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

IRC log for #mojo, 2015-02-22

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

All times shown according to UTC.

Time Nick Message
00:21 bobkare joined #mojo
00:26 bobkare_ joined #mojo
00:48 bobkare joined #mojo
00:54 bobkare joined #mojo
01:45 stephanj joined #mojo
02:08 cpan_mojo Mojolicious-Plugin-DSC 1.001 by Krasimir Berov - http://metacpan.org/release/BEROV/Mojolicious-Plugin-DSC-1.001
02:09 bobkare joined #mojo
02:31 good_news_everyon joined #mojo
02:31 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/AEjh
02:31 good_news_everyon mojo/master afbfefe Sebastian Riedel: another one-liner
02:31 good_news_everyon left #mojo
02:32 bpmedley As an fyi, if you want to add voice or sms to your application, then Mojolicious and Twilio are a match made in heaven.. :)
02:35 crab do tell
02:36 bpmedley Let’s say you have an ecommerce website and want to send a TXT or voice snippet when an item ships…
02:37 bpmedley https://www.twilio.com/sms/api
02:37 bpmedley As you can tell, the curl command can be easily converted to a Mojolicious or Mojolicious::Lite app.
02:39 bpmedley That may be too sporadic.  Want me to create a Mojolicious::Lite app with a REST api for sending a TXT?
02:41 bpmedley https://www.twilio.com/docs/api/rest/making-calls#example-1  <— the curl command initiates the call .. and then the TwiML routes/handles the call once connected.
02:43 bpmedley It’s a different paradigm for thinking of voice calls.  A POST request to dial and some XML to control the call flow.
02:45 bpmedley Does that make sense?
02:45 klapperl_ joined #mojo
02:45 jberger bpmedley: cool
02:46 jberger do they have a trial or demo or something?
02:46 bpmedley http://www.perl.com/pub/2011/12/building-telephony-applications-with-perl-and-twiml.html  <— More depth and uses Mojolicious.. :)
02:46 bpmedley I’m not sure; however, they do have the abliity to use a Sandbox app for testing, I believe.
03:17 Adura joined #mojo
03:35 kryptonite joined #mojo
03:36 good_news_everyon joined #mojo
03:36 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/AuL4
03:36 good_news_everyon mojo/master 0e272be Sebastian Riedel: handle constructors consistently
03:36 good_news_everyon left #mojo
03:42 good_news_everyon joined #mojo
03:42 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/AuLd
03:42 good_news_everyon mojo/master 7c6d324 Sebastian Riedel: better template examples
03:42 good_news_everyon left #mojo
03:46 aghbas joined #mojo
03:56 * tempire uses twilio
03:56 noganex_ joined #mojo
03:58 bpmedley tempire: Cool, what app?
03:59 tempire Replaces google voice for incoming and outgoing texts, and incoming calls
04:00 bpmedley Nice
05:44 bobkare joined #mojo
05:47 good_news_everyon joined #mojo
05:47 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/Au4J
05:47 good_news_everyon mojo/master cd309ef Sebastian Riedel: make Morbo restart a little faster
05:47 good_news_everyon left #mojo
05:49 sri i actually wanted to expose the daemon instance with a public accessor... so the morbo script can use $morbo->daemon->listen(\@listen) instead of $ENV{MOJO_LISTEN}
05:50 sri but correct behavior would be really hard to test... and i remembered recent differences with jberger about documented but untested code :S
05:51 sri so the morbo script stays ugly
05:53 sri i don't like this situation, jberger has weakened the rules
05:53 sri either we stick with only stuff that has tests and documentation is a feature, and i keep documenting very liberally... or we only we switch to everything that's documented is a feature... and i document *a lot* less
05:54 sri jberger: this one is primarily for you
05:55 sri (actually i would rip out quite a bit of documentation)
05:58 Grinnz sri, my opinion would be keep as it is, but indicate in the docs which are not considered features (mark experimental?)
