Camelia, the Perl 6 bug

IRC log for #mojo, 2011-08-19

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

All times shown according to UTC.

Time Nick Message
00:15 ispy_ joined #mojo
01:03 sri hmm, i might go a little crazy again soonish
01:03 sri just a heads up :D
01:03 sri had a little idea i'll try now
01:15 SmokeMachine joined #mojo
01:28 elb0w__ left #mojo
01:39 d4rkie joined #mojo
02:18 newbie joined #mojo
02:22 Alias joined #mojo
02:28 elb0w___ whats the proper way to die in a mojo route?
02:28 elb0w___ do I just return?
02:28 elb0w___ or die
02:32 GitHub128 joined #mojo
02:32 GitHub128 [mojo] kraih pushed 1 new commit to master: http://bit.ly/nN70U6
02:32 GitHub128 [mojo/master] renamed Mojo::DOM::Collection to Mojo::Collection and added a few new methods - Sebastian Riedel
02:32 GitHub128 left #mojo
02:33 sri a little crazy :)
02:34 elb0w___ like a fox?
02:34 sri but i think the advantages outweight the risks
02:36 perlrocks Twitter: "Made Mojo::DOM collections quite a bit more powerful. :) http://t.co/eyZ7vtz #mojolicious #perl" --kraih http://twitter.com/kraih/status/104381155566030848
02:38 t4nk665 joined #mojo
02:43 t4nk995 joined #mojo
02:43 t4nk995 left #mojo
02:45 GitHub121 joined #mojo
02:45 GitHub121 [mojo] kraih pushed 1 new commit to master: http://bit.ly/obGKpY
02:45 GitHub121 [mojo/master] better tests - Sebastian Riedel
02:45 GitHub121 left #mojo
02:46 GitHub168 joined #mojo
02:46 GitHub168 [mojo] kraih pushed 1 new commit to master: http://bit.ly/nfyYlX
02:46 GitHub168 [mojo/master] release preparations - Sebastian Riedel
02:46 GitHub168 left #mojo
02:48 xaka joined #mojo
02:49 perlrocks Twitter: "Mojolicious 1.81 by SRI - http://t.co/CQjXcHH" (ms) --cpan_new http://twitter.com/cpan_new​/status/104384544915984384
02:51 Alias left #mojo
02:59 elb0w____ joined #mojo
03:01 GitHub135 joined #mojo
03:01 GitHub135 [mojo] kraih pushed 1 new commit to master: http://bit.ly/nvrMxh
03:01 GitHub135 [mojo/master] better examples - Sebastian Riedel
03:01 GitHub135 left #mojo
03:14 alnewkirk joined #mojo
03:34 crab i
03:34 crab hi
03:39 alnewkirk hello
03:42 sri not 100% sure yet Mojo::Collection was such a good idea
03:42 * sri waves
03:43 sri but i can totally see it being useful for oneliners
03:47 sri perl -Mojo -E'c(@ARGV)->map(sub { "http://mojolicio.us$_" })->each(sub { g($_)->dom("title")->join("")->say })' / /perldoc
03:48 alnewkirk ummm
03:49 sri maybe standalone collections are pointless, not sure yet
03:55 alnewkirk :/ extjs has something similar which is IMO pointless
03:55 sri why is it pointless? in what context?
03:56 sri you do realize collection are absolutely fricking awesome inside Mojo::DOM right?
