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

IRC log for #mojo, 2016-03-24

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

All times shown according to UTC.

Time Nick Message
00:01 anonymous Grinnz, I'm sorry. Especially after seeing the list of Mojo modules you've created: https://metacpan.org/author/DBOOK
00:02 anonymous Both sri and Grinnz are good people. jberger, is this where you come in to make peace between these two individuals. Can it work. Look at all that Grinnz has created.
00:03 anonymous May God bless you jberger with similar wisdom given to King Solomon years ago.
00:04 anonymous Please enable peace.
00:06 jberger Would that i had that wisdom
00:08 jberger I want Grinnz in the fold too, but i can't see how to square that with sri's request that community members respect his wishes on how the distribution is packaged
00:09 jberger I'd hate to see that cause him to leave but by putting this policy in place we leave it in his court
00:12 anonymous joined #mojo
00:13 sri for the record, at no point did Grinnz attempt to get our blessing for a fork
00:15 sri in fact, it has been my impression that he has been evading all discussions
00:15 sri which is what made the whole thing escalate, we never got to talk
00:17 mattp_ Grinnz_: have you chosen to no longer maintain dom::tiny as a result of the negative reaction?
00:17 mattp_ im confused what the issue is if you are willing to maintain the managerial load
00:18 anonymous I wish King Solomon was here. What would King Solomon say? Grinnz, please make things right. Sri, you're bless with Perl kungfu. Is it possible for parts of Mojo / Mojolicous to run on Perl v5.8?
00:19 mattp_ wat
00:19 jberger 5.8 is close to EOL anyway
00:19 pink_mist close? it's been EOL for years
00:19 jberger I mean the toolchain isnt that far away from not supporting it
00:19 anonymous ok. yeah, it's nearly 7 years from 5.8.1.
00:20 anonymous jberger, likely doesn't make sense then
00:20 jberger There are things that we could really use in 5.16
00:20 sri someone could take over https://github.com/jamadam/mojo-legacy
00:21 sri if the goal is truly 5.8 support
00:21 sri and not just a strawman
00:21 mattp_ json::tiny pains me more than dom::tiny
00:21 mattp_ like the json namespace needs MORE fucking fragmentation
00:23 anonymous Grinnz, please make things right if possible.
00:24 jberger Especially today can i ask that we keep the profanity down?
00:24 jberger Just to calm frayed nerves (mine included)
00:25 mattp_ jberger: my bad
00:25 jberger It's OK, it is allowed
00:25 jberger I just would like to keep things calm
00:26 jberger I'm gonna go to the gym for a bit to clear my head
00:26 anonymous I have a lot of respect for you Grinnz. Also, for the team. Boy, I wish King Solomon would just appear. I'm certainly not qualified. Oh jberger, may you be blessed with the wisdom necessary.
00:26 jberger Biab
00:26 anonymous i'm out too. this has been a sad day.
00:27 jberger That's true
00:31 mattp_ sri: i dont follow your argument of a 'hostile' fork taking resources away from mojo
00:32 mattp_ surely some bug reports dont make it upstream, but that does not take time away from mojo
00:32 sri that bug would have been reported against mojo otherwise
00:33 sri that exact scenario happened with JSON::Tiny
00:33 sri which is in the irclog jberger linked to above
00:35 sri we might even have blessed a fork aiming specifically for perl 5.8 compatibility, but we were never even asked
00:35 sri Mojo::DOM58 or so would have been fine with me
00:35 sri just for the record
00:36 sri but at this point i think Grinnz is being dishonest, and perl 5.8 support is just a strawman argument after the fact
00:51 good_news_everyon joined #mojo
00:51 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/vaNfF
00:51 good_news_everyon mojo/master 17d7f84 Sebastian Riedel: fix another typo
00:51 good_news_everyon left #mojo
01:01 odc on that subject, it would be nice if JSON::PP could be made faster. That would pretty much remove the need for JSON::Tiny
01:01 haarg i tried to do that but didn't have much luck
01:02 odc haarg, you didn't see any place that could be optimized?
01:03 haarg i tried to adapt Mojo::JSON to add some of the options from JSON::PP, but the result wasn't really any faster than JSON::PP
01:04 odc ?? you mean the opposite right?
01:04 haarg no
01:04 odc >but the result wasn't really any faster than JSON::PP
01:05 odc but Mojo::JSON is faster than JSON::PP
01:05 haarg yes
01:05 haarg and once i added the extra stuff, it wasn't anymore
01:05 odc oh i see
01:05 sri it's really hard to do all those features in pure perl and stay fast
01:06 sri prolly impossible
01:06 haarg and there's some stuff that can't be reasonably added with the approach Mojo::JSON uses
01:06 odc i guess the real problem here is that JSON::XS should be in core
01:09 haarg you'd need a new name for it
01:10 odc sri, anyway. You already have like the most successful module on cpan. Whatever Grinnz does it won't do you much harm.
01:12 sri we also have very limited resources for maintaining mojolicious
01:13 sri any new fork that drains more will hurt
01:13 sri and i know there are people planning forks for more modules in the mojolicious dist
01:13 jberger Xs json in core would be amazing
01:14 * odc prepares to fork Mojolicious into Jomolicious to troll sri
01:14 sri odc: and remember, i just said i would have blessed a focused Mojo::DOM58
01:15 odc sri, what would be your conditions to bless Json::Tiny now?
