Camelia, the Perl 6 bug

IRC log for #mojo, 2012-03-04

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

All times shown according to UTC.

Time Nick Message
00:13 plu joined #mojo
00:15 sri joined #mojo
00:37 mire joined #mojo
01:37 jnap joined #mojo
02:40 ispy_ joined #mojo
02:43 jnap joined #mojo
02:52 jnap joined #mojo
04:24 tempire ping
06:02 kaare joined #mojo
06:54 mire joined #mojo
07:16 tempire I really hate LWP
07:17 tempire I know *why* this is here, but it makes me want to die a little: http://mojolicio.us/perldoc/LWP
08:36 tempire PING
09:16 kitt_vl joined #mojo
09:35 trek1s joined #mojo
09:39 crab what are you pinging?
09:41 d0as8 joined #mojo
10:00 sri PONG
10:01 d0as8 left #mojo
10:02 jr joined #mojo
10:04 sri tempire: http://www.washingtonpost.com/blogs/ezra-klei​n/post/why-an-mri-costs-1080-in-america-and-2​80-in-france/2011/08/25/gIQAVHztoR_blog.html # the free market will solve it!
10:10 d4rkie joined #mojo
10:25 vervain That one's tricky... there's an other reason for it too...
10:26 vervain US clinics will often sell on their machines every few years and buy the latest technology.  Guess where they get sold to.
10:30 tempire if the us had a free market health care system, it would
10:33 vervain Ya... I certainly can't choose my doctor/service.
10:36 tempire hmm
10:37 * tempire wishes he knew more about the irc protocol
10:57 Netfeed you really don't, it's a terrible protocol
11:00 TheAthlete joined #mojo
11:02 jamesw it is, and everyone does it differently
11:25 TheAthlete joined #mojo
11:26 fibo joined #mojo
11:27 tempire HEY
11:27 tempire I made a thing!
11:28 vervain Like a Swamp Thing? :-)
11:28 mojocasts joined #mojo
11:28 mojocasts joined #mojo
11:28 mojocasts almost
11:29 mojocasts http://moi.tempi.re/
11:29 tempire It's intended for noobs who are scared of irc, but need halp.
11:31 tempire the twitter auth should cut down on the spammyness of the last irc thing.
11:32 tempire it also enforces use of the twitter handle as the nick
11:32 tempire which I'm hoping will encourage noobs to be more friendly and aware of their actions.
11:33 tempire it uses eventsource, so everyone should be welcome to the party.
11:33 tempire haven't tried it in IE, though.
11:38 mojocasts joined #mojo
11:46 elpontificator joined #mojo
11:46 elpontificator hello
11:46 purl hi, elpontificator.
11:47 elpontificator neat
11:55 kraih joined #mojo
11:55 kraih :)
11:56 kraih hey, that's neat
11:56 tempire the ui will let you know if a message isn't successfully sent, after a 5 second timeout.
11:59 tempire :)
11:59 vervain I get '[Unprintable]ACTION' on the login even and other '/me' type commands.
11:59 vervain Client: XChat.
11:59 tempire really.
11:59 vervain This used to happen with the other widget too.
12:00 vervain ya... so earlier when 'kraih' seems to have done '/me likes'
12:00 vervain I see '[0001]ACTION likes'
12:00 sri looks like your client has a problem
12:01 vervain Actually: kraih: '[0001] ACTION likes' instead of the usual '* kraih likes'
12:01 vervain sri: I can't disagree.. .but it works normallly.
12:01 tempire I'm a little concerned about the twitter auth.  I only kind of understand it.
12:02 tempire if anyone out there has a good understanding of oauth, I'd like to talk.
12:02 vervain OAuth is a good thing to get up to speed on... It's coming along nicely from what I gather.
12:02 tempire It's much too complicated.
12:03 kraih 2 seems reasonable
12:03 sri oh
12:03 sri one message got lost
12:03 sri "kraih: oauth1 or 2?"
12:04 tempire does it have an exclamation point beside it?
12:04 sri nope
12:04 tempire what about the progress thingy?