03:56 sri *+s
03:57 sri i'm talking standalone usage only
03:58 alnewkirk :} i wasnt exactly sure how Mojo uses it but at-a-glance it looks like the extjs object MixedCollection which is basically a structure array with standard array manipulation methods attached
04:01 sri you might want to actually look up what mixedcollection really does
04:02 alnewkirk i have, i've written alot of extjs code of late, http://dev.sencha.com/deploy/ext-3.3.1/d​ocs/source/MixedCollection.html#cls-Ext.util.MixedCollection
04:03 sri it keeps a numeric index
04:04 sri of course it has to reinvent shit
04:05 sri (in addition to a key)
04:06 alnewkirk :/
04:06 alnewkirk whats this M::P::Geo plugin that just popped up on my screen
04:08 alnewkirk I find myself recommending Mojo::HTML alot lately
04:08 alnewkirk M::Dom i mean
04:17 GitHub195 joined #mojo
04:17 GitHub195 [mojo] kraih pushed 1 new commit to master: http://bit.ly/opPwBE
04:17 GitHub195 [mojo/master] added experimental filter method to Mojo::Collection - Sebastian Riedel
04:17 GitHub195 left #mojo
04:20 GitHub137 joined #mojo
04:20 GitHub137 [mojo] kraih pushed 1 new commit to master: http://bit.ly/pF3jlx
04:20 GitHub137 [mojo/master] better tests - Sebastian Riedel
04:20 GitHub137 left #mojo
04:21 sri i think the picture gets more clear now
04:25 alnewkirk this is probably another dumbass assumption on my part, ... i assumed url_for under a bridge would display the url from its origin, ... if that make any sense
04:41 alnewkirk sri, what do you think?
04:41 hernan joined #mojo
04:49 GitHub117 joined #mojo
04:49 GitHub117 [mojo] kraih pushed 1 new commit to master: http://bit.ly/nKmfeZ
04:49 GitHub117 [mojo/master] removed while and until methods from Mojo::Collection - Sebastian Riedel
04:49 GitHub117 left #mojo
04:52 GitHub5 joined #mojo
04:52 GitHub5 [mojo] kraih pushed 1 new commit to master: http://bit.ly/nGsKVT
04:52 GitHub5 [mojo/master] fixed typo - Sebastian Riedel
04:52 GitHub5 left #mojo
04:52 anaio joined #mojo
04:56 perlrocks Twitter: "Mojolicious 1.82 by SRI - http://t.co/8eKMjZ5" (ms) --cpan_new http://twitter.com/cpan_new​/status/104416404484276225
04:57 GitHub99 joined #mojo
04:57 GitHub99 [mojo] kraih pushed 1 new commit to master: http://bit.ly/oGK0e1
04:57 GitHub99 [mojo/master] fixed typo - Sebastian Riedel
04:57 GitHub99 left #mojo
05:00 alnewkirk joined #mojo
05:02 kaare joined #mojo
05:10 perlrocks Twitter: "Benchmarked perl webfrmks Dancer, Mojolicious and plain Plack with ab. Dancer ~ 200 requests/sec, Mojo ~ 200 req/s, Plack ~ 2000 req/s!" --truecrux http://twitter.com/truecrux​/status/104419832052723714
05:10 sri *facepalm*
05:12 xaka ok, time to move to Plack, ...
05:13 sri seriously
05:13 purl is that for real?
05:14 xaka it's ok that there is no session, helpers, templates, ..., ..., ..., but it does 2000/sec!
05:15 xaka so i can show it to my boss and get the bonus, hell yeah!
05:18 GitHub178 joined #mojo
05:18 GitHub178 [mojo] kraih pushed 1 new commit to master: http://bit.ly/r9pUbS
05:18 GitHub178 [mojo/master] renamed filter method in Mojo::Collection to grep - Sebastian Riedel
05:18 GitHub178 left #mojo
05:18 lammel2 joined #mojo
05:21 crab i benchmarked a hard disk, an ssd, and a ram disk with dd. hard disk: ~100MB/s, ssd: ~200MB/s, ram disk: 4GB/s! i'm going to store all my data in ram disks from now on.
05:21 perlrocks Twitter: "Mojolicious 1.83 by SRI - http://t.co/Wi6GTvd" (ms) --cpan_new http://twitter.com/cpan_new​/status/104422807567470592
05:23 sri /DEV/NULL IS WEB SCALE!!!1
05:23 xaka ok, time to move to ram disk from mojo, ...
05:23 xaka i can get one more bonus
05:23 xaka does is support http requests handling i hope?
05:24 xaka actually only notepad is web scale, write any result and you won!
