Camelia, the Perl 6 bug

IRC log for #mojo, 2012-08-29

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

All times shown according to UTC.

Time Nick Message
00:18 teejay joined #mojo
00:55 laouji joined #mojo
01:12 good_news_everyone joined #mojo
01:12 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/FtPpmw
01:12 good_news_everyone [mojo/master] fixed another typo in a quote - Sebastian Riedel
01:12 good_news_everyone left #mojo
01:17 jnap joined #mojo
01:46 good_news_everyone joined #mojo
01:46 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/9Fx9PA
01:46 good_news_everyone [mojo/master] a few more quotes - Sebastian Riedel
01:46 good_news_everyone left #mojo
01:55 d4rkie joined #mojo
01:58 scott joined #mojo
02:03 xaka joined #mojo
02:15 xaka1 joined #mojo
02:22 good_news_everyone joined #mojo
02:22 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/iUcOxA
02:22 good_news_everyone [mojo/master] removed all quotes again - Sebastian Riedel
02:22 good_news_everyone left #mojo
02:30 jnap joined #mojo
02:36 sri i have a feeling political correctness will become a hot topic... perhaps it is time to leave funny quotes behind :/ http://blogs.perl.org/users/joe_mcmahon1/2012/​08/why-im-considering-dropping-perlmonks.html
02:37 * sri has absolutely no idea which minorities might consider which quotes offensive
02:38 noganex joined #mojo
02:44 jnap joined #mojo
02:59 jwang sri: if memory serves, this particular issues has been mentioned on perl blogs quite a few times in the past.
03:00 sri jwang: i see it mentioned more and more frequently in my twitter stream
03:00 sri not just in the perl community
03:03 sri i actually think it's a good thing people are more aware of it, but i have no clue whatsoever where to draw the line
03:04 jnap joined #mojo
03:12 jwang I think this particular area has been a bit sensitive. The reactions have been getting bigger too
03:17 jwang it's like how you can't use master / slave for replication at a lot of companies now
03:23 good_news_everyone joined #mojo
03:23 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/z3Gttw
03:23 good_news_everyone [mojo/master] tweaked header examples - Sebastian Riedel
03:23 good_news_everyone left #mojo
03:24 sri i'm not even sure if "bite my shiny metal ass" is acceptable for examples :S
03:30 jnap bite my shiny metal ass is not in the same category as some outdated technology terms like master/slave that might be considered aggressive toward some people.
03:39 sri until robots become sentient...
03:49 mandreacchio joined #mojo
05:34 Britzel joined #mojo
05:38 laouji joined #mojo
05:39 memowe joined #mojo
06:06 d4rkie joined #mojo
06:25 cosmincx joined #mojo
06:47 spleenjack joined #mojo
07:00 Vandal joined #mojo
07:00 davido joined #mojo
07:37 omega how can I make a route that catches everything that doesn't match any other route?
07:41 batman_ joined #mojo
07:58 Kwa omega: Not sure if it's the recommended way -- https://groups.google.com/forum/?fromg​roups=#!topic/mojolicious/0wzBRnetiHo
07:59 omega ahh, ok, thanks :)
07:59 omega I couldn't get that to match just a / though
07:59 omega like make the catch all optional
07:59 omega but I just made two routes
08:01 pau4o joined #mojo
08:04 omega what is it in 5.10.0 that breaks mojo?
08:07 crab why do you think 5.10.0 breaks mojo?
08:09 crab if it's the 5.10.1 recommendation, it's because 5.10.0 is just broken, period.
08:10 crab oh, sorry. it's a requirement now. right.
08:18 mikegrb joined #mojo
08:19 dminus joined #mojo
08:19 omega just wondering, as current squeeze has 5.10.0
08:19 omega but I'll just get a wheezy vm
08:21 purl joined #mojo
08:21 tardisx joined #mojo
08:27 dminus joined #mojo
08:27 mikegrb joined #mojo
08:28 tardisx joined #mojo
08:31 purl joined #mojo
08:53 spleenjack joined #mojo
09:00 dminus joined #mojo
09:00 purl joined #mojo
09:01 mikegrb joined #mojo
09:01 nic omega: if you have 5.10.0 handy (as I did) just 'make test' in your Mojolicious download and it tells you which tests fail
09:01 nic I think several of them broke when I ran it
09:01 fhelmber_ joined #mojo
09:02 nic It be quite annoying that something as recent as 'squeeze' is that far behind with perl
09:02 tardisx joined #mojo
09:03 omega I'm guessing the smart match changes is something, get a lot of warnings about that at least
09:04 omega but only 2 tests failed, for ojo.t
09:04 marcus sri: I don't understand how this works - https://github.com/kraih/mojo/​blob/master/t/mojo/path.t#L16 doesn't parse return an arrayref?
