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

IRC log for #mojo, 2016-02-29

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

All times shown according to UTC.

Time Nick Message
00:02 janus joined #mojo
00:25 sri jberger: btw. re $exception->throw, it appears that $exception->rethrow is a thing in some languages
00:28 genio sri: I assume that's to not lose the original spot the error was thrown? to add the current spot to the stacktrace or something?  I'm having a hard time imagining a need for "rethrow"
00:29 sri $exception->throw would change the stack trace, while $esception->rethrow would keep looking the same
00:30 genio ah, that's the difference between   catch(Exception ex) { throw ex; }      and   catch(Exception ex) { throw; }    in C#
00:31 sri exactly, i'm looking at C# right now :)
00:31 sri http://blogs.msdn.com/b/jmstall/archive/2007/02/15/throw-vs-rethrow.aspx
00:31 jberger But of course in the perl context it isn't any different since there is no context attached
00:31 jberger python has some way to rethrow iirc
00:32 sri $exception->throw could wrape the old exception in a new one, while $esception->rethrow keeps it as is
00:32 jberger Or I'd there was a data field on the exception object ...
00:32 jberger Grrr *if
00:33 sri just throwing (pun intended!) this around finshing for ideas what to do with Mojo::Exception
00:33 sri for core we don't need more i guess
00:36 jberger I thought I was going to get a lot of open source time this weekend
00:36 jberger Yeah that didn't happen
00:36 jberger Nice weekend overall but different than I expected
00:36 sri guess attaching arbitrary data is really the only interesting direction for Mojo::Exception
00:37 sri have a hunch it will end up being $exception->more({...}), with shortcuts like Mojo::Exception->throw('Whatever!', {...})
00:38 jberger more?
00:38 sri more!
00:38 jberger use more 'Exceptions';
00:38 sri as in more information about the exception
00:39 jberger In my mind it was actually called data parallel to delay
00:40 sri doesn't really fit the context imo
00:41 sri although, C# actually does call it Data
00:41 sri ;p
00:41 sri and it is a bunch of key/value pairs
00:46 genio I just made the mistake of looking at metacpan.org/pod/Exception
00:48 sri there would actually be value for core in formalizing this pattern https://github.com/kraih/mojo/blob/master/lib/Mojo/Template.pm#L104-L106
00:48 sri we have two variations of that
00:48 sri https://github.com/kraih/mojo/blob/master/lib/Mojolicious.pm#L189
00:54 jberger I guess that could be called rethrow
00:54 sri best i can come up with so far is a class method that rethrows or returns the exception class
00:55 sri Mojo::Exception->rethrow($_[0])->throw($_[0])
00:55 sri Mojo::Exception->rethrow($_[0])->new($_[0])->inpsect->trace->verbose(1)->throw
00:55 * sri shrugs
00:56 sri wait, actually rethrow or create an exception and return it
00:56 sri that covers all use cases too
00:56 sri Mojo::Exception->rethrow(shift)->throw;
00:57 sri Mojo::Exception->rethrow(shift)->inspect->trace->verbose(1)->throw;
00:57 jberger Why wouldn't it just wrap and throw a new one?
00:57 sri ?
00:58 sri isn't wrapping like a super rare case?
00:59 jberger Why does it have to return, why couldn't it just take the non reference exception and throw it as a new object?