01:15 sri that it does something core can't provide
01:15 sri Mojo::JSON58? ;p
01:17 genio Molegacy   because   perl -Mole sounds fun
01:17 odc it provides speed and slight memory gains
01:17 odc heh
01:19 s1037989 Interesting...  I see your strawman concern.  Someone could create Mojo::JSON58 under the guise that it's for the purpose of running under 5.8, but a byproduct of this "worthwhile" fork would be that it doesn't have the rest of the Mojolicious codebase as a dependency, thereby getting the "feature" that many have begged for for a long time.  :/
01:20 odc jberger, JSON::Fast is available ;) We just need to fork Cpanel::JSON::XS now...
01:20 sri s1037989: i doubt they'd use a name like Mojo::DOM58, because it's just a strawman, and they really want a rebranding too
01:21 s1037989 A rebranding?  That seems unnecessary.
01:21 odc yeah, open source is marketing :(
01:21 sri i'm totally fine with modules like Minon::Backend::Pg91
01:22 s1037989 Yeah, sometimes that's straight necessary!
01:22 sri just do something the original can't and isn't interested in, talk to us, we agree on a name and it's all fine
01:23 s1037989 :D  Soooo simple!
01:23 sri i can't read minds, but if i had to guess i'd say that the true goal here is the old split mojolicious up into a million little distributions topic
01:24 sri but, people are dishonest about it, and make a huge mess for us
01:24 s1037989 Dumb!
01:24 sri and that's why i got mad earlier
01:24 s1037989 Something I've been curious about... I thought I read once that we shouldn't use the Mojo:: namespace, but a lot of people do.  What's your thoughts on that?
01:25 sri it's complicated, people should avoid namespaces core might want to use
01:25 sri generic stuff like Mojo::HTTP2
01:25 sri do not use that!
01:25 sri we want that
01:25 s1037989 So things like Mojo::Redis and Mojo::IRC are nbd because core would surely never get into that.
01:26 sri yea, it's best to ask here
01:26 s1037989 Ooh!  I'm gonna go reserve Mojo::HTTP2 right now!!  :D
01:26 sri ;p
01:26 s1037989 How about Mojo::Sendgrid https://gist.github.com/s1037989/c7e911fa61ed7d1651f9
01:26 odc makes me think I might release Mojo::URL::Shorten one day. I wonder if it's a good idea.
01:26 genio odc: You could help me by fixing WWW::Shorten with me
01:27 odc genio, oh you're the maintainer?
01:27 genio odc: just recently, yes. (matter of weeks)
01:27 odc genio, ok but how would i make it non-blocking?
01:28 genio odc: Currently it relies on a stupid LWP::UserAgent subclass that I'd like to get rid of.  But there are a lot of WWW::Shorten:: services I don't yet control, so I don't want to break everything.
01:28 genio I'm open to ideas, but maybe conversation about that shouldn't take place here as it isn't directly about Mojo
01:29 preaction why not Mojolicious::Command::nopaste?
01:32 odc preaction, because we use mojopaste here :p
01:33 preaction oh. i'm tired... you're doing _shortening_ not pasting...
01:33 odc tsss :)
01:34 genio :)
01:34 odc yeah, eventual it will be useful for that great irc bot i'm making
01:34 odc Mojo::IRC based of course
01:36 batman odc: have you looked at Mojo::IRC::UA?
01:37 odc batman, yeah but its functions do not really interest me
01:37 odc you think i should use it anyway?
01:37 batman Ok. Let me know if they could be improved or if somethings are lacking
01:38 odc ok
01:38 jberger odc: I'd take a mojopaste client to MCnopaste
01:38 batman Of course not. Why would you if the functionality doesn't interest you? :)
01:38 odc batman, well so far the thing that i miss is auto-reconnect
01:38 odc i'm stealing code from Convos :)
01:39 batman odc: I suck at auto-anything. Don't mind a PR though.
01:39 odc i'll see about that
01:39 batman :)
01:39 odc and while i'm here, i should mention that i'm looking for people interested in maintaining Mango
01:40 odc which i won't use at work anymore
01:40 odc maybe i'll contact that crazy russian fellow
01:42 good_news_everyon joined #mojo
01:42 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/vaNIA
01:42 good_news_everyon mojo/master 82fb2ce Sebastian Riedel: change a few words
01:42 good_news_everyon left #mojo
01:52 odc sri, so you've been working on Mojo almost everyday for several years now. How come you're not bored?
01:53 sri the web is a moving target, everyday something new
01:54 sri honestly, if i had more time i could do a hell of a lot more work, so much stuff on the todo list
01:54 sri only thing that annoys me is the politics of open source, like dealing with silly forks -.-
01:55 sri it's easier when your project is still small
01:55 odc hm. That's precisely why i'm trying to get out of webdev. Way too much new stuff all the time
01:55 odc sri, don't you miss he 90s?
01:56 sri i'm not that old, i started programming in the 00s
01:57 odc the 00s were good too. The amount of javascript was still limited
02:01 batman Grinnz_: really sucks that DOM::Tiny is hosted by sc. Makes it harder to follow up on changes.