09:04 marcus sri: or does the attr return itself?
09:06 Vandal joined #mojo
09:06 buu return $self->parts([split '/', $path, -1])
09:07 buu Huh
09:07 buu Autoboxing allowing you to call methods on arrayrefs?
09:08 marcus buu: guess the accessor is chaining on set
09:08 heytrav joined #mojo
09:10 buu marcus: ->parse returns a Mojo::Path object
09:10 buu How odd.
09:10 buu I'm sure this is all carefully explained someplace.
09:10 marcus buu: in Mojo::Base
09:11 buu "Create attributes for hash-based objects. An array reference can be used to create more than one attribute. Pass an optional second argument to set a default value, it should be a constant or a callback. The callback will be excuted at accessor read time if there's no set value.
09:12 marcus yeah, it's not really explained there...
09:13 buu 91     # Footer (return invocant)
09:13 buu 92     $code .= "  \$_[0];\n};";
09:14 buu If you get you get the value if you set you get the invocant.
09:14 buu Which is what you said anyway.
09:14 purl joined #mojo
09:21 dminus joined #mojo
09:21 mikegrb joined #mojo
09:22 tardisx joined #mojo
09:25 purl joined #mojo
09:35 mikegrb joined #mojo
09:35 dminus joined #mojo
09:35 tardisx joined #mojo
09:40 purl joined #mojo
10:20 vervain joined #mojo
10:27 daxim joined #mojo
10:28 Lucas1 joined #mojo
10:52 * vervain started a test websocket.t loop
10:52 d4rkie joined #mojo
10:58 andrefs joined #mojo
11:19 mire_ joined #mojo
11:22 ryozi joined #mojo
11:36 D4RK-PH0ENiX joined #mojo
12:40 d4rkie joined #mojo
12:48 batman can i have non-blocking bridges?
12:50 spleenjack joined #mojo
12:51 spleenjack1 joined #mojo
12:56 marcus air bridge.
12:56 cosmincx joined #mojo
13:06 batman a non-blocking bridge sounds like a bridge that always return true :P
13:10 marcus thereifixedit.com
13:10 CromeDome joined #mojo
13:13 jnap joined #mojo
13:13 spleenjack joined #mojo
13:17 gryphon joined #mojo
13:36 sri marcus: what exactly is not 100% clear?
13:36 sri the return value should be in the method description
13:37 marcus sri: I didn't find anything in Mojo::Base describing the return value of attributes on set.
13:38 sri the synopsis shows it
13:38 marcus nor do any of the examples in the synopsis chain.
13:38 sri http://mojolicio.us/perldoc/Mojo/Path#parse
13:38 marcus oh, they do
13:38 marcus you're quite right
13:38 marcus I was looking at the code in DESCRIPTION
13:39 daxim http://perlnews.org/2012/08​/plat_forms-deadline-2012/
13:46 good_news_everyone joined #mojo
13:46 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/AkBIig
13:46 good_news_everyone [mojo/master] mention attribute accessor chaining - Sebastian Riedel
13:46 good_news_everyone left #mojo
13:46 sri marcus: better?
13:51 good_news_everyone joined #mojo
13:51 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/FHj6IA
13:51 good_news_everyone [mojo/master] tweaked accessor description some more - Sebastian Riedel
13:51 good_news_everyone left #mojo
13:52 marcus sri: yes
13:52 sri \o/
13:52 sri argh, something broke Mojo::Template in 5.17.4
13:58 marcus Sue, flee!
14:02 * sri flees
14:09 vel joined #mojo
14:09 vervain FYI: no bad websocket.t tests yet.
14:10 sri vervain: thanks for testing
14:10 labrown joined #mojo
14:11 vervain np... I'm not counting iterations or anything... would it help if I did? Or setup anything specific environment wise?
14:12 sri don't think so, i get a failed test every 200-500 runs
14:13 sri it may be affected by side effects, like another cpu heavy test slowing it down
14:14 sri but i don't know for sure, it's way too rare
14:21 inokenty joined #mojo
14:29 Leandr joined #mojo
14:29 Britzel joined #mojo
14:35 vel joined #mojo
14:36 knshaum joined #mojo
15:33 pau4o left #mojo
15:54 Antisoche joined #mojo
15:57 zpmorgan joined #mojo
15:59 vervain I threw together a little script in case anyone else wants to run a loop to test websockets.t. http://pastebin.com/dRZT694J
16:00 sri vervain++
16:04 xaka joined #mojo
16:10 batman_ joined #mojo
16:16 sri new mongodb release looks pretty good
16:18 teejay joined #mojo
16:19 buu Anything exciting?