00:59 sri ummm
00:59 sri look at the links
00:59 sri we actually do stuff there
00:59 jberger Sorry I'm on my phone and if be having along if I weren't
01:00 jberger That was completely borked
01:00 sri CORE::die Mojo::Exception->new(shift) ->trace->inspect($self->unparsed, $self->code)->verbose(1);
01:00 jberger Sorry I'm on my phone and I'd be hacking along if I weren't
01:00 sri that's from Mojo::Template
01:00 sri ina $SIG{__DIE__}
01:01 jberger Ah the inspect is the problem
01:01 jberger Again
01:01 sri also the verbose
01:07 sri oh well, not worth it anyway, results in more code than before
02:03 KindOne joined #mojo
02:31 aborazmeh joined #mojo
02:32 mcsnolte joined #mojo
02:42 Rubes joined #mojo
02:55 good_news_everyon joined #mojo
02:55 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/v2V60
02:55 good_news_everyon mojo/master cc2a0d0 Sebastian Riedel: use shift consistently
02:55 good_news_everyon left #mojo
03:09 good_news_everyon joined #mojo
03:09 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/v2ViF
03:09 good_news_everyon mojo/master 158cd21 Sebastian Riedel: split the name only once
03:09 good_news_everyon left #mojo
03:27 cpan_mojo Statocles-0.071 by PREACTION https://metacpan.org/release/PREACTION/Statocles-0.071
03:37 inokenty-w joined #mojo
03:52 good_news_everyon joined #mojo
03:52 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/v2V1q
03:52 good_news_everyon mojo/master 0fcdf7f Sebastian Riedel: set the default at the top
03:52 good_news_everyon left #mojo
03:55 noganex_ joined #mojo
04:11 acamari joined #mojo
04:19 dragos_ joined #mojo
05:08 Rubes left #mojo
06:02 bwf joined #mojo
06:54 hernan604 joined #mojo
07:01 ashimema joined #mojo
07:05 dod joined #mojo
07:09 dod joined #mojo
07:15 McA joined #mojo
08:10 dvinciguerra_ joined #mojo
08:11 Vandal joined #mojo
08:14 osfabibisi joined #mojo
08:14 dod joined #mojo
08:26 bwf joined #mojo
08:36 AndrewIsh joined #mojo
08:43 trone joined #mojo
08:48 dod joined #mojo
08:49 bwf joined #mojo
09:19 jnbek joined #mojo
09:53 berov joined #mojo
09:54 bwf joined #mojo
09:56 bwf joined #mojo
10:09 trone joined #mojo
10:17 dvinciguerra joined #mojo
10:32 inokenty-w joined #mojo
11:14 kes joined #mojo
11:30 dvinciguerra_ joined #mojo
11:34 dvinciguerra joined #mojo
11:38 bwf joined #mojo
11:40 cpan_mojo Mojolicious-Plugin-Util-Endpoint-0.19 by AKRON https://metacpan.org/release/AKRON/Mojolicious-Plugin-Util-Endpoint-0.19
11:44 kaare joined #mojo
12:33 cpan_mojo Mojolicious-Plugin-PubSubHubbub-0.15 by AKRON https://metacpan.org/release/AKRON/Mojolicious-Plugin-PubSubHubbub-0.15
12:39 bwf joined #mojo
12:49 mcsnolte joined #mojo
12:55 bwf joined #mojo
13:15 asarch joined #mojo
13:18 bwf joined #mojo
13:18 upasana joined #mojo
13:20 esh joined #mojo
13:21 upasana_ joined #mojo
13:22 sri joined #mojo
13:23 stephan48 joined #mojo
13:24 crab joined #mojo
13:29 bwf joined #mojo
13:30 taichi joined #mojo
13:36 marty joined #mojo
13:44 neilhwatson joined #mojo
13:59 perlpilot joined #mojo
14:02 ramortegui joined #mojo
14:10 good_news_everyon joined #mojo
14:10 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/v2wF1
14:10 good_news_everyon mojo/master 42ff315 Sebastian Riedel: mention specifically that WebSockets are not available
14:10 good_news_everyon left #mojo
14:18 neilhwatson joined #mojo
14:23 kivilahtio joined #mojo
14:23 irqq joined #mojo
14:39 jb360 joined #mojo
14:43 orev joined #mojo
14:57 zivester joined #mojo
14:58 kivilahtio joined #mojo
15:53 dhg joined #mojo
15:59 gryphon joined #mojo
16:32 zivester joined #mojo
16:46 chienjo joined #mojo
17:20 Trelane I'm being a bear of very little brain today, and I was wondering if someone could explain mojo exception handling to me a little more
17:20 taichi joined #mojo
17:20 Trelane Is there some sort of $SIG{__DIE__} handling going on here?
17:21 Grinnz_ just evals, for the most part
17:21 Grinnz_ is there a particular instance you're having trouble?
