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

IRC log for #mojo, 2014-02-15

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

All times shown according to UTC.

Time Nick Message
00:00 sri only /foo//bar matches there
00:03 sri there's other things we already do, such as automatically prepending a / for ->route('foo')
00:13 good_news_everyone joined #mojo
00:13 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/336g1Q
00:13 good_news_everyone mojo/master b131e30 Sebastian Riedel: improved Mojolicious::Routes::Pattern to normalize more route variations
00:13 good_news_everyone left #mojo
00:15 bpmedley http://pastie.org/8734833  <-- AnyEvent::Pg::Pool  and http://pastie.org/8734835 <-- DBD::Pg  ....  Why is my AnyEvent code so much slower?   It's taking 0.5s per request.
00:17 batman "much slower" ... compared to?
00:18 bpmedley The DBD::Pg, which is running at 0.01 per request, I believe.
00:20 batman aha
00:20 batman maybe it takes a long time to initialize?
00:21 batman looks like you're initializing the sql socket on each GET instead of in your app...
00:21 bpmedley Dunno
00:28 good_news_everyone joined #mojo
00:28 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/CPYitg
00:28 good_news_everyone mojo/master 573bf89 Sebastian Riedel: more route normalization tests
00:28 good_news_everyone left #mojo
00:30 d4rkie joined #mojo
00:49 jb360 joined #mojo
00:50 bpmedley As an aside, I screwed up the DBD::Pg pastie and somehow messed up my original code.  I hope to have a working paste soon.
00:50 Stefan joined #mojo
01:22 bpmedley http://pastie.org/8734957 <-- For posterity
01:45 jack joined #mojo
02:04 * tempire jigs
02:04 sri sjn: i've added a Mojoconf ribbon to http://mojolicio.us
02:06 sri figured a ribbon would be the most sensible choice, since other sites could just steal it to help
02:06 sri design is stolen straight from github
02:12 KindOne joined #mojo
02:13 klapperl joined #mojo
02:21 sjn sri: nice!
02:22 sjn good idea
02:26 Stefan joined #mojo
02:35 Stefan joined #mojo
02:42 jnbek man I'd love to be able to afford to go to Norway....
02:45 jnbek I feel that being mostly of Norse descent that it's a must to visit the home country
02:52 jberger_ joined #mojo
03:02 KindTwo joined #mojo
03:26 good_news_everyone joined #mojo
03:26 good_news_everyone [mojo] kraih tagged v4.81 at ab4dc9c: http://git.io/UokErw
03:26 good_news_everyone left #mojo
03:27 cpan_mojo Mojolicious 4.81 by Sebastian Riedel - http://metacpan.org/release/SRI/Mojolicious-4.81
03:30 good_news_everyone joined #mojo
03:30 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/Ke5T3Q
03:30 good_news_everyone mojo/master 0239adb Sebastian Riedel: a few more route normalization tests
03:30 good_news_everyone left #mojo
03:36 czinczar joined #mojo
03:36 czinczar how can I make sure mojo is using kqueue for max speeds on freebsd?
03:36 czinczar I installed mojolicious with cpanp
03:38 sri https://gist.github.com/kraih/5551292
03:45 ka2u joined #mojo
03:50 mattastrophe joined #mojo
04:16 ka2u joined #mojo
04:20 russum mojocasts videos seem to be down ( media.mojocasts.com doesn't seem to host the files anymore )
04:28 jberger tempire ^^
04:33 jberger russum meantime if you hit the flash link it takes you to vimeo
04:34 russum yep, it also works on iTunes podcasts, thanks!
04:44 jberger FIX ALL THE THINGS!!!!