06:02 Grinnz the main sticking point was that people use features they see in the docs, but then if they aren't really considered features they have no way to know unless they look through test coverage or something
06:08 crab i think "only stuff that has tests and documentation is a feature" makes sense, but maybe i don't understand what we're talking about
06:18 sri Grinnz: nobody is going to go through the docs and search for untested stuff
06:19 sri it's simply unrealistic, people are lazy
06:20 sri only reason i document everything i come across right now is because there are little consequences to getting it wrong
06:20 Oleg joined #mojo
06:24 sri the real decision is if we want more docs, that might be wrong, or less docs that have a very high chance of being right
06:25 sri some specific things that come to mind would be daemonization and user/group switching
06:26 sri untestable, and therefore in question
06:26 sri those are just the most obvious, controlling server with signals would certainly be on the chopping block too
06:35 good_news_everyon joined #mojo
06:35 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/Auuo
06:35 good_news_everyon mojo/master c4cb9c7 Sebastian Riedel: mention Morbo signals
06:35 good_news_everyon left #mojo
06:35 sri like this for example
06:37 sri maybe the QUIT signal should be used for something better, i don't know, i document liberally because that's how it happens to work right now... without making guarantees
06:38 sri now, after jberger brought it up the problem has entered my thought process, and i don't like the uncertainty
06:39 sri jberger, crab, tempire, marcus, batman: so i'd like a vote, tests+docs=feature or docs=feature
06:39 sri (and keep in mind, i will document *a lot* less if it is the latter)
06:41 sri oh, and you have to find a solution for features in limbo, like the mentioned daemonize/setuidgid
06:42 kaare joined #mojo
06:42 sri for those unaware, jberger brought it up recently when i removed the accept mutex from the prefork server
06:43 sri and suggested such a big removal should happen in a major release
06:44 sri my point was of course that it was never a tested feature, his point was there are attributes/methods/options tied to it, which were documented and may be in use out there
06:44 sri and i'm talking to myself now :S
06:45 Mikey haha i'm combing through my code seeing if im impacted
06:46 Mikey i don't even remember why i did this but i monkey patch in a new _spawn altogether.
06:47 sri the situation with daemonize/setuidgid is actually really shitty, i've proposed recently to remove both since they are untested, and even here some folks were all "NONONONO I NEED DIS!!!1"
06:47 sri and to top it off, it's a very security relevant feature
06:48 sri with no hope of getting tests
06:50 Mikey i admire the philosophical purity of this project, but my practical side is kinda going 'wtf', i am with the "NONONO I NEED DIS!!!" group
06:51 sri Grinnz: think about that, marking a security feature experimental forever
06:52 sri well
06:52 sri </rant>
06:52 Mikey just cos it doesn't have tests?  are you going to mandate people run their tests as root?
06:52 Grinnz sri, indeed... but it's equally scary if it can be removed without notice
06:55 sri Mikey: i use linux capabilities
06:56 sri Grinnz: so what would you do?
06:56 Grinnz sri, perhaps just hold them to the same deprecation policies as tested features
06:57 sri "all options suck equally, i'll let sri decide and complain" ;p
06:58 Mikey lol
06:58 bobkare joined #mojo
06:58 dod joined #mojo
06:58 sri Grinnz: "them"?
06:58 purl "them" are Tony and Marty.
06:59 Grinnz sri, the untested methods/functionality
06:59 Grinnz that are documented
06:59 sri so, you say everything with documentation should be considered a feature?
06:59 Mikey maybe just call them 'untested', a new class of feature.
07:03 dod joined #mojo
07:09 good_news_everyon joined #mojo
07:09 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/AuaD
07:09 good_news_everyon mojo/master ec5c5cb Sebastian Riedel: use a little less code for keep_alive_requests check
07:09 good_news_everyon left #mojo
07:14 sri Grinnz: you may think that would result in what we have now with more backwards compatibility
07:14 sri but the reality is that less stuff would get documented
07:16 sri no docs for signals for example, first thing i would do
07:17 reneeb joined #mojo
07:21 davido___ joined #mojo
07:48 reneeb joined #mojo
08:06 irq joined #mojo
08:06 bobkare joined #mojo
08:07 sri hmm, the long version of -i is called --inactivity instead of --inactivity-timeout in the daemon and prefork commands, is that a typo?
08:10 melo joined #mojo
08:14 Vandal joined #mojo
08:16 bobkare joined #mojo
08:26 Gedge joined #mojo
08:38 cpan_mojo Mojo-HelloMojo 0.03 by Stefan Adams - http://metacpan.org/release/SADAMS/Mojo-HelloMojo-0.03
08:46 tempire I don't think it's practical to say that everything MUST be tested or be eradicated.
