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

IRC log for #mojo, 2016-03-26

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

All times shown according to UTC.

Time Nick Message
00:05 disputin joined #mojo
01:12 disputin joined #mojo
01:40 sri i actually like working with it https://github.com/kraih/mojo-pg/commit/eb027c60ab18406e04b44f323f357a3190914bff
02:00 commish joined #mojo
02:02 commish i have a banana pi running bananian/debian.  anybody know the proper way to start hypnotoad on this type of system?  i come from BSD so this linux thing is new to me.
02:02 commish sorry: start at boot.
02:03 commish automatically.
02:06 commish otherwise im going to just use a @reboot entry in crontab
02:26 meredith if i'm reading this right, it has sysvinit, so Toadfarm might be a nice clean way to set up
02:26 meredith https://metacpan.org/release/Toadfarm
02:26 meredith note Toadfarm::Manual::RunningToadfarm
02:28 commish yeah it's sysv and not systemd.
02:29 commish and is it just me or is the perl packaging a total nightmare :(
02:30 commish in openbsd you can rely on modules packaged as "p5-Package-Name" but this libpackage-name-perl thing in debian is awful.
02:31 meredith yeah i'm not a big fan of that style, i don't really use dist packages either.  some people set up private repositories with local packages in the same naming scheme :)
02:43 asarch joined #mojo
03:24 sri argh
03:24 sri i can't get over the ignorance https://www.reddit.com/r/perl/comments/4bvdny/die_if_fork_eq_hostile/d1dmrnw
03:28 sri just for the Mojo namespace i count about 35 distributions if you split everything up
03:28 sri who's going to pay for all that extra time?
03:28 noganex joined #mojo
03:40 kaare joined #mojo
04:02 preaction searching through HTML::TreeBuilder and it's deps, i've found about 5000 loc so far, plus it requires a C compiler (HTML::Parser). where are they living where less than 30kb of disk space requires one to write however much extra code to make TreeBuilder do what you actually want?
04:03 preaction sorry, i'm wrong. 9000 loc so far. forgot to actually get the xs, c, and h files
04:05 sri heh
04:05 sri figured the c and xs would add up
04:07 sri i'm actually wondering, is there any subcommunity that manages to actively maintain 35+ cpan dists?
04:08 preaction it's some weird mental thing. someone apparently can _unknowingly_ install something they won't use and that doesn't affect them in the slightest (like the three versions of HTML::Parser frontends that come with the HTML-Parser dist). but if they somehow _know_ they won't use some part of a dist, it's a _problem_
04:09 sri nobody seems to question that they have to install HTTP::Daemon when they install LWP::UserAgent
04:09 sri it's so puzzling
04:09 preaction the toolchain gang might come close to 35 dists. but they are also a huge sponge for PR and have 30 members on the github
04:09 sri maybe the generic names make it easier?
04:10 preaction also, a lot of the toolchain gang is basically dead code that nobody cares about
04:11 sri i count 27 dists https://github.com/Perl-Toolchain-Gang
04:12 sri one member per dist is not too bad
04:14 preaction i mean, i don't think anyone came to this conclusion logically, and so basically cannot accept any logical refuting of their claim that installing Mojolicious to get Mojo::DOM is abhorrent...
04:16 preaction and, having used a few of those other HTML tools, you could probably actually smack me upside the head every time I installed Mojolicious and it'd still be less painful to use Mojo::DOM...
04:19 sri think you right about the fact that logic can't really be applied here
04:19 sri *'re
04:20 sri not sure if we could guide them in a way that gets them to draw the right conclusions themselves though
04:28 sri maybe we just need a rails-ish majestic monolith pep talk :)
04:31 bpmedley I never considered the extra time involved to manage all the distributions.  That alone is enough to want to keep them together.  Thanks for the clarity.
04:32 sri it's insane when you think about the extra cost of developing something like http/2 support, that would touch maybe 30 of those dists
04:36 bpmedley http://us.battle.net/sc2/en/nova-covert-ops/
04:37 * sri is not much of an sc2 player anymore
04:39 sri still playing hearthstone every now and then though
04:55 cpan_mojo Statocles-0.074 by PREACTION https://metacpan.org/release/PREACTION/Statocles-0.074
05:16 melo joined #mojo
06:15 irqq joined #mojo
07:22 salva joined #mojo
07:46 dod joined #mojo
08:13 Vandal joined #mojo
08:30 * ribasushi rereads the DOM::Tiny conversation remains confused
