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

IRC log for #mojo, 2017-08-15

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

All times shown according to UTC.

Time Nick Message
00:06 veryrusty joined #mojo
00:10 gryphon_ joined #mojo
00:31 PryMar56 joined #mojo
00:39 tchaves joined #mojo
02:04 noganex_ joined #mojo
02:19 bwf joined #mojo
02:33 veryrusty joined #mojo
02:54 tchaves joined #mojo
02:55 bwf joined #mojo
03:18 stryx` joined #mojo
03:46 veryrusty joined #mojo
04:09 itaipu joined #mojo
04:13 stryx` joined #mojo
04:14 veryrusty joined #mojo
04:21 itaipu joined #mojo
04:21 preaction if you have some topic in mind, i likely could write about it
05:42 inokenty-w joined #mojo
06:08 mtj joined #mojo
06:47 veryrusty joined #mojo
06:57 Vandal joined #mojo
07:20 AndrewIsh joined #mojo
07:22 ribasushi joined #mojo
07:26 ribasushi joined #mojo
07:36 sri well, obvious topic would be role extensions for Mojo::UserAgent and friends
07:38 sri Minion could really use some blog posts too though
07:41 trone joined #mojo
08:02 CHYC joined #mojo
08:07 marcus I was pondering doing a role post.
08:08 petru joined #mojo
08:20 rshadow joined #mojo
08:34 dod joined #mojo
08:35 dod joined #mojo
08:42 veryrusty joined #mojo
09:04 marcm joined #mojo
09:10 dod joined #mojo
09:13 marcm joined #mojo
10:25 marcm joined #mojo
10:56 sri do it!
10:59 marcm joined #mojo
11:03 sri i really hope people are able to figure out the minion foreground feature, it's so damn useful
11:04 sh14 joined #mojo
11:04 sri minion could really use more docs
11:09 * ashimema hasn't dug far yet.. but in a brief look didn't understand minion foreground yet
11:24 dustinm joined #mojo
11:29 kes joined #mojo
11:29 mishanti1 We're having some interesting adventures with 'Internet Explorer Next Generation' AKA Chrome and playback of media content. We noticed a media player bugging out properly on chrome, but not Firefox, IE etc.
11:29 mishanti1 The difference being that Chrome always does two requests, first one with the header 'Range: bytes=0-' , then a "normal" GET without said header.
11:30 mishanti1 Is the "proper" way of handling this (pleasing Chrome) to have our Mojo return 206 instead of 200 when Range is specified? Anyone here ever run into this before?
11:30 marcm joined #mojo
11:51 sri yes, Mojolicious::Static should support Range: bytes=0-
11:51 sri it does not yet
11:54 sri or does it?
11:54 sri don't actua;;y know
11:56 pink_mist I'm pretty sure $c->reply->static supports it ... but whether that relies on Mojolicious::Static or not is something I don't know
12:04 mishanti1 I'll remember to look into Mojolicious::Static for this. Might be that we're could benefit from serving the content differently.
12:11 mishanti1 Seems like Chrome is also playing nice with media if you straight up disallow ranges. So if anyone run into this and have small enough media that chunking makes no sense then that is an option.
13:02 gryphon joined #mojo
13:05 gizmomathboy joined #mojo
13:17 Peppard joined #mojo
13:30 veryrusty joined #mojo
13:31 itaipu joined #mojo
13:35 vicash sri: i used the Minion foreground feature to debug a job. it was amazingly easy compared to the alternative of using logs
13:36 itaipu joined #mojo
13:50 petru joined #mojo
14:01 AndrewIsh joined #mojo
14:03 itaipu_ joined #mojo
14:07 mrEriksson Hello! I'm getting "Mojo::Reactor::Poll: I/O watcher failed: Can't call method "on" on an undefined value at ... when I try to do Mojo::IOLoop->client().
14:08 dod joined #mojo
14:08 mrEriksson Are there any additional stuff needed to get this going? Works like a charm on other systems. So the code it self should be good.
14:09 pink_mist perhaps your version of Mojolicious is too old?