12:04 kraih still appears to work
12:05 tempire oh, I see.  it got mixed up with a pong response.
12:06 tempire I'm relying on events being separate.
12:06 tempire sounds like a perfect opportunity to use drain
12:06 vervain joined #mojo
12:08 mojocasts joined #mojo
12:09 mojocasts was kicked by tempire: mojocasts
12:11 stephan48 tempire: got an url for me? i can test it in IE
12:11 tempire moi.tempi.re
12:11 tempire I'm almost sure the js will fail
12:14 stephan48 SCRIPT5009: "EventSource" ist undefiniert
12:14 tempire rock on
12:14 stephan48 ie 9
12:14 stephan48 no eventsource for ie it seems
12:15 stephan48 https://en.wikipedia.org/wiki/Server-sent_events
12:15 stephan48 say no IE
12:15 tempire I should probably care.
12:15 tempire but I don't.
12:16 stephan48 ;)
12:17 stephan48 maybe some sort of browser detect: please use a decent browser
12:33 lammel2 joined #mojo
12:37 abra joined #mojo
13:30 d4rkie joined #mojo
13:57 TheAthlete joined #mojo
13:58 sri btw. we've still not decided if we want to encourage cpanm -n
13:59 sri marcus, tempire, crab: yay or nay?
14:00 sri does "curl -L cpanmin.us | perl - -n Mojolicious" work?
14:03 sri (i mean portably)
14:03 stephan48 -n?
14:03 purl i guess -n is wrapping it in "while (<>){" and "}"
14:03 vervain Yes
14:03 vervain At least on top of a fresh perlbrew 5.14.2
14:03 hshong joined #mojo
14:04 vervain -n skips unit tests
14:04 sri "curl -L cpanmin.us | perl - -f -n Mojolicious" seems pretty fast, not the best first impression
14:04 sri umm
14:04 sri not the *worst* first impression
14:04 vervain I was wondering... :-)
14:04 stephan48 mh i am not sure if that would be good
14:04 Netfeed tempire: i'd suggest cinch for ircbots if you don't mind ruby
14:04 vervain But... encouraging -n isn't a good first impression either ( to folks who know )
14:05 stephan48 maybe just run a stripped down test set(just really basic tests ensuring perl is ok and so on)
14:05 stephan48 and full test set only when you really want it
14:06 sri what needs to be tested all the time?
14:07 stephan48 idk, but no tests also seem to be false
14:07 sri why?
14:10 stephan48 i cant explain but it just feels bad to have test and omit them
14:11 sri sounds like you've got no argument whatsoever then
14:11 stephan48 wont deny that
14:11 vervain If you were to put in a super fat explanation as to why -n is an 'ok' idea then it might be acceptable.
14:12 sri my argument is that a 10s installation is pretty damn pleasant for a first time user
14:12 vervain But if you have a code snipped as an example I'd still leave the -n out $0.02.
14:12 sri vervain: explain why it isn't
14:12 vervain Texts exist for a reason.  You dont' know what the user may or may not have done to their perl environment (compile options etc)
14:13 sri developers still run tests, cpantesters still run tests, just the end user won't
14:13 vervain They may have some totally wacked out frankenperl that your tests wouldn't pass in even though they pass e.g. cpan testers.
14:13 stephan48 vervain: thanks imeant something like that but failed in express it somehow
14:14 sri tests may or may not detect a frankenperl, for 99.999999999% that doesn't matter
14:14 sri should we run tests because one user may encounter a frankenperl?
14:14 ispy_ joined #mojo
14:14 vervain Then there's the encouraging '-n' just isn't 'the done thing'
14:16 sri i've not gotten a frankenperl report yet btw in all my years releasing cpan modules
14:17 hshong` joined #mojo
14:19 sri and you do realize that cloud installers using cpanm all do -n?
14:20 sri we won't just have perl developers installing mojolicious, ideally there will be many more folks bootstrapping mojolicious for some webapp
14:21 sri there is no way random devops can do something useful with failing tests
14:22 baton8 joined #mojo
14:23 sri teh few people that can send actually useful bug reports are the ones that know to leave out -n from the suggested oneliner
14:24 sri vervain: how does the majority benefit from leaving out -n?