05:27 GitHub71 joined #mojo
05:27 GitHub71 [mojo] kraih pushed 1 new commit to master: http://bit.ly/qWs22s
05:27 GitHub71 [mojo/master] documentation tweaks - Sebastian Riedel
05:27 GitHub71 left #mojo
05:43 Eugene joined #mojo
05:56 GitHub53 joined #mojo
05:56 GitHub53 [mojo] kraih pushed 1 new commit to master: http://bit.ly/pUyqsp
05:56 GitHub53 [mojo/master] documentation tweaks - Sebastian Riedel
05:56 GitHub53 left #mojo
06:06 sri http://mojolicio.us/perldoc/Mojo/Collection # still undecided if this module should go back into the Mojo::DOM namespace
06:08 sri (and in turn lose the c() constructor)
06:09 sri i better sleep over it
06:11 xaka i'm not sure that somebody will use it in perl often so if it was developed for DOM purposes, better move it into as part of cleanup
06:13 * xaka wonders if sri is interested in somebody opinion :)
06:13 minimalist wait a second ... how come plack is soooo much faster?
06:14 minimalist no I/O?
06:14 xaka no nothing
06:15 xaka somebody already asked the question here - what's better, use "die" or "render_exception"?
06:17 xaka the problem is that in production mode Mojo doesn't show anything, only "error happened" that is a bit useless for "die"
06:18 stephan48 log?
06:18 purl o/~ log, log, it's big, it's heavy, it's wood; log, log, it's better than bad, it's good! o/~ or flushed at startup
06:18 stephan48 create a log dir
06:19 stephan48 or drop in tghe dev exception template
06:22 sri minimalist: judging by the content of the tweet it seems rather safe to assume good old stupidity
06:22 sri xaka: why wouldn't anyone use it in perl?
06:22 sri the thing about method calls is that you can chain them almost infinitely without losing readability
06:23 sri try that with 2+ levels of normal map/grep
06:23 xaka it's for very specific tasks and map/grep "better" because you dont need import package, initiate it, depend on it and so on
06:24 sri so you did not pay attention?
06:24 sri http://mojolicio.us/perldoc/ojo#c
06:24 xaka for mathematics and difficult algos it's good, but how much such do we have in day by day?
06:25 sri the primary use case here is ojo
06:26 sri and the big question is if it makes quick hacks easier/more fun
06:26 xaka then is make sence ;) but again, if you need one liner, do you really want to download whole Mojo if it isn't installed yet? Mojo is Mojo
06:27 sri it is not supposed to be used on its own
06:28 koban joined #mojo
06:30 ispy_ joined #mojo
06:36 sri perl -Mojo -E'g("mojolicio.us")->dom("h1, h2, h3")->map(sub { $_->text })->grep(sub { /i/i })->join("\n")->say'
06:36 sri that is collections in action btw
06:39 xaka i've gotten "A::B::C does not exist, maybe a typo?" and then got in stuck till keepalive_timeout. May be better die than assuming delayed response?
06:40 xaka heuristic is very dangerous sometimes
06:41 xaka i could get the real error (typo) only when i ran it through "get" command, not test :( (because it stucked)
06:41 Firkin joined #mojo
06:43 sri maybe you should start assuming it is your fault
06:43 sri (because it clearly is here)
06:44 sri hint: tests can be run verbose
06:45 xaka i explicitly specified handler A::B::C and if it isn't found - i was waiting for an error of that
06:45 sri baah… why do i bother...
06:46 * sri resigns from user support again
06:47 xaka don't know, may be because sometimes it's not so user-friendly and as i said heuristics isn't good. This "may be" looks strange
06:48 sri prove -v is basic perl knowledge
06:48 xaka so your "may be" is about perl knowledge?
06:48 xaka when i write "use" in perl, it doesn't tell me "may be"
06:50 xaka if i dont use package - i dont write "use". if i use package - i write, make typo and get exception. that's how it should be
06:50 sri and if we enabled -v by default you people would be whining about it being too spammy
06:50 diegok http://d.pr/zbUm
06:50 diegok ^ In Riga :)
06:51 sri :)
06:53 marcus diegok: nice
06:53 sri marcus: Mojo::Collection of Mojo::DOM::Collection?
06:54 sri *or
06:54 marcus sri: will it have any DOM related methods? :)
06:55 * marcus reads back a bit
06:56 sri http://mojolicio.us/perldo​c/Mojo/Collection#SYNOPSIS # it's mostly about the c() constructor
06:58 sri basically, Mojo::DOM learned a few new tricks, and now i'm unsure if it's useful enough to allow standalone usage
07:02 sri i think it can be summed up as http://mojolicio.us/perldoc/ojo#c # yay or nay?