14:10 mrEriksson Shouldn't be, I've done good and created a cpanfile used on all hosts
14:11 mrEriksson The only difference is that this is the first redhat host, others are suse, debian and ubuntu
14:13 itaipu joined #mojo
14:14 pink_mist ah, well then the problem is probably that perl on redhat isn't what most people would refer to as perl
14:15 pink_mist try installing perl-core on redhat
14:15 pink_mist because that's what most people refer to as perl
14:23 itaipu joined #mojo
14:28 mrEriksson perl-core? Ok, thanks, I'll check that out
14:28 mrEriksson Currently I'm pulling in a newer version with perlbrew to see if that makes a difference
14:34 Grinnz make sure you reinstall any local modules you might have installed for the new perl, modules are generally not compatible across different perl versions and arches
14:36 mrEriksson Thanks for the hint
14:36 mrEriksson Waiting for perlbrew to finish, if that solves things, I guess it is related to sucky perl in rhel
14:37 mrEriksson On another note, IOLoop and Minion are two pieces of totally awesome software :)
14:38 Grinnz it's generally best to build your own perl for development regardless of how good the system Perl might be. it's still set up for the system to use
14:39 mrEriksson Yeah, I always use perlbrew for development
14:39 Grinnz you can use perl-build to build a Perl without the complications of perlbrew, useful for deployment
14:40 mrEriksson But what would be the best way to distribute something perlbased? What I've done so far is to use docker, which is perfect for microservices, but not as perfect in some other cases
14:40 mrEriksson Oh, need to check that out, thanks!
14:40 Grinnz i go with perl-build and carton for version pinning
14:42 mrEriksson I'll give that a look, thanks!
14:50 mrEriksson Erhm
14:51 mrEriksson So if use the wrong server address with IOLoop->client() you will get the exact same error as I did. Will not comment further on that :P
14:55 sri if you're deploying on suse linux you can add the perl development repo and get fairly new-ish perl modules for most of cpan as packages
14:57 sri that's how i deploy mostly, system perl on opensuse or sles, and devel:languages:perl packages
14:57 sri even for our enterprise distribution there's already postgres 9.6 packages
14:58 mrEriksson Might be worth checking out too, thanks
14:59 mrEriksson Mostly though, I do dev on sled and deploy in docker containers, since it is mostly webstuff anyways
14:59 foursixnine joined #mojo
14:59 sri http://download.opensuse.org/repositories/devel:/languages:/perl/SLE_12_SP2/
15:00 mrEriksson Am I detecting some hidden suse marketing? :-)
15:00 sri open suse marketing
15:00 mrEriksson Heh
15:01 sri but even if i wasn't working for suse, tumbleweed has won me over
15:02 mrEriksson Any particular reason?
15:02 mrEriksson I've used opensuse a couple of times but feelt it was a bit to bleeding edge
15:02 sri yea, it is very bleeding edge
15:03 sri but i've been using the same installation on my work laptop for almost 10 months now, without real problems
15:03 sri updating every other week
15:03 stryx` joined #mojo
15:04 mrEriksson for workstations, I've been using SLED for quite a few years now. Got fed up with thing always breaking at the wrong time and SLE has been very stable for most of the time. Except when snapshots fill the root drive
15:04 sri for a rolling distro it's very very stable
15:04 mrEriksson Agreed
15:05 mrEriksson I spend a good part of my time putting out fires, so need to make sure it works
15:13 marcm joined #mojo
15:13 sri fair enough
15:14 omgponies joined #mojo
15:14 sri SLE 12 is fairly good... just don't like 11
15:15 sri that one shipped with perl 5.10.0
15:15 sri :S
15:27 itaipu joined #mojo
15:49 Pyritic joined #mojo
16:15 mcsnolte joined #mojo
16:15 PryMar56 joined #mojo
16:15 Pyritic joined #mojo
16:31 petru joined #mojo
16:48 marty joined #mojo
16:58 itaipu_ joined #mojo
17:03 itaipu joined #mojo
17:08 rshadow joined #mojo
17:23 itaipu_ joined #mojo
17:36 mib_cprm7m joined #mojo
17:43 Pyritic joined #mojo
18:43 itaipu joined #mojo
18:52 itaipu joined #mojo
18:58 marcus I set up an unofficial mojo telegram group at t.me/perlmojo
19:14 jberger is there a way to escape in Mojo::DOM so that I can select a tag like dateTime.iso8601 and it won't interpret the iso8601 as a class
19:16 jberger hmm guess and ye shall find
19:16 jberger slash
19:16 jberger should we document that?
19:16 marcus wow, never realized Mojo::DOM and Mojo::ByteStream doesn't inherit from Mojo::Base
19:18 marcus so much for my plan of showing how to extend Mojo::ByteStream in my blog post :)
19:18 Grinnz i didnt know . was valid in tag names
19:19 Grinnz marcus: theyre not hash based objects so yeah :P
19:20 marcus Mojo::DOM::HTML is tho.
