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

IRC log for #mojo, 2015-09-05

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

All times shown according to UTC.

Time Nick Message
00:00 jontaylor joined #mojo
00:09 Kogurr joined #mojo
00:20 cpan_mojo XTaTIK-0.001002 by ZOFFIX https://metacpan.org/release/ZOFFIX/XTaTIK-0.001002
00:23 x86 joined #mojo
00:29 Janos joined #mojo
00:32 Janos hey there, developing a small web app with mojo, this app includes a perl script to parse/load a log file to the database and I was wondering if: 1- Implementing this script as a mojo command makes sense, this would allow me to reuse code and 2- Is there any tutorial on how to implement a mojo command ?
00:33 Grinnz Janos, mojo apps are already commands; you can add commands to them even
00:33 Grinnz oh you mean the separate script
00:33 Grinnz sure you could do that
00:33 Grinnz https://metacpan.org/pod/Mojolicious::Guides::Cookbook#Adding-commands-to-Mojolicious
00:35 Janos Grinnz, thanks a lot, I think that's exactly what I need
00:36 bpmedley Janos: I personally think that's a great approach.  With the mojo command everything would be together and organized.
00:37 mst Janos: your database access code shouldn't know your mojo app exists though - but once you've got your model factored out, there's no reason not to access it via a command
00:37 mst (or an extra script that loads your modules normally, depending)
01:20 mattastrophe joined #mojo
01:27 Janos mst, well I was kind of hoping to use the database plugin and the entire app conf in my mojo command, is that possible ?
01:32 mst Janos: Mojolicious::Plugin::Database is considered harmful, you're better off with a proper set of database access classes using DBIx::Connector and then passing config in to those during app setup
01:34 mst then you should be able to have your command do something like my $model = $app->model;
01:34 mst and just use $model
01:38 * jberger always uses DBIx::Connector too
01:40 Grinnz or Mojo::Pg/Mojo::mysql/Mojo::SQLite, which do the same sort of thing and were written by us
01:41 jberger Right, I mean if I'm going to use raw DBI anyway
01:42 jberger Mojo::Pg alikes have their own mechanism
01:45 Grinnz that ::Database plugin looks extremely useless
01:46 mst it's slightly more useful than Catalyst::Plugin::DateTime
01:46 mst but only slightly
01:46 Grinnz hah
01:51 jberger Janos: to answer your more direct question, yes once you have your db connection logic in your app, it makes perfect sense to reuse it in a script and the easiest way to do that in mojo is to make a command
01:51 mst so long as you don't mind going through the whole webapp setup process to get at the database
01:51 mst which I've always regarded as pretty horrible, but if it's fast enough you don't notice, shrug
01:52 mst it's a massive architecture smell but it probably works fine anyway
01:52 mst sometimes 'easiest' is better than 'good'
01:52 jberger As to the other thing, I don't know what to do as far as doling out "justice" for it, but remember we are optimizing for the people who otherwise are too intimidated to participate in open source since the culture can often be intimidating
01:55 jberger going forward at least, locker room culture has no place here, just as with language
01:55 Janos I agree that it's not optimal to init the web app just to access the database, so I will probably just handle my own connection, will check Mojo::Pg
01:56 mst jberger: hence my suggested clarification to the topic
01:56 jberger And if anyone tries to claim "ethics in video game journalism" I'll throw them out myself!
01:56 mst which is clearer, since anti-homophobia jokes really aren't locker room culture
01:57 Janos so last question is, can I use the config plugin from my custom command ? Or does that requires initializing the mojo application too ?