08:46 tempire For the sysadmin parts in particular.
08:46 tempire Even tests aren't guaranteed to be perfect.
08:47 tempire Since tests are always a best effort, the policy should reflect that.
08:48 tempire I do like movement in the direction of "not a feature if it's not tested", but that's a direction to always be moving in, not a constant state.
08:49 damaya joined #mojo
08:54 damaya joined #mojo
08:58 damaya exit
08:59 damaya joined #mojo
08:59 damaya alright, damn tmux
08:59 damaya Anyone have any deployment methods set up between git and hypnotoad / morbo?
09:00 damaya E.g., a push to master deploys app to production server, push to dev branch deploys to dev servers.
09:00 damaya Just trying to get a feel for what methods others might use. I was looking into Capistrano, but I absolutely hate, hate, hate Ruby.
09:03 bpmedley https://www.digitalocean.com/community/tutorials/how-to-use-git-hooks-to-automate-development-and-deployment-tasks  <— Looks promising, although lengthy.
09:04 bpmedley Would you consider a shell script or Mojolocious::Command to do the updates?
09:07 damaya I am using Gitlab, and for some reason a hook in one repo affects all of my repositories.
09:07 damaya And yes, I do know Gitlab is Ruby, but sometimes exceptions are made :D
09:08 damaya bpmedley: I'd definitely consider either.
09:08 bpmedley http://gitprep.org/  <— is that similar to Gitlab?
09:08 bpmedley How many servers are we talking about?
09:09 damaya Actually, it is pretty similar, but some features that are in Gitlab are missing, and I really need these features.
09:10 damaya I have 8 production servers, 2 development servers
09:10 bpmedley Any issues with using rsync + ssh keys?
09:14 damaya bpmedley, that's actually similar to my current set up, which is a post-receive hook on git server that rsyncs to remote servers, then I have a hook that restarts hypnotoad on production servers
09:14 bpmedley You can do the rsync and hypnotoad refresh in one script.
09:15 damaya Awesome, I didn't realize that. Do you have a set up like this?
09:16 bpmedley Are you familiar with how to run a remote command via ssh?
09:16 damaya yeah, -c flag
09:16 damaya well, that's how I've done it in the past... ahh man, that's a good idea!!
09:17 bpmedley You can also use a separate key that restricts the command.
09:18 bpmedley Are you sure it’s the -c flag?
09:18 damaya ohh no, I'm thinking su
09:18 damaya haha
09:19 damaya ssh user@server 'command'
09:19 damaya or I could use GNU parallel to do it
09:20 damaya parallel + ssh, all servers in one
09:20 damaya hmm
09:20 bpmedley There’s lots of ways, with your wizardr.. :)  All the hooks would confuse me, I would just use a separate script.
09:20 bpmedley *wizardry
09:20 damaya ok bpmedley, you've got me started on the right foot. Thank you so much!
09:21 damaya Right now it's a mess, and I still need to figure out what is causing Gitlab to apply a hook in one repo to all repos, because that's annoying. There's a bug for it on Github, but nobody came along saying how to fix it yet :/
09:21 bpmedley Also, somehting like: http://unix.stackexchange.com/questions/88957/finding-out-whether-rsync-modified-target
09:24 damaya I think I am going to script this out, get it all working, and write a blog post. I will credit you bpmedley and post it in here in case anyone else is interested. I should also write a blog post on my set up for init script for perlbrew + hypnotoad running as a user with no shell.
09:24 bpmedley Mojo blog posts are cool.. :)
09:25 damaya bpmedley: If you ever want a user on a development server, or something like that, let me know and I will hook you up.
09:25 damaya bpmedley: Or you want to host an app, anything like that.
09:25 bobkare joined #mojo
09:25 damaya bpmedley: Thanks for your help.
09:33 jontaylor joined #mojo
09:37 trone joined #mojo
10:32 Oleg joined #mojo
10:44 sh4 joined #mojo
10:52 Anon021 joined #mojo
10:55 sh4 joined #mojo
10:57 jberger sri: I didn't mean to stir up trouble :(
10:58 jberger I like tempire's sentiment here: "I do like movement in the direction of "not a feature if it's not tested", but that's a direction to always be moving in, not a constant state."