07:06 marcus sri: I liked judofyr's input about it working like Ruby enumerable.
07:06 marcus sri: if you just load it and add each method to your class.
07:07 marcus and you can iterate over whatever.
07:07 * sri doesn't follow
07:08 marcus sri: it's a mixin in Ruby tho, guess you can't support that without soiling the mojolicious inheritance
07:08 marcus sri: http://www.ruby-doc.org/co​re/classes/Enumerable.html
07:09 sri i know the ruby classes, but i don't see how that relates to the discussion
07:11 marcus sri: I'm just saying that your Collection class could easily be extended to support this functionality as well. But even if not, I think the c(@array) thing is nice.
07:11 marcus convenient
07:11 purl rumour has it convenient is still convenient
07:12 marcus I guess it's similar to https://metacpan.org/module/List::Enumerator :)
07:12 sri ok, i see :)
07:13 marcus sri: only problem I see is some people using $c for controller
07:13 marcus and being confused
07:14 sri that's ok i guess
07:17 sri important question for today is if we allow it to stay standalone or move it back into the Mojo::DOM namespace ;)
07:19 stephan48 i would let it go standalone if its useable from non DOM too
07:19 stephan48 if it depends on DOM move it back in
07:19 sri it doesn't
07:19 stephan48 doesnt depend?
07:19 stephan48 then standalone
07:20 arthas joined #mojo
07:24 xaka wooops, 1.81 haven't passed tests on my ubuntu: libmojolicious-perl-1.81/t/m​ojolicious/external/myapp.pl did not return a true value at t/mojolicious/external_lite_app.t line 19
07:24 sugar joined #mojo
07:24 rhaen Good morning everyone.
07:32 zakame_ hello all
07:32 purl It's a crazy world, but hello to you too!
07:32 zakame_ yep, crazy indeed
08:11 Drossel joined #mojo
08:15 ajgb joined #mojo
08:18 perlrocks Twitter: "Another exercise in learning Perl (using Mojolicious and IO::All) and JavaScript (jQuery-File-Upload) @ the same time: http://t.co/dEfEPYu" --zakame http://twitter.com/zakame/​status/104467353491869696
08:19 xaka sri: just found that 1.77 builds ok and starting from 1.78 it always fails on t/mojolicious/external_lite_app.t
08:23 Eugene joined #mojo
08:23 diegok Hm, I'm having a strange error using UA. This happen sometimes, just $ua->get('https://...') on a just created UA: http://pastebin.com/DG1gQ4Gv
08:24 diegok ^ sorry http instead of https
08:28 marcus sri: I like the collection, it fits into the mojo lego mindset
08:42 rhaen marcus: will you show some lego stop motion movies on the next yapc then? :)
08:44 marcus rhaen: maybe I will write some async software for my NXT using Mojo!
08:48 marcus https://metacpan.org/module/NXT
08:48 marcus seems like a good web sockets example!
08:48 spleenjack joined #mojo
08:51 cosmincx joined #mojo
08:51 rhaen I am still fighting the async things in mojo.
08:52 marcus bringing http://distillery.s3.amazonaws.com/media/2011​/08/07/eb0b63c0ef3140dfa4f446204fb3ba70_7.jpg to the talk and controlling it would be the ultimate live demo
08:52 rhaen I haven't looked into it that deep - I am using async stuff with node but it feels more "natural" to me in JS
08:52 marcus so much room for failure
08:52 rhaen you areplaying with LEGO NXT?
08:52 marcus yepp
08:52 rhaen on Linux? With Perl?
08:53 marcus on mac, with NXC so far
08:53 marcus but I want to try that perl module
08:53 rhaen huh, how is it? Is it fun?
08:53 marcus yupp, fun
08:53 rhaen really really fun?
08:54 rhaen are you building small robots or are you using the nxt brick for "sensor" stuff?
08:55 marcus robots so far, but I just got it
08:55 marcus built up the standard robot you see there and played around with controlling it and moving the sensors
08:55 marcus next project is quite ambitious, a robot to bring us water from the water coolr
08:55 marcus cooler
08:56 rhaen cool! :) I am dreaming of this gadget for a longer time. It's very expensive, tho.