14:25 vervain I personally just feel it's bad to encourage newbies to do such things... unless they fully understand why it many be acceptable in this case'
14:26 vervain Or Mojolicious needs to figure out how to enable tests for devs/cpanm but disable them for cpanm.
14:26 sri you're just assuming they understand test failures?
14:26 sri how do we know they won't just throw away mojolicious and give up?
14:26 vervain No... I'm assuming that they should see them if they exist.
14:26 vervain You don't.
14:26 vervain But catering to children isn't wise either $0.02.
14:27 sri beginners != children
14:27 vervain btw -= childred !<-> age
14:27 vervain :-)
14:27 stephan48 when tests fail how do you know they end with something useable and dont throw it away?
14:27 stephan48 when tests are omitted*
14:27 sri judging by the small number of reports we get i think we can assume that most just give up
14:28 vervain Or that -n will lead to them falling down some rabbit hole later because it wasn't caught earlier.
14:28 vervain I thought that -n was to 'speed' things up not avoid failed tests?
14:28 sri but then they contact us because they've already invested in mojolicious
14:28 sri and we can tell tem to run tests
14:28 vervain Do you get failed tests in the wild often?
14:28 sri we don't know
14:29 sri that's the point
14:29 sri during the last few years we've not gotten a single one that wasn't on cpantesters
14:30 sri some duplicates, but that's it
14:32 sri vervain: what makes you believe a newbie would contact us after the installation oneliner fails?
14:33 sri i for one have no evidence whatsoever for that argument
14:33 vervain Maybe I fail to relate with your target market.
14:33 TheAthlete left #mojo
14:34 sri who would use a suggested installation oneliner besides newbies?
14:34 vervain (target in this particular case that is)
14:34 sri everybody who actually knows what it does can remove -n
14:35 vervain And think about those who know... are they not going to raise a brow because it's there... I would have.
14:36 sri so we make the installation suck for newbies to look better in the eyes of old timers?
14:37 vervain The installation doesn't suck.
14:37 sri if that was the goal why don't we split up mojolicious into 50 small packages?
14:37 vervain Huh?
14:37 sri 1 minute sucks compared to 10 seconds
14:40 sri if we split up mojolicious it would look even more professional
14:40 sri installation would take 10 minutes, but fun installation is obviously not the goal ;p
14:40 Netfeed enterprise!
14:40 purl well, enterprise is the name of a CVN in the United State Navy or the name of a fictional starship in a certain popular scifi TV show or a neologism for boring business dataprocessing in ex-cobol shops. or a new Trek show or Star Trek: Crusade or a tech industry buzzword that translates to "loud as fuck" or the prestel server in london or equivalent to "bloated, overly complicated shit"
14:41 sri the only real argument we have against it right now seems to be "it looks unprofessional"
14:42 vervain I'd use 'irresponsible' not 'unprofessional'
14:42 sri irresponsible sounds like there was a real argument behind it
14:42 sri but it's just show
14:43 vervain What is just show?
14:43 sri or did i miss evidence that has proven tests on installation to serve a real purpose?
14:44 ispy_ joined #mojo
14:46 vervain I think that you hold a possition of influence... and encouraging the use of -n (to my mind) is a bad thing to influence.
14:46 sri but why? :)
14:46 pfig joined #mojo
14:47 vervain Because newbs will start using it to get around installation failures even when they really shouldn't.
14:47 sri force install happens all the time already
14:47 sri may i remind you of the countless EV reports we got ;p
14:48 sri even marcus messed that one up
14:48 sri heck, at one point i messed it up
14:49 vervain Like I originally said... put a big "This skips tests for a super speedy wicked cool install" and my particular complaint would be addressed.
14:50 sri you want -n without -n?
14:50 sri %-)
14:50 stephan48 whats so cool at 10 sec install Mojo already beats most of all other frameworks with 1min install...
14:50 stephan48 +?
14:50 purl + is a modifier, you need to give it something to modify
14:50 vervain No... I'm just saying if you want '-n' in the cut and paste on liner.. then put in the warning with significant visual authority.