10:58 fhelmber_ joined #mojo
10:59 jberger not everything needs to be an exact matter of policy
10:59 jberger I encouraged removal of those methods
10:59 jberger but less painfully if possible, and it was possible
11:00 jberger if it has to codified in policy I would say "removal of public api which does not meet the standard of "feature
11:00 jberger sry
11:02 csson joined #mojo
11:03 jberger "though removal of public api which does not meet the standard of "feature" may occur be at any time, in practice this should me removed via the same deprecation policies as for features if possible."
11:04 jberger again, as I mentioned then, I'm not trying to impede progress
11:04 jberger and my first response was not to change policy or deprecate but to cut a major release
11:08 jberger "though removal of public api which does not meet the standard of "feature" may occur at any time, in the interest of compatibility they should me removed via the same deprecation policies as for features if possible."
11:08 jberger and then move on
11:09 jberger I like having lots of documentation, I think that is preferred
11:10 jberger I might encourage a note in the docs for code that we know is likely to change, if we know that in advance
11:22 amon joined #mojo
11:31 bobkare joined #mojo
11:53 kaare joined #mojo
11:57 marmez joined #mojo
11:58 marmez left #mojo
13:05 berov joined #mojo
13:26 till joined #mojo
13:29 irq_ joined #mojo
13:29 till is it possible to say: render default template for this route as if it had action "bar" WITHOUT actually calling action "bar"
13:29 till I could do $foo->get->to(template => 'foo/bar') but I don't want to hardcode the template path
13:33 pink_mist if you only want to specify which template to render in ->bar, then you need to call ->bar
13:34 pink_mist otherwise you'll want to specify it elsewhere.
13:38 till example: http://pastebin.com/7dK2QagY
13:39 till I would like to avoid the to->(template...)
13:56 romel hi guys. what's the most proper place for defining minion tasks?
14:01 jberger till: why do you want that?
14:01 bpmedley till: http://mojolicio.us/perldoc/Mojolicious/Controller#render <— See the template option, is this what you want?
14:02 jberger romel: simply stated, they need to be defined in the startup
14:03 jberger practically stated, I've been toying with where to do those lately
14:03 jberger for now mostly I have them in plugins
14:05 romel yeah, i've been thinking about ploguins too
14:05 romel thanks
14:05 romel ploguins lol
14:22 gryphon joined #mojo
14:25 cpan_mojo Perl-Box 0.03 by Gianluca Casati - http://metacpan.org/release/FIBO/Perl-Box-0.03 (depends on Mojolicious)
14:28 punter joined #mojo
14:37 n2plato joined #mojo
15:10 bobkare joined #mojo
15:17 meredith i made a small plugin that uses mojo::loader to load everything under MyApp::Task, using the last part of the package name (decamelized) as the task name, and the (coderef) result of package->new->register as the task code
15:27 meredith use Mojo::Loader;
15:27 meredith oops, heh
15:28 meredith https://gist.github.com/merrilymeredith/e1be3789c4b66435747b
15:28 meredith there we go.  idk about the interface because i literally made one actual task after doing this, but it felt more organized
15:33 jberger personally I might have had register return a hash of tasks
15:34 jberger sub register { return { task1 => \&task1, task2 => \&task2 } }
15:35 meredith yeah that'd probably be better for grouping by topic rather than file-per-task
15:35 jberger then again, at that point, it isn't hard to see how to go from there to task-specific plugins
15:35 mst sub register { return { task1 => $_[0]->curry::run_task1, ...