02:01 mattp_ shadowcat? if theres not a github mirror there can be
02:01 batman Harder = not something I would do
02:02 sri and issues are on rt.cpan, which is the worse part
02:05 batman rt--
02:06 mattp_ ya, I never got why schmorp got so much heat for shitting on the RT situation. i sympathized with his point
02:06 odc rt is open source and in perl. So RT++ :D
02:07 mattp_ odc: thats well and good. it shouldnt be a forced imposition
02:10 TonyC it's a default, you can opt out
02:11 odc TonyC, how?
02:11 jb360 joined #mojo
02:11 asarch joined #mojo
02:12 TonyC you can set bugtracker under resources in your META.yml
02:13 TonyC https://metacpan.org/source/SRI/Mojolicious-6.57/META.yml like in this one
02:15 odc oh that
02:15 odc i thought there was an option i pause or something
02:15 odc in*
02:21 bwf_ joined #mojo
02:30 jberger TonyC: that's not an opt out
02:30 jberger That's meta data that users can choose to use
02:30 jberger But the bug queue still exists in rt
02:31 jberger And people who are in this channel encourage users to just use it as a quick way to file bugs even if we have used that "opt out"
02:33 mroy joined #mojo
02:34 TonyC all my dists use rt.cpan.org, so it's not something I've had to deal with
02:35 mroy jberger, batman, sri: Am I allowed to make Mojo::IOLoop::HoboCall for IOLoop interface to MCE::Hobo?
02:35 TonyC if you have new tickets being created on an RT queue where you've pointed bugtracker elsewhere, have you tried mailing rt-cpan-admin to fix it?
02:37 Rubes joined #mojo
03:02 mroy Am going to hold off on making Mojo::IOLoop::HoboCall.
03:06 inokenty-w joined #mojo
03:07 mroy left #mojo
03:16 lluad joined #mojo
03:20 sri talking about splitting up distributions https://twitter.com/TobyKurien/status/712536641391484929
03:20 sri hahaha
03:25 mroy joined #mojo
03:25 mroy Are folks allowed to make plugins for Mojolicious?
03:25 mattp_ mroy: of course
03:25 mroy ok. thanks.
03:28 mroy left #mojo
03:31 noganex joined #mojo
03:38 Rubes joined #mojo
03:39 jberger mroy: please do
03:40 jberger This situation has is completely different
03:40 jberger s/has//
03:44 mroy joined #mojo
03:46 jberger Ummmm you really don't want to check the #mojolicious hashtag at work
03:47 mroy jberger, thank you.
03:48 mroy The RTB advertisement has rather interesting adds and wish there is a way to disable it.
03:48 mroy https://client00.chat.mibbit.com/?channel=%23mojo&server=irc.perl.org
03:49 mroy This has no effect: https://www.rtbhouse.com/optout-page/
03:54 jberger mroy: mibbit is usually thought of as a stopgap irc client
03:54 jberger Have you thought about trying a full irc client?
03:55 jberger What os are you on?
03:55 mroy am happy to know there's another way. I'm on Mac OS X.
03:55 jberger What was the native on i was using ....
03:56 jberger https://www.codeux.com/textual/
03:57 jberger There are dozens of irc clients though
03:57 jberger Most free
03:57 mroy nice. thanks.
03:57 jberger Textual isn't free, but at least a while back you could essentially use the free trial forever
03:58 mroy $4.99 is okay for me to move off of mibbit. Am not liking the ads.
03:59 jberger Sure
04:00 Rubes joined #mojo
04:00 jberger There's a mojo based in browser irc client that batman and marcus wrote
04:00 jberger Called convos
04:01 jberger It's most useful if you have a server you can get to from the Web and have it running persistently
04:02 jberger I use it now personally
04:02 jberger But it can be non trivial to set up
04:02 mroy I will try Textual first.
04:02 jberger There's an ongoing effort at a facelift for it
04:03 jberger I think that's a good idea coming from something like mibbit
04:03 mroy batman++, marcus++ for Mojolicious-based app.
04:03 mroy It's not fun having mibbit in a windows with crazy ads at Panera.
04:03 jberger Once irc becomes indispensable part of your day then a persistent irc connection is really handy
04:04 jberger Hahaha no i can imagine not!
04:04 mroy Will try convos too, later.
04:06 s1037989 Convos++
04:07 Rubes joined #mojo
04:07 mroy My server machine with the Linux VM under Hyper-V is down. This is due to a Windows update and AMD graphics driver looping into a black screen. Hyper-V is cool. I was testing Windows 10 for the first time.
04:09 abra joined #mojo
04:15 mroy left #mojo
04:19 bwf_ joined #mojo
04:25 mroy joined #mojo
04:26 mroy left #mojo
04:27 mroy joined #mojo
04:30 mroy Am off of mibbit and now using the Textual Client for Mac OS X. Thank you jberger. Zero ads. :)
04:31 jberger mroy: yay!
04:32 mroy Now, I can see as well after increasing the font size.
04:33 melo joined #mojo
04:34 mroy left #mojo
04:38 commish joined #mojo
04:44 jbaumgartner joined #mojo
04:44 jbaumgartner hello
04:44 jbaumgartner is there a way to use the quote method from DBD::Pg when using Mojo::Pg?
04:49 sri http://mojolicious.org/perldoc/Mojo/Pg/Database#dbh
04:49 jbaumgartner perfect thanks!