16:19 sri http://docs.mongodb.org/manual/release-notes/2.2/
16:20 sri a lot actually
16:21 inokenty I really like the "#{ EXPRESSION }" ruby feature. Is there any chance it will be implemented in Perl? =)
16:21 buu Heh, not likely. Obviously you can do $foo=expression; "$foo" or similar.
16:22 buu And there's hacks which emulate it.
16:22 inokenty perl -E'say "@{[1+1]}"' ?
16:22 sri yes
16:22 inokenty Looks ugly =)
16:23 sri i use that every now and then
16:23 inokenty And what about performance of this operatin?
16:23 inokenty *operation
16:23 sri go benchmark it ;)
16:25 sri the python version is neat too
16:26 sri "%s" % 1+1
16:27 inokenty Like sprintf )
16:27 inokenty But shorter
16:27 sri ye
16:29 buu Oh hey, cool. http://docs.mongodb.org/manual/​reference/projection/elemMatch/
16:33 inokenty http://www.hastebin.com/voqujowipo.pl
16:34 inokenty Slow
16:35 sri i somehow doubt it will end up being a bottleneck in a real app :)
16:35 buu heh
16:37 xaka TTL Collections - at last!
16:37 hide joined #mojo
16:41 sri inokenty: http://www.hastebin.com/raw/buqihovido
16:42 * sri didn't think sprintf was so damn fast
16:42 inokenty ( O_o)
16:42 inokenty That's ... awesome!
16:44 jnap joined #mojo
16:45 inokenty Now I love sprintf =)
16:47 buu woah
16:49 fhelmber_ joined #mojo
16:50 marcus that is not really fair ..
16:50 marcus https://gist.github.com/3515534
16:53 sri marcus: it's not about fairness
16:54 sri marcus: you made the hack completely pointless ;p
16:56 sri the normal vs sprintf comparison is intersting though
16:58 sri hmmm... looks like every single one of our smartmatch use cases will break now
17:04 * tempire supports the funny quotes regardless of what the hipsters are hand waiving about
17:05 tempire it's not about "right" behavior.  there's no such thing.  it's about people and their expectations.
17:05 tempire best you can do is try not to give people a reason to be marginalized
17:06 vel joined #mojo
17:06 tempire but that's never a guarantee, no matter what you do
17:08 tempire as such, it's probably a good idea to avoid anything that hints toward objectification of women, since that's what people are most sensitive about right now.
17:09 tempire correction, I meant "best you can do is try not to give people a reason to *feel* marginalized"
17:12 sri not being a native english speaker it's pretty much impossible for me to know which futurama quote might offend which minority
17:13 tempire just stay away from comments about women's bodies.
17:13 sri i will always stumble over slang i don't know
17:13 Vandal imho normal slower because of init new var
17:14 tempire trying to eliminate offense is like trying to eliminate side effects in a command line program
17:14 ryozi joined #mojo
17:15 sri i just don't think funny quotes are worth the headache anymore for me :/
17:15 tempire this is coming from twitter comments?
17:16 * tempire demands urls
17:16 sri nothing specific, just stuff that started adding up over the months
17:17 tempire would you say you feel marginalized by people assuming you mean something negative?
17:17 sri i really don't want some "mojolicious is sexist" blog posts just because i failed to grok some slang in a futurama quote
17:18 alester joined #mojo
17:19 tempire interesting that the popularity of public criticism is causing an atmosphere of fear
17:19 * tempire supports sri in the matter no matter what he chooses
17:19 sri i would be less afraid if the quotes were german
17:22 mayfield joined #mojo
17:31 tempire you could pre-empt it by making a blog post about how mojolicious includes futurama posts, and you're concerned that because you're not a native speaker, you're afraid someone is going to get offended unintentionly.
17:31 tempire but then again, it could just draw attention.
17:32 sri yea
17:32 tempire I see it similar to micro-benchmarking posts.
17:32 tempire some people are going to do them.  others are going to believe them.  nothing you can do about it.
17:41 tempire You could put into the coding guidelines that not everyone is a native english speaker, so if one of the quotes has been misunderstood, message someone privately and it will be addressed.
17:41 tempire or maybe an asterisk at the bottom
17:42 tempire the people who don't find that acceptable reveal themselves as those who are more interested in being offended that finding solutions.
