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

IRC log for #mojo, 2014-09-06

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

All times shown according to UTC.

Time Nick Message
00:23 d4rkie joined #mojo
00:38 sri jberger: welcome to the cool kids table \o/
00:45 d4rkie joined #mojo
00:45 jnbek oo can i sit at the cool kids table?! :D
00:45 d4rkie joined #mojo
01:12 klapperl_ joined #mojo
01:28 * tempire is the cool kids table
01:29 tempire the japanese emoticons site is among the greatest things ever
01:48 sri should this stuff be mentioned in the routing guide or not? https://gist.github.com/anonymous/746d4807bff7984db043
01:49 sri it is already implied, since the descriptions of the lite functions tell you that they are just wrappers around the methods
01:50 sri http://mojolicio.us/perldoc/Mojolicious/Lite#get
01:52 sri hmm, the more i think about it the less i like it
01:53 sri for now i think it might be better just to limit it to the stuff we actually recommend
01:53 sri and add more details if problems pop up
01:59 sri hahahaha... so silk road was brought down by an ip address leaked by bad php code http://www.wired.com/2014/09/the-fbi-finally-says-how-it-legally-pinpointed-silk-roads-server/
02:00 sri our url_for protects against that stuff!
02:11 woz joined #mojo
02:13 D4RK-PH0ENiX joined #mojo
02:34 davido_ joined #mojo
02:42 woz joined #mojo
02:58 noganex_ joined #mojo
03:06 GabrielV_ joined #mojo
03:10 basic6 joined #mojo
03:22 woz joined #mojo
03:27 sri muhahahaha
03:28 sri you remember how i made a streaming psgi proposal in 2011? so it looks like odds are that's where ruby is heading https://github.com/tenderlove/the_metal
03:46 pete What's the mechanism for streaming?
03:46 pete That you pass back an object that can be queried instead of a string as the body?
03:48 sri no, you push chunks of data whenever you like
03:48 sri exactly like mojolicious does it :)
03:49 sri $c->write and $c->finish
03:51 sri http://mojolicio.us/perldoc/Mojolicious/Controller#write
03:53 woz joined #mojo
03:56 sri my psgi-ified proposal (without wrapper objects) https://gist.github.com/kraih/2011353
04:00 sri (summary: mojolicious gets it very right, psgi is screwed)
04:03 sri the rack 2.0 proposal also makes a few rookie mistakes though
04:04 sri they capture the essence of node.js, but forgot about the importance of certain events
04:04 sri (because they are used to blocking i/o)
04:05 sri specifically drain and close events... which i've been trying to get into psgi for two years... to fix the current version of streaming
04:08 sri guess i should stop rehashing this... psgi is a lost cause, since miyagawa doesn't care anymore
04:15 woz joined #mojo
04:25 irq joined #mojo
04:39 jberger_ The most vocal proponent now is sawyer
04:42 sri i only know him from the few occasions where he was cursing at me
04:49 jberger_ I have had many meaningful conversations with him. He does have a certain vocabulary as you see in his talks
04:49 jberger_ Maybe not "many" but at least a few
04:50 jberger_ I met him in person at yapc in Madison 2012
04:50 jberger_ A very nice guy
04:50 preaction joined #mojo
04:52 sri well, he's pretty much the reason there has been bad blood between dancer and mojolicious :)
04:54 sri i made peace with sukria after the whole "you stole our idea" thing... sawyer kept the fight going afterwards
04:55 preaction th.. they came up with a completely original idea that had never been done before by anyone, nor anything similar?
04:57 sri yea, it was awkward... and quickly forgotten by most... but i think sawyer makes snippy remarks about mojolicious to this day
04:58 preaction eh, i never heard of any animosity between the three, four, or 17 projects (in Perl)
04:59 preaction but obviously the only good web framework is Flea
05:00 preaction and that only barely beats out webgui
05:00 sri only thing about it that's a little sad is that we may have a hard time in amsterdam because of it
05:01 preaction that sounds way too high school for me...
05:01 sri and that's why you're not sitting at the cool kids table!
05:02 preaction awww
05:02 preaction i'll build my own cool kid's table, with blackjack and hookers! in fact, forget the table and the blackjack!
05:02 preaction ah... screw the whole thing...
05:03 sri apparently tempire has decided to become a table earlier
05:03 preaction that's very zen
05:04 * jberger_ sits down at the tempire
05:09 * jberger_ wonders how a framework could claim to be an original concept when it also purports to be a port of another language's framework
05:09 * jberger_ is not trying to start anything though
05:12 sri so odd... everywhere i look i see references to hapi.js now
05:16 preaction oh, backend. the description sounded like every frontend web framework ever
05:16 preaction i mean, i guess it still sounds like every web framework ever...