05:03 Rubes joined #mojo
05:07 Rubes joined #mojo
05:16 irqq joined #mojo
05:45 punter joined #mojo
05:51 Rubes joined #mojo
06:18 good_news_everyon joined #mojo
06:18 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/vaN2Z
06:18 good_news_everyon mojo/master 1b0383f Sebastian Riedel: more consistent description
06:18 good_news_everyon left #mojo
07:16 dod joined #mojo
07:18 kivilahtio joined #mojo
07:21 dod joined #mojo
07:41 Vandal joined #mojo
07:50 sugar joined #mojo
08:11 csroli joined #mojo
08:15 dod joined #mojo
08:34 trone joined #mojo
08:37 AndrewIsh joined #mojo
08:48 cpan_mojo Stenciller-Plugin-ToMojoliciousTest-0.0103 by CSSON https://metacpan.org/release/CSSON/Stenciller-Plugin-ToMojoliciousTest-0.0103
09:07 melo joined #mojo
09:10 melo1 joined #mojo
09:12 Rubes joined #mojo
09:35 Rubes joined #mojo
09:41 osfabibisi joined #mojo
09:44 irqq joined #mojo
09:57 dustinm joined #mojo
10:23 cpan_mojo Mojolicious-Plugin-QuickPg-1.0 by AZJADFTRE https://metacpan.org/release/AZJADFTRE/Mojolicious-Plugin-QuickPg-1.0
10:27 dustinm joined #mojo
10:46 melo joined #mojo
11:09 dvinciguerra joined #mojo
11:44 mitya joined #mojo
11:57 CHYC May I please ask for some advice on non-blocking Mojo::Pg in a before_dispatch? The following appears to work for me, but I'm not sure if it's idiomatic: http://pastebin.com/3GDH1v1X
11:58 CHYC Is this the correct way, or is there some way to block within the hook until the stash is ready?
12:20 melo1 joined #mojo
12:34 jberger CHYC: i don't remember if non blocking hooks are supported
12:35 jberger But certainly non blocking under directives are
12:35 jberger Could that code go in an under?
12:36 Kripton joined #mojo
12:51 neilhwatson joined #mojo
13:20 mrEriksson joined #mojo
13:24 cpan_mojo Mojolicious-Plugin-QuickPg-1.1 by AZJADFTRE https://metacpan.org/release/AZJADFTRE/Mojolicious-Plugin-QuickPg-1.1
13:36 CHYC jberger: I hadn't thought of using an under. I'll give that a try. Thanks!
13:52 bwf_ joined #mojo
13:52 PryMar56 joined #mojo
13:58 mcsnolte joined #mojo
14:06 melo joined #mojo
14:41 punter joined #mojo
14:54 asarch joined #mojo
15:04 Kripton joined #mojo
15:18 Adura joined #mojo
15:20 CHYC jberger: Thanks for the tip. I have something that looks much nicer, but unfortunately it gives a 500 error (undeclared variables). Is there something that I'm doing wrong? http://pastebin.com/BGiVMHuH
15:21 jberger CHYC: in the under you should return 0
15:21 jberger Then call continue when done
15:27 CHYC Oh thank you so much for that, it works perfectly.
15:52 mcsnolte thanks again jberger for the MKE.pm mojo talk!
15:52 jberger I'm sorry you didn't get to see it
15:52 jberger how are things on your end?
15:52 mcsnolte np! i have the slides :)
15:52 jberger yes you do
15:52 jberger :D
15:53 mcsnolte and my co-workers are filling me in on all the cool stuff!  everyone loved it!
15:53 mcsnolte things are just fine on my ned, thanks :)
15:53 mcsnolte *end
15:54 osfabibisi joined #mojo
15:59 jberger I was really happy with the conversation afterwards
15:59 jberger the Q&A was kinda terse but I think people were just processing
16:00 jberger once we got back into the pizza the discussion started flowing and I (we) had a great time
16:02 good_news_everyon joined #mojo
16:02 good_news_everyon [mojo] kraih pushed 2 new commits to master: https://git.io/vaxYa
16:02 good_news_everyon mojo/master 0dd4803 Andreas Koenig: Fix two test names to match the test result
16:02 good_news_everyon mojo/master 0c10cd4 Sebastian Riedel: Merge pull request #940 from andk/master...
16:02 good_news_everyon left #mojo
16:05 Rubes joined #mojo
16:07 jabberwok dissecting the Mojo::Pg documentation: should Mojo::Pg::Database explain that: "Queries are first prepared and then executed. For any parameter to query() which is a hash reference containing the key 'json', that parameter's value will be converted, using Mojo::JSON's to_json(), to JSON text before being prepared."
16:08 sri jberger: found another fork policy https://bitcoin.org/en/posts/hard-fork-policy
16:09 jberger that's an even stronger environment though
16:09 jberger literal currency is harder to fork and maintain value than mindshare
16:10 sri i find it interesting to see another environment where forks are such a big problem
16:10 jberger indeed
16:11 good_news_everyon joined #mojo
16:11 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/vax3a
16:11 good_news_everyon mojo/master 6dd61db Sebastian Riedel: just call it a policy
16:11 good_news_everyon left #mojo
16:12 Adura Bitcoin Core is owned by Blockstream, not wanting a fork serves its financial interests.
16:13 Adura And by fork, I mean Bitcoin's blockchain.
