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

IRC log for #mojo, 2014-09-11

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

All times shown according to UTC.

Time Nick Message
00:03 bobkare joined #mojo
00:13 hesperaux joined #mojo
00:14 neyasov_ joined #mojo
00:21 KindOne joined #mojo
00:29 woz joined #mojo
00:34 KindOne joined #mojo
00:35 sri hmm, i was just looking into turning render_not_found and render_exception into helpers... but that won't work so well
00:36 sri since people will have custom controller classes with those methods overloaded, which means i can't actually free up those method names and have to keep all internal uses to $c->render_exception/$c->render_not_found
00:37 sri so the result would be a mess
00:37 sri this has to be done in a major release as a breaking change i believe
00:38 woz joined #mojo
00:43 d4rkie joined #mojo
00:46 miked joined #mojo
00:47 sri jberger: looks fine
00:56 sri pretty simple patch, not sure if it's the right thing to do though https://gist.github.com/anonymous/77488e9da95f583d2728
00:57 sri the ability to redefine exception handling with "helper render_exception => sub {...};" is neat, but the method would have to stay hidden, even though it doesn't exist anymore
01:06 Guest joined #mojo
01:06 klapperl_ joined #mojo
01:08 Averna joined #mojo
01:10 sri perhaps this is an opportunity to do something radical :)
01:15 neyasov_ joined #mojo
01:27 sri since we have no examples for nested helpers in core yet
01:27 sri https://gist.github.com/anonymous/022d60d1103afb8942c4
01:28 sri this might be an opportunity ;p
01:28 sri $c->reply->not_found
01:28 sri $c->reply->exception(...)
01:29 sri "reply" is an arbitrary choice, could just as well be $c->respond_with->not_found
01:30 sri point is, the render_exception/not_found methods stay as placeholders, we get nested helper examples, and those helpers are very easy to replace for the whole app
01:32 * sri pokes jberger, tempire, marcus, batman and crab
01:36 jberger I get the idea
01:36 tempire hmm
01:36 * tempire feels like a pandora's box was just opened
01:36 jberger Do you eventually remove the controller methods then?
01:36 jberger At a breaking release for instance?
01:36 sri dunno, can only be done in a major release though
01:37 jberger I'm not sure i like the long term look of a core class method calling to a plugin, even if it is a core plugin
01:38 sri not sure either
01:39 jberger But in the short term I'm ok with that
01:39 sri it's better than turning those two methods into helpers though ;p
01:39 sri should we decide the feature is worth it
01:40 jberger Especially if there are more things that a helper namespace might do now that you don't have to worry about namespace collisions
01:42 sri that reminds me to prefix those helper calls in Mojolicious::Controller
01:45 sri sub render_exception { shift->helpers->reply->exception(@_) }
01:45 sri now you can even have a reply method in your controller and everything works
01:45 good_news_everyon joined #mojo
01:45 good_news_everyon [mojo] kraih created reply_helpers (+1 new commit): http://git.io/--sXzA
01:45 good_news_everyon mojo/reply_helpers 8ec15ea Sebastian Riedel: added some nested helpers as examples
01:45 good_news_everyon left #mojo
01:45 sri better to look at in a branch
01:45 sri love that we can fully qualify helpers with $c->helpers now, makes using them in core much less risky
01:47 sri also not sure yet if nested helpers should be documented as "reply->exception" or "reply.exception"
01:47 sri for the =head2
01:58 mattastrophe joined #mojo
02:15 neyasov_ joined #mojo
02:29 laouji joined #mojo
02:46 Eke- joined #mojo
02:53 noganex joined #mojo
03:05 woz joined #mojo
03:05 basic6_ joined #mojo
03:07 sh4 joined #mojo
03:10 sujithm joined #mojo
03:24 neyasov_ joined #mojo
03:27 denny joined #mojo
03:35 woz joined #mojo
03:54 neyasov_ joined #mojo
04:00 neyasov_ joined #mojo
04:13 tempire sri: reply->exception
04:15 woz joined #mojo
04:27 irq joined #mojo
04:36 neyasov_ joined #mojo
04:46 woz joined #mojo
04:58 jamesaxl joined #mojo
05:14 laouji joined #mojo
05:21 preaction joined #mojo
05:21 preaction_ joined #mojo
05:43 sujithm joined #mojo
05:44 neyasov_ joined #mojo
05:51 sugar joined #mojo
05:54 Eke- joined #mojo
06:04 tempire BTDUBS
06:04 tempire I love the idea of embracing nested helpers more and more
06:07 D4RK-PH0ENiX joined #mojo
06:28 basiliscos joined #mojo
06:28 mr-foobar joined #mojo
06:30 dod joined #mojo
06:32 dod joined #mojo
06:32 batman jberger: seriously? how could you vote +1 and still wonder about your "special" emit_safe() case?