01:57 mst in fact the entire point is to -insult- locker room culture
01:57 mst hence my proposing 'no innuendo' since that can't be misunderstood
01:57 jberger mst: yeah I can see that, especially knowing that you are actually very consistent in that regard
01:59 mst Janos: so far as I can tell, you would -probably- get away with Mojolicious::Plugin::Config->load($file_name) provided you aren't relying on app-> in the config file
01:59 jberger The problem is that people who aren't familiar with your stance don't know that
02:00 jberger And they are who I care about most
02:00 mst which is fine, hence why I proposed a tweak that would ban my comment anyway
02:00 Janos mst, thanks, I will try that
02:01 mst I appreciate the goal of the rules, I'm just trying to help refactor them so they actually cover the things needed to achieve that goal
02:02 jberger The more I think about it, the g rating already covers it pretty well
02:02 jberger And in that context, I suppose a strike is warranted
02:04 mst given the number of innuendos that get snuck into g-rated things to stop the parents from getting bored, I think 'no innuendo' is clearer, but I've already apologised for it being interpreted a different way than intended so either way, I guess
02:05 sri mst: stop twisting words
02:07 noganex joined #mojo
02:14 jberger Janos: the idea that it's a lot of architecture to load depends quite strongly on how often this is going to run vs how much time should be spent developing it
02:14 mst also, honestly ... check how long it takes. mojo app startup is often fast enough that you won't notice the difference anyway
02:15 mst so long as you do your best not to code yourself into a corner so if it -does- matter later you can switch, it's probably far from the most pressing issue
02:29 mst was kicked by sri: 1 week temp ban (second strike)
02:31 ZoffixMobile joined #mojo
03:03 Repaster joined #mojo
03:04 Grinnz_ joined #mojo
03:06 Zoffix left #mojo
03:13 Grinnz joined #mojo
03:14 Grinnz joined #mojo
03:52 jberger nn all
03:53 sri nn
04:07 melo joined #mojo
04:16 sri if anyone here thinks the g-rating rules are wrong, please feel free to tell me in private
04:19 disputin joined #mojo
04:40 disputin joined #mojo
04:59 disputin joined #mojo
05:19 asm35 joined #mojo
06:30 kaare joined #mojo
07:08 cpan_mojo W3C-SOAP-WADL-0.007 by IVANWILLS https://metacpan.org/release/IVANWILLS/W3C-SOAP-WADL-0.007
07:26 Vandal joined #mojo
08:01 berov joined #mojo
08:40 iloveyou joined #mojo
09:00 amon joined #mojo
09:34 iloveyou joined #mojo
09:46 sh4 joined #mojo
09:51 KindOne joined #mojo
09:53 iloveyou joined #mojo
09:57 KindOne joined #mojo
10:04 KindOne joined #mojo
10:22 KindOne joined #mojo
10:35 iloveyou joined #mojo
11:02 dotandimet joined #mojo
11:30 stephen joined #mojo
11:45 absolut_todd joined #mojo
11:55 cpan_mojo JSON-Validator-0.52 by JHTHORSEN https://metacpan.org/release/JHTHORSEN/JSON-Validator-0.52
11:58 jontaylor joined #mojo
12:21 meshl joined #mojo
12:25 x86_ joined #mojo
12:52 akhasanov joined #mojo
13:02 akhasanov joined #mojo
13:03 mattastrophe joined #mojo
13:08 asm35 joined #mojo
13:29 iloveyou joined #mojo
13:40 mattastrophe joined #mojo
13:40 mattastrophe_ joined #mojo
14:00 PryMar56 joined #mojo
14:02 asarch joined #mojo
14:35 kaare joined #mojo
14:38 asm35 joined #mojo
14:44 kaare joined #mojo
15:34 asarch joined #mojo
15:38 ajr_ joined #mojo
15:41 iloveyou joined #mojo
15:49 x86 joined #mojo
16:07 jnbek joined #mojo
16:16 thowe joined #mojo
16:36 jontaylor joined #mojo
16:36 iloveyou joined #mojo
16:56 davido_ joined #mojo
17:18 iloveyou joined #mojo
17:36 sri marcus, batman, crab, tempire, jberger, Grinnz: proposal for a more refined Code of Conduct https://github.com/kraih/mojo/wiki/Code-of-Conduct
17:37 sri it is the same used by the django project, and i like it very much https://www.djangoproject.com/conduct/
17:38 sri especially parts like "This isn’t an exhaustive list of things that you can’t do. Rather, take it in the spirit in which it’s intended - a guide to make it easier to enrich all of us and the technical communities in which we participate."
