Camelia, the Perl 6 bug

IRC log for #mojo, 2013-05-07

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

All times shown according to UTC.

Time Nick Message
00:31 jberger the more I work on this ODM for my mango project, the more I understand DBIC
00:46 shmuel joined #mojo
00:54 sri batman: btw. you might want to check Mojo::Redis with mojolicious master... a lot of ioloop internals changed
01:09 ka2u joined #mojo
01:29 egopro joined #mojo
01:49 cooper joined #mojo
01:56 Meiermann joined #mojo
02:41 d4rkie joined #mojo
02:41 good_news_everyone joined #mojo
02:41 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/q_4xLA
02:41 good_news_everyone mojo/master be51a3b Sebastian Riedel: added schemes attribute to Mojo::URL
02:41 good_news_everyone left #mojo
02:42 sri that's the solution i've talked about before, obviously there's some interest, so lets try it
03:10 ahutton anyone used NYTProf with Hypnotoad?
03:10 sri what would be the point?
03:11 ahutton sri: we normally run tests under hypnotoad instead of the daemon because production uses hypnotoad
03:12 sri hypnotoad is just a pool of daemons basically
03:13 ahutton sri: I get your point... if hypnotoad were limited to one worker it would be essentially the same as running as the daemon
03:13 * sri profiles with daemon
03:14 ahutton sri: if you profile with NYTProf, do you know how to get NYTProf to provide code coverage? Devel::Cover is not building properly in my environment it seems.
03:15 sri afraid i don't understand the question
03:16 ahutton sri: I saw some talk when I was google searching that people would use NYTProf and parse the output to produce code coverage reports, but I didn't really see any examples of this. I was wondering if you did anything like that?
03:16 sri no
03:16 ahutton sri: ok :)
03:26 inokenty joined #mojo
03:44 preflex_ joined #mojo
03:56 fildon__ joined #mojo
04:04 sh4 joined #mojo
04:30 * sri is not sure yet if Mojo::URL->schemes will stay
04:44 Kulag joined #mojo
05:05 buu Does Test::Mojo do something to prevent stderr from being sent by actions its executing?
05:06 buu I get a bunch of prelimary warnings, as expected, but then when it executes the actual 'post_json_ok' test I don't see any more stderr
05:07 buu Hrm, ok, it was throwing an exception before the warnings.. but I don't see that anywhere
05:08 ynonp joined #mojo
05:11 Kulag joined #mojo
05:13 xaka joined #mojo
05:16 ka2u joined #mojo
05:19 basiliscos joined #mojo
05:19 Drossel joined #mojo
05:23 egopro joined #mojo
05:23 ka2u joined #mojo
05:38 Kulag joined #mojo
05:42 ver joined #mojo
05:45 mire_ joined #mojo
06:05 denisboyun joined #mojo
06:19 Kulag joined #mojo
06:35 Mike-PerlRecruiter_ joined #mojo
06:43 dpetrov_ joined #mojo
06:51 ynonp joined #mojo
06:51 dod joined #mojo
06:52 Kulag joined #mojo
07:03 dod joined #mojo
07:10 Kulag joined #mojo
07:11 rihegher joined #mojo
07:18 Kulag joined #mojo
07:23 Vandal joined #mojo
07:30 Kulag joined #mojo
07:36 Drossel joined #mojo
07:40 rihegher left #mojo
07:51 SmokeMachine joined #mojo
07:53 batman sri++ # http://git.io/q_4xLA
07:54 d4rkie joined #mojo
07:54 batman sri: i didn't know that i used any internals of the Mojo::IOLoop in Mojo::Redis ...
08:13 fhelmber_ joined #mojo
08:23 d4rkie joined #mojo
08:23 ObseLeTe joined #mojo
08:25 jzawodn joined #mojo
08:29 maxhq joined #mojo
08:32 Vandal how make Mojolicious::Plugin::I18N not to change urls?
