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

IRC log for #mojo, 2016-04-12

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

All times shown according to UTC.

Time Nick Message
00:57 asarch joined #mojo
01:19 tchaves joined #mojo
01:21 tchaves joined #mojo
01:51 woz joined #mojo
01:54 jberger mgrimes: ping
01:55 jberger I'm looking at Log::Any::Adapator::Mojo and I noticed you filed https://rt.cpan.org/Public/Bug/Display.html?id=111631
02:00 ivi joined #mojo
02:07 inokenty-w joined #mojo
02:13 noganex joined #mojo
02:37 tchaves joined #mojo
02:42 disputin joined #mojo
02:44 noganex_ joined #mojo
02:53 woz joined #mojo
03:35 Crisewng joined #mojo
03:39 melo1 joined #mojo
04:55 woz joined #mojo
05:58 Guest-guest joined #mojo
06:01 woz joined #mojo
06:11 cpan_mojo Mojolicious-Plugin-EDumper-0.00003 by MCHE https://metacpan.org/release/MCHE/Mojolicious-Plugin-EDumper-0.00003
06:23 Vandal joined #mojo
06:31 salva joined #mojo
06:35 cpan_mojo Swagger2-0.77 by JHTHORSEN https://metacpan.org/release/JHTHORSEN/Swagger2-0.77
06:50 dod joined #mojo
06:56 dod joined #mojo
06:57 batman BinGOs: yay! M::P::CGI pass :)
07:03 woz joined #mojo
07:03 AndrewIsh joined #mojo
07:23 trone joined #mojo
07:28 mortisha joined #mojo
07:30 woz joined #mojo
07:43 osfabibisi joined #mojo
08:12 osfabibisi joined #mojo
08:34 jasanj is it good to write a class base on Mojo::Collection ? I mean use Mojo::Base 'Mojo::Collection'
08:35 inokenty-w joined #mojo
08:37 jasanj i have a class contained a list of class, i thought it seems like a collection with some special method
08:51 woz joined #mojo
08:53 cpan_mojo Mojolicious-Plugin-RoutesAuthDBI-0.300 by MCHE https://metacpan.org/release/MCHE/Mojolicious-Plugin-RoutesAuthDBI-0.300
08:53 Bender *yarg*
08:55 woz joined #mojo
08:57 Crisewng left #mojo
09:04 melo joined #mojo
09:07 inokenty-w joined #mojo
09:14 cpan_mojo Mojolicious-Plugin-RoutesAuthDBI-0.301 by MCHE https://metacpan.org/release/MCHE/Mojolicious-Plugin-RoutesAuthDBI-0.301
09:14 Bender -.. .- .... ..- -
09:20 stephan48 is mche on IRC?
09:20 Guest-guest here
09:22 Guest-guest its i am
09:23 stephan48 interesting plugin!
09:23 Guest-guest :)
09:25 stephan48 it does automatically detect the DB design and exposes routes for it?
09:26 stephan48 I am currently writing a frontend Plugin for M::P::A myself(just needs tests, a name and a release), which provides a simple login+auth dropin
09:30 Guest-guest Postgres based Access.pm and Admin.pm as example
09:31 Guest-guest Sqlite soon
09:38 stephan48 ah, now I get it
09:39 stephan48 will look at it a bit more when I am at home
09:39 stephan48 looks definitively interesting!
09:40 Guest-guest ok
09:47 Guest-guest at that db design the most difficult is refs between tables controllers, actions, routes, roles, users
09:48 arthas joined #mojo
09:50 stephan48 yea
09:55 woz joined #mojo
10:29 woz joined #mojo
10:31 kaare joined #mojo
10:53 dvinciguerra_ joined #mojo
10:58 woz joined #mojo
11:11 tchaves joined #mojo
11:23 woz joined #mojo
11:24 tchaves joined #mojo
11:31 disputin joined #mojo
11:40 woz joined #mojo
11:43 tchaves joined #mojo
11:45 punter joined #mojo
11:54 woz joined #mojo
12:03 woz joined #mojo
12:10 woz joined #mojo
12:26 cpan_mojo Mojolicious-Plugin-RoutesAuthDBI-0.302 by MCHE https://metacpan.org/release/MCHE/Mojolicious-Plugin-RoutesAuthDBI-0.302
12:26 Bender DAAAAAHUUUT!
12:27 mgrimes jberger: yes that's me. if I remember correctly it is just changes to the tests.
12:32 woz joined #mojo
12:33 woz joined #mojo
12:55 ramortegui joined #mojo
13:00 woz joined #mojo
13:06 marty joined #mojo
13:10 memowe joined #mojo
13:24 disputin joined #mojo
13:36 asarch joined #mojo
13:39 PryMar56 joined #mojo
13:45 woz joined #mojo
13:48 jb360 joined #mojo
13:54 woz joined #mojo
13:54 disputin joined #mojo
13:56 jberger mgrimes: it is more than that
13:56 jberger it seems like the tests at first
13:57 jberger but the crazy hacks used to inject package and line numbers is wayyyyy broken