17:21 bradjm joined #mojo
17:21 Trelane Yes, but give me a second to explain
17:22 Trelane We have an application that has it's own $SIG{__DIE__} handler (the merits of this are probably out of scope of this conversation) and it's not handling the Mojo::Exceptions right - i.e. when something re-throws the exception it's attaching a stack trace from the point it was thrown
17:23 Trelane I suspect this is because it's not recognising the Mojo::E stack trace is there
17:23 Trelane So I dove into the code and started adding Data::Dumper statements everywhere
17:23 Grinnz_ that's just stored in the object. perl doesn't have any standard format for "stack traces"
17:24 Trelane And I finally figured out that the exception is a) A Mojo Exception b) sometimes when I dump it it has frames and sometimes it doesn't
17:25 Trelane From my reading of the source code it should always get frames if ->throw was called, but not if ->new is called without calling other things afterwards
17:25 Trelane so, given that the error is a standard DBIx::Class exception - i.e. something unrelated to Mojo, I'm wondering how it became a Mojo::Exception
17:26 Trelane Also, who do I have to cross their hand with silver to get decent exception handling into the Perl core! ;-)
17:31 sri i would kill for decent exceptions in Perl core
17:32 lluad joined #mojo
17:34 sri something interesting just came up in #perl
17:35 captbipto1 has anyone come across where they installed mojo and the mojo command itself didn't do anything at all?
17:35 sri someone got confused because one example uses "say" without actually enabling the feature
17:35 sri captbipto1: no
17:36 jberger Trelane: yes there is such a handler
17:36 sri we were planning a documentation conventions section for Mojolicious::Guides to introduce core signatures
17:36 orev yeah, I still consider say to be a wildly new feature that only the bleeding edge people use...
17:36 sri maybe we should do that already for 5.10 features
17:36 jberger Trelane: https://github.com/kraih/mojo/blob/master/lib/Mojolicious.pm#L188
17:36 jberger but yeah its basically for the reason of ENOWAYELSETOPROVIDENICEEXCEPTIONHANDLING
17:37 jberger sri: don't we document that we enable 5.10 features?
17:37 Grinnz_ enable it, but not use it in examples
17:37 sri jberger: yes, but not that examples won't show that
17:38 sri jberger: http://mojolicious.org/perldoc/Mojo/IOLoop#SYNOPSIS
17:38 sri that uses say, but doesn't actiavate the feature
17:38 sri it's implied
17:39 Trelane Why are you using say there anyway?
17:39 sri it looks good
17:40 dhg joined #mojo
17:41 Trelane jberger: where is such a handler documented?
17:43 jberger Trelane: its a pretty low-level thing, I'm not sure its explicitly documented as such
17:44 Trelane no problem.  Wielding grep I see that there's one in Mojolicious and one in Mojo::Template
17:44 jberger the closest thing I could probably present as documentation is the documentation of the exception pages system: http://mojolicious.org/perldoc/Mojolicious/Guides/Tutorial#Built-in-exception-and-not_found-pages
17:44 jberger it would be rather hard for it to generate those pages in that way without it
17:44 Trelane they both seem to leave things the heck alone if they're already objects, but if not they turn them into Mojo::Exceptions
17:44 sri note that exception handling changed quite a bit in the last cpan release
17:44 Trelane from what I read though they should both have traces in them
17:45 taichi joined #mojo
17:45 jberger yeah, Trelane are you seeing changes in the behavior, Mojo::Exception has been on our mind a lot lately
17:45 Trelane sri: oh, really?  I should probably stop grepping the github checkout when comparing to installed behaviour then
17:45 sri and in fact, if you want to propose more changes, now is the perfect time
17:45 Trelane okay, first things first...let me work out what version we have, and if upgrading changes anyhting
17:51 good_news_everyon joined #mojo
17:51 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/v2rSm
17:51 good_news_everyon mojo/master ea5d85f Sebastian Riedel: add CONVENTIONS section to Mojolicious::Guides
17:51 good_news_everyon left #mojo
17:51 sri ok, first rty
17:51 sri http://mojolicious.org/perldoc#CONVENTIONS
17:52 sri patches welcome, content and placement, all is up for discussion
17:52 jberger I do like the first bit
17:53 preaction looks good, and that seems the best place for it, but prepare to be disappointed in the uptake :(
17:53 jberger the second bit is a little odd
17:53 sri preaction: that's ok, we just need something to link to when the question pops up ;)
17:53 jberger I might mention that any use of Mojo::Base enables the above
17:53 jberger and that you can use it in scripts with -strict
17:54 Trelane That statement also might be considered a little dated very soon
17:54 jberger as you wrote it I bet I'd see people doing 'package My::Class; use Mojo::Base -strict; use Mojo::Base -base;'
17:54 Trelane "Mojolicious uses modern Perl"
17:54 jberger Trelane: oh, what do you see as "modern perl" in the near future?