17:42 sri takes the fun out of having quotes in the first place imo
17:43 tempire to me, it just provides an out in case someone does complain.  it says, "hey, we were thoughtful, it's just that you jumped to conclusions"
17:48 davido I've never even had the time to watch futurama.  For me to adhere to the coding guideline of including a futurama quote, I would have to hit google up for one first.
17:54 mire_ joined #mojo
17:56 andrefs joined #mojo
17:57 sri heh, speaking of the devil http://www.modernperlbooks.com/mt/2012/08/you-​dont-get-to-choose-how-other-people-feel.html
18:04 vervain Political Correctness meets Technology... whooda thunkit.
18:04 vervain http://www.oook.cz/screenshots/takeittux2.png
18:08 vervain Bring back the BOFH I say ;-)
18:13 vervain Groovy article though.
18:16 tolid_air joined #mojo
18:29 * sri wonders if we should just get rid of all smartmatch
18:37 osprey_ joined #mojo
18:38 osprey_ Hi
18:38 purl hi, osprey_.
18:40 osprey_ I use Mojo::UserAgent for downloading some files and I get 411 - Length Required with proxy
18:41 osprey_ 1 sec, need test with old mojo
18:57 tempire I've never used it.
18:57 tempire (smartmatch)
18:58 tempire it seems to lean on the wrong side of being obvious
18:58 tempire "it acts like this...except in this situation, and that situation, and sometimes this situation"
18:59 good_news_everyone joined #mojo
18:59 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/ftuw3g
18:59 good_news_everyone [mojo/master] no more smartmatch - Sebastian Riedel
18:59 good_news_everyone left #mojo
19:00 sri tempire: problem is smartmatch will break completely in 5.18
19:00 breakEM joined #mojo
19:00 sri http://www.xray.mpe.mpg.de/mailing-lis​ts/perl5-porters/2012-08/msg01102.html # that's the new table atm
19:00 sri none of our use cases is in there anymore :S
19:01 osprey_ heh when I set in header Referer and use proxy I'm fail http://pastebin.com/1EB687eg
19:01 tempire right...though I didn't read the proposed updates.  I've already written it off, so the discussion is just proof that I was right :)
19:01 osprey_ without $header{Referer} all ok
19:02 sri it's mostly matching multiple values [1,2,3] and ($foo // '') eq 'bar'
19:03 sri $foo ~~ 'bar' just looks better :)
19:04 sri osprey_: \%header
19:05 osprey_ sri: Thanks!
19:48 osprey_ left #mojo
19:51 tempire eak!
19:51 tempire http://mojolicio.us/perldoc/Mojo/Base <- missing 'paws' attribute
19:51 tempire in Cat class
19:53 davido Smartmatch is a can of worms.  "1ringy-dingy" ~~ @array (where @array = (4, 3, 2, 1)) will force a numeric comparison on "1ringy-dingy", so you get 1==1 where you're expecting "1ringy-dingy" ne '1'.  This and many other special cases will haunt you with ~~
19:54 davido Hmm... is a can of worms PC? ;)
19:57 tempire let's light up a fag and think about it.
19:58 batman_ joined #mojo
19:58 tempire ^ cigarette in british slang, in case anyone doesn't know
19:58 * tempire hides
19:59 batman hi.
20:00 batman i must confess something......
20:00 batman i've stopped using quotes around my hash keys.
20:01 * tempire never uses syntax unless it's necessary or helps in readability
20:01 batman :)
20:01 * tempire also uses hash rockets instead of commas most of the time
20:03 sri HASH ROCKETS!
20:03 * jzawodn never heard 'em called that before.  makes sense though
20:03 * tempire is also sad when he has to use a for loop instead of a map or reduce
20:07 Netfeed batman: isn't that good practice?
20:07 batman what is good practice?
20:07 purl i guess good practice is use warnings and use strict
20:07 batman i think not using perltidy is the best practice.
20:07 good_news_everyone joined #mojo
20:07 good_news_everyone [mojo] kraih pushed 2 new commits to master: http://git.io/atNvDw
20:07 good_news_everyone [mojo/master] fixed Mojo::Base synopsis example - Sebastian Riedel
20:07 good_news_everyone [mojo/master] fixed script mode - Sebastian Riedel
20:07 good_news_everyone left #mojo
20:08 davido Good practice: Use your head.
20:08 Netfeed batman: skipping quotes around hash keys
20:08 batman if i would write "perl best practices #2" i would write one page: write readable code in you own style, to force others understand how to read code on cpan
20:08 batman Netfeed: i have no idea. but i'm starting to like it
20:09 Netfeed :)
20:10 davido I noticed that the cert advisories for Perl are mostly based on PBP.