06:35 batman i think emit() is a very bad idea, since it might break the flow of the emitter.
06:35 batman now i have to do eval { $self->emit(...); }; all over the place where i need to make sure i execute the rest of the code
06:36 preaction joined #mojo
06:36 batman i think of events mostly as a decoupled thing: something broadcast a message, and it's up to the receivers what they do with it.
06:37 batman i don't care if one of the receivers go and kill them self, and they don't have to worry what i'm going to do next.
06:38 batman with this change, the coupling between receiver and sender is very tight. it's like a instead of screaming, i'm throwing them all a hand grenade while we're all inside a bus.
06:42 KCL joined #mojo
06:45 neyasov_ joined #mojo
06:45 neyasov__ joined #mojo
06:49 woz joined #mojo
06:55 dp_ joined #mojo
06:58 neyasov__ joined #mojo
06:59 rawler joined #mojo
07:15 sujithm joined #mojo
07:24 trone joined #mojo
07:42 woz joined #mojo
07:48 irq joined #mojo
07:50 denis_boyun joined #mojo
07:50 Vandal joined #mojo
07:52 trone joined #mojo
07:59 cpan_mojo Mojolicious-Plugin-CBOR 0.04 by DSHADOW - http://metacpan.org/release/DSHADOW/Mojolicious-Plugin-CBOR-0.04
08:08 arthas joined #mojo
08:08 punter joined #mojo
08:20 basiliscos joined #mojo
08:24 sujithm joined #mojo
08:46 edestler joined #mojo
08:48 aleksey joined #mojo
08:57 disco joined #mojo
08:57 disco hi there
08:57 purl hola, disco.
08:58 disco new to mojo, and though having lots of questions ;) nice to find this place.
09:03 disco what is the best way to run mojolicious on a production server ? using mod_proxy and starting automatically morbo ? (I would like to keep using mojolicious::lite version).
09:04 Nei did you read https://metacpan.org/pod/Mojolicious::Guides::Cookbook#DEPLOYMENT
09:07 disco Nei: ye.
09:07 disco Nei: yes
09:07 disco Nei: actually I was wondering what meant "for bigger application", in the "hypnotoad" part.
09:09 Nei I suppose if you are already running apache you may want to use mod_proxy or maybe mod_proxy_uwsgi
09:10 disco indeed. the only issue I see is if there is something "blocking" in the mojo/morbo, the app will not be available. or is this wrong ?
09:10 Nei write non-blocking code then
09:11 disco :-)
09:12 disco and to start automatically morbo ? what is your method ? rc.local ? own rc.d script ?
09:14 Nei sure, just write a systemd unit for it
09:16 berov joined #mojo
09:18 cpan_mojo Mojolicious-Plugin-SemanticUI 0.05 by Krasimir Berov - http://metacpan.org/release/BEROV/Mojolicious-Plugin-SemanticUI-0.05
09:25 disco Nei: ok, thank you
09:41 diegok disco: you want hypnotoad for production
09:42 woz joined #mojo
09:51 batman marcus, sri: any comments on my coment?
10:13 woz joined #mojo
10:37 aleksey joined #mojo
10:45 woz joined #mojo
10:52 ver joined #mojo
10:53 woz joined #mojo
10:56 cpan_mojo Mojolicious-Plugin-BootstrapHelpers 0.015.000 by CSSON - http://metacpan.org/release/CSSON/Mojolicious-Plugin-BootstrapHelpers-0.015.000
11:06 cpan_mojo Mandel 0.23 by Jan Henning Thorsen - http://metacpan.org/release/JHTHORSEN/Mandel-0.23 (depends on Mojolicious)
11:07 batman jberger: Mandel can now have ->fresh->cats ;)
11:14 neyasov__ @batman thanks!
11:15 batman neyasov__: didn't know you where here :)
11:15 batman neyasov__: i'm trying to understand https://github.com/jhthorsen/mandel/issues/18 now... writing a unit test
11:16 neyasov__ @batman ok
11:17 batman neyasov__: do you mean it should have an empty list in the database or in perl?