17:55 preaction eh, it does. we're not talking 5.8 practices here, using bless and rubbing sticks together like cavepeople
17:55 Trelane Well, use 5.0024 would turn on postderef
17:55 sri jberger: make a full sentence!
17:55 pink_mist 5.2.400? 0_o
17:55 Grinnz_ Trelane: modern perl as much as can be expected when it still runs on 5.10
17:56 Trelane Grinnz_: yes, but that's not what it actually says ;-)
17:56 pink_mist Trelane: no you need to specifically enable that feature by name afaik; it won't be used simply by doing 'use 5.024;'
17:56 sri note that i said "modern Perl", and not "Modern Perl"
17:56 sri (tm)
17:57 sri "Mojolicious uses a modern dialect of Perl exclusively..."
17:57 Trelane Someone could argue when you mean "modern Perl" you mean "features offered by the version of Perl no longer supported four major releases of Perl ago"
17:58 Trelane ...but that someone would sound like an asshole
17:59 Trelane ho hum, upgrading Mojolicious didn't magically solve the issue, I guess I'll have to look into it
18:02 good_news_everyon joined #mojo
18:02 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/v2rQe
18:02 good_news_everyon mojo/master d7ef0b7 Sebastian Riedel: call it a dialect
18:02 good_news_everyon left #mojo
18:02 sri ok, next try http://mojolicious.org/perldoc#CONVENTIONS
18:02 jberger sri++ I like that a lot
18:03 ribasushi dialect carries a tad of a stigma 'won't run on my stock stuff from RH"
18:03 ribasushi subset is much more neutral
18:03 sri i like subset
18:04 Trelane s/will enable them/will wherever they are used enable these features/
18:04 ribasushi ( you can even throw in "strict subset" ) for correctness, and for catch-eye effect from "strict"
18:05 Trelane (I say this because it's not clear *where* the features will be enabled otherwise)
18:06 good_news_everyon joined #mojo
18:06 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/v2r7W
18:06 good_news_everyon mojo/master 2ccdcaf Sebastian Riedel: subset sounds better than dialect
18:06 good_news_everyon left #mojo
18:07 sri hird try http://mojolicious.org/perldoc#CONVENTIONS
18:07 sri gotta throw in some commas there
18:07 Grinnz_ the commas look fine :P
18:07 jberger and a semi-colon or two
18:08 jberger ;-P
18:08 Trelane aaaaaaaaargh
18:08 Trelane So, here's a thing that sucks in our code.  We're doing $exeption->can('trace') to see if we've got a method that will return us a stack trace
18:08 jberger s/exclusively\K, and/;/
18:09 Trelane ...because that's what Moose::Exception and compatible classes do
18:09 Trelane ...this doesn't work out so well when that resets your stack trace
18:09 sri lol
18:10 jberger sri: the trace method could bail out if already traced
18:10 sri you have to to do ->DOES or ->isa if you want to duck type like that i think
18:10 Trelane I offer this in the "dumb things that people will do that you may want to consider working around even though you shouldn't have to" category
18:10 jberger not sure that's generally needed but it certainly seems like you wouldn't need to trace twice
18:10 jberger sri: good point re isa
18:11 Trelane sri: If we're duck typing you don't do that.  If it quacks like a duck (i.e. it ->can('quack') then it is a duck
18:11 sri didn't i ask what ->inspect and ->trace should do if called twice?
18:11 Trelane ...this is how you end up shooting the hunter with the bird call however
18:11 sri like 2 days ago ;p
18:12 jberger sri: really don't remember that question, though there were a bunch of questions being asked
18:12 Trelane that was saturday.  I was taking small people to gymnastics and horse riding.
18:12 sri anyway, calling ->inspect/->trace twice seems currently untested, so you can still propose changes
18:13 Trelane I should propose that calling it multiple times return the same stack trace each time
18:13 Trelane but that's not the mojo way where we return self
18:14 trone joined #mojo
18:14 Trelane ...and anyway, that's not going to return a Devel::StackTrace object
18:18 sri patch is as simple as https://gist.github.com/anonymous/3c4ffb1d725a411b16db
18:19 sri jberger, Trelane: now is the time to change it
18:20 Trelane The returning the stack trace seems like a hack
18:21 Trelane Having it die if you try and set the stack trace a second time seems sensible
18:21 sri Trelane: i did not understand your last few sentences
18:21 sri we return the invocant, always
18:21 PryMar56 joined #mojo
18:21 Trelane There were two proposals if ->trace was called twice (1) death (2) returning the result of ->frames each time
18:22 sri ewww
18:22 Trelane While (2) seems like it might help, it's inconsitent and hacking
18:22 Trelane I suggest (1)
18:22 sri 2 is never ever going to happen
18:22 sri that's horrible!