05:21 sri yea, i'm having trouble finding something that makes it special, aside from verbosity
05:24 sri i guess composability might be it
05:24 sri the whole packs thing
05:28 sri https://medium.com/@_expr/the-pursuit-of-hapi-ness-d82777afaa4b
05:30 sri configuration centric describes it very well
05:30 cpan_mojo Test-Mojo-Trim 0.06 by CSSON - http://metacpan.org/release/CSSON/Test-Mojo-Trim-0.06
05:32 sri the caching support is kinda funny, basically our helpers and you can attach caching rules to them
05:34 howitdo joined #mojo
05:37 preaction configuration over code is one reason why i hate a lot of java
05:37 preaction if i wanted a configuration language with the power of a programming language, i'd use a programming language...
05:37 preaction_ joined #mojo
05:38 sri obviously i'm not a fan either :)
05:39 sri i find support for caching at the core of the framework intriguing though
05:42 preaction_ i dunno. SQLAlchemy also has, at its core, a caching mechanism, and all it does is cause invalidation problems
05:42 preaction_ different problem domains, sure, but it has left me wary of deep-down caching
06:02 howitdo joined #mojo
06:13 mr-foobar joined #mojo
06:33 avkhozov sri: May be we need return command_cursor instead of cursor. Since it is made in python driver. http://api.mongodb.org/python/current/api/pymongo/collection.html#pymongo.collection.Collection.aggregate
06:35 avkhozov I was just confused by the presence of count method, which does not work as described in the documentation, if its cause in aggregation result.
06:42 KCL_ joined #mojo
06:55 Vandal joined #mojo
07:08 jamesaxl joined #mojo
07:16 avkhozov_ joined #mojo
07:58 berov joined #mojo
08:23 batman jberger: i'm getting "http://pastie.org/pastes" from "mojo nopaste pastie".
08:23 batman want an issue on github?
08:54 voegelas joined #mojo
09:16 sh4 joined #mojo
09:18 basiliscos joined #mojo
09:30 sanya_com_ua joined #mojo
09:32 sanya_com_ua is it possible to force render exception page on delayed response if action died for some reason?
09:55 marcus sanya_com_ua: yes
09:56 sanya_com_ua how? subscribe to error event?
10:00 GabrielVieira joined #mojo
10:17 marcus sanya_com_ua: yes
10:25 d4rkie joined #mojo
10:39 d4rkie joined #mojo
10:39 bodgix joined #mojo
11:17 dp_ joined #mojo
11:36 mr-foobar joined #mojo
12:05 jberger_ batman: hmmmm did they change their "api" such as it was
12:05 batman no idea :/
12:06 jberger_ Sure, i guess
12:18 jberger_ sanya_com_ua: you might also want to look at the relatively new delay helper
12:19 sanya_com_ua and use catch method?
12:20 jberger_ The delay helper already has an error handle attached
12:21 Akron joined #mojo
12:21 jberger_ http://mojolicio.us/perldoc/Mojolicious/Plugin/DefaultHelpers#delay
12:26 sanya_com_ua thanks. i'll use this new helper after upgrade to newer mojo
12:33 Jonneh joined #mojo
12:35 sanya_com_ua btw, this new helper and delay's error event doesn't solve problem with fatal errors in actions delayed by timer, not by ioloop-delay.
12:39 jberger_ It does if the timer is inside the delay :-)
12:59 cpan_mojo Mojolicious-Plugin-NYTProf 0.09 by Lee Johnson - http://metacpan.org/release/LEEJO/Mojolicious-Plugin-NYTProf-0.09
12:59 jberger_ batman: it seems to be working for me
12:59 jberger_ http://pastie.org/9531764
13:01 jberger_ batman: are you perhaps not sending any content?
13:04 cpan_mojo Test-Mojo-Most 0.05 by Lee Johnson - http://metacpan.org/release/LEEJO/Test-Mojo-Most-0.05
13:05 sanya_com_ua jberger_: it's not clear for me how to place timer inside delay to catch fatal errors in timer cb
13:11 jberger_ sanya_com_ua: http://pastie.org/9531783
13:11 jberger_ this waits 2 seconds, then might die or render randomly
13:15 sanya_com_ua thanks, i have tried catch error in timer callback. didn't think that we can delay next step with timer :)
13:17 jberger_ kinda cool eh?
