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

IRC log for #mojo, 2014-01-28

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

All times shown according to UTC.

Time Nick Message
00:06 sri yea, looks like it is broken and doesn't report Z_BUF_ERROR http://pastie.org/8673657
00:11 sri maybe it's a platform thing, can anyone try it on something else than os x?
00:17 sri hmm, this is problematic
00:18 sri i can prevent zip bombs, but there is no indicator that one has actually been prevented
00:27 d4rkie joined #mojo
00:28 d4rkie joined #mojo
00:34 sri oh well, opened a ticket
00:41 btyler_ joined #mojo
00:49 good_news_everyone joined #mojo
00:49 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/e9Q15g
00:49 good_news_everyone mojo/master f077d47 Sebastian Riedel: fixed a few compression bugs in Mojo::Transaction::WebSocket and Mojo::Content
00:49 good_news_everyone left #mojo
01:15 marty joined #mojo
01:23 dvinciguerra joined #mojo
01:27 ka2u joined #mojo
01:36 laouji joined #mojo
02:06 good_news_everyone joined #mojo
02:06 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/QfpRMQ
02:06 good_news_everyone mojo/master 5d18678 Sebastian Riedel: use _fallbacks as a function instead of method
02:06 good_news_everyone left #mojo
02:18 zM joined #mojo
02:23 laouji joined #mojo
02:26 laouji joined #mojo
02:31 klapperl joined #mojo
03:09 good_news_everyone joined #mojo
03:09 good_news_everyone [mojo] kraih tagged v4.71 at 669eb6a: http://git.io/hrm_Ig
03:09 good_news_everyone left #mojo
03:11 sri upgrade... or i'm going to zip bomb your hypnotoads!
04:27 laouji joined #mojo
04:46 zivester joined #mojo
04:48 crab noo! don't zip-bomb me!
05:00 ka2u joined #mojo
05:33 irq joined #mojo
05:44 btyler joined #mojo
05:47 laouji joined #mojo
06:31 marcus joined #mojo
06:47 Kulag joined #mojo
06:50 basiliscos joined #mojo
07:10 Kulag joined #mojo
07:15 Vandal joined #mojo
07:31 dpetrov_ joined #mojo
07:48 dod joined #mojo
08:01 jack joined #mojo
08:04 trone joined #mojo
08:08 laouji joined #mojo
08:10 dod joined #mojo
08:12 egopro joined #mojo
08:43 denisboyun joined #mojo
08:47 irq joined #mojo
08:54 maxhq joined #mojo
08:57 odc joined #mojo
08:59 Negor joined #mojo
09:13 rawler joined #mojo
09:34 hummeleBop joined #mojo
09:42 cleverfox joined #mojo
10:23 KindTwo joined #mojo
10:30 dvinciguerra joined #mojo
10:40 ver joined #mojo
10:51 Vandal joined #mojo
10:55 trone joined #mojo
11:12 ladnaV joined #mojo
11:16 rem_lex|pivo joined #mojo
11:31 bowtie_ joined #mojo
11:44 d4rkie joined #mojo
12:00 d4rkie joined #mojo
12:15 tl joined #mojo
12:22 arpadszasz joined #mojo
12:24 zM joined #mojo
12:57 nicomen what was the verdict again for using Test::Mojo with external sites?
13:06 maxhq joined #mojo
13:06 kwa nicomen: I remember one conversation where someone wanted to bind to a host for all tests because of some cookie problem he had. That the one you're talking about?
13:06 nicomen almost
13:09 kwa In that case, sri said something like why bother binding to a specific physical interface, and told the guy how to use http headers to fake the domain.
13:09 nicomen turns out one can just use Test::Mojo->new() and get_ok seems to work ;)
13:10 kwa Ah, yeah. It doesn't matter about the url. It uses fully qualified urls just fine. :)
13:10 kwa I'd imagine it'd be quite nice to hack up a test script for Nagios.
13:11 dvinciguerra joined #mojo
13:11 kwa http://pastebin.com/BHXsZanx <-- I'm missing something here. Is there an attribute I can set to allow simultaneous requests from a client? This example from the same machine takes 10s, from two clients on separate IPs it takes 5s.
13:13 moritz kwa: don't use morbo when you have several clients
13:13 moritz kwa: use a production webserver, like hypnotoad
13:14 kwa moritz: I did the same test using hypnotoad. Lemme try again.
13:16 kwa moritz: Yeah, same issue.
13:17 denis_boyun joined #mojo
13:22 dpetrov_ joined #mojo
13:29 maxhq1 joined #mojo
13:39 inokenty joined #mojo
13:39 r0b3rt joined #mojo
13:46 nic kwa: yeah, feeding into nagios from ua tests is easy; but I hate nagios so trialing similar for zabbix
13:49 dpetrov_ joined #mojo
13:56 russum joined #mojo
13:57 denis_boyun_ joined #mojo
14:05 bd kwa: works here
14:13 kwa bd: from morbo or hypnotoad? To test you're opening two web connections at the same time from the same machine and they both finish at the same time? :/
14:13 mire_ joined #mojo
14:16 bd kwa: i pasted your .pl into /tmp/foo.pl, ran morbo -l http://*:3002 foo.pl, in two terminals wget -O- -q localhost:3002 => both wgets completed in something over 5 seconds
14:16 kwa bah, cheers :)
14:17 kwa hmm, wondering if it's firefox limiting it
14:17 bd some proxy in between?
14:18 bd and did you trie appending some query parameter, maybe to fool a proxy?
14:19 bd kwa: I do have libev-perl instaleld though, no idea if it is a loop "feature"
14:19 kwa Turns out it's firefox only opening one connection at a time. Testing from FireFox and Chrome yields expected results. Cheers bd, was you mentioning terminals which led me to that.