15:35 mst that way you still have the object to keep state in
15:37 jberger package MyApp::Tasks; use Mojo::Base 'Mojolicious::Plugin'; has tasks => sub { {} }; sub register ($self, $app) { $app->minion->add_task($_ => $self->task->{$_} for keys $self->task->%* } 1;
15:37 meredith i'll clearly take another glance at this next time i update dependencies anyway, heh
15:37 jberger meredith: so you know, I'm actually saying this as much for me as I am for you :-)
15:39 jberger mst: an interesting though
15:39 jberger thought
15:40 meredith *nods*  it's something that i only did in a pinch because i didn't like the idea of stuffing more add_task lines in app setup somewhere
15:40 Mikey joined #mojo
15:40 meredith currying would be handy too
15:41 * meredith afk for like 9 hours -- flying DCA -> LAX
15:44 mst meredith: http://p3rl.org/curry works fine with Mojo
15:48 user_8524 joined #mojo
15:49 meredith oh, handy
15:49 meredith i feel so functional
15:51 mst meredith: I also wrote http://p3rl.org/Object::Tap for when you're using non-Mojo stuff and still want ->tap :)
15:51 mst (though Safe::Isa is still cooler)
15:54 meredith i've actually used object-tap before, but i've been using object-util lately
16:01 mst oh, huh, he's basically cloned ::Tap and Safe::Isa and then added a bunch more stuff
16:01 mst ... and then he's used B::Hooks::Parser to lexically inject them rather than the simple freaking approach so it won't fatpack
16:02 mst so now I'm probably going to have to, effectively, re-fork my own work to get a usable version
16:02 mst YAY
16:02 mst that or I need to send him a patch
16:03 mst ooh, no, it only does that if it can
16:03 mst so the behaviour changes randomly depending on your perl version and what's installed
16:03 * mst cries
16:03 mst I love tobyink's stuff to use, but there's so much crack sometimes
16:04 meredith haha, wow. never looked into it
16:04 meredith yeah he's fun, sometimes too much fun
16:05 mst plus the extra $_with_traits and $_extend stuff basically sugars a pattern I've concluded is almost always the wrong idea
16:06 mst and $_new is basically just use_module(...)->new which I think is usually clearer
16:06 mst yeah, ok, this is cute, but Safe::Isa+Object::Tap contain all the actually safe to use parts, I think
16:07 mst except perhaps $_clone though I tend to turn that functionality into a method on the classes I'm writing
16:07 mst sub but { ref($_[0])->new(%{$_[0]}, @_[1..$#_]) }
16:07 mst meredith: note, not saying you should stop using it at all
16:08 mst just personally concluding that it doesn't gain me enough over my dep-free code to engage with the crack :)
16:08 sri meredith: you might want to upgrade to 5.81 and use the new Mojo::Loader api
16:08 jberger she mentions that in her post
16:09 mst esp. since it won't fatpack at all
16:09 meredith nah i know how it is.  honestly i mostly use stuff that would be covered by object-tap and safe-isa anyway, i haven't scrutinized and it's good to
16:09 mst oh, also, 'use Devel::Dwarn;' exports a $Dwarn
16:10 mst which is equivalent to $thing->$_tap(sub { warn $_[0]->$_dump })
16:16 sri so, was --inactivity instead of --inactivity-timeout a typo in the daemon and prefork commands?
16:40 zivester joined #mojo
16:40 jberger sri: ?
16:40 * jberger looks
16:42 jberger it looks like --inactivity-timeout would be more consistent
16:43 sri yes, i don't know why it would be called --inactivity, which is why i'm asking
16:44 jberger sounds like typo to mean then
16:44 jberger *me
16:44 meredith w
16:44 jberger v
16:50 jontaylor joined #mojo
17:30 damaya_ joined #mojo
17:36 damaya joined #mojo
17:41 asarch joined #mojo
17:56 * sri wonders if there are cases where someone might want to use --multi-accept with the daemon command
18:35 alnewkirk joined #mojo
18:45 good_news_everyon joined #mojo
18:45 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/A2Pl
18:45 good_news_everyon mojo/master c5c359a Sebastian Riedel: a lot of new tests
18:45 good_news_everyon left #mojo
18:51 noganex joined #mojo
19:00 cpan_mojo Mojolicious-Plugin-DSC 1.002 by Krasimir Berov - http://metacpan.org/release/BEROV/Mojolicious-Plugin-DSC-1.002
19:20 good_news_everyon joined #mojo