16:29 sri oh, docker without virtualbox on os x now https://blog.docker.com/2016/03/docker-for-mac-windows-beta/
16:31 disputin joined #mojo
16:31 Kripton joined #mojo
16:34 jberger I thought docker used lxc
16:34 jberger does osx have something like lxc now?
16:35 sri seems it is using xhyve
16:36 sri which is built on Hypervisor.framework
16:36 sri so, might be something native
16:43 csroli left #mojo
16:46 sri jberger: oh, we just recently talked about citusdb, now it's all open source :) https://www.citusdata.com/blog/17-ozgun-erdogan/403-citus-unforks-postgresql-goes-open-source
16:46 jberger woah
16:46 * jberger reads
16:47 sri actually pretty cool what they've done, everything as normal postgres extensions
16:48 melo1 joined #mojo
16:49 Rubes joined #mojo
16:54 sri they have a very simple architecture
16:55 sri master node just keeps cluster state, and you can have multiple masters, and all your data is distributed across a pool of worker nodes only the masters talk to
16:56 sri and you decide the replication factor of your data
16:57 jberger Phil21: ^^
16:58 CandyAngel sri: What sort of situations would garner consent for forking Mojolicious (or parts of it)?
16:58 sri CandyAngel: Mojo::DOM58 for perl 5.8 support for example
16:58 sri we already have that with Minion::Backend::Pg91
16:59 CandyAngel So legacy support stuff is okay?
16:59 CandyAngel (generally speaking I mean)
16:59 sri think so, if the module name reflects it
16:59 CandyAngel Okie dokie
16:59 sri with a name like Mojo::DOM58 we get marketing value out of it
17:00 sri that's a win for us
17:01 sri jabberwok: that doc patch doesn't really reflect the style of the surrounding docs
17:03 sri but i'll leave it open, so other can chime in
17:05 sri is "PostgreSQL argument" a real term?
17:08 sri would be much cleaner to make a new json example below the non-blocking one and start with an entirely separate description
17:09 sri and include ->expand
17:09 sri if you wanted to explain the json stuff
17:12 disputin joined #mojo
17:13 oalders joined #mojo
17:18 dod joined #mojo
17:21 irqq joined #mojo
17:39 Rubes joined #mojo
17:42 dvinciguerra joined #mojo
18:02 Atog joined #mojo
18:04 jabberwok sri: very good. some of these bits i have had to read the source to see what's really going on, preferably others will not have to.
18:05 sri also nto a fan of the type cast mention
18:05 sri that part is a bit too complicated to just casually mention imo
18:06 sri you also have ::jsonb type casts
18:06 Atog Hi all
18:10 Atog I'm playing with mojolicious (already loving it!) figuring what the best way i can find to stream my audio collection to a browser.
18:11 Atog Got one qeustion: is it possible to access resources that have a '#' character in his name?
18:13 jabberwok it's not really fair to use ::json in the first example, specifically one describing "json round-trip" without giving the reader a way to look that up on google, at least, is it?
18:13 sri # means fragment, those don't reach the server
18:13 jberger Atog: if those need to make it to the server, they need to be url encoded
18:13 sri jabberwok: fair enough
18:15 Atog oh nice, so i guess i can use a hook to encode it before it hits the router
18:15 Atog (just reading about them)
18:20 sri jabberwok: does that work for you? https://github.com/kraih/mojo-pg/commit/6efec707ac0d63dde6a6e14d2d73b96998d5df5a
18:20 pink_mist Atog: no, your web browser literally doesn't send them to the server at all
18:21 pink_mist Atog: unless they're already url-encoded for your browser
18:22 sri jabberwok: actually, maybe using a ::jsonb type cast would make it more obvious
18:23 sri http://mojolicious.org/perldoc/Mojo/Pg/Database#query
18:24 Atog oh, so i only have to urlencode my requests
18:24 Atog i'm really that rusted :)
18:26 jabberwok sri:   Absolutely.  =tip of the hat=
18:27 sri jabberwok++
18:28 sri man, postgresql is so exciting
18:29 al237 joined #mojo
18:29 sri always something cool that happens
18:32 trone joined #mojo
18:41 al237 I saw that you were having issues with bountysource. Is it still a problem? How is Gratipay?
18:43 sri nope, bountysource seems to be operating normally again
18:44 sri gratipay seems more reliable though
18:44 al237 ok, thanks
18:45 sri personally, i like services dealing with money to be open about problems
18:46 sri when gratipay had its meltdown it was very easy to get information, with bountysource it was total silence until they resumed work
18:47 al237 I know you shouldn't judge a book by its cover, but Gratipay's web design is horrible, and for that reason I prefer bountysource.
18:48 sri both services have some growing up to do, that's for sure
18:48 al237 I guess you are not a fan of Bitcoin?
18:48 al237 no mention of it anywhere
18:49 sri not really, hard for germans to convert bitcoin into real money
18:49 sri think i do have an address listed on gratipay actually
18:49 sri haha, they don't show it anymore
18:50 al237 that is, if you need to convert it. but it still has some value. many services accept it
18:50 sri hmm, i've not thought about paying services with bitcoin
18:51 sri oh my, gandi.net actually accepts bitcoin
18:51 jberger expedia does as do a few online stores
18:51 al237 i use gandi. i also recommend airvpn, who accept it.
18:51 genio bitcoin is still a thing?
18:51 sri gandi.net is interesting because the mojolicious domains are there
18:52 al237 i don't know if bitcoin is a wise investment. but one thing is for sure, it's not going away.