19:20 marcus Grinnz: yeah, the distinction makes sense, I just never considered it.
19:21 marcus no roles for Mojo::File either then.
19:21 Grinnz Mojo::Bytestream->Mojo::Base::with_roles(...)->new should work, fwiw
19:22 Grinnz it doesnt actually use the hash part of the object for applying roles, so
19:22 sri Mojo::File::with_roles should exist, it's an oversight that it wasn't added
19:25 petru joined #mojo
19:31 jberger so I'm almost done porting XMLRPC::Fast to Mojo (something I've been tinkering with on-and-off for the past few days) and I've come to the last stumbling block
19:32 jberger it defines a tag for base64 encoded data
19:32 jberger and I'm wondering how I should represent that
19:32 jberger in perl space
19:32 sri marcus: now it works
19:33 sri looks like the github bot is slow today
19:33 Grinnz Variable::Magic, String::Tagged?
19:33 Grinnz not sure what you mean exactly by 'represent that' so just guessing
19:33 jberger Variable::Magic sounds overkill
19:33 marcus sri++
19:34 jberger I just mean, should I wrap it in an object, say Mojo::XMLRPC::Base64 or should I use Mojo::ByteStream for that (I don't like that idea) or maybe a reference to a string scalar
19:35 sri https://github.com/kraih/mojo/compare/477514a381189ea3e92dbe496b9ec6498fad9b08...7749d4c4035ed993b28c0f98f81879a20745f31c
19:35 Grinnz an object would be the most generic way
19:39 good_news_everyon joined #mojo
19:39 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/v7FVs
19:39 good_news_everyon mojo/master a6b83c3 Sebastian Riedel: with_roles should be available for all objects
19:39 good_news_everyon left #mojo
19:40 good_news_everyon joined #mojo
19:40 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/v7FVE
19:40 good_news_everyon mojo/master 7749d4c Sebastian Riedel: forgot to add the tests
19:40 good_news_everyon left #mojo
19:40 Grinnz there it goes :P
19:48 itaipu joined #mojo
19:55 marcm joined #mojo
20:00 mib_t3k0zi joined #mojo
20:03 marcus https://gist.github.com/e1575625777cbe649383b0eb844dd28c yay
20:07 itaipu joined #mojo
20:16 arcanez why you take all the moose? :(
20:19 mishanti1 Ok, roles seems neat.
20:20 preaction because there wasn't any squirrel
20:20 mishanti1 Might please our die-hard .NET-guy.
20:22 marcus First we take the Moose, then we take the Bullwinkle
20:23 PopeFelix joined #mojo
20:25 PopeFelix Someone please check me here. If I do Mojo::Useragent->new->get( $url => sub {}), I've got to do everything I want to get done inside of that sub {}, or close over something external to that sub and wait for the IOLoop to finish, correct?
20:26 PopeFelix I hope that made sense.
20:27 Grinnz you don't want to do that in one line, because the Mojo::UserAgent object will go out of scope before the transaction completes
20:28 PopeFelix Oh, sure. That was just for the sake of compactness
20:28 Grinnz all the code after you call ->get() with a callback will run before the request completes. only the code in the callback will run after you get a response
20:28 Grinnz if that helps
20:28 PopeFelix right, that's what i understood
20:29 * PopeFelix is trying to update a big application to run asynchronously.
20:29 Grinnz regarding "wait for the ioloop to finish", that entirely depends whether the ioloop is already running when you make this request
20:29 Grinnz if you're already in a running ioloop, that "waiting" just happens by itself
20:29 PopeFelix Grinnz, point taken
20:30 Grinnz otherwise, you need to start the ioloop until your response callback is called
20:30 Grinnz which $delay->wait is convenient for sometimes
20:30 PopeFelix ok
20:30 Grinnz or $future->get, etc
20:30 PopeFelix sure
20:38 Pyritic joined #mojo
21:33 gryphon joined #mojo
21:49 sri jberger: holy shit, trump outed himself as a neo-nazi
21:49 sri http://www.politico.com/story/2017/08/15/full-text-trump-comments-white-supremacists-alt-left-transcript-241662
21:49 sri "TRUMP: Okay, what about the alt-left that came charging at us – excuse me – what about the alt-left that came charging at the, as you say, the alt right? Do they have any semblance of guilt?"