08:38 Vandal no, not urls, templates
08:40 jamesw joined #mojo
08:40 Vandal i get it
08:47 suy joined #mojo
09:08 nelio joined #mojo
09:18 jpn joined #mojo
09:43 egopro joined #mojo
09:46 mrphilov1 joined #mojo
09:51 mrphilov joined #mojo
10:02 ladnaV joined #mojo
10:24 Kulag joined #mojo
10:54 denisboyun joined #mojo
10:56 hrupp joined #mojo
11:26 bowtie joined #mojo
11:40 mtths joined #mojo
12:07 inokenty left #mojo
12:44 marty_ joined #mojo
12:52 rem_lex| joined #mojo
13:05 abra joined #mojo
13:18 gryphon joined #mojo
13:18 moltar joined #mojo
13:18 Kripton joined #mojo
13:20 rl joined #mojo
13:36 ynonp joined #mojo
13:38 ObseLeTe joined #mojo
13:57 ynonp joined #mojo
14:09 ObseLeTe joined #mojo
14:12 Bender joined #mojo
14:14 bluescreen joined #mojo
14:46 sh4 joined #mojo
14:53 btyler joined #mojo
14:58 Kripton joined #mojo
15:04 inokenty joined #mojo
15:09 Kripton joined #mojo
15:09 ka2u joined #mojo
15:13 xaka joined #mojo
15:14 marty joined #mojo
15:15 tianon sri: I'm definitely interested in keeping Mojo::URL->schemes (obviously), but will be understanding if you don't ;)
15:22 sri tianon: the big question is if there's something smarter we could be doing instead
15:31 btyler joined #mojo
15:39 sri i might actually have made a mistake with Mojo::URL->data
15:48 denisboyun joined #mojo
15:48 sri shit, i think i really messed up
15:51 jberger time for a breaking release?
15:55 sri i don't have time to write a 4.0 announcement atm :S
15:55 sri but we need to make a breaking change
15:57 sri i suppose we may have to leave the cpan version broken for some time :(
15:58 jberger how broken is it?
15:59 sri very, ->data needs to be gone
15:59 sri and parsing rules changed completely
16:00 jberger :-/
16:00 sri mailto:sri@example.com?foo=bar has to result in <scheme>:<path>?<query>
16:00 * marty loves breaking changes cuz it "usually" means things are progressing without baggage.   weeeee!   btw, o/
16:00 sri damn url spec, i completely misinterpreted it
16:01 sri -- for everyone who let me do it though :/
16:02 ka2u joined #mojo
16:10 good_news_everyone joined #mojo
16:10 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/qsixGw
16:10 good_news_everyone mojo/master d4436f3 Sebastian Riedel: removed schemes and data attributes from Mojo::URL
16:10 good_news_everyone left #mojo
16:10 sri this is how it's supposed to be
16:11 sri master is now 4.0 though :S
16:14 yko hoho, 4.0
16:14 yko it feels like only yesterday we had 0.99xxx and almost no documentation around
16:14 sri :)
16:15 yko time runs so fast.
16:15 yko I've actually got married somewhen around 3.55 :D
16:16 sri oh, congrats!
16:16 yko hehe, thanks
16:20 batman sri: what is the result of that? looks like it will parse redis: and mongodb: stuff again..?
16:20 sri it does
16:20 batman yay!
16:21 batman good_news_everyone++ # ;)
16:22 sri it now just follows RFC 3986 rules
16:22 batman ok
16:23 Adurah Mojolicious 4? Looks like you've lost the Asians for a while. :P
16:23 whitebook joined #mojo
16:23 sri i would very much appreciate it if some of you would review the code for RFC 3986 and http://url.spec.whatwg.org compliance though
16:24 sri so we don't end up with a similar mess again
16:28 sri *crickets*
16:28 ObseLeTe joined #mojo
16:29 * jberger is reading the spec
16:30 jberger its not as nice as some of the others you have pointed out :-
16:30 jberger :-/
16:30 sri the whatwg spec is one of the worst i've ever seen
16:30 sri http://www.ietf.org/rfc/rfc3986.txt # the RFC is not too bad
16:31 sri just a bit bloated
16:31 jberger whatwg section 5.1 seems to be the most readable
16:31 jberger but still, ick
16:32 basiliscos joined #mojo
16:32 tianon looks like mailto: can also have a fragment (according to whatwg)
16:32 sri they all can, yea
16:32 tianon tasty lol
16:35 sri but you're right, that should be tested
16:35 good_news_everyone joined #mojo
16:35 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/h9hw1A
16:35 good_news_everyone mojo/master ff4464a Sebastian Riedel: more Mojo::URL tests
16:35 good_news_everyone left #mojo
16:37 jberger should the mailto addressee be in the userinfo section?
16:37 jberger mailto:userinfo@host
16:37 jberger ?
16:38 good_news_everyone joined #mojo
16:38 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/-Qy3cw
16:38 good_news_everyone mojo/master 501c637 Sebastian Riedel: use more realistic data in Mojo::URL tests
16:38 good_news_everyone left #mojo
16:38 sri jberger: mailto has no //, so it has no authority
16:38 jberger oh ok
16:38 sri therefore it can't have userinfo
16:38 jberger gotcha,
16:38 jberger man this is hard to follow
16:39 sri the rules are pretty simple once you grok them
16:39 jberger oh there I see that one
16:39 sri of course i prolly missed a dozen exceptions from the rules as usual :/
16:40 jberger sri: could I convince you to use /x for the official regex?