17:39 sri things like ffs and wtf, or swearing in a self deprecating way would be ok-ish again
17:40 sri in other regards it is more strict
17:40 sri "Be careful in the words that you choose. We are a community of professionals, and we conduct ourselves professionally. Be kind to others. Do not insult or put down other participants."
17:42 sri anyone here is of course welcome to comment
17:47 preaction looks good to me. in another channel, we adopted this network's reasonable person principle, but the django one seems a bit more explicit
17:51 Grinnz i think that code of conduct looks very thorough and reasonable
17:53 preaction i don't think clarity of the rules is the problem really. no rules could possibly cover 100% of all cases. if they could, we could get a computer to enforce them.
17:53 mishanti1 sri: I think the CoC looks good. Explicit enough that people should be able to figure out if something they say might be breaching it.
17:53 jberger preaction: true, and yet the clairity of the G rating rules kept coming up
17:53 jberger so this is take 2
17:53 preaction often, it's someone's reaction when they are told they have broken a rule is the problem.
17:53 jberger a proposed take 2 I should say
17:54 preaction right, but as far as i can tell, that's a repeated refrain of a single person looking to find a way around having to say "sorry"
17:55 jontaylor i like this proposal better because while its trivially easy to avoid being unkind to people, it was only a matter of time before I tripped up on the no swearing thing, especially the abbreviated versions
17:55 jberger that may have been the context, but the questions have come from others
17:55 jberger preaction: ^^
17:56 preaction okay, then i agree that this version seems a lot clearer
17:56 mishanti1 Yeat, several of us did not understand what was acceptable and not before. Now it is pretty clear. :)
17:56 mishanti1 Yeat? What?
17:56 preaction the network-wide RPP has also worked for me
17:56 mishanti1 Typos, I has them.
17:58 iloveyou joined #mojo
17:58 Topic for #mojo is now ???? cheers | http://mojolicio.us | http://irclog.perlgeek.de/mojo/today | https://github.com/kraih/mojo/wiki/Code-of-Conduct
17:58 sri for now it is an experiment, if it works out the CoC will go into the guides
17:59 * sri likes especially that the intention of the CoC is highlighted so clearly
18:01 mishanti1 Indeed.
18:02 kaare joined #mojo
18:09 trone_ joined #mojo
18:10 iloveyou joined #mojo
18:17 meredith ++
18:18 meredith I also appreciate the express mention of the spirit of the code, as I generally regard "testing the boundary because it isn't clear" not in the spirit of codes even if they -don't- have a bit like that.
18:21 Adura joined #mojo
18:21 meredith And thank you for addressing the comment last night, I saw it at the time and was very disappointed. I detached my session for a while.
18:33 tempire I dig it.
18:34 tempire I would suggest we add one thing.
18:35 tempire "Don't be mean or passive aggressive"
18:46 Adura Australians are welcome back to the channel!
18:46 x86 joined #mojo
18:47 preaction er... they weren't before?
18:48 x86_ joined #mojo
18:48 Kogurr joined #mojo
19:01 iloveyou joined #mojo
19:02 bpmedley I just wanted to say thank you for fostering a community.  I'm glad to help, as much as possible.
19:03 sri tempire: that should already be covered
19:03 sri bpmedley++ # your suggestion made me go look for other CoCs
19:04 bpmedley Anytime!
19:10 sri was also interesting to read the open code of conduct github has adopted
19:11 sri it's based on this one, but they tried to be more specific, which failed miserably
19:11 sri http://todogroup.org/opencodeofconduct/
19:11 batman sri: I really like the CoC!
19:12 batman I don't have the best of internets here, so it took some time to go through it :)
19:14 batman I agree what tempura suggested should be covered, but if you (tempire) find a place where it should be (missing?), I'm not against it.
19:17 jberger I always agree with tempura
19:18 jberger btw, tempire, I showed my wife IceJJFish yesterday and she flipped out
19:18 tempire Because she's in love?
19:18 jberger then went immediately to message some friends
19:18 tempire It's inevitable
19:18 jberger exactly
19:21 jberger so here's something crazy
19:21 jberger I'm trying these scaleway servers
19:21 jberger and nginx failed to install
19:22 jberger because there is literally no ipv6 available, not even localhost
19:22 jberger which the debian packagers consider to be a highly non-standard configuration, except I didn't set it up that way!