18:22 sri ->trace returns the invocant
18:22 Trelane I also suggest the name is something other than ->trace
18:22 Trelane ...but the ship might have sailed on that
18:23 sri you're two days too late for that
18:23 Trelane So, yeah, I agree it should go boom if you try to recreate the stack trace
18:24 Trelane that makes little to no sense
18:24 sri then i vote -1 on changes
18:24 sri you're too focused on your specific case, and not considering the bigger picture
18:25 sri jberger: anything to add?
18:25 Trelane ...So what I suggest is (a) Nothing changes in the method name or what it returns but (b) if you call it twice there's an error message
18:26 haarg Trelane: Moose::Exception isn't really meant be a public interface, so staying away from isa checks is a good idea
18:26 sri calling ->trace twice in Mojo::Exception is not something fatal
18:27 Trelane haarg: what would you suggest instead?
18:27 Trelane sri: but isn't that an accessor?
18:27 sri no
18:27 sri http://mojolicious.org/perldoc/Mojo/Exception#trace
18:28 Trelane sri: Why would you want to alter the stack trace?  Is there a situation where that's useful?
18:32 sri for a rethrow?
18:33 sri i don't know really, we looked around and C# allows it
18:37 jberger I'd say the only change to calling these methods twice is perhaps they shouldn't recalculate the results
18:37 jberger but I don't think we can support the api of the other exception modules because as you say they aren't the mojo chainable style
18:38 jberger so it probably doesnt matter
18:39 Trelane Wouldn't you explictly not want to alter the stack trace for a rethrow
18:39 sri you can just not call ->trace
18:40 sri it is *NOT* and accessor
18:41 sri the actual stack trace is stored in ->frames
18:41 jberger frames is the accessor
18:41 jberger oh sri beat me to it
18:44 good_news_everyon joined #mojo
18:44 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/v2rjX
18:44 good_news_everyon mojo/master 2a96a28 Sebastian Riedel: trace is not the right name
18:44 good_news_everyon left #mojo
18:45 Trelane sri: I understand the concept of mistakenly calling ->trace is wrong, because it's not the accessor (frames is.)  What I don't understand is why you'd ever want to set the frames more than once with trace, and therefore why re-calling it isn't an error
18:48 sri say we allow ->throw to be called on objects, like Mojo::Exception->new('Whatever')->trace->verbose(1)->throw, and we want to keep ->throw behavior consistent, so it calls ->trace again automatically, silently ignoring the call would be good here
18:49 sri jberger: i think that's kinda important for your proposal
18:50 sri so, you might want to keep an eye on this so it doesn't mess with your plans
18:50 sri ->trace generating errors could be very bad for future extensions
18:51 dhg left #mojo
18:52 sri i guess calling ->inspect liberally might be more useful going forward
18:52 Trelane hmm.
18:52 sri and we prolly also want to keep ->trace and ->inspect kinda consistent
18:52 Trelane You make good points, but I don't quite see the logic here, and let me explain why
18:53 jberger I don't see any real reason to prevent inspect being called twie
18:53 jberger twice
18:53 Trelane I don't see how changing the exception object once it's been created makes any sense
18:53 jberger and I really don't see throwing an exception
18:53 sri jberger: but should it regenerate the context information?
18:54 Trelane Either when the object is created is where the "problem" happened. i.e. where you called trace the first time
18:54 jberger I understand the idea of not regenerating it, but again, I don't see any reason to prohibit it
18:54 Trelane or it's where you called ->throw
18:55 Trelane maybe what sri is suggesting makes sense to say you don't call ->trace from within throw if it's already been called
18:55 dod joined #mojo
18:56 jberger Trelane: right
18:56 jberger that's how I see it now
18:56 sri mojo core needs none of this btw.
18:56 sri it's all just for 3rd party use
18:56 jberger IF it had been the way you generate the trace AND access it I would agree
18:57 jberger but the api cannot be made consistent with Moose::Exception so I don't see any reason to make the code less general in a vain attempt to do it partway
18:57 sri jberger: agreed
18:57 sri isn't Throwable used more these days?