14:51 sri stephan48: we can do better
14:51 sri if good enough was good enough we'd be using catalyst
14:51 stephan48 i dont see 1min as a problem and for me its perfectly ok
14:51 vervain stephan48: we are not the droids he is looking for. :-)
14:52 stephan48 ye seems so
14:53 stephan48 maybe we should just add a failing test and break a feature, somewhat simple then throw a amount of newbies at it with and without -n and see what they do upon test failure and when they see the broken feature
14:53 stephan48 (and if they care of it takes 1 min or 10 sec for mojo to install)
14:54 sri that's a flawed test
14:54 vervain Better yet... put up a curl target for |sh... anyone who trusts that get's what they get in any case... and you're not exposing newbs to -n directly. :-)
14:54 sri realistically you would compare mojolicious and say rails installation
14:54 sri rails does not run tests either
14:55 sri or express.js if you want to be hip, no tests there either
14:56 sri no newbie comparing frameworks would bother figuring out why a mojolicious test failed
14:56 sri that's the users we target with the oneliner
14:57 sri *not* perl people, just random folks looking for a web framework
15:00 vervain Is there a reason you would not be willing to put in the 'this skip the tests for convenience' note?
15:00 sri it's distracting
15:00 sri the point of a oneliner is to be simple
15:00 ispy_ joined #mojo
15:01 vervain So we are trying to cater to children... shiney objects may get in their way. :-)
15:02 sri that's the attitude that keeps perl out of the php market
15:02 vervain Here's my thing... anyone you are catering for who will be lead astray by the things this may prevent...
15:02 vervain ... will very quickly be bitten by perl itself.
15:03 vervain That's why RoR was so popular... it _actually_ constrains people... perl will never do that.
15:03 Netfeed i thought it was because a 15 min video + good marketing
15:03 sri well, there is no way to avoid those people if you want a bigger marketshare
15:04 vervain Well... nto complete constraint... but they make it such a PITA to do anything off the rails... that it may as well be a constraint.
15:06 vervain Or maybe this is where your 'third' DSL version comes in.
15:09 black joined #mojo
15:12 fibo joined #mojo
15:14 vervain Maybe you could turn it to your advantage...
15:14 vervain Use the 'distracting' message about skipping tests as a way to brag about the extensive test suite that they can choose to skip.
15:14 vervain And tell them... You're choosing between around a minute vs around 10 seconds.
15:28 jnap joined #mojo
15:40 amirite joined #mojo
15:52 kitt_vl joined #mojo
15:56 jnap joined #mojo
16:01 crab vervain: you still haven't answered the principal question: what's the point?
16:07 marcusramberg joined #mojo
16:07 marcusramberg Ponycorns on the iPad \m/
16:07 jamesw you need to have a \cA at the end of CTCPs
16:08 crab you killed him
16:08 jamesw heh
16:11 sri moar ponycorns
16:11 * sri kinda likes the netty architecture
16:13 sri https://github.com/netty/netty # the java overabstraction is quite insane and makes it impossible to find anything
16:13 sri but i like tha channels and codecs concepts
16:15 sri something like that would be very nice for useragent and daemon
16:16 sri at least channels
16:20 marcus sri: guess you know my position on -n
16:21 sri marcus: you're in favor of it?
16:21 marcus sri: yes
16:21 sri i think tempire actually came up with the idea
16:21 sri and crab didn't say anything yet i think
16:24 vervain crab: surely sticking with convention doesn’t' _need a point_... it's the contra that needs the point... when why it's a question at all.
16:24 sri the contra has a point, that's the point :)
16:25 vervain And if you express that point to the reader then I don't have a complaint.
16:27 crab surely sticking with convention doesn't need a point? wtf?
16:28 crab so you'd just keep doing the same stuff the same way forever without asking why?
16:28 vel joined #mojo
16:29 vervain no... and I'm not saying that... but perl modules have standards... one of which is to run tests on installation.
16:29 vervain If there wasn't a quesion about it being right this conversation would have never happened.
16:30 sri that's not true either
16:30 crab vervain: do you know how many people install perl modules using debian or rpm packages?