19:20 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/A2FO
19:20 good_news_everyon mojo/master f7eca60 Sebastian Riedel: test cpanify command too
19:20 good_news_everyon left #mojo
19:26 Oleg joined #mojo
19:26 sri catalys keeps getting closer and closer with lines of code
19:26 sri *+t
19:27 sri mojolicious is now almost exactly 1k lines bigger than catalyst
19:34 maze joined #mojo
19:38 mst at some point I'm going to have to go in and delete half of it again
19:38 sri don't you dare ;p
19:39 sri gonna be so much fun once catalyst core is bigger than mojolicious
19:40 sri think i can cheat a little and make mojolicious 200 lines smaller
19:40 Grinnz lol
19:41 sri there is still some constant data stored in code, like http response codes
19:56 disputin joined #mojo
20:00 riche sri: not even sure it's a fair fight given it looks like catalyst depends on hella lot more than core - maybe the catalyst count should include all noncore deps
20:02 mst riche: that's not the game we're playing here
20:03 riche okay dokey
20:06 riche sri: and thanks for making fun of me running 5.14 on cygwin64.  Found a super nasty cygwin bug and a few others.  anyways now I have 5.20.1.  PEER PRESSURE WORKS
20:16 cpan_mojo Ado 0.80 by Krasimir Berov - http://metacpan.org/release/BEROV/Ado-0.80 (depends on Mojolicious::Plugin::OAuth2)
20:18 berov I am running Mojo on cygwin on WInXP too
20:19 berov perl 5.14
20:19 berov not always tough :D
20:24 mst running Web::Simple on a Nexus 7 was fun
20:24 sri riche: with deps would be too easy ;)
20:25 sri for starters, Moose core is twice as big as Mojolicious
20:26 sri literally, it's about 17k lines
20:27 sri btw. we now have 92.5% test coverage
20:36 marmez joined #mojo
21:25 Grinnz joined #mojo
21:52 sri hmm, looks like mojolicio.us is under attack from china
21:52 berov sri: 92.5% is more than enough! Above this is either very hard or nonsense
21:54 mst sri: maybe you've ended up as one of their torrent/youtube redirect targets
21:55 sri constant ssh connects, from many different ip addresses
21:55 sri all going through chinatelecom-ic-306387-ffm-b10.c.telia.net though
21:56 mst ah
21:56 preaction that reminds me, i should move the ssh port on my machine...
21:57 mst one of the funny parts of running Netcraft's SSL Survey was getting people claiming we were trying to break into their systems
21:58 mst YOU'RE THE ONE WHO PUT SSH ON FOUR FOUR THREE YOU FUCKWITS
21:58 Grinnz ... ssh on port 443? lol
21:58 preaction to get around stupid proxies probably
21:58 mst I'm ok with them doing that
21:58 mst it was the complaining when the SSL survey tested the port for openness that annoyed me
21:59 sri lol
22:00 pink_mist http://www.rutschle.net/tech/sslh.shtml <-- this is some fun tech (not something I'd put on production though)
22:24 disputin joined #mojo
22:27 tempire I need photos of mojoconf2014
22:27 tempire What is Bjornfan Pettersson's handle?
22:29 denny joined #mojo
23:11 jberger tempire: I think it is just bjornfan, and I don't see him on here much
23:11 jberger much more active on Twitter
23:18 Averna joined #mojo
23:19 sri ooooooooh, signature subs may soon be faster than normal subs :o http://code.activestate.com/lists/perl5-porters/218173/
23:20 * tempire has been using them for everything
23:21 tempire almost
23:22 preaction i like the fake signature thing too
23:23 preaction davem++
23:26 Grinnz fancy
23:27 Grinnz gonna have to start changing all my $self = shift; to my ($self) = @_; soon :P
23:29 sri only 688 results for "= shift;" in mojolicious :S
23:29 Grinnz hehe
23:30 sri damn... that 688 lines gone with signatures :O
23:30 sri *+is
23:30 Grinnz well, when you bump the perl req to 5.22 :P
23:30 sri still experimental in 5.22
23:30 Grinnz 5.24!
23:31 sri we can have an experimental fallback with 5.20+ though
23:31 sri Mojo::Base would enable it
23:32 sri i'm a little disappointed that there is still no polyfill module on cpan :(
23:32 sri or we would already be using signatures
23:36 riche jberger: are you still considering act
23:36 riche i mean mct
23:36 riche or was it just a fleeting thought
23:40 absolut_todd joined #mojo
23:51 gryphon joined #mojo
23:52 punter joined #mojo

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