18:58 sri still a lot of Exception::Class use out there too
18:58 genio I'm trying to get ownership of the Exception, Exceptions, and SimpleException module names (all in the same dist).  Still no response from the current owners.  So many different things going on that I'd like to make that one at least be close to the others
19:00 sri jberger: would be nice though if we could settle the question of what should happen if ->inspect/->trace get called twice
19:00 sri i think the options are a) inspect/trace again, b) do nothing
19:00 Grinnz Throwable and Exception::Class are the ones i know of
19:00 jberger +1 for (a)
19:00 Grinnz Throwable being usable for both Moo and Moose now
19:01 jberger since I see no reason to prohibit it
19:01 sri jberger: you don't think that would be bad for combo methods? (methods with implied ->inspect->trace)
19:01 jberger basically I don't see why a consumer of the exception would ever call them again
19:01 sri jberger: like say $e->throw
19:02 jberger sri: I must not understand the problem
19:02 jberger throw also calls die so we are going to leave the context anywya
19:03 sri sub throw { die (ref $_[0] ? $_[0] : shift->new(shift))->trace->inspect }
19:03 jberger we discussed that before
19:04 jberger I thought we decided that if you want to throw your own very-custom exception that you shouldn't use the shortcut method
19:04 sri $e->trace would retrace and reinspect
19:04 sri i don't remember a decision
19:04 sri didn't we say we leave that for the future?
19:05 sri life could be so nice if there were exceptions in Perl core :S
19:06 sri honestly, i didn't even mind the autodie way, if only it wasn't so slow https://metacpan.org/pod/autodie#EXCEPTIONS
19:07 jberger starting here: http://irclog.perlgeek.de/mojo/2016-02-26#i_12107345
19:08 jberger as long as there is an opening for ->throw($msg, $data) if we add data, which we said was fine, then I don't see any reason to call throw on an instance
19:08 jberger because then its just a shortcut for die and its more characters anyway
19:09 sri jberger: guess i'm still thinking too hard about how to integrate the rethrow of existing exceptions pattern into the api https://github.com/kraih/mojo/blob/master/lib/Mojolicious.pm#L189
19:10 sri anyone remember why we use CORE::die?
19:11 sri i vaguely remember there was an edge case with overloaded die
19:11 jberger I'm sure there are tons of edge cases with overloaded die
19:11 captbipto1 left #mojo
19:12 jberger BEGIN{ *CORE::GLOBAL::die = sub { warn $_[0] } }
19:12 genio o.O
19:22 disputin joined #mojo
19:23 odc joined #mojo
19:24 cpan_mojo Mango-1.27 by ODC https://metacpan.org/release/ODC/Mango-1.27
19:33 good_news_everyon joined #mojo
19:33 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/v2oYK
19:33 good_news_everyon mojo/master 86e7266 Sebastian Riedel: more commas
19:33 good_news_everyon left #mojo
19:35 sri for the record, even more than core exceptions, i would like to see a signatures polyfill module on cpan :/
19:35 jberger me too
19:36 jberger at least we are likely to get a __SUB__ polyfill soon
19:57 pink_mist could take a look at how Data::Munge's rec() works for a __SUB__ polyfill maybe?
19:58 pink_mist though you said it's already on the way, so never mind I guess
20:02 jberger https://rt.cpan.org/Public/Bug/Display.html?id=111466
20:03 jberger I've already had verbal confirmation that this is acceptable
20:03 jberger oh he even added a comment on it saying so
20:04 pink_mist ah, nice =)
20:04 pink_mist leonerd++
20:11 cpan_mojo App-txtnix-0.03 by MDOM https://metacpan.org/release/MDOM/App-txtnix-0.03
21:11 good_news_everyon joined #mojo
21:11 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/v2ool
21:11 good_news_everyon mojo/master fcd3137 Sebastian Riedel: use a subheading
21:11 good_news_everyon left #mojo
21:12 sri looks better in context, doesn't it? http://mojolicious.org/perldoc#CONVENTIONS
21:16 jberger yeah
21:16 jberger and especially if we want to do a "Variables" subheading
21:16 jberger like we've discussed before
21:17 jberger for $app, $c, $r conventions
21:17 * sri doesn't actually remember that
21:17 jberger $bytes vs $chars
21:19 jberger http://irclog.perlgeek.de/mojo/2015-08-12#i_11047030 for example
21:19 jberger $t, $ua
21:21 jberger I found another example, but it would probably derail the idea :-P
21:21 jberger it does contain a good memory of purl though
21:21 * jberger wipes away a tear
21:28 cpan_mojo Test-Mock-Time-v0.1.6 by POWERMAN https://metacpan.org/release/POWERMAN/Test-Mock-Time-v0.1.6
22:32 bpmedley joined #mojo
22:40 good_news_everyon joined #mojo
22:40 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/v2ohJ
22:40 good_news_everyon mojo/master 464b583 Sebastian Riedel: more commas
22:40 good_news_everyon left #mojo
22:42 Grinnz_ sri: you overdid it a bit on the Tutorial one :P
22:42 sri oh noes!