04:45 * jberger considers submitting to yapc na
04:46 jberger I'm not sure I can do it and mojoconf and I really want to do mojoconf
04:54 jnbek joined #mojo
05:02 zM joined #mojo
06:14 mattastrophe joined #mojo
06:20 zM joined #mojo
06:41 Vandal joined #mojo
07:22 ka2u joined #mojo
07:45 abra_ joined #mojo
08:04 marcus jberger: Do mojoconf and yapc eu instead? ;-)
08:27 ka2u joined #mojo
08:42 firnsy joined #mojo
08:47 dvinciguerra joined #mojo
08:49 Stefan joined #mojo
08:52 sh4 joined #mojo
09:06 ladnaV joined #mojo
09:13 marcus joined #mojo
09:13 Vandal joined #mojo
09:13 marcus joined #mojo
09:31 inokenty joined #mojo
09:44 ka2u joined #mojo
09:51 maxhq joined #mojo
10:39 basiliscos joined #mojo
10:43 denisboyun joined #mojo
10:43 mire_ joined #mojo
11:35 gnephiak joined #mojo
11:37 gnephiak joined #mojo
11:39 mattastrophe joined #mojo
11:39 rem_lex| joined #mojo
11:53 ladnaV joined #mojo
11:58 dvinciguerra joined #mojo
12:00 Stefan joined #mojo
12:02 Vandal joined #mojo
12:52 jack joined #mojo
12:53 bowtie joined #mojo
13:11 meshl joined #mojo
13:20 paladinn joined #mojo
13:20 bowtie_ joined #mojo
13:40 d4rkie joined #mojo
13:45 dvinciguerra joined #mojo
13:49 Stefan joined #mojo
13:57 d4rkie joined #mojo
14:09 punter joined #mojo
14:11 hesco left #mojo
14:39 mire_ joined #mojo
14:44 GabrielVieira joined #mojo
15:17 bpmedley http://pastie.org/8736398 <-- What am I doing wrong?  I get these error messages at the end of the "wrk" process.  I'm wondering if "wrk" is closing the connections at the very end of the benchmark.
15:17 bpmedley The goal is to run X number of queries against mango in a non-blocking fashion.
15:18 bpmedley Sorry, against MongoDB using Mango.
15:27 asarch joined #mojo
15:35 rem_lex|pivo joined #mojo
15:43 D4RK-PH0ENiX joined #mojo
15:45 firnsy joined #mojo
16:01 HtbaaPi grr, I must be missing something. I've got nginx (https) in front of hypnotoad (with proxy => 1), but all redirects caused by my mojolicious app redirect to http instead of https
16:01 jberger_ joined #mojo
16:03 HtbaaPi my mojolicious app is simply using redirect_to('route_name')
16:03 HtbaaPi ah! Looks like I've got it! X-Forwarded-HTTPS was set to 0
16:03 HtbaaPi passing $proxy_add_x_forwarded_for seems to fix this
16:21 davido__ joined #mojo
16:55 s1037989 joined #mojo
16:56 bpmedley https://gist.github.com/brianmed/9021900 <-- What do ya'll think of this as a benchmark utility?  The goal is to have X number of independant slots that are hitting a url.  I had some issue with wrk, so I wrote this.
17:04 tempire russum: mojocasts fixed
17:06 jberger_ tempire++
17:06 jberger_ MOAR MOJOCASTS!
17:08 tempire MOAR TIME!
17:20 ka2u joined #mojo
17:26 GabrielV_ joined #mojo
17:54 denisboyun joined #mojo
18:06 meshl joined #mojo
18:11 bpmedley http://pastie.org/8736807 <-- Is this expected behaviour?  Am I doing something wrong?
18:15 jberger_ bpmedley, that code is all kinds of wrong, unfortunately
18:15 bpmedley In what way?
18:15 jberger_ first of all, why sleep?
18:16 jberger_ sleep is blocking
18:16 jberger_ if you really want to pause for 15s you should use a timer
18:16 bpmedley I'm trying to find out why sometimes "wrk" fails.
18:17 jberger_ then you are using delay/steps except you dont actually
18:17 bpmedley Hrmm.  Let me get a better example for the problem.
18:19 meshl joined #mojo
18:21 jberger_ http://pastie.org/8736828
18:21 jberger_ bpmedley ^^
18:23 bpmedley http://pastie.org/8736834 <-- Here is a better example of the issue.  If an incomplete transaction happens, then I get that error in the logs.
18:24 bpmedley Sorry my 1st example was not indictive of the issue.
18:25 jberger_ ok, but you are still not actually using the delay, right?
18:25 bpmedley Yes, I do believe the delay is in use.
18:26 jberger_ it is not
18:26 jberger_ you never call begin
18:27 bpmedley You don't have to, one sec.
18:27 jberger_ sure, it "works" but its not actually doing anything
18:27 bpmedley http://mojolicio.us/perldoc/Mojo/IOLoop/Delay#steps
18:27 jberger_ with only one step (as your current example is) you don't actually need it
18:27 bpmedley ... Sequentialize multiple events, the first callback will run right away ...
18:28 jberger_ sure
18:28 schetchik joined #mojo
18:28 jberger_ but why use it, if you dont USE it?
18:28 bpmedley I'm trimming down a larger example program to try and get a minimal test failure.