19:24 jberger maybe I should just forego the reverse proxy on systems this small
19:24 bpmedley nginx failed due to no ipv6?  Is that correct?
19:26 bpmedley Are specifying ipv6 in nginx.conf?
19:30 jberger Installing via apt failed!
19:30 bpmedley I see.  You're trying to install nginx via apt and it fails due to network config of the box?
19:31 jberger Exactly
19:32 batman jberger: toadfarm ftw :)
19:35 jberger batman: this may be just the reason I need to try it out
19:36 Grinnz why would installation of nginx depend on network config o_0
19:36 Grinnz how odd
19:40 jberger Indeed, the ticket on the matter basically makes a case that the case is rare enough to not care about
19:40 jberger But sadly that's where I am, unless I can figure out why it isn't there in the first place
19:41 batman :)
19:42 batman It's so simple with the new DSL
19:43 Craftsmanship joined #mojo
19:44 Craftsmanship If I want to build my own logger, is it best to just leave that as a plugin? does anything special happen outside of the request response cycle that I might want the logs for?
19:58 asm35 joined #mojo
20:01 ladnaV joined #mojo
20:02 bpmedley jberger: $ cat /proc/net/if_inet6, if you would run that
20:03 batman Craftsmanship: What kind of logging do you want?
20:03 batman Have you searched CPAN for the existing log modules? Like AccessLog and Logf?
20:04 jberger bpmedley: no such file or directory
20:04 batman I might have missed out on the details in your question...
20:04 batman jberger: That's the best file.
20:04 Craftsmanship Yes, I stole the guts out of AccessLog already
20:04 * batman is rambling :)
20:05 bpmedley jberger: Dang.  From reading, that shows if your linux currently supports ipv6.
20:05 batman Craftsmanship: ok. Did you look at toadfarm::p
20:05 batman Oops!
20:05 Craftsmanship which is why my logging is, so far, trapped witin the request cycle
20:05 jberger this is debian jessie
20:05 jberger though compiled by the host for this architecture
20:05 batman Toadfarm::Plugin::AccessLog?
20:05 jberger I wonder if they have an image with nginx assembeld
20:06 Craftsmanship i'm subscribing to $log's message ... but i'm not sure how often i have to do it, or if i'm going to miss anything good if I only have logging in the request cycle
20:06 batman (I think that is the name)
20:06 bpmedley jberger: $ lsmod | grep ipv6
20:06 batman Craftsmanship: If you subscribe to the log event on startup(), then you won't miss out on anything
20:07 jberger bpmedley: you're going to like this one
20:07 jberger -bash: lsmod: command not found
20:07 bpmedley $ whereis lsmod.. you  may need root for that
20:07 batman jberger: Let me know when you <3 Toadfarm ;)
20:07 Craftsmanship batman: I'm stuffing things into $tx->{some-key} and then json'ing it to a log file, so i'm not sure what to do at startup :S
20:07 jberger bpmedley: I am root
20:07 jberger (I am groot!)
20:08 batman Craftsmanship: Can you show some code?
20:08 bpmedley How did you login to root?
20:08 batman ...and explain what you're trying to achieve
20:08 jberger Craftsmanship: storing in the $tx seems hairy, could you store it in the session
20:08 jberger not session
20:08 jberger stash
20:08 batman jberger++
20:08 Craftsmanship stash?
20:08 Craftsmanship for logging stuff?
20:08 batman Yes
20:08 Craftsmanship i'm trapping warn's
20:09 batman Oh my...
20:09 Craftsmanship so i can log them.
20:09 Craftsmanship it's also half of Carp::Always
20:10 batman I still don't understand what you're trying to achieve
20:10 batman You lost me before the initial question.
20:10 batman Sorry :/
20:10 bpmedley jberger: Perhaps install / reinstall module-init-tools?
