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

IRC log for #mojo, 2013-11-29

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

All times shown according to UTC.

Time Nick Message
00:47 dvinciguerra joined #mojo
00:52 good_news_everyone joined #mojo
00:52 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/M5lKgw
00:52 good_news_everyone mojo/master e4679a5 Sebastian Riedel: more consistent boolean examples
00:52 good_news_everyone left #mojo
00:55 laouji joined #mojo
01:24 d4rkie joined #mojo
01:35 athenot joined #mojo
02:25 laouji joined #mojo
02:32 laouji joined #mojo
03:30 good_news_everyone joined #mojo
03:30 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/zlY6wA
03:30 good_news_everyone mojo/master 6870b3b Sebastian Riedel: more WebSocket compression tests
03:30 good_news_everyone left #mojo
03:56 d4rkie joined #mojo
04:26 laouji joined #mojo
04:29 marty joined #mojo
04:34 beyondcreed joined #mojo
04:56 preflex_ joined #mojo
05:17 tba_ joined #mojo
05:55 laouji joined #mojo
05:59 athenot joined #mojo
06:52 dod joined #mojo
06:58 dod joined #mojo
07:03 dostioffski joined #mojo
07:04 dostioffski left #mojo
07:13 Vandal joined #mojo
07:33 xtetsuji joined #mojo
07:43 dod joined #mojo
07:45 basiliscos joined #mojo
08:03 laouji joined #mojo
08:11 laouji joined #mojo
08:20 nicolaas joined #mojo
08:34 trone joined #mojo
08:53 fhelmber_ joined #mojo
09:03 mire__ joined #mojo
09:16 maxhq joined #mojo
09:49 ver joined #mojo
10:04 akiym joined #mojo
10:16 keedi joined #mojo
10:22 laouji joined #mojo
10:48 d4rkie joined #mojo
10:59 ver joined #mojo
11:02 d4rkie joined #mojo
11:11 mire__ joined #mojo
11:41 batman can i check if STDOUT is open to /dev/null ?
11:41 batman or maybe check if STDOUT is not connected to pipe or...uhm...STDOUT :)
11:42 moritz I only know -t for testing if it's connected to a terminal
11:42 batman me as well... i was hoping it got a new fileno when re-opened, but it still returns 1
11:47 batman i see stat(STDOUT) changes
11:51 batman a bit ugly, but i can check if gid of /dev/tty != gid of stat(STDOUT)
12:38 dvinciguerra joined #mojo
13:04 athenot joined #mojo
13:07 ryozi joined #mojo
13:28 themage joined #mojo
13:30 hahainternet batman: what exactly are you trying to do?
13:31 jberger_ ForkCall has one failing report in CPANTesters. The error that is thrown is something I have never seen before!
13:31 jberger_ http://www.cpantesters.org/cpan/report/ea930492-584e-11e3-b6c1-e1339e5baa37
13:32 jberger_ "Variable "$args" is not available"
13:32 hahainternet that's uh
13:32 hahainternet unique
13:32 jberger_ "Variable "$cb" is not available"
13:32 jberger_ yeah
13:32 jberger_ whaaaa
13:32 jberger_ I suspect the tester has some kind of eval cage
13:33 jberger_ so I'm not too worried about the test itself, but the error is .... interesting
13:33 hahainternet it appears to be an internal perl error
13:33 hahainternet so yeah i'm going to agree it's probably running in some weird eval
13:34 batman hahainternet: i'm setting up Log::Log4perl automatically and i want it to NOT log to screen, but log to file if STDOUT>/dev/null
13:34 hahainternet batman: i don't know any syscall that would give you what you want directly
13:34 hahainternet but all the information you need is likely in /proc/$$/fd
13:34 hahainternet i assume you know
13:34 batman hahainternet: this is not cpan code, so i will hack it using stat(STDOUT)
13:35 batman jberger_: that was super weird
13:35 hahainternet not really sure what that responds with batman
13:36 batman hahainternet: stat() ?
13:36 purl hmmm... stat() is far more useful anyway
13:36 hahainternet batman: yeah i've never statted a filehandle from inside perl to read its destination :)
13:37 D4RK-PH0ENiX joined #mojo
13:38 hahainternet batman: readlink /proc/$$/fd/$fdno
13:38 hahainternet is probably much more useful to you
13:44 batman ah! clever.
13:50 moritz and totally linux-only :-)
13:59 punter joined #mojo
14:02 sh4 joined #mojo
14:05 jberger_ joined #mojo
14:15 fhelmbe__ joined #mojo
14:47 dod joined #mojo
14:51 mire__ joined #mojo
14:51 batman How do we compete with node on the code can run on client and server side?
14:51 batman Is there any Perl magic for that?
14:52 nicomen let's make it!
14:52 batman http://nerds.airbnb.com/isomorphic-javascript-future-web-apps/
14:59 diegok batman: that's on my dark side for the last year at least :-/
15:01 sri batman: https://github.com/fglock/Perlito/
15:04 asarch joined #mojo
15:04 sri one language for everything is bullshit imo though
15:07 * sri wonders if the new perl one-liner book is any good http://shop.oreilly.com/product/9781593275204.do
15:07 crab most important question: how long is it? ;-)
15:12 hahainternet batman: i don't see any benefit for code running on client and server side