18:29 jberger_ ok, well we will come back to this
18:29 jberger_ in the meantime, how long does the db request take?
18:30 schetchik left #mojo
18:30 bpmedley http://pastie.org/8736850 <-- More minimal example
18:30 jberger_ if it's longer than the transaction timeout then it could be a problem
18:30 bpmedley Sorry for the confusion, I'm sending a GET request to Mojo and then the nc is not waiting for the response.   So, it's an incomplete HTTP transaction.
18:31 bpmedley The db read takes less than a second.
18:31 bpmedley Very small dataset.
18:34 hypnotoad joined #mojo
18:35 hypnotoad hi
18:37 bpmedley hypnotoad: Morning
18:39 s1037989 joined #mojo
18:41 skullbocks joined #mojo
18:42 jberger_ it seems that the transaction is going out of scope
18:43 jberger_ is it weakened somewhere?
18:43 jberger_ bpmedley: perhaps you should subscribe to the transaction's finish event, set a flag and check it before rendering
18:43 bpmedley Hrmm
18:44 cpan_mojo Mojolicious-Plugin-GoogleAnalytics 1.005 by ADAMJS - http://metacpan.org/release/ADAMJS/Mojolicious-Plugin-GoogleAnalytics-1.005
18:45 jberger_ bpmedley: http://pastie.org/8736887
18:46 jberger_ sri, does that look right? ^^
18:49 bpmedley jberger_: I tried it and does work.  What about this version: http://pastie.org/8736896
18:49 Stefan joined #mojo
18:50 jberger_ bpmedley: that should probably also work, I'm just a little confused about why the transaction object goes away
18:51 stokachu so i figured out the websocket issue, how do i force a websocket to ignore verify_ssl
18:51 jberger_ bpmedley: https://github.com/kraih/mojo/blob/master/lib/Mojolicious/Controller.pm#L19  <--- there is a strong reference to the tx
18:52 jberger_ oh wait!
18:52 * jberger_ digs
18:52 stokachu using anyevent i can set the verify_tls to 0 and the websocket connections happily talk to wss
18:54 jberger_ bpmedley: https://github.com/kraih/mojo/blob/master/lib/Mojolicious.pm#L117
18:56 bpmedley Hrmm.   I'll comment that out.
18:56 jberger_ bpmedley: ha! http://mojolicio.us/perldoc/Mojolicious/Controller#tx
18:56 jberger_ "Note that this reference is usually weakened, so the object needs to be referenced elsewhere as well when you're performing non-blocking operations and the underlying connection might get closed early."
18:56 jberger_ hi, I'm back
18:56 * jberger_ headdesk
18:56 purl *thunk*
18:56 * jberger_ pushes purl down the stairs
18:56 * purl dodges and laughs as jberger_ loses balance and crashes down the stairs.
18:57 jberger_ stokachu: ok, sorry, we were working on something else
18:57 stokachu jberger_: oh no worries
18:57 jberger_ I'm not familiar with your issue
18:58 stokachu jberger_: so https://gist.github.com/battlemidget/8948917 was my last attempt at making a websocket connection to a wss server
18:58 bpmedley jberger_: Gaw!!  Thanks.
18:58 stokachu i can do it with anyevent::websocket::client by setting the handle to not verify_tls
18:59 stokachu i think if i can do the same here then it'll work but i dont see an option in the documentation for websocket to disable ssl checking
18:59 stokachu i also try IO::Socket::SSL::set_defaults without any luck
19:05 Lee_ joined #mojo
19:05 jberger_ stokachu: does the verify callback offer a hook to ignore? http://mojolicio.us/perldoc/Mojo/UserAgent#ca
19:05 jberger_ I'm out of my comfort zone here
19:06 stokachu lemme read up on that
19:06 bpmedley stokachu: I can't seem to access the code.
19:06 stokachu bpmedley: the gist?
19:06 purl i think the gist is write the SQL for both queries, and see how they differ, then try to make some DBIC queries
19:06 bpmedley Yeah, I wonder if I'm having a network issue.
19:06 stokachu bpmedley: https://gist.github.com/battlemidget/8948917
19:06 stokachu thats the link it gives me in the page
19:07 bpmedley Seems my DNS is down.
19:07 stokachu ah
19:07 hypnotoad battlemidget is an offensive name
19:07 stokachu yea i tried the IO::Socket::SSL::set_defaults( SSL_verify_callback )
19:08 stokachu but i dont remember what i tried, should i just make it 0?