18:53 Atog what's the proper way to export a function (url_encode for example) to be used in a data template?
18:53 Atog i'm reading about helpers, and i can wrap it there
18:54 punter joined #mojo
18:54 jberger Atog: personally I wouldn't export it, I tend to use function in a template via Fully::Qualified::symbol()
18:55 Atog many thanks, will read about it
18:57 jberger (Note that that was pseudocode)
18:58 Atog oh, i get it
18:58 good_news_everyon joined #mojo
18:58 good_news_everyon [mojo] kraih pushed 1 new commit to master: https://git.io/vaxbS
18:58 good_news_everyon mojo/master fc2b337 Sebastian Riedel: mention the Bitcoin address too
18:58 good_news_everyon left #mojo
18:58 sri i'll just renew the domains with bitcoins once there's some
19:08 Phil21 any modern best practices for storing encrypted data in a database these days?
19:09 Phil21 (e.g. storing passwords you need to retrieve the plaintext of)
19:09 Phil21 like a phone password for a user
19:09 sri good question, i'd also like to know that :)
19:09 pink_mist sure, the best practice is: don't :P
19:09 Phil21 pink_mist: have to
19:10 Phil21 storing secrets is a thing :)
19:10 Phil21 obviously there are attack profiles here that suck, but it's unavoidable for some use cases
19:10 sri on the other hand, a hash works for phone passwords too
19:10 Phil21 eh, bad example
19:10 Phil21 the actual use case is storing passwords to use on industrial equipment
19:11 Phil21 it's a step up from the typical environment using the same password on * for 2 decades :)
19:11 Phil21 so "give me console I can tap into console on device #45345"
19:11 Phil21 er, password
19:12 Phil21 although I would argue using hashes for phone passwords is more difficult than it seems UIx wise ;)
19:12 sri oh, old stuff where you can't use key pairs?
19:12 Phil21 (having done it before)
19:12 Phil21 but that's a different argument
19:12 Phil21 yeah
19:12 Phil21 plenty of old shit like that still out there
19:13 sri guess everyone is still using bcrypt for that
19:13 Phil21 bcrypt is fine for storing pw hashes, yeah
19:13 Phil21 but that's one-way
19:14 Phil21 or another more succinct use case people here may relate to: storing snmp communities
19:14 Phil21 need to get the plaintext to build your polling, but sucks to have them all the same/sitting around in plaintext
19:15 Phil21 anyways, I'll continue to google
19:16 sri no wait, i didn't mean bcrypt... meant aes :)
19:18 sri hmm, pgp functions in postgresql look interesting
19:20 sri http://www.postgresql.org/docs/9.5/static/pgcrypto.html#AEN171380
19:22 sri guess pgp is encouraged with postgresql these days
19:33 abra joined #mojo
19:54 HtbaaPi joined #mojo
19:54 cpan_mojo Clustericious-1.18 by PLICEASE https://metacpan.org/release/PLICEASE/Clustericious-1.18
20:12 asarch joined #mojo
20:15 tyldis Phil21: libsodium is a good starting point
20:16 Rubes joined #mojo
20:58 PopeFelix joined #mojo
21:00 PopeFelix https://github.com/PopeFelix/Mojo-UserAgent-Mockable/blob/master/t/022_record_nonblocking.t#L30
21:00 PopeFelix I'm sure I must have had a good reason for putting that in there, but I can't remember why. Is there a reason I need that?
21:03 bpmedley Atog : https://github.com/brianmed/sparky <-- You might like this
21:04 jberger bpmedley: I don't see how that is related
21:05 jberger PopeFelix: I think that was supposed to just be ->stop
21:05 bpmedley jberger : Sparky can stream audio
21:05 jberger and you don't need to check if/unless is running in a script
21:05 jberger ah
21:05 jberger nice
21:06 PopeFelix jberger, but why am I stopping the IO loop at all? It seems like I start it once, but then try stopping it twice.
21:07 bpmedley Phil21 : How do the users authenticate?
21:09 dvinciguerra joined #mojo
21:13 disputin joined #mojo
21:17 sri thanks for the bitcoins, lets try paying gandi with them :)
21:18 sri ah, they use bitpay.com
21:23 PopeFelix https://github.com/PopeFelix/Mojo-UserAgent-Mockable/tree/fix-022_record_nonblocking - I'd appreciate it if folks could check this out and run the tests.
21:28 Rubes joined #mojo
21:35 Atog hey, many thanks bpmedley!
21:40 Atog heh, looks like what i'm trying to do with even a bunch of pluses
21:40 Atog got it already working for the supported hml5 codecs, now i'm trying to transcode and serve flac to ogg
21:41 Atog honeslty i never thought i would do that much on an evening
21:41 Atog i'm really loving that stuff
21:43 bpmedley Atog : Sorry for the confusion, are you referring to sparky?
21:44 mrEriksson joined #mojo
21:45 bpmedley PopeFelix : Files=27, Tests=204, 101 wallclock secs ( 0.16 usr  0.06 sys + 16.89 cusr  1.75 csys = 18.86 CPU)
21:45 bpmedley PASS
21:46 PopeFelix bpmedley, Thanks
21:46 PopeFelix bpmedley, what platform?
21:48 bpmedley OS X 11
21:48 PopeFelix bpmedley, thanks.