16:30 sri distributions package modules, active perl packages anyone?
16:30 crab do you know how many documents recommend that over installing from CPAN?
16:31 crab there are plenty of people who actually refuse to install modules from CPAN and if they need something that there isn't a package for, they make a package with dh-make-perl or whatever.
16:31 sri i'd go as far as to say installation via cpan shell is not the norm
16:31 vervain Indeed... but I don't see the correlation with what we're discussing.
16:31 crab you said it's a "standard" to run tests on installation
16:32 vervain I know plenty of people who won't even use a module if it's not in their distribution.. that's a trust issue.
16:33 marcus vervain: yet how can they trust it if it doesn't run tests ?
16:33 marcus :>
16:33 vervain GDAmmit... you KNOW what I'm saying... it's standard to run tests when installing modules wiht cpanm!!!! f)*)($*#!
16:33 crab there's solid experiential evidence that the faster and easier the installation process is, the further people get with looking at it the first time
16:33 marcus vervain: http://www.openswartz.com/2012/01/​31/stop-running-tests-on-install/
16:33 crab vervain: what benefit do we derive from it?
16:34 vervain Isn't there a way to use an '_' for version numbers that get's you cpantester tests but then you can remove tests on 'release' versions?  I thought I read something like that somewhere.
16:34 marcus vervain: miyagawa actually wanted to make cpanm -n the default.
16:34 sri marcus: oh, that blog post is good
16:35 crab yes, it is
16:36 crab most test suites aren't written to find real problems on foreign systems at install time anyway
16:36 crab and 99 times out of 100 in my experience, a test failure from a new user in such circumstances gives you no useful information on what went wrong
16:36 vervain I agree with all of this... I think there should be a way to not run them locally as well... Just tell the damn user that's what you are doing and why.
16:36 crab why?
16:37 vervain I'm sorry...but I find this unbelievable... what vanity!
16:37 crab i strongly believe that every _word_ leading up to installing and trying something (not just mojo, anything) should pull its weight.
16:37 crab vanity?
16:37 purl Sors immanis et inanis, rota tu volubilis, status MALUS, vana salus semper dissolubilis, obumbrata et velata michi quoque niteris; nunc per ludum dorsum nudum fero tui sceleris. or what people who dont get any resort to to look cool
16:38 vervain I personaly believe you are afraid of what the works 'skipping tests' will say to the user.
16:38 vervain You don't trust the user.
16:39 vervain Which is why I don't personally know who is being catered to here.
16:39 GitHub40 joined #mojo
16:39 GitHub40 [mojo] kraih pushed 1 new commit to master: http://git.io/udrAhA
16:39 GitHub40 [mojo/master] recommend faster installation - Sebastian Riedel
16:39 GitHub40 left #mojo
16:39 Netfeed the admin at work refuses to install stuff from cpan, everything must go via apt-get
16:39 crab maybe if you actually listened to what people told you, you would know
16:40 sri lets see what happens
16:40 crab but i'm not very motivated to continue with this conversation after you started throwing insults around.
16:45 metys quit
16:45 metys f**k, sorry...
16:46 sri sudo make me a sandwich
16:47 crab metys: now you have to stay around and entertain us to atone for your mistake.
16:51 vervain I apologize... apparently the feeling of 'not being heard/understood' is mutual.
16:51 metys crab: I would like to, but currently I don't have time for that, sorry :-)
16:52 * crab cries bitterly
16:54 marcus sri: btw, wouldn't it be better to do that as 'curl -L cpanmin.us | sudo perl - -n  Mojolicious' ?
16:55 sri marcus: would it?
16:55 marcus less typing, and curl don't need root privileges.
16:55 * sri shrugs
16:56 vervain On perlbrew installs sudo perl may not run the right one
16:56 sri we had it without sudo for a few months i think, which caused more problems
16:56 sri but i wouldn't mind trying again
16:56 marcus hmm, maybe I should buy a https certificate for cpanmin.us, and we could recommend curl https://cpanmin.us  for protection again man in the middle attacks.