16:40 xaka joined #mojo
16:40 jberger I know you like a tight ship, but its rather hard to read
16:41 * jberger can deal with is as-is though
16:41 sri it's the official regex copy and pasted from the spec
16:41 sri it never changes
16:41 jberger oh
16:41 jberger ok nm them
16:41 jberger then
16:41 ynonp joined #mojo
16:41 tianon page 50, btw :)
16:42 tianon er, 51
16:42 tianon "Appendix B.  Parsing a URI Reference with a Regular Expression"
16:43 jberger tianon, got it, thanks
16:43 jberger I had still been reading the whatwg spec
16:43 jberger yeah, this spec is better
16:43 tianon yeah, I was looking through that and was confused at part of that regex too, then found the official from the RFC, which is very nice :)
16:44 tianon way better than whatwg :)
16:45 sri whatwg has done great work for the html5 spec, but the url one is just terrible
16:45 jberger I will look some more later today
16:45 jberger I have some things scheduled
16:45 jberger compliance++
16:46 jberger as always
16:46 tianon which in this case leads to usability++ :)
16:48 tianon so doesn't this mean that file:/// urls have an authority of the empty string, but should still stringify to file:/// ?
16:49 tianon basically empty string vs undefined?  so line 202 should to be if defined?
16:49 tianon *should be
16:54 tianon the same could probably be said of fragment and query, but I'd think they're much less important than file:/// getting the slashes it's supposed to have :)
16:58 tianon so along those same lines, would it be important for an input of "file:///?#" to come back out exactly as it went in?
17:02 Britzel joined #mojo
17:03 whitebook1 joined #mojo
17:06 denisboyun_ joined #mojo
17:07 labrown joined #mojo
17:08 good_news_everyone joined #mojo
17:08 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/UpavUA
17:08 good_news_everyone mojo/master 82af89a Sebastian Riedel: removed Mojolicious::Plugin::PoweredBy
17:08 good_news_everyone left #mojo
17:09 sri tianon: hmm, file:/// is a problem :S
17:22 sri tianon++ # well spotted
17:22 * sri thinks he has a solution
17:22 sri there are actually a few more bugs
17:31 tianon sri: glad I could help :)  seriously, you do good work, and any little bit I can contribute back, I look for :)
17:33 xaka joined #mojo
17:35 sri the github bot is a bit late, but i've fixed it \o/
17:35 sri https://github.com/kraih/mojo/commit/e7​65d3805da454e06350d64b1b8f80ea6fd81adb
17:35 sri empty authority is an interesting case
17:36 sri since you can have an empty authority without a scheme
17:36 sri since a path can't start with //
17:37 sri (relative that is)
17:37 dod joined #mojo
17:38 sri //foo/bar is not a relative URL with authority foo and path /bar
17:38 trone Hi all. I have a mojolicious app and don't get what are make continously get action that was logged as 'GET /robots.txt (Anonymojo).' … someone knows?
17:39 sri s/not//
17:39 sri (relative to a scheme)
17:46 sri crab: i've removed the X-Powered-By header completely
17:48 xaka joined #mojo
17:52 whitebook joined #mojo
17:54 sri jberger, tempire, marcus, crab: you ok with the code name Top Hat?
17:56 good_news_everyone joined #mojo
17:56 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/CvT7Iw
17:56 good_news_everyone mojo/master 788f833 Sebastian Riedel: this is a major release
17:56 good_news_everyone left #mojo
17:57 sri if there's anything else you want to get rid of, now it the time
17:57 jberger sri: I like it!
17:57 jberger perl -CSAD -Mutf8=:charnames -E 'say "\N{TOP HAT}"'
17:58 jberger I don't really expect you'll like it, but I just want to ask one more time:
17:59 jberger $ws->send_json({...}) or $ws->send( json => { ... })
17:59 sri it's not that i don't like it
17:59 sri the problem is receiving json
17:59 jberger its how to specify the OPCODE
18:00 jberger $ws->on( json => sub { } )
18:00 jberger but how to know
18:00 jberger hmmmm
18:00 jberger yeah, 5.0
18:00 jberger :-)
18:00 sri anyway, it's not a breaking change
18:00 jberger true
18:00 sri no need to hurry :)
18:01 jberger were we going to do something about content negotiation defaults?