08:30 ribasushi everything above is semi-reasonable
08:31 ribasushi but then this puts words into Grinnz' mouth: "Mojo::DOM58, focused solely on Perl 5.8 compatibility"  http://cpanratings.perl.org/dist/DOM-Tiny#12742
08:31 ribasushi whereas afaiui a large motivation is "DOM without the mojo"
08:31 ribasushi please clarify?
09:01 sh3 joined #mojo
09:32 cpan_mojo Mojo-Webqq-1.7.6 by SJDY https://metacpan.org/release/SJDY/Mojo-Webqq-1.7.6
11:31 kes joined #mojo
12:11 Kripton joined #mojo
12:23 bpmedley joined #mojo
12:45 cpan_mojo Mojolicious-Command-migration-0.14 by LIKHATSKI https://metacpan.org/release/LIKHATSKI/Mojolicious-Command-migration-0.14
12:46 jberger Well there's a rather generic name used for a specific purpose
12:59 Kripton joined #mojo
13:26 asarch joined #mojo
14:09 Kripton joined #mojo
14:38 sugar joined #mojo
15:19 Kripton joined #mojo
15:22 jabberwok Attempting to follow the cookbook for adding commands. I'm probably missing something obvious?  http://www.pastebeest.com/D0
15:24 jabberwok I assume the example's Application should have the "use lib 'lib';" and "use MyApp::Command::spy;" lines ?  still, nothing happens.
16:01 lluad joined #mojo
16:15 jberger jabberwok: what happens when you run ./myapp.pl help
16:15 jberger Do you see your command in the list?
16:17 jberger Also try ./myapp.pl eval -V '\@INC'
16:30 disputin joined #mojo
17:05 dod joined #mojo
17:07 dod joined #mojo
17:08 jabberwok jberger: both those produce no output at all
17:13 sri so, any thoughts on this little change? https://github.com/kraih/mojo-pg/commit/eb027c60ab18406e04b44f323f357a3190914bff
17:13 sri think what i like least is that there's so many ways to deal with json now
17:14 sri $pubsub->json('foo') vs $db->query('...', {json => {...}}) vs $db->query(...)->expand->hash
17:15 sri no to mention $c->on(json => sub {...}) in mojolicious
17:16 sri oh and $tx->res->json
17:17 sri $ua->get(... => json => {...}) too
17:17 sri hmm, guess it's moot... there's so many ways already ;p
17:18 jberger jabberwok If neither of those produces output something else is wrong
17:19 jberger What does your application script look like?
17:24 jberger Also it really should be the application script that sets up your library paths
17:25 jberger Oh wait, that is your application script!
17:25 jberger Well that's not gonna work :p
17:25 jberger That should be lib/MyApp.pm
17:27 jberger http://mojolicious.org/perldoc/Mojolicious/Guides/Growing#Simplified-application-script
17:27 jberger And that's your script
17:27 jberger jabberwok: ^^
17:32 PryMar56 joined #mojo
17:50 Kripton joined #mojo
17:58 jabberwok i don't even like "the simpsons" but this is a "D'Oh!" occasion.  Thanks
18:02 jabberwok perhaps the "1;" line at the end of "# Application" listing in http://mojolicious.org/perldoc/Mojolicious/Guides/Cookbook#Adding-commands-to-Mojolicious should be changed to "Mojolicious::Commands->start_app('MyApp');"
18:10 good_news_everyon joined #mojo
18:10 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/vVvDZ
18:10 good_news_everyon mojo/master 788c3f5 Sebastian Riedel: mention @INC
18:10 good_news_everyon left #mojo
18:16 disputin joined #mojo
18:17 disputin joined #mojo
18:36 disputin joined #mojo
19:06 cpan_mojo Mojo-Pg-2.25 by SRI https://metacpan.org/release/SRI/Mojo-Pg-2.25
19:17 jabberwok cheers sri
19:21 jberger jabberwok: why would you add the start_app at the bottom of the class file?
19:27 sri and of course there is no response :S https://www.reddit.com/r/perl/comments/4bvdny/die_if_fork_eq_hostile/d1dmrnw
19:28 jberger I don't expect there to be one
19:28 jberger its reddi
19:28 jberger t
19:28 jberger we made our point, future readers will see it too
20:20 dvinciguerra joined #mojo
20:32 noganex_ joined #mojo
20:43 ribasushi replied on the reddit thread
20:43 ribasushi Grinnz: any word on my question from this morning?
20:47 sri ribasushi: the topic is over
20:49 ribasushi sri: there is a glaring inconsistency, I wanted to clarify what happened: did Grinnz do a 180, or did he decide to drop it
20:49 ribasushi it's important for me to know
20:49 jberger its in the logs
20:50 jberger starting here: http://irclog.perlgeek.de/mojo/2016-03-25#i_12237080
20:50 * ribasushi will reread again after dinner, I may have missed a point somewhere, in which case mea culpa
20:51 jberger and much more close to the resolution is here: http://irclog.perlgeek.de/mojo/2016-03-25#i_12239006
20:52 ribasushi yes, this was my understanding too, but nowehere in the log was there a "sole focus on 5.8 compat" as stated in the review