14:19 bd welcome.
14:19 purl Welcome to the Internet, where men are men, women are men, and kids are cops.
14:19 bd lol
14:19 kwa :)
14:22 dvinciguerra joined #mojo
14:25 kwa As an extra for those who are interested. Turns out that trying to load the same page at the same time on Firefox and Chrome restricts the concurrent connections to the server. Adding cruft to the query string yields expected results.
14:25 kwa I guess it throttles it for when you open several tabs to the same page by accident, or on purpose in some javascript hell.
14:32 bd kwa: does ff open an http 1.0 or 1.1 connection?
14:34 kwa bd: 1.1
14:36 ka2u joined #mojo
14:36 bluescreen joined #mojo
14:46 btyler joined #mojo
14:48 x86 joined #mojo
14:51 arpadszasz joined #mojo
14:55 cleverfox joined #mojo
15:12 sh4 joined #mojo
15:13 marty joined #mojo
15:13 marty joined #mojo
15:28 cleverfox joined #mojo
15:36 tl joined #mojo
15:48 nic sri: The documentation for Mojolicious::new mentions calling ${mode}_mode
15:48 nic Is that still the case or is it deprecated?
15:48 sri oh noes!
15:48 nic MAJOR BUG??
15:51 good_news_everyone joined #mojo
15:51 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/_7WnZQ
15:51 good_news_everyone mojo/master 2b20041 Sebastian Riedel: remove feature that does not exist anymore from documentation
15:51 good_news_everyone left #mojo
15:51 nic ah, thanking you
15:51 sri nic++
15:53 nicomen kwa: I think it's common for browsers to share connections/results if the URLs match
15:53 maxhq joined #mojo
15:54 kwa nicomen: Yeah, it makes sense. Didn't immediately jump out at me when I was trying to debug why it didn't DWIM.
15:57 kwa Forced me to upgrade to 4.71 at least, which led me to read up on the switch from secret to secrets.
15:58 d^_^b my google seems lacking today, how can i include a local js file in my template?
15:59 d^_^b s/google/googlefoo/
16:00 d^_^b oh nvm found static file
16:05 priodev joined #mojo
16:06 btyler_ joined #mojo
16:06 arpadszasz_ joined #mojo
16:13 maxhq1 joined #mojo
16:33 tempire joined #mojo
16:39 priodev joined #mojo
16:46 ka2u joined #mojo
16:49 btyler joined #mojo
17:19 beyondcreed joined #mojo
17:21 sivoais joined #mojo
17:24 fhelmber_ joined #mojo
17:28 priodev joined #mojo
17:33 sri oh shit i'm stupid
17:33 sri of course there's a workaround to detect zip bombs
17:37 ka2u joined #mojo
17:44 good_news_everyone joined #mojo
17:44 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/UkbZbw
17:44 good_news_everyone mojo/master 3cd31a2 Sebastian Riedel: fixed bug in Mojo::Transaction::WebSocket that prevented decompression errors from being handled gracefully
17:44 good_news_everyone left #mojo
17:44 sri i just check if it consumed the input buffer completely \o/
17:53 sri it works since i'm only dealing with complete websocket messages
17:59 sivoais joined #mojo
18:00 sivoais joined #mojo
18:04 dod joined #mojo
18:46 dotandimet joined #mojo
18:55 sivoais joined #mojo
18:57 AirDisa joined #mojo
19:08 ka2u joined #mojo
19:16 maxhq joined #mojo
19:18 maxhq1 joined #mojo
19:26 maxhq joined #mojo
19:39 KindOne joined #mojo
19:41 dotandimet joined #mojo
19:42 mire_ joined #mojo
19:44 mire__ joined #mojo
19:46 maxhq1 joined #mojo
19:51 dotandimet joined #mojo
20:02 maxhq joined #mojo
20:21 maxhq1 joined #mojo
20:22 smammy joined #mojo
20:39 lukep joined #mojo
20:44 maxhq joined #mojo
21:33 irq joined #mojo
21:44 TitanOfOld joined #mojo
22:09 sri hmmmm.... node.js uses a very interesting trick to implement process.nextTick
22:10 sri could make mojolicious apps a little more responsive too
22:23 sri nope, not working out for us... the fact that we have stopped event loops ruins it
22:24 sri process.nextTick  runs whenever the current chunk of javascript returns control to the c++ code underneath... so basically whenever the current event callback you're in returns
22:28 good_news_everyone joined #mojo
22:28 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/vNgKZA
22:28 good_news_everyone mojo/master 8c1e908 Sebastian Riedel: detect event loop consistently
22:28 good_news_everyone left #mojo
22:44 rem_lex| joined #mojo
23:05 sri so signatures in perl5 is not happening... who would have thought
23:05 sri i'm going to invest more in perl6 now
23:08 dotandimet joined #mojo
23:11 torbjorn what kind of signatures?
23:12 sri subroutine signatures
23:12 purl subroutine signatures are documented in http://perlcabal.org/syn/S06.html
23:14 torbjorn oh right
23:20 AirDisa joined #mojo
23:20 denis_boyun joined #mojo
23:28 ka2u joined #mojo
23:38 * sri stumbled over an interesting node.js idiom that applies for mojolicious as well
23:39 sri Mojo::IOLoop->timer(0 => sub { Mojo::IOLoop->timer(0 => __SUB__) });
23:42 sri guess this oneliner brings the point better across
23:42 sri perl -Mojo -E 'Mojo::IOLoop->timer(0 => sub { say "CHUNK OF WORK ", time; Mojo::IOLoop->timer(0 => __SUB__) }); Mojo::IOLoop->start'
23:46 sri i guess it doesn't really explain why stuff might block the event loop to a novice ;p
23:52 ka2u joined #mojo

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