13:58 jberger so anyway, I'm thinking that rather than submit my patch, I'm going to release as a new module
13:59 jberger I almost don't think it would be responsible for upstream to merge my patch at this point, I change the format too much
14:00 mgrimes I've never noticed the line number issue.
14:01 jberger read the code, its terrifying
14:01 jberger it is also dependent on the mechanism that the default message subscriber uses to emit the message
14:02 jberger and the stack levels involved
14:04 disputin joined #mojo
14:04 jberger but mostly I'm worried about changing the log format, because if people are using monitoring tools that depend on that format, I really wouldn't want to break them
14:04 jberger but that is also why my version isn't going to change the message at all
14:04 jberger it is just a bridge
14:04 jberger mgrimes: https://github.com/henryykt/Log-Any-Adapter-Mojo/compare/master...jberger:fixes
14:07 mgrimes Yikes. That is ugly.
14:11 mgrimes I typically don't include the line/filename, so it wouldn't impact my usage.
14:12 mgrimes I'd have to do some digging to understand how the two compare.
14:14 mgrimes The existing maintainer doesn't seem particuarly responsive.
14:18 mcsnolte joined #mojo
14:20 mgrimes What about adding a flag to toggle munging of package/line or only munging in earlier Mojo versions. If someone is updating to a current Mojo seems they would expect changes in log output.
14:23 disputin joined #mojo
14:26 jberger mgrimes: given the current state of monitoring via logging (think logstash) I don't want to be responsible for that
14:26 jberger I will certainly work with the author to backport, but I don't see how my changes could be responsibly merged
14:27 jberger even if the author gave me comaint I would still fork
14:27 jberger I'd fix his as much as possible
14:30 preaction also, there's still Log4perl
14:31 jberger preaction: I've already got the code
14:32 jberger .... now I'm doubting myself
14:32 preaction if you want to toss me on co-maint right away to try to avoid the disappearing maintainer problem, i'd be amenable to that
14:33 mgrimes Well, I'd probably switch to your module Joel, given the lack of response. Although, I'll try sending a bug report/PR through GH first.
14:35 preaction now i'm thinking it'd be neat to have a mojo-modules group (github org or something) that could adopt useful mojo ecosystem dists
14:35 jberger preaction: sorry, I don't have perms
14:35 jberger I'm saying I already have the patch code
14:35 preaction you said you were forking though
14:36 jberger meaning there's now reason to move to log4perl
14:36 preaction ahhh
14:36 jberger s/now/no/
14:36 preaction right, but if you're releasing as a new module, you've got perms on that new module
14:37 jberger true, I'm happy to give you perms on that
14:37 preaction and if you're hesitating because you don't want to just perpetuate the problem, i was offering to help spread the load on that
14:37 jberger I'm just balancing the choice of "is it responsible to change the log format" vs "not forking if you don't have to"
14:38 jberger I guess I should contact the author, the choice is moot if he doesn't want the patch
14:38 jberger the question gets harder if he doesn't respond
14:39 mgrimes maybe he would be open to comaint for you or a mojo-module group
14:40 abra joined #mojo
15:03 zivester joined #mojo
15:12 woz joined #mojo
15:31 jberger hmmmm his cpan email bounced and his last twitter post was sept
15:57 asarch joined #mojo
16:00 woz joined #mojo
16:16 cpan_mojo Log-Any-Adapter-MojoLog-0.01 by JBERGER https://metacpan.org/release/JBERGER/Log-Any-Adapter-MojoLog-0.01
16:17 jberger I'll still work with him to port upstream if he wants
16:18 jberger but I'm not hopeful to find him
16:18 jberger and takeover process will/would take a while
16:24 jberger opened a PR with the relevant commits https://github.com/henryykt/Log-Any-Adapter-Mojo/pull/1