21:50 pink_mist I'm not surprised in the least; disappointed, but not surprised
21:51 sri i'm disgusted
22:00 rshadow joined #mojo
22:08 marty Trump is a megalomaniac, no doubt about that. And the neo-nazi group are evil ignorant haters.  However, in all fairness, it was the alt-left crowd that tried to shut down the neo-nazi group from exercising their constitutional right to demonstrate.
22:09 marty The alt-left crowd can be quite intolerant as well.  SJW and all that.
22:10 haarg wow
22:10 Grinnz this is probably not a productive discussion to start
22:11 haarg the only thing i say to that won't break the code of conduct is "wow"
22:14 marty Not really a discussion.  Everybody hates neo-nazi's.   The only question is, if they should be censored then why stop there, Muslims can be kinda crazy, and feminists and mormons, don't forget the mormons.
22:14 haarg wow
22:15 sri wow
22:19 kivilahtio_ joined #mojo
22:20 sri just to be clear, if anyone here has the slightest bit of sympathy for the "alt-right" movement, please leave this community immediately and never come back
22:21 marty But please remember, the 1st amendment to the constitution specifically states "the right of the people peaceably to assemble".  peaceably being the operative word.
22:22 marty alt-right is a catch all phrase.  That group was definitely neo-nazi and KKK.  The true scum of the earth.
22:22 Grinnz and it sucks for anyone who was peaceful there. but you should really stop while you're behind
22:24 marty There were very few peaceful people there, from both sides.  The video evidence is compelling in that regard.
22:24 haarg those damn alt-left.  they did so much property damage to that dude's bumper.  violence comes from both sides you know.
22:25 marty The crazy car guy is a seperate issue.  He should in jail or worse for the rest of his life. Everyone agrees with that.  but that is not related to anything. just a crazy individual
22:26 haarg wow
22:43 jantore joined #mojo
22:46 sri marty: that statement was absolutely not ok
22:46 marty What statement?
22:48 sri distancing "the crazy individual" from the organized hate groups
22:49 marty If I recall, a individual muslim drove his van into a crowd in France.  Does that mean be blame all Muslims?
22:49 Grinnz muslims are not a hate group. please stop with the false equivalences
22:50 sri not all nazis are bad, seriously?
22:50 marty I dont think the neo-nazi's planned that attach or had any forethought about it.  I think that crazy guy decided to do it by himself in a heat of rage.  There is no evidence that it was pre mediated.
22:51 marty I'm not saying that at all, sri
22:51 haarg just one guy.  a single individual nazi.
22:51 haarg not like those other ones
22:52 haarg the good nazis
22:52 marty Do you have any evidence that the neo-nazi group planned that attack?
22:52 Grinnz why does that matter?
22:53 marty I mean comon, there is ample evidence that ISIS and others are organized terrorist groups.  Then, yes, of course, individuals who identify as that group must be considered as part of the group and their actions taken into account.
22:54 jberger JHFC
22:54 marty well ya know, evidence is usually needed to link an individual to a group.  Otherwise I could choose any random act of violence and claim they belonged to the "Grinnz" group
22:54 zen was this the channel where if you asked a perl question it was considered offtopic?
22:54 sri marty: and i don't give a fuck about your 1st amendment, spreading nazi ideology is forbidden here, and that's good!
22:54 Grinnz why does that matter?
22:55 jberger I step away for an hour with my brother and he puts himself in the ... shoes ... of ...
22:55 sri here being germany
22:55 Grinnz neonazis are a hate group, they don't have to be linked to the driver to be a hate group
22:55 pink_mist zen: I think that was #perl
22:55 Grinnz that is literally their ideology
22:56 marty I agree, sri.  I think the sensitivities of the German people should be taken into account with regard to anything nazi related.
22:56 zen politics is for puppets
23:00 marty sri, but to be clear, nobody is spreading nazi ideology.   I'm simply stating that censorship, even for the most hated groups, can be a very slippery slope.
23:00 Grinnz there's a reason that slippery slope is the name of that fallacy
23:01 Grinnz we are humans, we are capable of nuance (usually)
23:01 jberger Nazis, including people at that rally encouraged violence. When one of their less stable members takes it into their own hands the group isn't without blame
23:02 jberger I don't like the notion of the antifa, I don't think answering violence and hate with violence is good. But there is no equivalence there.
23:03 jberger Nazis want innocent/uninvolved people who are unlike them to die and antifa, for that reason, want to punch them. No equivalence
23:04 haarg someone please explain to me when punching nazis became a bad thing
23:05 marty If it was premeditated I would agree.  But there is no evidence that the group promoted the car crash.