20:10 davido er, cert recommendations.
20:12 batman who cares?
20:12 purl not me
20:12 batman purl: that's why you're so awesome :D
20:12 purl batman: sorry...
20:15 * tempire just realized the perfect way to return errors from objects
20:16 tempire can't believe I've never thought to use overload 'bool' => sub {()};
20:20 Leandr joined #mojo
20:23 arthas joined #mojo
20:34 good_news_everyone joined #mojo
20:34 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/Sxmjkw
20:34 good_news_everyone [mojo/master] documentation tweaks - Sebastian Riedel
20:34 good_news_everyone left #mojo
20:35 sri we still use smartmatch in two places, but odds are they will keep working, not sure if they should be left alone
20:35 sri https://github.com/kraih/mojo/blob​/master/lib/Mojo/Collection.pm#L31
20:52 marcus sri: I think they should.
20:54 good_news_everyone joined #mojo
20:54 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/5b9C_w
20:54 good_news_everyone [mojo/master] fixed error handling examples - Sebastian Riedel
20:54 good_news_everyone left #mojo
20:55 sri marcus: mkay
21:13 davido http://perlnews.org/2012/08​/plat_forms-deadline-2012/
21:15 tabbi joined #mojo
21:17 tabbi hi to all! is there a simple way to protect a subdirectory of the public directory from access by all? i'd like to have this directory only accessible for logged in users. saying $r->bridge('/thedirectory'​)->to('auth#checklogin'); doesnt work.
21:19 perlite joined #mojo
21:38 daniel_ joined #mojo
21:38 daniel_ hi.  has anyone ever used mojo with a mediatemple gs?
21:39 njlg_ joined #mojo
21:41 daniel_ joined #mojo
21:42 njlg_ joined #mojo
21:42 njlg_ joined #mojo
21:43 tolid_air joined #mojo
21:44 daniel_ "CGI is supported out of the box and your http://mojolicio.us/perldoc/Mojolicious application will automatically detect that it is executed as a CGI script."  that could be all i need
21:44 * daniel_ tries
21:46 rem_lex| joined #mojo
21:47 tabbi left #mojo
21:56 njlg_ joined #mojo
21:57 vervain joined #mojo
21:59 lukep joined #mojo
21:59 mollusk joined #mojo
22:02 mollusk Is there a good or preferred way to mix Mojolicious apps with plain socket connections?  I made a basic app starting with POE that listens for socket connections and takes text input, then I made it into a Mojolicious app which is super, but I'd like to listen for POE style connections without forking
22:04 sri mollusk: you would have to fix POE::Loop::EV first
22:05 sri aside from that it's no big deal http://mojolicio.us/perldoc/Mojoli​cious/Guides/Cookbook#Event_loops
22:05 mollusk thanks for the reference!
22:06 sri should someone fix POE::Loop::EV i will add a recipe for it to the cookbook too
22:08 sri someone remind me to punch xantus next time he shows up, apparently he has a maint bit for POE::Loop::EV :o
22:08 mollusk for me Mojo::IOLoop and IO::Async as documented are probably just fine, but I'll take a look at POE::Loop::EV if I get a chance
22:29 mollusk left #mojo
22:37 daniel_ joined #mojo
22:50 Averna joined #mojo
23:00 jzawodn joined #mojo
23:03 good_news_everyone joined #mojo
23:03 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/5a0NrA
23:03 good_news_everyone [mojo/master] fixed small multipart bug - Sebastian Riedel
23:03 good_news_everyone left #mojo
23:08 good_news_everyone joined #mojo
23:08 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/mHSIBQ
23:08 good_news_everyone [mojo/master] fixed a few test descriptions - Sebastian Riedel
23:08 good_news_everyone left #mojo
23:17 mandreacchio joined #mojo
23:44 davido_ joined #mojo
23:45 vervain sri: 5000 good tests 0 bad.  And this was while working all day so the system was under various load ( although not heavy )
23:45 vervain What me to load it up over night and see if the tests are affected?
23:45 sri vervain: interesting, what os?
23:45 sri 5000 should have been more than enough
23:46 vervain Debian squeeze: Linux albus 2.6.32-5-amd64 #1 SMP Sun May 6 04:00:17 UTC 2012 x86_64 GNU/Linux
23:46 * sri is on os x with perl 5.16.1
23:46 vervain I'm on perl 5.14.2
23:47 sri maybe it's some mountain lion specific bug
23:47 sri also the first time i've compiled perl with clang
23:48 vervain Oh... and Intel(R) Core(TM) i7-2640M CPU @ 2.80GHz
23:55 zivester joined #mojo

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