13:19 jberger_ think of delays as flow control in the non-blocking script and things start to fall in place
13:28 lipizzan joined #mojo
13:33 cpan_mojo Test-Mojo-Trim 0.07 by CSSON - http://metacpan.org/release/CSSON/Test-Mojo-Trim-0.07
13:44 Akron joined #mojo
13:50 cpan_mojo Mojolicious-Plugin-OAuth2 1.2 by Marcus Ramberg - http://metacpan.org/release/MRAMBERG/Mojolicious-Plugin-OAuth2-1.2
14:01 batman marcus++
14:08 Zx3 joined #mojo
14:39 woz joined #mojo
14:42 jberger_ batman: did you see my message? I can't replicate your problem
14:43 batman jberger_: i'm running this: MOJO_REDIS_DEBUG=1 MOJO_IRC_DEBUG=1 CONVOS_REDIS_URL=test prove -lv t/chat.t 2>&1 | mojo nopaste pastie
14:44 batman it works with "gist"
14:48 jberger_ joel@ultrabook:~/Programs/Dist/Galileo$ prove -lv t/basic.t 2>&1 | mojo nopaste pastie http://pastie.org/9531959
14:49 batman weird :/
14:49 jberger_ are you sure you are getting output?
14:50 jberger_ when I do mojo nopaste pastie and then Ctrl+d I get a warning and the link you see
14:50 batman yeah, as i said, it works when i use "gist" instead of "pastie"
14:50 jberger_ oh right
14:50 jberger_ hmmmmm
14:51 jberger_ gist does a slightly different read than most, but when reading from a pipe it is darn near identical
14:55 jberger_ batman: I take it that that is the convos repo?
14:55 batman yeah
14:56 * jberger_ clones
14:59 cpan_mojo Mojolicious-Plugin-SemanticUI 0.01 by Krasimir Berov - http://metacpan.org/release/BEROV/Mojolicious-Plugin-SemanticUI-0.01
15:00 jberger_ batman:
15:00 jberger_ joel@ultrabook:~/Programs/Build/convos$ MOJO_REDIS_DEBUG=1 MOJO_IRC_DEBUG=1 CONVOS_REDIS_URL=test prove -lv t/chat.t 2>&1 | mojo nopaste pastie http://pastie.org/9531984
15:01 jberger_ oh, hehe
15:01 batman MOJO_REDIS_DEBUG=1 MOJO_IRC_DEBUG=1 CONVOS_REDIS_URL=test prove -lv t/chat.t 2>&1 | mojo nopaste pastie ====> http://pastie.org/pastes
15:01 jberger_ mine wasn't probably a valid test
15:02 batman it should be the same... both commands give output
15:02 jberger_ yeah, but I didn't have all the modules installed
15:02 batman cat t/chat 2>&1 | mojo nopaste pastie ====> http://pastie.org/9531990
15:03 batman cat t/chat.t 2>&1 | mojo nopaste pastie ====> Not Found
15:03 batman it's just a pipe... the output shouldn't matter
15:03 jberger_ the second one didn't work??
15:04 batman https://gist.github.com/jhthorsen/c710b7a3c896bb44e007
15:05 batman is there any environment variables that might screw things up?
15:05 jberger_ how odd, I just git it too, with your convos test
15:06 batman maybe the input text has too many lines..?
15:06 batman wild guess...
15:06 purl rumour has it wild guess is that it is just poorly written
15:06 jberger_ could be
15:06 purl hmmm... could be is not sufficient! I want fatal or nothing.
15:09 jberger_ ok, so your `cat t/chat.t` makes no darn sense
15:09 jberger_ try it with MOJO_USERAGENT_DEBUG=1
15:12 batman :/
15:12 jberger_ batman: OHOHOH
15:12 jberger_ you can't paste it in the browser either!
15:12 batman ?
15:12 batman that is just awesome :)
15:12 jberger_ go to pastie.org and try to paste the contents of t/chat.t
15:13 jberger_ ok then, not my bug!
15:13 jberger_ :-)
15:15 jberger_ and I see the same behavior as the nopaste for your test output too
15:15 woz joined #mojo
15:16 batman pastie--
15:16 jberger_ it redirects back to pastie.org/pastes for some reason
15:16 jberger_ yeah
15:17 jberger_ batman: http://paste.ubuntu.com/8270981/
15:18 batman yeah, just drop it
15:18 jberger_ yuppers
15:18 jberger_ anyway, thanks for the report!
15:23 batman you're welcome :)
15:24 jberger_ I'm just glad to see that someone is using it really :-)
15:27 batman it's really sweet when i need to pipe something to some paste service
15:27 batman i like it more now that i can the paste, using gists
15:28 jberger_ btw, gist can do multiple files in one paste
15:28 jberger_ not from a pipe obviously
15:29 batman cool
15:29 GabrielVieira joined #mojo
15:30 jberger_ it does flatten folders though, cause gist doesn't let you do real paths
15:30 cpan_mojo WWW-Crawler-Mojo 0.01 by Sugama Keita - http://metacpan.org/release/JAMADAM/WWW-Crawler-Mojo-0.01
15:37 jberger_ ^^ has no real files?