20:10 Craftsmanship put interesting things in a structure, write that strucure to a line in a log file
20:11 preaction Craftsmanship: use Mojo::Log::Log4perl
20:11 jberger bpmedley++ ok that helped
20:11 jberger no mention of ipv6 though
20:11 jberger is that a module that can be built?
20:11 bpmedley # modprobe ipv6
20:11 bpmedley It may already be installed.
20:11 Craftsmanship preaction: I don't know how to tell log4perl to log serialised structs, is that a thing it can do?
20:11 bpmedley Just not loaded.
20:12 preaction Craftsmanship: you can JSON encode them or you can write an appender that does what you want
20:12 bpmedley Did modprobe work?
20:12 Craftsmanship (I don't even really know how I would tell log4perl to "here, hold this hash")
20:12 Craftsmanship also, afaik log4perl doesn't have an idea of a "request"
20:12 preaction it has logging contexts
20:12 bpmedley Craftsmanship: Are you building a context aware Mojo logger?
20:13 Craftsmanship context?
20:13 preaction log4perl can do contexts. context like "the current request"
20:13 Craftsmanship i'm building something that logs one hash-ref per request
20:13 jberger bpmedley: it didn't print out anything
20:13 bpmedley Hrmm.  I see, forgot I mentioned context.
20:13 jberger so I assume that's no
20:13 bpmedley Please try the lsmod again.
20:14 iloveyou joined #mojo
20:14 Craftsmanship bpmedley: i'm not 100% sure what "context aware" means in this ... case
20:14 bpmedley Craftsmanship: I was confused.  My bad.
20:15 Craftsmanship understood
20:15 preaction Craftsmanship: so what you're saying is that a request could end up trying to log multiple hashrefs, but you only want one to actually be logged?
20:16 jberger oh!
20:16 bpmedley jberger: Would you try the lsmod again, please?
20:16 Craftsmanship I want all the things to go into the hash, and then for it to be saved at the end of the request
20:17 jberger bpmedley: now it shows up
20:17 Craftsmanship I think this might be un-good if I start doing some websockets type stuff
20:17 preaction Craftsmanship: that doesn't sound like a log. that sounds like a helper
20:17 bpmedley # cat /proc/net/if_inet6
20:17 Craftsmanship preaction: it does need a helper
20:17 jberger bingo
20:17 Craftsmanship to allow the application to stick things into the hash
20:17 bpmedley You may need to reboot.
20:17 preaction Craftsmanship: that sounds like the stash, and some event handling to flush it when the request is done
20:17 preaction so it's not logging at all
20:17 Craftsmanship wat.
20:17 Craftsmanship how is that not logging?
20:19 preaction mostly because it's stateful, requiring multiple pieces of information to be collected before an entry can be saved
20:19 Craftsmanship and context aware logging isn't that?
20:20 preaction no. saving bits of data to be included with future log lines is not that. each line is still its own thing
20:20 bpmedley jberger: Did you consider reboot the VM?
20:20 jberger I'd like to try without, so that I can do this all as part of an ansible playbook
20:21 bpmedley jberger: The reason I say you may need to reboot is that if the module-init package wasn't installed, then you may have a partially configured system.
20:22 bpmedley However, if you want to continue.  Do you have localhost setup for ipv6?
20:22 jberger it is in the hosts file
20:23 jberger and now when I ifconfig it has ipv6 information on lo
20:23 bpmedley Ok, does nginx install?
20:23 jberger trying now
20:23 jberger like a charm
20:23 jberger bpmedley++
20:23 bpmedley $ grep ipv6 /etc/modules
20:24 bpmedley Cool.
20:24 Craftsmanship preaction: yep. one log entry per request.