23:05 marty wow
23:05 jberger marty don't be naive
23:05 jberger They use violent rhetoric constantly
23:05 marty jberger, I'm not.  do you have evidence?
23:05 jberger Fuck off
23:06 marty you guys, no profanity please
23:06 haarg but he's being so reasonable
23:06 Grinnz do you understand nazi ideology? whether they are linked to the car crash is completely irrelevant
23:06 haarg please provide this evidence that nazis are bad.  aren't you interested in having a reasoned conversation?
23:07 marty You could accuse every president of the US for, not only using violent rhetoric, but real violence also.
23:07 pink_mist what the hell
23:07 jberger No whatabouts.
23:08 jberger Nazis have a defined ideology and history that I don't need to debate.
23:08 pink_mist marty: please stop. your continued insistence that neo-nazis are not in and of themselves promoting violence is insane and utterly wrong.
23:08 marty ok.  well it's off topic now.   bottom line is, nazi's are evil, kkk is evil, they should be shunned and condemed at every possible moment, but, the constitution protect everyone, even if we do not like them.
23:08 Grinnz no it doesn't
23:09 Grinnz it protects those who are not infringing on the rights of others
23:09 marty Grinzz, agreed.   Demonstrating is everyones right.
23:10 haarg why can't you see both sides of the debate?  on the one side, you have nazis.  and on the other, you have anti-nazi protesters getting run over by a nazi.  they're really both the same when you think about it.
23:11 marty To much emotion.  I'm out.  Peace and love to all!
23:11 jberger Demonstrating peaceably, but I claim a group that continually spews hate and violent speech cannot peaceably assemble and certainly they didn't here
23:12 haarg "too much emotion" when someone condemns a nazi murderer
23:20 jberger Dont think anyone who follows me on Twitter can be too surprised by my reaction
23:24 sri marty: watch this https://www.youtube.com/watch?v=P54sP0Nlngg
23:24 sri around minute 4 you can see the "crazy individual" with his friends and their fasces shields
23:25 marty It is very interesting how some topics can be so divisive even between people of similar political views, ie, my views on social issues are usually considered far left.  Interesting
23:26 sri anyone who makes excuses for nazis is a bad person in my book
23:30 haarg if you think your political views are far left
23:30 haarg you should probably talk to more people
23:30 sri and a reminder of how dangerous free speech and fascism can be (made by your own government in 1947) https://www.youtube.com/watch?v=23X14HS4gLk
23:31 marty Well I guess I'm the only one who thinks this way, oh... except the most liberal institution in the US.   https://www.vox.com/policy-and-politics/2017/8/12/16138326/aclu-charlottesville-protests-racism
23:31 haarg i'm not sure if you know what liberal means
23:31 haarg or "left"
23:32 marty haarg, more than you know
23:33 marty I am probably more left than most in this room
23:33 marty at least on social issues  :)
23:36 marty It's easy to hate nazi's   (inglorious bastards anyone?) :)   But when events like this happen, and emotions run high, that is the time to step back and see the forest through the trees.
23:41 haarg a film reference as some kind of cultural touchstone to explain how people don't like nazis?
23:42 haarg is this copy/pasted from somewhere?
23:49 marty Well I hope we are all still friends.   :)
23:51 jberger Defend free speech all you want, but don't straw an equivalence between those protesters and their messages with the ones from the counter protesters
23:51 jberger *draw
23:52 marty what?   Who's doing that?
23:53 haarg you.
23:53 haarg you did.
23:53 haarg explicitly
23:53 marty I agree with the ACLU.   If you think it's wrong, then so be it.
23:54 marty haarg, wrong
23:54 pink_mist marty: if you think you didn't, then you must not have read your own lines.
23:55 jberger https://irclog.perlgeek.de/mojo/2017-08-15#i_15021321 and the line after
23:56 jberger Even the term alt-left is alt-right dog whistle
23:56 marty oh ya.  Of course.  They have equivalence concerning their rights to demonstrate.  thats all
23:57 haarg just kick it past those goal posts.  wait they seem to be getting a bit farther away.  hmm, i can't quite see them anymore.  but still, kick it past them.
23:57 marty That is the position of the ACLU.  That is all I was saying.
23:58 marty haarg.  Go read up on the ACLU position.  If you agree with that, they we agree, if not, then we do not.

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