15:38 * jberger_ guesses an accidental dzil release
15:39 sri avkhozov_: you mean like this? :) https://github.com/kraih/mango/commit/00c221a75b5c7e5fcaf6a1b9fc1be20c23675364
15:40 denis_boyun joined #mojo
16:02 cpan_mojo WWW-Crawler-Mojo 0.02 by Sugama Keita - http://metacpan.org/release/JAMADAM/WWW-Crawler-Mojo-0.02
16:18 jamesaxl joined #mojo
16:21 woz joined #mojo
16:33 sri actually... after looking through some of the official mongodb drivers... i have no idea what multiple cursors are supposed to look like
16:33 sri so i'll just split Mango::Cursor and Mango::Cursor::Query for now
16:34 sri https://github.com/kraih/mango/commit/f117238867e1fdf0b73dc8fffcaa5af0f5c1b67b
16:38 cpan_mojo Mango 1.10 by Sebastian Riedel - http://metacpan.org/release/SRI/Mango-1.10 (depends on Mojolicious)
16:42 basiliscos joined #mojo
17:00 cpan_mojo Rex-JobControl 0.0.4 by Jan Gehring - http://metacpan.org/release/JFRIED/Rex-JobControl-0.0.4 (depends on Mojolicious::Plugin)
17:14 lipizzan joined #mojo
17:19 basiliscos joined #mojo
17:19 punter joined #mojo
17:22 woz joined #mojo
17:22 dod joined #mojo
17:26 cpan_mojo Mojo-IRC 0.12 by Jan Henning Thorsen - http://metacpan.org/release/JHTHORSEN/Mojo-IRC-0.12
17:42 sri this is scary... based on minion https://metacpan.org/pod/Rex::JobControl
17:48 sri (because minion is still very experimental)
17:55 basiliscos joined #mojo
18:14 Adura joined #mojo
18:29 axlrose joined #mojo
18:38 woz joined #mojo
18:42 KCL joined #mojo
19:09 jnbek joined #mojo
19:25 berov Hi all, When generating a new plugin , '<% %>' eats the line above.
19:25 berov Example: http://paste.ubuntu.com/8272689/
19:25 berov see '=head2 register'
19:31 berov this results in invalid POD. Everybody should edit his newly created plugin, but is this behavior of <% %> expected in general?
19:36 punter joined #mojo
19:42 woz joined #mojo
20:02 sri berov: no
20:13 woz joined #mojo
20:52 sri yea, was a regression from an optimization i made
20:54 marcus_ joined #mojo
20:58 good_news_everyon joined #mojo
20:58 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/VzI3rQ
20:58 good_news_everyon mojo/master ded109b Sebastian Riedel: fixed bug in Mojo::Template where newline characters could get lost
20:58 good_news_everyon left #mojo
21:16 punter joined #mojo
21:18 woz joined #mojo
21:22 alnewkirk hey guys, I've noticed that my CI pipeline broke twice recently, caused by missing Mojolicious distributions
21:22 alnewkirk first SRI/Mojolicious-5.33.tar.gz, then SRI/Mojolicious-5.36.tar.gz
21:24 alnewkirk I gather that those distro have been deleted from CPAN, is this normal to your release operation, or merely a coincidence?
21:35 sri anyone got any experience with highlight.js? https://highlightjs.org/
21:35 sri it looks like a pretty decent replacement for prettify.js
21:45 sri hmm, looks like highlight.js can't handle mixed languages so well... like lite apps with __DATA__
21:47 sri i suppose it could be cool if someone added real support... which might not be so hard, based on the TM2/atom packages
21:47 sri doesn't look so hard https://github.com/isagalaev/highlight.js/blob/master/src/languages/django.js
22:08 jberger_ re: Rex::JobControl, I was about to be impressed with their app construction, and then ... https://github.com/RexOps/rex-jobcontrol/blob/master/t/basic.t
22:14 actuallymarcus jberger_: *done*
22:14 jberger_ ?
22:16 actuallymarcus jberger_: I meant the test ;)
22:17 actuallymarcus got object, *done*
22:17 jberger_ Ah yes, a very useful (single) test,  that one
22:18 jberger_ They could test the snot out of an app like that
22:23 woz joined #mojo
22:39 bodgix left #mojo
22:43 disputin joined #mojo
23:07 sri argh
23:07 sri i've actually hacked together a highlight.js language file for mojolicious... but i can't get the python build tool to work
23:08 sri or should i say python3 and java build tool -.-
23:09 sri i give up... for future reference https://gist.github.com/anonymous/0c710ee29cb4104ec79a
23:32 jamesaxl joined #mojo
23:33 woz joined #mojo
23:37 woz joined #mojo
23:41 jberger_ why does a javascript library need a python3 + java build tool?
23:54 sri for building... duh!
23:54 sri how else would you build your javascript libraries?
23:59 * jberger_ can't decide between running away screaming and falling to the ground sobbing

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