21:49 Rubes joined #mojo
21:50 Rubes joined #mojo
21:51 sri oh noes, this tells me i should be programming ruby :O http://www.wfplsiu.com
21:52 odc i'm not surprised
21:52 PopeFelix The local uni is teaching PHP. :\
21:56 Atog bpmedley, was talking about mycode. Now i'm gonna try sparky.
21:58 s1037989 I'm supposed to use Ruby, too!!  :'(  Actually, no: Keep using that f-in language. Unless you can't accomplish your goals with your current language, you're setting back progress by starting with a new language.
21:59 s1037989 I've already established Perl and Mojolicious for my projects.  So I'm grandfathered in!  :D
22:01 bpmedley Atog : that's really cool, is your code available for perusal?
22:04 jberger we're so old they don't even mock us anymore :s
22:08 Atog http://hastebin.com/epeyolafob.coffee
22:08 Atog nothing fancy at all and just a mashup of thing i'm looking and trying around
22:09 Atog but hey it's nice to see things working as intended
22:13 bpmedley Atog : Code works.. :)
22:15 sri paying gandi with bitcoins worked, transaction took 30 mins, but otherwise very easy
22:21 Atog being only used to plain old cgi since now i'm kinda overwhelmed on what can i do now
22:22 Atog i'll defenitelly rewrite anything useful i have and will keep toying
22:22 bpmedley Atog : Have you tried morbo?
22:23 Atog yeah, i'm using morbo
22:23 Atog just begun with mojolicious this evening!
22:23 Atog i readed about hypnotoad, but i don't feel like i still need to dig more
22:25 Atog i cleared my doubts for sure about moving to python for django
22:25 Atog i'm staying with perl
22:25 bpmedley Atog++
22:33 s1037989 Atog++
22:34 jberger \o/
22:34 jberger Atog++
22:34 sri \o/
22:43 Atog i'm very fond of anyevent, it's nice to see how i can fit it here
22:43 Atog i'll check ioloop, for sure
22:48 s1037989 ioloop++
22:50 sri ioloop got a lot of improvements recently, profited a lot from the scaling to 20+ cores experiments
22:53 sri pretty pretty curve https://github.com/kraih/mojo/issues/925#issuecomment-191905975
22:57 s1037989 What's the y-axis?  Concurrent connections?
22:57 sri worker processes
22:57 sri x is rps, y is workers
22:57 s1037989 30,000 processes??  Like unique Linux PIDs, e.g.?
22:58 sri umm
22:58 sri the other way around ;p
22:58 s1037989 Yeah I didn't think I understood that right.
22:58 sri x=workers, y=rps
22:59 s1037989 Ok, that more understandable.  :D
22:59 Atog that's cool, i'm realizing how i won't even need a webserver at all for the things i do just a kind of gateway app
23:00 s1037989 Ok, next question, why does it flatten at 12 core?  I'm just curious, I don't know anything about multiple cores or concurrency or smart things in general.  :D
23:00 s1037989 Wait, *workers*, not cores.
23:00 sri because the test box has 12 cpu cores
23:00 Rubes joined #mojo
23:00 sri it shows how hyper threads are worth a lot less than real cores
23:00 s1037989 Got it!  :D  I thought I saw you say 20+.  But, yes, this image is 12 cores.  Got it!
23:01 sri yea, we couldn't find a 20+ test box -.-
23:01 s1037989 I see, I see!  That makes sense!  12 cores, rps increases drastically, over 12 it kinda tanks.  So it's a great demonstration of the value.
23:01 sri but the number suggest that 20+ is not a problem at all
23:01 sri *numbers
23:02 s1037989 How much time would you need on a box with 20 cores for the test?
23:02 sri test script and setup is all in the issue, anyone with access could do it
23:02 s1037989 And can you do it from a VM?
23:03 sri vm is not so good
23:03 s1037989 Ok, good to know.  I see why getting access would be so tough then.  :D
23:04 sri that shell script has everything needed https://github.com/kraih/mojo/issues/925#issuecomment-191667399
23:05 sri yea, as it turns out very few people have 20+ core boxes standing around idling :)
23:05 jberger ummmmmm
23:06 s1037989 :D  We should be getting a new 24 (32?) core in a couple of months.  Maybe I can borrow it for a couple minutes during unboxing!  :D
23:09 bpmedley https://github.com/brianmed/livetutorial <-- Would someone try installing this pls and letting me know if it's worth adding more content?
23:11 s1037989 Can't load application from file "vhost.pl": Can't load application from file "live_tutorial/script/live_tutorial": Can't locate load_class.pm in @INC
23:12 bpmedley Weird..hrmm
23:13 bpmedley s1037989 : Can you run vhost.pl directly?
23:13 s1037989 $ ./vhost.pl   ?
23:13 bpmedley I think so.
23:13 s1037989 Can't find application class "LiveTutorial" in @INC
23:14 bpmedley perl -Ilive_tutorial/lib:photo/lib vhost.pl
23:14 s1037989 Can't find application class "LiveTutorial" in @INC
23:14 s1037989 Shit.  Hold on.
23:18 s1037989 Success!
23:19 bpmedley s1037989++
23:19 bpmedley What was the issue?
23:19 s1037989 Ignore those first issues.  It was me not understanding my own environment, that everything expects Perlbrew.