11:17 batman https://github.com/jhthorsen/mandel/commit/9036e48e2a10d9c12c0dabd00516cfff0dab9bea#diff-96021bf6a21034dad0140503732fb154R19
11:17 batman ^ it returns an empty array-ref in perl...
11:18 neyasov__ in database
11:18 batman oh.
11:18 batman why?
11:19 batman i'm trying to add as little as possible to the database until you actually have data...
11:19 batman i think it's the same for HasMany
11:20 D4RK-PH0ENiX joined #mojo
11:21 batman neyasov__: i'm commenting on the issue...
11:24 woz joined #mojo
11:30 oetiker joined #mojo
11:38 sujithm joined #mojo
11:52 neilhwatson joined #mojo
11:55 jberger batman : yay fresh cats!
11:55 batman :)
11:56 jberger And for the record i voted against removing emit_safe from the class
11:56 batman but you voted +1 for replacing emit_safe() with emit()
11:56 batman i figured out the unit test, though it made me question Mango, or how i use it... i wonder if it makes a new connection to the database for each operation, until the connection pool is full
12:03 Dandre joined #mojo
12:15 woz joined #mojo
12:28 lipizzan joined #mojo
12:35 forke_ joined #mojo
12:37 basiliscos joined #mojo
12:41 batman neyasov__: any feedback about Mandel is much appreciated
12:49 Shaeto joined #mojo
12:52 Shaeto before 4.90 version i used this call $tx->req->headers->accept(['multipart/related; type=application/dicom+xml']); for client to set Accept header, anything changed since 4.90 ? now i see 'Accept: ARRAY(XXXX)' in the http request header
13:01 Nei https://metacpan.org/pod/Mojo::Headers#accept the docs would suggest it needs a plain old string
13:02 Shaeto yes seems there is no arrays anymore :)
13:02 Shaeto thanks
13:04 Nei may be "Removed list context support from header method in Mojo::Headers." in Mojolicious 5.0
13:10 Shaeto seems yes
13:15 woz joined #mojo
13:15 berov Can I make websocket route to gracefully be served via GET and/or POST or I need to make a separate route pointing to the same piece of code?
13:19 * moritz didn't even know there were websocket routes
13:29 * jberger_ now knows that moritz doesn't read his blog :(
13:31 berov :)
13:33 sri batman: i do not believe the use case you describe for emit_safe is a good one, it only makes me more sure removing it is a good idea
13:34 sri "here's an event, i don't care if you die"... that does not at all sound like it encourages good development practices
13:35 jberger_ berov: a websocket is a specialized get request, and though you might be able to piggy back then I'd you try hard enough, i don't see what it buys you for the trouble
13:35 jberger_ Grrr: .. piggy back them if you...
13:36 berov may be I have to read more - I was just thinking I can reuse this way the action - laziness.. what to do  :)
13:36 aleksey joined #mojo
13:37 berov will make separate actions and reuse only pieces of code in them
13:38 batman sri: ok, so the decoupling and broadcast way of thinking is plain wrong?
13:38 berov jberher_: thanks
13:38 sri batman: expecting failure at such a high level is wrong imo
13:39 batman so if i do use autodie; open my $FH, '<', 'some-file' inside one of the events, then another one that sends the message to syslog is _not_ supposed to happen just because the file could not be opened?
13:39 disputin joined #mojo
13:40 sri the original idea behind emit_safe was that it wouldn't take down the server, that was all
13:40 batman oh.
13:40 batman i've always used to to make sure _all_ listeners get the messag
13:40 batman +e
13:40 purl 2.71828182845905
13:42 sri jberger, tempire, batman, marcus, crab: so what are we thinking about this branch? https://github.com/kraih/mojo/compare/reply_helpers
13:44 sri jberger_: btw. there is actually some value in keeping render_exception/not_found too, they allow per controller logic, while helpers are app wide :)
13:45 sri and for both to work they do need different names
13:46 mattastrophe joined #mojo
13:48 aleksey joined #mojo
13:55 marty joined #mojo
14:08 marcus +1 for reply_helpers from me.