19:11 bpmedley_ joined #mojo
19:15 bpmedley stokachu: I found this: http://learnitcorrect.com/blog/websocket-is-great-but-not-the-origin-policy.html?utm_source=feedburner&amp;utm_medium=twitter&amp;utm_campaign=Feed%3A+hackernewsyc+(Hacker+News+YC)
19:16 bpmedley Not sure it's applicable; however, you are getting a 403 status code.
19:16 stokachu ok thanks ill read on that now
19:28 jberger_ stokachu: I think setting the default verify callback to sub{ 1 } is wahat you want
19:28 jberger_ I can't find a clean easier way
19:34 jberger_ bpmedley: oh and for completeness
19:34 jberger_ in your actual code, you do use $delay->begin, right?
19:35 stokachu jberger_: ok cool going to try that ina few
19:35 stokachu thx all
19:35 GabrielVieira joined #mojo
19:36 jberger_ stokachu: it looks like it might not be too hard to add a pass_through tls_verify_callback parameter to Mojo::UserAgent and Mojo::IOLoop::Client
19:36 jberger_ but you would have to propose that to sri (probably with code and tests)
19:53 punter joined #mojo
20:20 beyondcreed joined #mojo
20:25 punter In a route handler, does the data start travelling towards the user as soon as $self->render is encountered? Or does the sub need to complete its execution?
20:25 punter easy to check, sorry
20:27 moritz punter: you can write or write_chunk
20:28 punter moritz: No, I actually want it to send as soon as possible, so that I can do a "cleanup" routine afterwards
20:29 punter moritz: So, my question still holds - does render send data immediately
20:29 punter ?
20:36 dvinciguerra joined #mojo
20:40 mattastrophe joined #mojo
20:53 thowe joined #mojo
21:00 bpmedley punter: Not exactly.  Are you wanting to do long polling or somesuch?
21:00 bpmedley jberger_: Do you want to see the full code?
21:01 punter bpmedley: No, I don't really remember what I wanted to do. Sorry
21:02 punter It was something similar to a cleanup handler, which I would place inside the around_action hook, but I just wanted to make sure that the user wouldn't have to wait for 10 seconds while the around_action block was executing the cleanup handler
21:03 punter What's a good way of doing that, bpmedley ?
21:03 bpmedley punter: The long polling thing?
21:03 punter bpmedley: no, the cleanup handler that I wrote two messages ago (without mentioning your name, sorry)
21:04 bpmedley punter: Hrmm, can you paste some code with a Mojolicious::Lite app and we go from there?
21:04 punter bpmedley: Does Mojolicious::Lite have around_action hooks?
21:04 punter ok I'll try
21:05 bpmedley Yes, my understanding it has all the capabilties of a full app.
21:05 bpmedley http://mojolicio.us/perldoc/Mojolicious/Lite#hook
21:06 s1037989 joined #mojo
21:06 czinczar joined #mojo
21:21 czinczar anyone got a mojo website I can browse to? exta point for  a forum I can join and post to
21:21 bpmedley https://github.com/brianmed/InfoServant-Feeds
21:21 bpmedley http://infoservant.org
21:25 rjb360 joined #mojo
21:33 punter I have a forum
21:33 punter But it's alpha-version still. I'll give you the credentials in private message
21:33 punter czinczar (see above 2 messages)
21:37 rjb360 left #mojo
21:51 stokachu jberger_: ok thanks ill see if i can get something together for that
21:55 jberger_ stokachu: I don't get final say-so, sri needs to be on board
21:55 jberger_ prepare enough for a useful patch, and a test or two, but don't spend hours or anything, I can't promise inclusion
21:56 jberger_ I'm guessing your patch probably only needs to be about 3 lines of additions (sans tests)
21:56 jberger_ bpmedley: sure
21:57 jberger_ czinczar: you have asked to see a mojo-based forum several times, I'm curious why you are curious
21:57 bpmedley http://pastie.org/8737358  <-- A bit different than before - I now know what my ignorance was.
21:58 bpmedley https://github.com/TechEmpower/FrameworkBenchmarks/blob/master/mojolicious/app.pl  <-- Goal is to fix this
21:58 bpmedley $self->render( text => JSON::XS::encode_json( { message => 'Hello, world!' } ), format => 'json' );    ###  :(
22:02 jberger_ bpmedley: how many queries are there likely to be?