16:56 sri :D
16:56 marcus sri: note that I added the sudo to perl command
16:57 sri ooh
16:57 sri nono
16:57 sri that's a problem then
16:57 marcus why? Works fine here.
16:57 sri yea, but it asks for the password much later
16:57 marcus only problem I can see is if you don't have sudo
16:57 sri without context
16:57 purl well, without context is always priceless
16:57 marcus sri: yes, once curl is done
16:58 marcus not *much* later.
16:58 sri it's rather unintuitive to be asked for a password out of nowhere
16:58 sri i think cpanm even has a sudo flag we could theoretically use
16:58 marcus sri: yes, that asks for sudo for installation only
16:58 sri -S
16:59 crab we should make "curl https://mojolicio.us|sh" work.
16:59 marcus curl -L cpanmin.us | perl - -n -S  Mojolicious
16:59 crab or something like that, anyway.
17:00 vervain crab: lol... I suggested that earlier.
17:00 marcus actually, it's up to miyagawa now
17:00 marcus he now hosts the redirect
17:01 crab get.mojolicio.us
17:02 sri what should the script look like?
17:02 sri shebang+marcus oneliner?
17:03 marcus you could probably do user agent detection for curl ;)
17:03 marcus sri: yeah, makes sense.
17:11 sri marcus: http://get.mojolicio.us/ :)
17:12 sri "curl get.mojolicio.us | sh" <- works
17:13 vervain And in my opinion that's a win.  Those who know will understand the catering.. those who don't or don't care get what they want.
17:13 GitHub122 joined #mojo
17:13 GitHub122 [mojo] kraih pushed 1 new commit to master: http://git.io/7sxkOg
17:13 GitHub122 [mojo/master] simplified installation oneliner - Sebastian Riedel
17:13 GitHub122 left #mojo
17:18 Foxcool joined #mojo
17:18 pfig joined #mojo
17:19 sri weird
17:19 sri i try to tweet the oneliner and get "forbidden" errors
17:29 vel joined #mojo
17:57 brambles joined #mojo
18:11 dog_chocolate joined #mojo
18:13 dog_chocolate HI, these mojolicious::sessions cookies, it seems a bit unsecure storing info on the client, does anyone know if the signing of the cookie factors in the expiry date so that the client cannot keep unexpiring their own cookie?
18:22 dog_chocolate To log someone in I'm effectively following one of the mojo examples I;ve seen.  If they successfully authenticate then drop the user_id into a session cookie with a reasonable expiry date.  Then if on each request they can present me with a valid session cookie with a user_id field I am considering them auth'ed for that user_id.
18:22 dog_chocolate It feels a bit dirty.
18:23 dog_chocolate On the face of it I may be worrying about nothing, the only hole I can see is if a user keeps extending the expiry to keep themselves logged in.
18:24 dog_chocolate Or if the cookie's eavesdropped but that would apply even if the logged in status were stored server side.
18:44 sri expiry date is stored and signed in the cookie as well
18:44 sri a user can't extend it without knowing the secret
18:45 dog_chocolate It sounds secure, I'm probably worrying about nothing
18:46 sri https://github.com/rails/rails/commit/b​83965785db1eec019edf1fc272b1aa393e6dc57 # ohoh, teh dramaz!
18:46 marcus mm
18:46 marcus was reading that as well
18:48 xaka joined #mojo
18:52 sri http://homakov.blogspot.com/201​2/03/egor-stop-hacking-gh.html # damn
18:53 marcus #1 & #2 on hackernews. ;)
18:55 xaka joined #mojo
19:23 tempire I support the curl | sh install.
19:23 tempire If we're going to go simple, let's go all the way.
19:24 sri wat?
19:24 purl Watt?  Watt?  I can't hear you, there's this buzzing noise where my brain should be.
19:24 tempire Old timer perl people are going to complain no matter what we do, because we use too many rainbows.
19:24 sri it's already done :)
19:24 sri curl get.mojolicio.us | sh
19:25 tempire Rock on.
19:25 purl The rock activates and lumbers straight over you. LOSE 10 HIT POINTS.