14:12 Nei thanks to mojo+websockets: http://anti.teamidiot.de/static/nei/teddy_tiger.png
14:15 woz joined #mojo
14:20 sh4 joined #mojo
14:24 sri jberger_: actually scratch that, we don't really use controller specific render_exception/not_found methods
14:24 Nei also thanks for the follow-up on my question
14:25 Nei unsubscribing the error on the reactor seems to work well
14:25 sri you're making an irc client too? :o
14:25 Nei no I make something better, it's a multi-protocol client
14:26 Nei actually it's just a front-end to irssi so it supports icb, xmpp, silc and irc
14:26 Nei and if you like the smell of text you can still ssh into it
14:26 Nei the client part is pure browser-side javascript so it doesn't need to be hosted
14:27 sri marcus: any idea what to do with the methods in the future? remove in 6.0?
14:27 Nei it probably supports twitter via twirssi as a result of that but I haven't tried
14:32 * sri is very +1-ish on reply_helpers too, only problem is what to do with render_exception/not_found in the future
14:40 mr-foobar joined #mojo
14:43 Akron joined #mojo
14:50 batman +3 on reply_helpers from me :)
14:51 batman Nei: can't you rather join the Convos team?
14:52 Nei convos looks pretty interesting, I was very much wondering if it would be possible to steal/share frontend code
14:53 batman i want you to contribute to the backend
14:54 Nei but I also think irssi is a good client. html5 is nicer on mobile. and the goals of the projects are different; I really want the other protocols, not "just" irc. but the backlog storage in irssi is sufficient for a few users, I don't need redis nor is the goal mass hosting/logging like irccloud (where they have the advantage of logging every channel only once)
14:54 batman they are not different. we just haven't gotten there
14:55 batman we've selected to store the raw irc messages intead of using irssi format, since irssi does not have all the types we need
14:55 batman we want other protocols than irc as well, but we haven't gotten there.
14:55 Nei I can use all my channel oper, stalker, etc scripts, the completions, the plug-ins
14:55 batman we also want to make a backend without redis, but we also haven't gotten there....
14:55 Nei that sounds like a great vision but it's so much work to reinvent the wheel :)
14:56 batman instead of using irssi..?
14:56 Nei irssi is lacking one thing, that's message tags / message meta information, but I'm working on fixing that
14:57 marcus Net::Convos::Connection::Irssi? :-o
14:57 Nei I like that I can continue using irssi and the web frontend in parallel. but that doesn't mean convos approach of not using irssi is inferior
14:57 marcus s/Net:://
14:57 batman marcus: Convos::Core::Connection::Irrsi :)
14:57 marcus batman: sounds righter
14:57 batman ;)
14:58 Nei another advantage is that I do not have to host anythinng, just running irssi is enough
14:58 batman Nei: we are going to support using Convos as an irc proxy, but we also haven't gotten there.
14:58 Nei yes but that's useless, irc proxies are inferior
14:58 batman you always have to host something
14:58 Nei see the whole bnc/znc stuff
14:59 marcus Apparently convos works nicely behind znc now.
14:59 Nei (or the irssi "irc proxy" module)
14:59 Nei it always needs hacks for backlog transfer, @time metadata annotations, etc
14:59 Nei and most importantly it is *limited to irc*
14:59 batman i would not mind Convos::Core::Connection::Irrsi, but i won't be hacking on it
14:59 Nei which is the one thing I find irssi really did well, the generailsation so that it works for different types of servers
15:00 Nei you need to abstract away the raw data
15:00 Nei that's why storing raw irc data doesn't cut the bill for me
15:00 * batman wish he could remember what irssi is not storing...
15:00 Nei maybe LeoNerd's object remote transfer irc client has some ideas in the right direction on that
15:01 batman i think you're missing the point of what convos is, and what it might be.
15:01 Nei I tell you what, irssi doesn't store the nick name separately out of the box at the moment
15:01 sri batman: you any idea what should happen with render_exception/not_found in the future? (6.0)
15:01 batman sri: i don't get the question :(
15:02 Nei well, you wanted to convince me of convos so tell me more about it;) but in the end it might just turn out that convos is not what I want/need (which is my current feeling about it, although it looks similar on surface)
15:02 sri batman: it would just be sub render_exception { shift->helpers->reply->exception(@_) }
15:03 sri so, theoretically we could just remove it in 6.0
15:04 Nei this thing is extremely light-weight, rendering, connectivity, protocols, it's all done by irssi. I just need to forward the displayed messages via websocket. so just a few lines of mojolicious::lite code basically
15:04 batman sri: yeah, i would remove it in 6.0
15:04 sri i see
15:04 batman sri: REMOVE ALL THE THINGS! :)
15:05 batman Nei: i won't try to convince you. you seem pretty set.