22:02 bpmedley I'm not 100% sure in their benchmarks
22:03 jberger_ JSON::XS is certainly faster than Mojo::JSON
22:03 jberger_ then again, I have no idea why that is a useful test
22:05 bpmedley http://www.techempower.com/benchmarks/#section=data-r8  <-- The graph, basically
22:07 jberger_ all that test does is benchmark the speed of the JSON encoder
22:07 bpmedley I agree.
22:08 jberger_ why not do: get '/json' => sub { shift->render( text => q[{'message':'hello world'}], format => 'json' };
22:08 KindTwo joined #mojo
22:09 bpmedley I'm trying to get the query piece working better.  They are using blocking DBI.  I hope that can be improved.
22:09 jberger_ will they accept a non-blocking MongoDB solution, it looks as if they use mysql
22:09 bpmedley query => db
22:10 bpmedley http://www.techempower.com/benchmarks/#section=environment&amp;hw=i7&amp;test=db  <-- MongoDB is listed
22:11 jberger_ ah cool
22:11 stokachu jberger_: ok, np, thanks again for the help
22:13 czinczar cuz i wana build one :)
22:18 jberger_ bpmedley: http://pastie.org/8737397
22:19 jberger_ using JSON::XS would be faster
22:19 jberger_ building the JSON in a string would be event faster still
22:19 hypnotoad joined #mojo
22:20 jberger_ maybe ....
22:20 jberger_ JSON::XS is pretty fast
22:21 bpmedley jberger_: I show a syntax error?
22:21 jberger_ oh, maybe
22:21 jberger_ I can't actually run it of course
22:21 * jberger_ tries perl -c
22:21 bpmedley Do you have mango?
22:22 bpmedley Or, MongoDB and Mango working?
22:22 jberger_ http://pastie.org/8737404
22:22 jberger_ missing ) on line 15
22:22 jberger_ bpmedley: I don't have any data
22:23 jberger_ I guess I could generate it
22:23 bpmedley my $oid   = $mango->db('test')->collection('foo')->insert({id => 42, randomNumber => 2702});
22:23 hypnotoad if you expect the transaction to goes away, y not increase timeout?
22:24 bpmedley hypnotoad: What code line are you referring to?
22:24 hypnotoad 14
22:24 jberger_ hypnotoad: client might disconnet
22:24 bpmedley hypnotoad: Which pastie?
22:25 jberger_ bpmedley: unless $tx->is_finished
22:25 bpmedley Oh, how does that increase the timeout?
22:25 hypnotoad ok, thought problem was Mango using only 5 connections and backlog growing too big
22:26 hypnotoad and then connections times out
22:26 bpmedley That could be a factor; however, I don't think so.
22:27 hypnotoad but i beginner with Mango
22:27 jberger_ I'm not sure how many simulataneous requests can be going
22:27 jberger_ I see in this that there will be at least 501 queries
22:27 jberger_ https://gist.github.com/msmith-techempower/8365319
22:28 bpmedley In DBD::Pg it's one async query at a time.. :(
22:28 hypnotoad i see https://github.com/kraih/mango/blob/master/lib/Mango.pm#L497
22:28 bpmedley hypnotoad: Thanks!  More tweaking.. :)
22:30 hypnotoad maybe look backlog https://github.com/kraih/mango/blob/master/lib/Mango.pm#L528
22:30 * jberger_ is reading the actual rules now
22:41 basiliscos joined #mojo
22:47 marcus joined #mojo
22:50 beyondcreed joined #mojo
22:53 bpmedley hypnotoad: I mis-read your question.  The reason not to is because I believe the problem was an incomplete HTTP transaction.
22:55 jberger_ bpmedley: current state: http://pastie.org/8737474
22:57 bpmedley Are those find_one subs concurrent?
22:57 jberger_ yes
23:01 * jberger_ starts an actual gist
23:01 jberger_ https://gist.github.com/jberger/9026436
23:02 hypnotoad faster no? $coll->find_one({id => $id} => {randomNumber => 1} => sub {
23:02 bpmedley Line 17.. looks_like_number
23:02 bpmedley Moment
23:03 GabrielV_ joined #mojo
23:13 jberger_ bpmedley: thoughts?
23:13 purl Moonlight shines through the dark night / clouds move overhead, casting shadows / dancing in the firelight
23:14 * jberger_ punches purl
23:14 purl Watch it, jberger_, I've got chunks of tougher guys than you in my stool!
23:14 bpmedley Few more minutes, please.. sorry.