22:42 Grinnz_ i would only use the middle comma there
22:42 sri but yea, i was really not sure about that one
22:43 sri after object?
22:43 Grinnz_ yea
22:43 pink_mist yeah
22:45 pink_mist the first comma isn't that bad, but the last one should definitely go imo
22:46 jberger I might not use any of those :-P
22:47 good_news_everyon joined #mojo
22:47 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/v2ojD
22:47 good_news_everyon mojo/master b15eb7b Sebastian Riedel: less commas
22:47 good_news_everyon left #mojo
22:49 good_news_everyon joined #mojo
22:49 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/v2KeI
22:49 good_news_everyon mojo/master a8ba9c0 Sebastian Riedel: that one title actually had a test
22:49 good_news_everyon left #mojo
22:50 pink_mist hah
22:50 * pink_mist is unsure about the comma after the irc channel
22:51 Grinnz_ lol
22:52 sri i think that one is fine
22:53 Grinnz_ I kind of think that first comma should be the end of a sentence actually
22:53 Grinnz_ in the DON'T PANIC section
22:53 sri semicolon? -.-
22:54 Grinnz_ perhaps
22:54 sri jberger taught me well
22:54 preaction that one with the irc channel's not fine, actually, unless you also have a comma after "IRC channel". the commas circumscribe a subordinate clause, which means you can just remove the clause entirely and the sentence makes sense
22:54 Grinnz_ but yeah that comma should go :P
22:54 Grinnz_ the second one
22:54 preaction so, "join the official IRC channel, #mojo on C<irc.perl.org>, to get all the latest updates"
22:54 sri NOOOOO!
22:55 preaction that could also be a parenthetical aside, but eh, six of one, half-dozen of the other
22:55 Grinnz_ you could keep the first one as a comma, if you put "so" after it, maybe
22:55 preaction ... "circumscribe a subordinate clause"... how much caffiene did i just drink?
22:55 sri never gonna use a semicolon, just gonna make it a point
22:56 Grinnz_ parenthesize the C<#mojo> on C<irc.perl.org> part?
22:56 preaction you could, yes
22:56 preaction or commas around it, either way
22:57 good_news_everyon joined #mojo
22:57 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/v2Kvo
22:57 good_news_everyon mojo/master 89e90fa Sebastian Riedel: apparently this should be two sentences
22:57 good_news_everyon left #mojo
22:57 preaction depends on how far outside the current thought you want it to look
23:03 good_news_everyon joined #mojo
23:03 good_news_everyon [mojo] kraih tagged v6.51 at 3bb1671: https://git.io/v2KfP
23:03 good_news_everyon left #mojo
23:04 good_news_everyon joined #mojo
23:04 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/v2Kfb
23:04 good_news_everyon mojo/master a834d61 Sebastian Riedel: bump version
23:04 good_news_everyon left #mojo
23:11 cpan_mojo MojoX-UserAgent-Throttler-v1.0.1 by POWERMAN https://metacpan.org/release/POWERMAN/MojoX-UserAgent-Throttler-v1.0.1
23:13 genio Curiosity question here:  https://github.com/kraih/mojo/blob/master/t/mojo/base.t#L5-L6   why is FindBin being used in some tests like this?
23:14 Grinnz_ to load https://github.com/kraih/mojo/tree/master/t/mojo/lib
23:15 genio oh, yea, I should have thought about that some more
23:28 good_news_everyon joined #mojo
23:28 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/v2Kte
23:28 good_news_everyon mojo/master 6b978f9 Sebastian Riedel: link to feature pragma
23:28 good_news_everyon left #mojo
23:36 disputin joined #mojo

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