15:06 Nei on your website it says it's the best way to use irc
15:07 batman Nei: right, because if we wrote "best way to do voice convesation", it would be a lie
15:07 batman the README is for end users, not developers
15:07 Nei do you have a public roadmap?
15:08 batman we don't have a wiki page that says anything about where we want to be.
15:08 batman we can't set the ambitions to high, since we simple don't have the time to create -all-the-things-
15:09 Nei ;)
15:09 batman so what we do, is simply make what me and marcus need work the best.
15:09 batman at the moment that is an IRC frontend.
15:10 Nei yea. and I think that's pretty cool (although there does seem to be a surge in irc web clients recently)
15:10 batman the only limit is that all the things need to be non-blocking.
15:10 Nei but at the same time I have a job and a family and not much time, so I can tell you already I won't have time to port silc to perl
15:11 batman but i think you misunderstand: there's a huge gap between what me and marcus are making and what convos can support.
15:11 Nei nor would I probably find the time to make a *good* jabber connection for convos despite the protocol modules being plenty on cpan
15:11 sri allright, looks like we have +1 from marcus and batman for reply_helpers, jberger and tempire seemed +1-ish too
15:11 r0b3rt joined #mojo
15:11 batman Nei: but why not make a connection for convos that support irssi as backend?
15:12 Nei but I dislike bitlbee simply for that I think bridges through irc "feel" ugly, even if it's a solid piece
15:12 sri jberger's only problem was long term use of a method calling a helper, we can address that by removing the methods in 6.0
15:12 batman sri: or you could name the helpers render_exception and render_not_found :)
15:12 sri batman: but we want nested helpers!
15:13 sri one of the reasons for doing this was having a nested helper in core as example :)
15:13 tempire yes, +1 for me
15:13 batman am i included in "we"? if so, i guess i'm +1 as well ;)
15:13 sri batman: oh, and of course that can't be done before 6.0 at all
15:13 batman Nei: [whatever] -> [irssi] -> [convos] -> [frontend]
15:13 Nei sure that sounds like an interesting challenge, but again I don't know how involved it is (plus the fact that I already have a pure-JS irssi frontend with a few lines of mojolicious::lite running which does not need any additional services running)
15:14 sri batman: okok... ME ;p
15:14 batman hehe
15:14 Nei so I don't know what convos could add to that
15:14 sri no wait, jberger and tempire liked having nested helpers too
15:14 sri anyway, we have a plan
15:14 Nei tell me what you think it would add
15:15 sri everyone likes the prefix $c->reply->*?
15:15 batman Nei: i'm almost certain that you will save time creating a ::Connection::Irssi module, compared to making all the thing we have...
15:15 batman Nei: but i understand that it's difficult to understand. (i often feel the same way)
15:15 batman luckily i got marcus to set me straight
15:15 woz joined #mojo
15:15 Nei yes, this is difficult to understand :D
15:15 batman sri: +1 on "reply"
15:16 sri allright, i don't have any other name proposals anyway :D
15:16 batman Nei: it's very easy to explain why it's difficult to understand: you're outside the box, while i'm inside
15:16 good_news_everyon joined #mojo
15:16 good_news_everyon [mojo] kraih merged reply_helpers into master: http://git.io/69B-Mg
15:16 good_news_everyon left #mojo
15:17 batman Nei: but if you have some time, look at https://github.com/Nordaaker/convos/blob/master/lib/Convos/Core/Connection.pm and see how we hook into Mojo::IRC to emit events to the frontend/database
15:17 batman like here: https://github.com/Nordaaker/convos/blob/master/lib/Convos/Core/Connection.pm#L372
15:18 batman i need to refactor the connection.pm module to Connection/IRC.pm but i wouldn't mind doing that.