20:24 jberger bpmedley: nothing output
20:25 bpmedley jberger: http://www.hpc.mil/index.php/2013-08-29-16-03-23/networking-overview/2013-10-03-17-24-38/ipv6-knowledge-base-ip-transport/enabling-ipv6-in-debian-and-ubuntu-linux
20:25 preaction Craftsmanship: the point i'm trying to make is that a logger object is generally not what is going to solve your problem, as loggers generally do not ever hold on to information waiting to write it later
20:25 bpmedley Step 2b)
20:25 jberger ah
20:27 Craftsmanship preaction: sure. I'm just trying to figure out of things will be happening at startup, that are logged, that I might not want to miss (because this thing is per-request, and there isn't a request happening at startup)
20:27 preaction that's what Mojo::Log is for
20:27 preaction so if you replace your logger object, you won't get what you want. which is why you should not do what you want with a logger object
20:28 batman Craftsmanship: You can app->log at startup
20:28 Craftsmanship Right, my plugin isn't a Mojo::Log, it just calls ->on(messaage ... )
20:28 Craftsmanship so it can include logged things in its "stuff that happened this request" struct
20:31 batman Craftsmanship: do you want to prefix the messages..?
20:32 Craftsmanship !paste
20:33 bpmedley Craftsmanship: I see.  You want your log dynamically adjust to the request; however,  you only want the output to be in one place and possibly one line.  Is that correct?
20:34 Craftsmanship where should I be pasting perl?
20:34 melo joined #mojo
20:35 bpmedley I use pastie.org
20:36 stephen joined #mojo
20:37 Craftsmanship damn, i swear a lot in comments, one sec
20:42 Craftsmanship http://pastie.org/private/qra8gknnjifok3kap5zkw
20:43 Craftsmanship It's Mojolicious::Plugin::AccessLog and Carp::Always wedged together
20:45 preaction why? what are you trying to achieve with this?
20:45 Craftsmanship one log entry per request
20:45 Craftsmanship with /stuff/ in it.
20:46 Craftsmanship an access log that I don't need to parse back out with a regex
20:46 preaction what? access log formats are well-known, and just about every tool in the world knows how to parse them
20:46 Craftsmanship mine doesn't
20:47 preaction so use mine that does then, ygrok, or use logstash, or use flume
20:47 meshl joined #mojo
20:47 Craftsmanship adding random stuff to those (like, just perl warnings) seems like it would be quite a chore
20:48 preaction are you trying to correlate the request with the other logging it generates?
20:48 Craftsmanship that's a big part of it, sure.
20:49 preaction because when the log level is debug, that already happens, because logs are ordered
20:49 Craftsmanship I'd also like to be able to include timing stuff from backend queries and things.
20:50 preaction you can do that with just regular logging. log everything, and then build reports based on the logs
20:51 Craftsmanship so, i can get a request ID or something to stick in the logs then/
20:51 Craftsmanship and then i have to tie it all back together based on that, right?
20:52 preaction i'm not at work, so i can't go look at my logs, but that is what i do, yes
20:52 Craftsmanship and then i just have to grab all the ... start/finish time stamps for each event, and what they relate back to
20:52 preaction i also have a lot of logging in other places, access logs for the database, access logs for other apis, all being stored in a single database for reporting
20:53 Craftsmanship so - why is this easier than just having one struct with the info in it?
20:53 preaction because i didn't have to write any code like what you wrote
20:53 bpmedley Craftsmanship: Is there a bug?  What are we looking for?
20:53 Craftsmanship aparently I posted it so i could be told it's a bad idea.
20:54 bpmedley Craftsmanship: Hrmm.  I'm not sure it's a bad idea, or not.  From what I understand it's a unique solution to your situation.
20:54 preaction it's a lot of extra work for very little extra benefit, yes
20:54 Craftsmanship preaction: I have no logging infra, and no army of parsers and reports
20:54 Craftsmanship so i have all that work ahead of me still
20:54 bpmedley preaction: What's wrong with one line vs multiple lines of logs?  Is that the situation?
20:55 preaction i installed logstash. that's my logging infra and army of parsers. then i installed kibana for reports
20:56 bpmedley Is logstatsh free?
20:56 Craftsmanship bpmedley: having it all in one structure means not having to tie it all back togheter
20:56 preaction bpmedley: the issue is storing the state of the request for later writing to a log. the issue is hijacking die and warn to collect what they do, but only during requests
20:56 Craftsmanship that's all.
20:57 bpmedley Craftsmanship: That makes sense.
20:57 Craftsmanship preaction: seems like warnings and fatals in production would be an interesting thing to log, no?