15:13 hahainternet i think node.js is essentially nothing worth noting
15:14 hahainternet also moritz it is totally linux only, but i don't know of any cross platform way to check
15:14 sri libuv the foundation of node.js is very well done
15:14 sri a proper binding would be a huge win for perl
15:15 hahainternet i thought there actually was a proper binding, maybe i was thinking python
15:15 hahainternet oh, libev vs libuv
15:15 hahainternet doh
15:30 batman hahainternet: are you serious? not having to duplicate code, just because the server runs perl and the client runs js..?
15:34 hahainternet batman: i can't think of much code that i do share between server/client
15:34 hahainternet could you give an example?
15:34 beyondcreed joined #mojo
15:34 batman hahainternet: do you write much javascript using ws and/or ajax?
15:35 * batman got the x240 today
15:36 batman the trackpoint sucks and the keyboard takes some getting used to
15:36 hahainternet batman: a little using both
15:36 batman hahainternet: ok. do it a lot and you will probably notice stuff getting duplicated
15:36 batman but it sure is fast compared to the x201 :)
15:37 hahainternet batman: websockets perhaps, as you'd write serialize/deserialize on both ends i expect
15:37 hahainternet but even so, it's a relatively minor bonus in my eyes
15:37 hahainternet javascript is fairly quick, but also fairly horrifying
15:37 batman i hate repeating myself. even for the little things
15:38 batman i think javascript is awesome. i think i would only swap perl for js and maybe go.
15:38 hahainternet Go is quite nice, but perl has DWIM/TMTOWTDI
15:38 hahainternet and neither JS nor Go easily compete with that
15:38 Adurah_ Perl and JS are good because I know them most.
15:38 hahainternet p5-mop / p11 / p6 is all fantastic looking
15:38 hahainternet so i'm quite excited about perl's future
15:39 hahainternet i've been writing some python and a little ruby recently
15:39 hahainternet and every time i do i can't help but think 'this is just a gussied up version of perl with exactly the same problems'
15:39 batman hahainternet: probably because you haven't done enough js.
15:39 hahainternet batman: i wrote javascript games for a living for some time
15:40 batman a new js universe opened up once i poked inside the jquery code
15:40 hahainternet it's just a language, a relatively ugly language imo
15:40 batman hahainternet: oh. i misread your "a little using both" comment
15:40 hahainternet yeah that was ws/ajax primarily
15:40 * batman need some dinner
15:41 hahainternet ajax i tend to use JSON so encoding/decoding is all implicit
15:41 hahainternet websockets too i guess but that's less implicit
15:41 hahainternet one neat javascript thing: gnome 3's desktop is extensible via it
15:41 hahainternet that's extremely neat and a good choice of language
15:42 hahainternet but if you asked me to list the positives of node.js over Go I would be at a loss
15:50 * sri thinks that if you share a lot of code between client and server side, odds are your app is designed badly
15:53 hahainternet i can imagine a few cases where that might not be true
15:53 hahainternet games for example
15:53 hahainternet where identical bounds checking and interpolation etc would be done on both sides
15:54 hahainternet even so, i'd be hard pressed to favour js over Go, with the added bonus that reimplementations in different languages are less likely to have identical bugs
15:56 sri i've not seen such a sophisticated html5 game yet
15:57 sri but i agree about switching languages, it's a normal part of scaling up
16:09 diegok sri & hahainternet: we have a big app that is mostly client-js (backbone) + mojo as API, but we should serve same'ish app as static html's to bots. We don't need to run all same code, but we have some libs both sides and we would really enjoy sharing templates.
16:09 sri don't need js to use mustache templates on both sides
16:16 dsteinbrunner anyone work with https://metacpan.org/release/Jemplate
16:16 dsteinbrunner ?
16:23 rem_lex|pivo joined #mojo
16:42 hrupp joined #mojo
16:55 mire__ joined #mojo
16:59 Dandre left #mojo
17:02 Dandre joined #mojo
17:03 rem_lex joined #mojo
17:32 tba joined #mojo
18:02 asarch joined #mojo
18:46 Adurah joined #mojo
19:08 lammel2 joined #mojo
19:12 Mike-PerlRecruiter_ joined #mojo
19:29 jberger__ joined #mojo
19:31 KindOne joined #mojo
19:40 mire__ joined #mojo
19:58 hahainternet joined #mojo
20:18 hummeleBop joined #mojo
20:34 jberger_ joined #mojo
20:35 espent joined #mojo
21:06 tba_ joined #mojo
21:11 meshl joined #mojo
21:44 trone joined #mojo
21:58 cottoo joined #mojo
23:00 jnbek^dt joined #mojo
23:11 jberger_ joined #mojo
23:45 jnbek^dt joined #mojo
23:52 jberger_ async post #2 has been <s>fixed</s> updated

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