16:42 Grinnz_ jberger: the dist appears to have two licenses defined in meta
16:42 jberger Grinnz_: yes I'm just mentioning that to leont
16:42 jberger in #toolchain
16:43 jberger mbtiny did that unexpectedly
16:56 cpan_mojo Log-Any-Adapter-MojoLog-0.02 by JBERGER https://metacpan.org/release/JBERGER/Log-Any-Adapter-MojoLog-0.02
16:56 mgrimes jberger: the more i think about it, the best outcome would be to modify the original, if the author is reponsive or give comaint
16:56 jberger mgrimes: I'm ok with that
16:56 jberger at least to fix its problems
16:56 jberger then we would have to debate changing the format
16:58 mgrimes i think the mojo community is used to breaking changes
16:58 trone joined #mojo
16:59 jberger mgrimes: true, but given log monitoring this might actually be the way that people detect breakages
16:59 mgrimes more broadly, it would be great to see the Perl community embrace semantic versioning. node/npm does this really well.
17:00 jberger mgrimes: yeah
17:00 jberger I could discuss that at length, but its lunch time :-P
17:01 genio and the 5 is a fixed number. :)  *runs and hides*
17:02 mgrimes :)
17:26 dod joined #mojo
17:31 Dada joined #mojo
17:42 Dada Hi. I was looking for a way to get every parameters from a post request, and I found this : https://github.com/kraih/mojo/wiki/Request-data, so I tried the first example (my @params = $self->param; etc), but it doesn't work and it yells some errors (for instance: "Use of uninitialized value $name in string eq at /usr/local/share/perl/5.18.2/Mojo/Parameters.pm line 46"). I found another way to get every param (@{$self->req->params->names}).
17:42 Dada But still I'm wondering why the first approach didn't work.. Any ideas?
17:44 bpmedley Dada: you might try params->names
17:45 Grinnz_ that wiki page is outdated. https://github.com/kraih/mojo/wiki/Upgrading#parameter-name-listing
17:45 woz joined #mojo
17:47 Dada oh right, I didn't see that. Bad google suggesting outdated wiki in 1st position!
17:47 Dada thanks guys :)
17:48 Grinnz_ the nice thing about wiki pages is they are easy to update
17:48 trone_ joined #mojo
17:58 woz joined #mojo
18:06 Grinnz_ Dada: updated the other wiki page
18:08 meshl joined #mojo
18:11 Dada great, thanks
18:16 asarch joined #mojo
18:59 woz joined #mojo
19:17 go|dfish joined #mojo
19:21 vicash batman: hi. in Mojo::IOLoop::ReadWriteFork, if I do not provide a $fork->on(read => sub{}) does the STDERR/STDOUT automatically go to /dev/null ?
19:21 batman vicash: no. STDOUT/ERR is still emitted
19:22 vicash batman: can I provide "> /dev/null" in the arguments list then ?
19:23 batman i think so.
19:23 batman you can also do program => sub { close STDOUT; close STDERR; exec @program; }
19:23 vicash interesting. i will try that!
19:24 batman at least in theory :)
19:31 batman vicash: here are the details: https://metacpan.org/source/JHTHORSEN/Mojo-IOLoop-ReadWriteFork-0.17/lib/Mojo/IOLoop/ReadWriteFork.pm#L278
19:32 batman this means that if you don't specify any "program_args" it should be just exec $program, meaning you can do all kinds of crazy redirects like "app_a | app_b 2>/dev/null"
19:32 batman just be careful about what kind of input you take then, so you don't end up with "app_a; rm -rf / | app_b 2>/dev/null"
19:34 vicash absolutely.
19:54 jberger joined #mojo
19:56 vicash batman: didn't have to do anything except not provide a $fork->on(read => sub{}). that solved the problem for me in the simplest way.
19:57 batman vicash: cool
19:57 vicash thanks!
19:58 batman you're very welcome :)
20:05 woz joined #mojo
20:11 punter joined #mojo
20:24 marty joined #mojo
20:32 go|dfish joined #mojo
20:34 woz joined #mojo
20:45 mtths joined #mojo
21:03 go|dfish joined #mojo
21:26 woz joined #mojo
21:37 lluad joined #mojo
21:48 woz joined #mojo
22:00 woz joined #mojo
22:16 lluad joined #mojo
22:32 meshl joined #mojo

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