20:57 preaction having it all in one structure, preformatted, also means that you only get what you already wrote
20:57 Craftsmanship yep, I get what I log.
20:57 Craftsmanship but that's fairly common.
20:58 Craftsmanship I believe i have all the things that AccessLog has
20:58 preaction it is, but not this way. also, you're preventing it from actually dying
20:58 bpmedley Craftsmanship: The only thing I see that might be wrong is that if your request dies you don't get the log written to disk, I assume.
20:59 Craftsmanship bpmedley: you don't mean perl's die there, right?
20:59 bpmedley Correct.  Sorry, if the process is terminated for any reason.
21:00 Craftsmanship I think if someone murders the application most loggers will suffer that one,
21:01 Craftsmanship I think things not fataling properly is a somewhat of an issue
21:01 bpmedley Gaw.. :).  Not necessarily, what if at T0, T1, T2, and T3 you write into your log.  However, at T4 there is an error and the process dies.  Your on finish won't execute, will it?
21:02 bpmedley A logger that writes to disk immediately will log at T0, T1, T2, and T3.  If there is an error at T4, then you have the log from T0 - T3.
21:02 bpmedley Is that correct?
21:07 Craftsmanship I don't have a "write the log" thing yet... so i don't have that bug :)
21:10 bpmedley Totally.. :)  So, the question is: would anything that happens at startup be missed?
21:12 Craftsmanship I can imaigne that I might have stuff happening at startup that I would want to log at some point
21:12 Craftsmanship but idk
21:13 bpmedley Sure, that makes sense.  I'm just trying to understand the question.  Not sure I can help at that level, though.
21:14 Craftsmanship I guess I could register at startup, and log that with some kind of "this_isnt_from_a_request => 1" type thing
21:15 bpmedley My apologies, I don't know the initialization part of Mojo as much as I would like.  I didn't mean to waste any of your time, I was just clueless.
21:16 Craftsmanship I don't know about it either. Seems like you have a solid understanding of the question though... so there's that.
21:17 Craftsmanship I'll leave the existing logging alone, and run mine along side, so i can use the stuff reasonable people use too.
21:26 giancarlos joined #mojo
21:35 KindOne joined #mojo
22:11 iloveyou joined #mojo
22:22 bpmedley https://bitbucket.org/snippets/bpmedley/qRBBp <-- Is this an accurate way to approach event based logging at the controller context level?
22:27 Craftsmanship do I get "Nothing has been rendered, expecting delayed response" even when I explicitly say ->render_later?
22:30 Grinnz you actually have to use render_later to get tht
22:30 Grinnz render_later just disables automatic rendering
22:31 Grinnz allowing for a render in a delayed callback
22:31 Craftsmanship odd.
22:31 Craftsmanship the message sounds like it's making an assumption, or guessing something.
22:33 Grinnz https://metacpan.org/pod/Mojolicious::Plugin::DefaultHelpers#delay
22:33 Craftsmanship I know
22:34 Craftsmanship sounds like it's saying i've forgotten something, i just wanted to check if I had.
22:34 Grinnz it would be weird to call render_later and then not render anything later, that would mean the client would get no response
22:34 bpmedley Craftsmanship: I believe the assumption is that the non-blocking write routines will be used at a later time.
22:35 Grinnz no, just a normal message indicating that the dispatch cycle has ended but it's waiting for a delayed response instead of automatic rendering
22:35 Craftsmanship I see.
22:49 iloveyou joined #mojo
22:52 meshl joined #mojo
23:23 PopeFelix joined #mojo
23:24 akhasanov joined #mojo
23:34 giancarlos joined #mojo
23:35 sri http://mojolicio.us/perldoc/Mojolicious/Guides/FAQ#What-does-Nothing-has-been-rendered-expecting-delayed-response-mean
23:46 asarch joined #mojo
23:59 asarch joined #mojo
23:59 good_news_everyon joined #mojo
23:59 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/vZeon
23:59 good_news_everyon mojo/master f2507d9 Sebastian Riedel: add code of conduct to Mojolicious::Guides::Contributing
23:59 good_news_everyon left #mojo

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