23:14 jberger_ np
23:18 bpmedley Ok; I think I see what makes your concurrent.  You're not using steps; simply utilzing the active event counter.  Is that correct?
23:20 jberger_ $delay is used for concurrent requests
23:20 jberger_ steps is used for sequential requests
23:21 jberger_ the delay can be used inside any step to make sequential batches of concurrent requests
23:21 bpmedley They're both delay objects, right?
23:21 jberger_ steps is a a delay method
23:23 bpmedley However, what makes the "finish" event emit is when the active event counter goes to zero, right?
23:23 jberger_ yes
23:24 bpmedley Technically, do we have ot use a $delay object?  Can we just blast out find_one calls until we get $q results?
23:25 jberger_ I suppose you could
23:25 bpmedley I don't think it will be any faster.  Yours is probably easier to understand.
23:25 jberger_ what would matter is if they wanted an ordered set
23:26 bpmedley Right, then you would probably use steps, correct?
23:26 jberger_ but then I would modify the code to use $end->({...})
23:26 jberger_ no, just collect everything in the finish callback
23:27 bpmedley I see, yeah, that would be a good way.
23:28 jberger_ my over-reliance on steps is something that I only recently started to understand and break away from
23:28 jberger_ I have even modified one of my blog posts to remove it
23:28 bpmedley Right, they serialize too much, possibly.
23:29 jberger_ its not about serialization, its that steps is only needed when you have to wait for a previous non-blocking step before moving on to another one
23:29 bpmedley http://pastie.org/pastes/8737523/text <-- I must be doing something wrong
23:30 jberger_ yeah that seems slow
23:30 jberger_ is that your code or mine?
23:30 bpmedley Yours
23:30 punter Can anyone here install Mojolicious::Plugin::UnicodeNormalize? It fails on all of my two Ubuntu VMs (one is fresh with perlbrew & 5.18, the other has been used a lot but should pose a problem, with ordinary 5.14) - however all the cpantests have succeeded somehow
23:30 punter Mystery
23:31 punter It's the tests that fail. I'll post a bug.
23:32 bpmedley punter: Do you just want someone to verify if the tests pass?
23:32 punter bpmedley: Yes.
23:32 punter I'd guess such a useful module would have a userbase here already
23:32 jberger_ bpmedley: note that I changed the structure of the query from id => $num to _id => $num
23:33 jberger_ adding the leading underscore
23:33 jberger_ is your dataset erroring out?
23:35 hypnotoad is no better? $coll->find_one({_id => $id} => {randomNumber => 1} => sub {
23:37 bpmedley http://pastie.org/8737539  <-- With latest code from gist
23:38 bpmedley Woops.  That is without the _id mod.  It's faster to error, lets keep it that way.. :)
23:38 bpmedley hypnotoad: I don't understand.
23:38 hypnotoad limit fields return
23:38 bpmedley 643 requests in 30.01s, 714.85KB read   <-- With _ removed from _id.
23:39 bpmedley Hrmm.  I only have two fields in the db.  However, I will try it.
23:41 hypnotoad maybe app->mango->max_connections(1)
23:42 hypnotoad MongoDB not very good concurrency
23:43 bpmedley Do you know how many concurrent async requests it supports?
23:44 jberger_ bpmedley: more update, including a populate helper
23:48 bpmedley http://pastie.org/8737564 <-- Somewhat better
23:50 hypnotoad is MongoDB 2.4?
23:51 bpmedley mongo-10gen-server-2.4.9-mongodb_1.x86_64
23:51 jberger_ more updates
23:52 hypnotoad maybe try MongoDB from CPAN
23:52 jberger_ bpmedley: http://pastie.org/8737571
23:52 jberger_ hypnotoad: ha!
23:53 bpmedley What kind of box do you have?
23:53 jberger_ a big one :-)
23:54 jberger_ core i7, 32GB
23:54 hypnotoad maybe bpmedley have raspberry pi :)
23:54 bpmedley lol
23:54 bpmedley Is that with the look_like_number fix?
23:54 jberger_ yes
23:55 jberger_ oh, I forgot to push to gist
23:55 jberger_ ok NOW its updated
23:55 jberger_ :-P
23:56 hypnotoad you do not limit fields
23:57 jberger_ ?
23:58 jberger_ what do you mean?
23:58 hypnotoad not sure, maybe better $coll->find_one({_id => $id} => {_id => 0} => sub {
23:58 jberger_ why?
23:58 hypnotoad then return document only {randomNumber => 123}
23:59 beyondcreed joined #mojo

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