20:53 ribasushi hence me asking
20:54 sri i have answered https://www.reddit.com/r/perl/comments/4bvdny/die_if_fork_eq_hostile/d1eilrk
20:57 sri and mojolicious is pretty damn successful at it, perhaps you should consider that we are right
20:59 sugar joined #mojo
21:01 sri hmm, i might have missed the inflammatory bit the first time reading it
21:01 sri "move fast and break stuff"
21:02 sri that's meant to stir up drama isn't it?
21:03 sri honestly, i don't value your opinion enough to bite, sorry
21:10 ribasushi no, that was not intended to "stir up drama", it was added because it was relevant
21:10 jberger in your opinion, but we have already answered that exact question so what purpose does it serve?
21:10 jberger you get your say again, well good, that's taken care of
21:11 jberger we know your opinion
21:11 jberger (/me really needs to take a cool down)
21:11 ribasushi there was a complaint "nobody answered on the reddit thread" just couple hours ago
21:11 ribasushi now there is a complaint "someone answered"
21:11 jberger him
21:11 jberger we wanted to hear from HIM
21:12 ribasushi which you probably will soon enough
21:12 * jberger goes and does anything else
21:12 * ribasushi goes for that dinner
21:22 sri literally look at JSON::Tiny, people can't even keep up with merging patches from upstream
21:22 sri nothing backs up your opinion
21:22 sri where are the super successful super fragmented web frameworks?
21:23 sri (maintained by a small team of volunteers)
21:28 sri it's great people like Mojo::DOM, but it is a result of the way we manage mojolicious
21:30 sri i guess outsiders don't see the interconnectedness, how Mojo::DOM grows with Test::Mojo or Mojolicious::Plugin::TagHelpers
21:31 sri the different parts are pushing each other forward
21:34 dvinciguerra joined #mojo
21:37 sri bottom line, it's pointless to argue for mojolicious to be split up, because it can't exist that way
21:37 sri it would be too expensive
21:44 sri i can with certainty say that the few people donating so i can put in a few extra hours every month do get a lot more bang for the buck because mojolicious is managed efficiently
21:49 sri still think ribasushi is trolling, but collecting these thoughts for future reference
22:10 sri to illustrate, here's one recent refactoring https://github.com/kraih/mojo/commit/85504faa5d435193d9c98ebd6136e2c73ec44e88
22:12 sri that would have been at least 5 separate dists
22:13 sri honestly, i prolly wouldn't have found the correct way to stop accepting connections
22:14 sri and this refactoring resulted in a dozen smaller ones that made the ioloop much more efficient
22:15 sri fixing just as many bugs, like one where stop_gracefully had to use a resurring timer to avoid a timing problem
22:15 sri s/s/c/
22:15 sri these things have ripple effects across the whole toolkit
22:17 sri this is why we can do what we do with just 8000 lines of code, about the same number as the catalyst core dist
22:17 sri it's efficient
22:18 * sri steps off the soapbox
23:08 jberger Hear hear
23:18 s1037989 Amen!
23:34 pink_mist re convos and it needing Redis ... iirc a guy called FatalNIX was working on a pure perl thing that was at least meant to be able to replace Redis .. I should poke him and ask him how that's coming along
23:37 CandyAngel On a lighter note, this is probably the fabric I will be making my Mojolicious dress with: http://tinyurl.com/jhgs7qs
23:38 CandyAngel (I still haven't figured out why I associate that colour with Mojolicious)
23:38 sri was about to question the color :)
23:38 dvinciguerra joined #mojo
23:43 sri it could be on the sticker though https://www.stickermule.com/marketplace/10014-mojolicious-night-sky
23:44 s1037989 pink_mist: why no redis for convos?
23:48 pink_mist s1037989: there were some comments about it being a shame it needed a backend DB service; if one could get a pure perl alternative, that's just mean it's a regular perl dependency
23:48 pink_mist *that'd
23:50 s1037989 Cool! That would be good! Thanks!
23:51 pink_mist don't thank me, and especially don't thank me *yet* :P rather thank FatalNIX, if what he has is actually usable :P (he's even in here...)
23:52 sri i think a file backend would be better
23:53 sri think the biggest problem convos had initially was that it was supposed to be a hosted solution shared by multiple people, instead of something individuals run on their server for personal use
23:54 pink_mist yeah, that's my biggest issue with it too
23:58 sri batman should be on the right track now though, from what i've seen

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