18:02 btyler joined #mojo
18:05 good_news_everyone joined #mojo
18:05 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/TI8E4w
18:05 good_news_everyone mojo/master e765d38 Sebastian Riedel: fixed a few authority handling bugs in Mojo::URL
18:05 good_news_everyone left #mojo
18:05 ka2u joined #mojo
18:17 wingfold joined #mojo
18:32 whitebook joined #mojo
18:37 Mike-PerlRecruiter_ joined #mojo
18:47 alester joined #mojo
18:52 gtodd would such a thing be useful -->  Mojo::Plugin::HTML5Boilerplate (add h5bp templates to a site)
18:53 gtodd be suitable for a plugin?  ... it seems more like using Mojo::Plugin would be not really needed not sure how much of an advantage it would be .... maybe for deployment/management etc.
18:56 bluescreen_ joined #mojo
19:04 inokenty left #mojo
19:09 whitebook joined #mojo
19:12 dbr joined #mojo
19:12 yko gtodd: maybe that should be a command?
19:16 suy joined #mojo
19:19 yko http://bit.ly/12eeiN9 <-- time ago I used to bootstrap controllers with a single command
19:20 yko gtodd: ^^ for (probably a bit outdated) example
19:22 yko did anyone already create something like `mojo generate bootstrap` to fetch latest bootstrap template and turn it into a default layout?
19:41 xaka joined #mojo
19:57 whitebook joined #mojo
20:02 good_news_everyone joined #mojo
20:02 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/pdw37w
20:02 good_news_everyone mojo/master b87ec42 Sebastian Riedel: use regex from RFC 3986 unchanged
20:02 good_news_everyone left #mojo
20:02 sri if we say it's the official regex it really should actually be -.-
20:04 tianon heh, indeed :)
20:05 dpetrov_ joined #mojo
20:11 nelio joined #mojo
20:14 dod joined #mojo
20:28 ObseLeTe joined #mojo
20:38 tianon joined #mojo
21:00 * sri considers removing Mojolicious::Plugin::RequestTimer as well
21:00 sri it depends on mojolicious internals, making it not such a great example
21:03 sri in addition we would start with no active hooks by default
21:09 jpn joined #mojo
21:18 whitebook joined #mojo
21:20 good_news_everyone joined #mojo
21:20 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/8TkzOA
21:20 good_news_everyone mojo/master de6955d Sebastian Riedel: removed Mojolicious::Plugin::RequestTimer
21:20 good_news_everyone left #mojo
21:21 SmokeMachine joined #mojo
21:21 good_news_everyone joined #mojo
21:21 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/Y2V44w
21:21 good_news_everyone mojo/master efe5c63 Sebastian Riedel: updated plugin list
21:21 good_news_everyone left #mojo
21:22 davido joined #mojo
21:22 tempire I find myself expecting json arrays returned by UserAgent to be in collecdtions
21:22 tempire *collections
21:22 sri hehe
21:30 sri out of the other plugins there is now only PODRenderer that still kinda depends on internals
21:31 sri which is only Mojolicious::Renderer->_bundled(), to get a bundled template
21:33 sri don't think there's a cleaner solution though
21:34 sri since that bundled template also shares a partial template with the built in 404/500 pages
21:39 ka2u joined #mojo
21:44 good_news_everyone joined #mojo
21:44 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/JKp2KQ
21:44 good_news_everyone mojo/master 36c543c Sebastian Riedel: stop timer more consistently
21:44 good_news_everyone left #mojo
22:15 mire_ joined #mojo
22:16 Gedge joined #mojo
22:39 gtodd yko: thanks for that reference
22:39 gtodd good idea re: command
22:47 mrphilov joined #mojo
23:15 jpn joined #mojo
23:25 good_news_everyone joined #mojo
23:25 good_news_everyone [mojo] kraih pushed 1 new commit to master: http://git.io/H5fvIw
23:25 good_news_everyone mojo/master 911fcc8 Sebastian Riedel: removed MOJO_EXCEPTION_VERBOSE environment variable
23:25 good_news_everyone left #mojo
23:29 * sri wonders if 4.0 should accept PLACK_ENV as an alternative to MOJO_MODE
23:30 sri MOJO_MODE would have a higher precedence
23:31 * sri still doesn't like abstraction leaking out... but there is no way to make it work through Mojo::Server::PSGI... just Mojolicious
23:46 ka2u joined #mojo
23:50 jberger http://en.wikipedia.org/wiki/​Dunning%E2%80%93Kruger_effect
23:50 * jberger wants to go off topic for a min
23:51 jberger when writing a code sample for a Perl job, using CPAN is cool right?
23:53 jberger I could have written something like Algorithm::Combinatorics::combinations, but why? https://metacpan.org/module/Algori​thm::Combinatorics#subsets-data-k
23:53 jberger oops A::C::subsets

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