23:20 s1037989 So the only real issue I had was I had to install Mojolicious::Plugin::AccessLog
23:20 bpmedley aha.. I wonder if that should be optional
23:20 bpmedley What do you think of the content and presentation?
23:21 s1037989 Can't locate object method "new" via package "LiveTutorial::Chapter::Ch02::Startup"
23:21 s1037989 I like the TOC!
23:22 bpmedley Do you have the -I when you start vhost.pl?
23:22 s1037989 What's different about it versus liveperl.us?
23:22 s1037989 I just ran $ bash runme.sh
23:22 bpmedley LiveTutorial is meant to run from your local box
23:22 s1037989 Ah!  Nice!
23:22 s1037989 Yeah, I think the whole liveperl thing is great!
23:23 bpmedley Thanks.  Imagine the scenario where you have a unix like os and want to learn web dev.. that's the goal with this.
23:24 s1037989 Yes, I actually bumped batman a while ago looking for liveperl.us.  I remember he tweeted it or something.  I wanted to connect this to others at work to introduce them to Mojo.
23:24 s1037989 All of Ch01 worked great, but Ch02 gave the error ^
23:25 bpmedley Hrmm.. do the blue links up top work well for you?
23:26 s1037989 I think it's absolutely tremendous!  People at work tell me that they're interested to try programming but don't know how to start.  Setting up your environment is important, but it shouldn't be a prereq to introduce people.  They just want to dive in and see if it resonates with them.
23:27 s1037989 Yeah, the blue link that pops up the "lecture" functions just fine!
23:28 bpmedley Thanks.  Will you try "$ perl -Ilive_tutorial/lib:photo/lib vhost.pl daemon" ?
23:28 s1037989 Can't find application class "LiveTutorial" in @INC
23:29 s1037989 bash runme.sh works fine tho
23:29 bpmedley perl -Ilive_tutorial/lib -MLiveTutorial -e 1
23:30 s1037989 This works: $ perl -Ilive_tutorial/lib -Iphoto/lib vhost.pl daemon
23:30 s1037989 Perl        (v5.22.1, linux)   Mojolicious (6.57, Clinking Beer Mugs)
23:31 bpmedley So, ch02 is accessible with a couple -I ?
23:31 s1037989 Sorry, no.  I should've realized why you were asking...  :D
23:32 s1037989 What I did just allows running the daemon
23:32 bpmedley When do you get the error?
23:33 s1037989 TOC -> Chapter 2 -> Photo App.  Then I get a Mojo exception page
23:33 s1037989 This route: http://localhost:3000/Ch02/Startup
23:33 bpmedley Gaw
23:35 Atog well, let's leave it for today
23:35 Atog thank you everyone, have an nice day/night!
23:35 jzawodn joined #mojo
23:35 Atog left #mojo
23:38 bpmedley s1037989 : Will you pull and try again, pls?
23:38 s1037989 Same. :(
23:39 s1037989 Can't locate object method "new" via package "LiveTutorial::Chapter::Ch02::Startup" at live_tutorial/lib/LiveTutorial.pm line 15
23:39 s1037989 Different, actually.
23:40 s1037989 Try moving line 15 to npm?
23:41 bpmedley :-0
23:42 s1037989 Ah ha!
23:43 s1037989 It's asking for Startup, but it's on disk as startup.pm
23:43 s1037989 bpmedley: ^
23:43 bpmedley Eek
23:45 bpmedley Cool, perhaps a pull and retry?
23:46 s1037989 Success!
23:46 bpmedley Thanks!!
23:46 s1037989 No -- not entirely.
23:47 s1037989 The page is fine, the left pane is fine, the right gets a Mojo exception.
23:47 bpmedley Cool stuff.. what can we fix next.. :)
23:47 s1037989 can't opendir /home/stefan/Git-Forks/livetutorial/photo/albums: No such file or directory at /home/stefan/Git-Forks/livetutorial/photo/script/../lib/SiteCode/Albums.pm line 14,
23:48 s1037989 Also, the browser tab title shows: "LivePerl - [% c.sample.title() %]"
23:50 s1037989 I created the directory photos/albums: Success!
23:50 s1037989 bpmedley:
23:50 bpmedley Thx, I need to have that automated somehow
23:50 s1037989 Can git not handle empty dirs?
23:51 bpmedley I'm probably doing something wrong
23:51 s1037989 I personally don't know.  I've never had it capture empty dirs.
23:51 s1037989 Curious if it's possible.  Surely...  :/
23:52 osfameron git only tracks files, not directories
23:53 osfameron so you can add a dummy file like `.keep` or whatever, to make sure it keeps a directory
23:53 s1037989 osfameron: Thanks!  Good to know!
23:54 bpmedley !.gitignore can be added to the .gitignore file, as well..
23:54 bpmedley I pushed a fix, I hope
23:55 s1037989 bpmedley: All good!  And the title is fixed as well.
23:56 bpmedley Thanks, the help is much needed
23:57 s1037989 bpmedley: I deleted the clone and cloned brand new.  Full success, including the existence of the albums dir.
23:57 bpmedley Nice!!
23:58 s1037989 Questions...  On Hello World, this seems weird: "Please complete the questions below."
23:58 s1037989 What questions?
23:59 bpmedley Lines 6-7
23:59 s1037989 Oh!  Haha!  Great!  Ok, and that answers my second question.  :D

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