19:25 sri not sure about curl detection
19:26 sri it would mean that the website would be inaccessible from curl
19:26 tempire Can't we use parallel tests with this method?
19:27 sri we could
19:27 tempire I'm on an iPad, otherwise I would try out the time diff
19:29 tempire We need to cover windows users as well.  Do they have a curl-like command?
19:34 sri with cygwin they do i suppose
19:34 sri there are also prebuilt packages for activeperl
19:35 sri and strawberry users usually know what they want and don't need a oneliner
19:36 tempire good enough, then
19:36 sri (if strawberry is still alive)
19:36 tempire I thought strawberry was the new hipness
19:36 sri it's been a year and they have no 5.14 i think
19:36 tempire ah
19:36 sri naah, activeperl has been catching up i heard
19:36 tempire wait, that's not true
19:36 tempire 5.14.2
19:37 sri ah, must be new
19:38 sri argh
19:39 sri TWITTER Y U RUIN MY ONELINER!
19:39 tempire you gotta quote it
19:40 sri how?
19:42 tempire https://twitter.com/#!/elpontificator
19:42 tempire without tests, I can install mojolicious in 6.574 seconds
19:42 tempire with parallel tests, 15 seconds.
19:43 tempire 6 seconds almost feels like something went wrong.
19:43 sri tempire: the link is still shortened
19:44 tempire the link itself, but not the display
19:44 sri depending on client
19:45 sri some clients show the shortened link
19:45 sri which doesn't work
19:45 tempire wellst
19:45 tempire then they are broken.
19:45 tempire problem solved!
19:45 purl glad i could help
19:45 sri :D
19:45 tempire which clients?
19:46 tempire it works in the web and yoriofurofurofo
19:46 sri facebook forward, for example
19:48 sri damn, 6 seconds is fast
19:50 sri 10 seconds without tests and 21 with here
19:50 tempire it's like you're in 2007
19:50 sri :(
19:51 * tempire gives sri 3 seconds
19:51 sri \o/
19:51 sri must be the sea cable between here and the cpan
19:52 tempire 4 seconds of sea cable
19:52 tempire sea cabal tax
19:52 sri :/
19:53 tempire I love that github thread
19:53 tempire it really is like reddit invaded
19:54 tempire the real problem I have with no tests is that it puts everyone on a level playing field, and downplays how awesome a 15/20 second installation is *with* tests.
19:55 tempire that's not a real reason against it.
19:55 tempire but it does make me sad
19:55 * sri agrees
19:55 * sri cries
19:57 tempire the internet seems to agree that xchat is broken in its incorrect displaying of /me
19:57 tempire just not sure how, though.
19:57 tempire since limechat /me displays just fine
19:58 sri didn't someone earlier say that there was a char missing or so?
20:06 tempire ok, fixed.
20:06 sri the rails bug is a good reminder that too many protections do in fact give a false sense of security
20:07 mojocasts joined #mojo
20:07 * mojocasts has logged in from twitter, and friggin loves ponycorns!
20:07 * tempire is fixed
20:07 * sri pokes mojocasts
20:08 * mojocasts gets all these irc windows confused
20:08 malware joined #mojo
20:12 tempire ah!  /login_form!
20:12 kraih joined #mojo
20:12 * kraih has logged in from twitter, and friggin loves ponycorns!
20:12 kraih lalala
20:12 arrizer joined #mojo
20:12 * arrizer has logged in from twitter, and friggin loves ponycorns!
20:13 mire joined #mojo
20:13 * sri goes to get cookies
20:18 LoonyPandora joined #mojo
20:18 * LoonyPandora has logged in from twitter, and friggin loves ponycorns!
20:19 mojocasts joined #mojo
20:19 * mojocasts has logged in from twitter, and friggin loves ponycorns!
20:19 mojocasts test 1
20:19 mojocasts test 2
20:20 xomaa joined #mojo
20:20 * xomaa has logged in from twitter, and friggin loves ponycorns!
20:25 pfig joined #mojo
20:25 ZadYree joined #mojo
20:25 * ZadYree has logged in from twitter, and friggin loves ponycorns!
20:26 ZadYree Hai!