15:18 batman that's something i need to do anyway
15:19 sujithm joined #mojo
15:19 Nei I think I "have" a lot; http://anti.teamidiot.de/static/nei/teddy21.png themes. http://anti.teamidiot.de/static/nei/teddy24.png html colors, http://anti.teamidiot.de/static/nei/teddyx.png side by side, http://anti.teamidiot.de/static/nei/teddy8.png mobile+multiple items sharing a window
15:19 sri allright, and it's on the "future" list https://github.com/kraih/mojo/issues/674
15:21 Nei and the "database" is already built into irssi or available as log files which can be loaded with logresume too
15:21 Nei nevertheless I have some outstanding issues (mostly JS related), like supporting remote tab completion
15:22 Nei btw channels do not start with # and & only
15:23 Nei on ircnet channels start with !<5 character ID here> and some channels start with +
15:23 Nei and that is for irc of course, silc and xmpp doesn't require channels to start with anything special
15:25 ignacio_ joined #mojo
15:25 batman sri: ok if i commit this to master: https://gist.github.com/095e57026ddf9bf8434a
15:26 good_news_everyon joined #mojo
15:26 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/EJz8nw
15:26 good_news_everyon mojo/master e58eded Sebastian Riedel: updated some links in the documentation
15:26 good_news_everyon left #mojo
15:27 sri batman: nope, very much not ok
15:27 sri breaks backwards compatibility
15:27 batman ok.
15:28 sri that's why i opened the issue ;)
15:28 batman i see
15:29 sri we would also use $c->helpers->reply->* to allow people to use reply and an action name
15:29 sri s/and/as/
15:30 Nei for example one "irssi" would be connected to more than one "server", Convos::Core::Connection - Represents a connection to an IRC server -- so that's already the first issue I see. how is it going to be represented
15:30 batman sri: ah! of course :)
15:40 sri Nei: btw. the recipe for the new behavior makes sense to you? http://mojolicio.us/perldoc/Mojolicious/Guides/Cookbook#Exceptions-in-events
15:41 Nei yes thank you very much for the improvements also in the docs, very helpful for a newbie like me
15:41 sri \o/
15:42 disputin joined #mojo
15:45 sri i think being able to make all event exceptions fatal with Mojo::IOLoop->singleton->reactor->unsubscribe('error') is a nice win for debugging :)
15:46 bodgix_wrk joined #mojo
15:47 sri batman: this is actually very well written, in case you've not looked yet ;) https://www.joyent.com/developers/node/design/errors
15:48 Nei glad we could come to an agreement even though I lack the technical expertise;)
15:48 sri emit_safe makes programmer errors look not fatal, and that's bad imo
15:48 batman sri: *saved to instapaper*
15:49 sri the distinction between operational errors and programmer errors makes sense
15:49 batman sri: i'm just glad my comment was read.
15:49 sri batman: of course :)
15:49 batman and i will not be mad about the change for very long.
15:49 batman i mean... maybe half a year, but not any longer
15:49 batman ;)
15:49 * sri puts a lot of thought into this kinda decision
15:51 sri the fact that we can easily explain the behavior with a cookbook recipe is a huge win
15:51 batman i agree. it's often correct if it's easy to document (and test)
15:52 sri documentation driven development!
15:53 batman :)
15:58 good_news_everyon joined #mojo
15:58 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/LABtyA
15:58 good_news_everyon mojo/master 45f925e Sebastian Riedel: use $c instead of $self
15:58 good_news_everyon left #mojo
16:00 d4rkie joined #mojo
16:12 good_news_everyon joined #mojo
16:12 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/Vd0O7w
16:12 good_news_everyon mojo/master 9c43af5 Sebastian Riedel: mention that crashing is not just for development
16:12 good_news_everyon left #mojo
16:15 woz joined #mojo
16:29 sri almost forgot, there is one thing about nested helpers me may need to think about
16:30 sri when you define "foo.bar" and "foo", the helper "foo" gets precedence
16:30 sri i'm not sure how it should be
16:31 dod joined #mojo
16:31 sri s/me/we/
16:32 * sri heads off for his daily run
16:39 ignacio_ joined #mojo
16:40 tempire hmm
16:41 tempire there should be a big fat warning
16:45 irq joined #mojo
17:05 sri tempire: make a proposal
17:11 marty joined #mojo
17:14 marty joined #mojo
17:15 woz joined #mojo
17:18 jamesaxl joined #mojo
17:18 sri neat :) http://justinhileman.info/article/git-pretty/git-pretty.png
17:30 marty joined #mojo
17:32 good_news_everyon joined #mojo
17:32 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/Jx_ooA
17:32 good_news_everyon mojo/master bcd0744 Sebastian Riedel: get rid of a warning
17:32 good_news_everyon left #mojo
17:44 basiliscos joined #mojo
18:03 disputin joined #mojo
18:05 sri tempire: i have no idea where and how to mention it
18:08 good_news_everyon joined #mojo
18:08 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/BZlt6w
18:08 good_news_everyon mojo/master 737b3bb Sebastian Riedel: slightly more consistent descriptions
18:08 good_news_everyon left #mojo
18:15 woz joined #mojo
18:36 mattastrophe joined #mojo
18:41 KCL_ joined #mojo
19:15 woz joined #mojo
19:27 _eugen joined #mojo
19:34 Ptolemarch joined #mojo
19:34 lipizzan joined #mojo
19:35 bluescreen joined #mojo
19:43 mattastrophe joined #mojo
19:47 preaction joined #mojo
20:04 woz joined #mojo
20:05 sri maybe this would be a solution https://gist.github.com/anonymous/9a48ac3ad8200c14fbeb
20:05 Zx3 joined #mojo
20:07 disputin joined #mojo
20:09 sri of course that could get very annoying if you do it intentionally https://gist.github.com/anonymous/c6fe30dfe7834d736a23
20:09 sri perhaps a simple warning in the docs would be better
20:11 sri i guess we are back to...
