: brew install git
22:15 jberger does homebrew handle existing apps well?
22:16 sri have not had problems yet
22:17 sri was a bit sceptical of a second openssl, when i needed it a few months back, but it's all good
22:17 * tempire doesn't clone from other git repos due to the sheer ugliness of their interfaces
22:17 * sri is waiting for postgres 9.4 to hit homebrew :O
22:19 sri so, i guess upgrades are the weak spot of postgres https://github.com/Homebrew/homebrew/pull/35093
22:19 irq joined #mojo
22:23 preaction isn't that usually true for databases?
22:36 woz joined #mojo
22:40 sri hmmmmm https://github.com/jvinet/pg-jsonpath
22:44 Grinnz_ databases are hard to upgrade... best practice is usually to dump it before upgrade, and re-initialize from scratch using the dump in the new version
23:26 Kripton joined #mojo
23:34 oetiker batman: the problem ist that 0.9 is > 0.10 so first cpan is unhappy with you and secondly cpan++ will not upgrade 0.9 to to 0.10
23:41 jberger preaction++ # mention in mstpan
23:41 * jberger is jealous
23:41 preaction go me!
23:42 * preaction is embarassed
23:50 neyasov____ joined #mojo
23:53 sri oh my, javascript the good parts second edition :o https://www.youtube.com/watch?v=PSGEjv3Tqo0#t=0
23:55 preaction is that blue hair?
23:55 preaction oh, no, backdrop and lighting
23:55 sri haha, didn't notice that... i've just been looking at the weird stage the whole time :)
23:55 preaction the thumbnail i see makes it stand out a lot more
23:56 preaction it's shockingly bright blue. like a sky blue
23:59 sri muhahahaha... "WeakMap has the worst name ever...because it has the word weak in it...and people don't want weak stuff in their programs"

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