20:26 ZadYree Hai!
20:33 isomorphisms joined #mojo
20:33 * isomorphisms has logged in from twitter, and friggin loves ponycorns!
20:33 isomorphisms holita
20:34 tempire hai
20:34 * sri waves
20:44 lukep joined #mojo
20:45 tempire how can I modify the session from a test?
20:46 tempire guess I have to make a new cookie jar
20:53 dog_chocolate sorry, asking all these basic questions.  But is there an easy way to specify a global ep template header/footer without having to specify them on each template?
20:54 dog_chocolate Was hoping I'd find something in EPRenderer, nearest I've found is prepend/append but not sure those are the right place to be doing it
20:56 ispy_ joined #mojo
21:06 tempire dog_chocolate: I used to do that with tt
21:07 tempire but I've since changed my ways 'cause you lose flexibility
21:07 tempire 'cause there's always that one thing that you don't want to have the global header/footer
21:07 tempire and then it's a big to-do
21:19 ttyS1 joined #mojo
21:19 * ttyS1 has logged in from twitter, and friggin loves ponycorns!
21:19 ttyS1 nice
21:19 ttyS1 nice
21:20 cstamas no feedback for me ^ on the web
21:21 cstamas will the source made available? thx
21:22 dog_chocolate tempire, presumably you could detect the condition and set the header dynamically vis appends/prepends, i.e. if(cookie==mobile) {... }
21:23 dog_chocolate my main worry is that appends/preepends gets evalled or something and isn't very efficient, that there might be a more recommended way of doing it
21:23 dog_chocolate that I can;t find in the docs (still trying to wrap my head round mojolicious atm =) )
21:28 abra joined #mojo
21:43 tempire dog_chocolate: I'd recommend using layouts, if it wasn't clear
21:43 tempire cstamas: what /me message did you send?
21:45 newbie joined #mojo
21:54 mojocasts joined #mojo
21:54 * mojocasts has logged in from twitter, and friggin loves ponycorns!
22:09 cstamas tempire: the whole windows was "dark" I mean no feedback/update at all
22:10 cstamas ttyS1 was me
22:10 tempire what browser?
22:10 purl browser is broken or http://www.gfd-dennou.org/li​brary/cc-env/OLD/httpd_cern/
22:10 cstamas firefox 10.0.2
22:11 cstamas (actually iceweasel)
22:12 * cstamas trying with chromium now
22:14 ttyS1 joined #mojo
22:14 * ttyS1 has logged in from twitter, and friggin loves ponycorns!
22:14 ttyS1 here again
22:14 cstamas with chromium it is ok
22:16 ttyS1 joined #mojo
22:16 * ttyS1 has logged in from twitter, and friggin loves ponycorns!
22:25 tempire woah.
22:25 tempire new firefox inspector is uber nice
22:29 cstamas :)
22:30 * cstamas got more and more into the web business because of perl and mojolicious
22:31 cstamas but it is more of a hobby for me, irl I am a sysadmin
22:31 marcus Irl?
22:31 purl Irl is In Real Life
22:31 marcus Isn't this real? :-/
22:32 sri yes it is neo
22:32 sri now go back to your cubicle
22:32 * marcus eats some random pilla.
22:33 marcus Pills even.
22:33 cstamas :)
22:33 marcus Bought hunger games as an audio book today.
22:34 jnap joined #mojo
22:36 * tempire consumes the pilla
22:43 jnap joined #mojo
22:59 batman joined #mojo
23:11 marty hmm.  Does anyone know how to call a template block from within an included file?  Ie: the template block is defined in one file, then a second file is %= include(d) and we try to call the template block from the included file.  Hope that makes sense.
23:30 newbie joined #mojo
23:36 ispy_ joined #mojo
23:36 jnap joined #mojo
23:43 newbie joined #mojo
23:46 jnap joined #mojo
23:57 mojocasts joined #mojo
23:57 * mojocasts has logged in from twitter, and friggin loves ponycorns!
23:57 mojocasts ok, moi works in firefox now
23:57 sri mojocasts++
23:57 mojocasts also, the lost message problem that sri had earlier is solved

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