20:11 sri *crickets*
20:12 preaction better than *cicadas*
20:15 sri *cicadas*
20:16 preaction agh!
20:17 genio That'll teach you to mention things!
20:18 preaction personally, i dunno. i can see both sides, and when that happens i tend to err on the side of flexibility, which means not bugging people when they're doing something that could be flexible but also could be problematic
20:18 woz joined #mojo
20:18 preaction and if it becomes a commonly-complained about problem, adding a note or warning or whatever
20:21 genio grr.  Test::Pod::Coverage wants me to cover   use constant FOO => 'bar';
20:37 bobkare joined #mojo
20:40 denis_boyun joined #mojo
20:40 bluescreen_ joined #mojo
20:41 berov sri: "(18,41,06) tempire: there should be a big fat warning" I like big fat warnings
20:42 sri you're too late, no i've decided to wait until the problem actually comes up
20:42 sri s/no/now/
20:42 sri and i don't think you can actually support tempire before he has shown his proposal
20:43 * preaction gets blamed!
20:43 * sri blames preaction
20:43 jfried joined #mojo
20:50 bobkare joined #mojo
21:04 bobkare joined #mojo
21:45 hesperaux joined #mojo
21:47 disputin joined #mojo
21:57 cpan_mojo Mojo-APNS 0.05 by Jan Henning Thorsen - http://metacpan.org/release/JHTHORSEN/Mojo-APNS-0.05
22:00 batman tempire: i wonder what the pack('n', 32) is all about :)
22:00 batman https://github.com/jhthorsen/mojo-apns/blob/master/lib/Mojo/APNS.pm#L224
22:00 tempire lulz
22:00 tempire you tell me
22:00 batman the ways of APNS are mysterious :)
22:00 tempire It's been 40 decades since I looked at pack
22:01 batman haha, it's just a static number. i kind of get chr(0) (not really though), but why it's followed by another static value is beyond me :/
22:01 batman tempire: but you look som young!
22:01 batman like captain america!
22:01 batman ;)
22:02 tempire acacia juice
22:02 batman hehe
22:02 * batman gets his beauty sleep
22:02 batman enjoy Mojo::APNS :)
22:03 disputin joined #mojo
22:10 bodgix joined #mojo
22:11 ignacio_ joined #mojo
22:13 good_news_everyon joined #mojo
22:13 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/sTfgpw
22:13 good_news_everyon mojo/master e51f30f Sebastian Riedel: moved around a few sections in the guides
22:13 good_news_everyon left #mojo
22:21 good_news_everyon joined #mojo
22:21 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/lDOAZw
22:21 good_news_everyon mojo/master d30925b Sebastian Riedel: IRIs are very common these days
22:21 good_news_everyon left #mojo
22:21 sri when i first wrote that few people really knew what that was about :)
22:22 sri then came the emoji fonts
22:49 good_news_everyon joined #mojo
22:49 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/tysBtg
22:49 good_news_everyon mojo/master 42633b1 Sebastian Riedel: deprecations go first
22:49 good_news_everyon left #mojo
22:50 disputin joined #mojo
23:03 mattastrophe joined #mojo
23:04 bodgix left #mojo
23:08 bodgix joined #mojo
23:48 Mikey joined #mojo
23:49 Claudio joined #mojo

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