09:00 marcus I got some peeps to bring a kit from the US as thanks for fixing their arduino installation
09:06 xaka fck fck fck, 2am and i finnaly found hidden bug and sent pull request
09:06 xaka time to go sleep
09:06 * marty waves
09:07 marcus xaka: nite
09:08 xaka the most magical part of bug is that when you run tests - everything works, but when you build Debian package (debuild), only one test fails
09:08 xaka https://github.com/kraih/mojo/pull/203/files - the only what i did to fix it
09:09 random joined #mojo
09:12 marcus xaka: returning self instead of the return value from run?
09:13 xaka marcus: exactly
09:14 marty xaka: 2am?  Are you located on west coast of US?
09:14 xaka marty: california
09:14 purl well, california is druggy druggy druggy druggy druggy druggy druggy druggy. or smoggy and crowded. or the land of the shiny happy people or degenerating or home of one of the most beautiful stretches of coast in the US (PCH from Santa Barbara to Santa Cruz) or (: california people) or the granola state or the land of fruits and nuts or on the Left Coast
09:14 * marty lives in Oregon
09:16 xaka why it fails only when i build package - don't know, really magic (paranormal activity, part 4)
09:21 marcus xaka: what about make disttest?
09:23 zakame lol paranormal
09:30 koban joined #mojo
09:36 gizzlon joined #mojo
09:36 nicomen joined #mojo
09:36 preflex joined #mojo
09:38 gizzlon hey
09:39 gizzlon does anyone know if this woul free memory: $data[0]->{big_chunk_of_text} = undef;
09:40 gizzlon would*
10:00 gizzlon left #mojo
10:00 Alias joined #mojo
10:03 gizzlon joined #mojo
10:04 gizzlon bah, never mind
10:16 random joined #mojo
10:44 tabbi joined #mojo
10:45 tabbi left #mojo
10:47 kaare joined #mojo
12:20 perlrocks Twitter: "Selection of materials on Mojolicious (version 2.0, will be added): - Run Mojolicious / PSGI application: min ... http://t.co/HFU12cN" (ru) --TenmeiFuemieh http://twitter.com/TenmeiFuem​ieh/status/104528109658832896
12:28 perlrocks Twitter: "@cstrep Locale::Maketext::Lexicon now support Mojolicious tags http://t.co/e6nID6b - thanks to @cstrep for the patch" --clintongormley http://twitter.com/clintongorm​ley/status/104530097847009280
12:31 Foxcool joined #mojo
12:34 ispy_ joined #mojo
12:39 dekimsey joined #mojo
12:43 visiogirl joined #mojo
12:47 cybergggirl joined #mojo
12:52 spleenjack1 joined #mojo
13:12 SmokeMachine joined #mojo
13:13 xxtjaxx joined #mojo
13:14 xxtjaxx Hi! Can I use the rendering/templates from inside my Plugin?
13:30 marcus xxtjaxx: yes, PodRenderer plugin does that
13:34 xxtjaxx marcus: ah found another way using proper plugin arch. just get $app passed in use it as a class variable and do a $content = $app->render_content('mytemplate'); :D
13:38 pinklady joined #mojo
13:40 perlrocks Twitter: "Selection of materials on Mojolicious (version 2.0, will be added): - Run Mojolicious / PSGI application: min ... http://t.co/KWOHxLJ" (ru) --TsukubaAsunawu http://twitter.com/TsukubaAsun​awu/status/104548287549407233
14:05 koban left #mojo
14:18 GitHub11 joined #mojo
14:18 GitHub11 [mojo] kraih pushed 1 new commit to master: http://bit.ly/o3Xf4f
14:18 GitHub11 [mojo/master] fixed small test portability issue - Sebastian Riedel
14:18 GitHub11 left #mojo
14:32 GitHub171 joined #mojo
14:32 GitHub171 [mojo] kraih pushed 1 new commit to master: http://bit.ly/nBPy2W
14:32 GitHub171 [mojo/master] added experimental first method to Mojo::Collection - Sebastian Riedel
14:32 GitHub171 left #mojo
14:34 sri i guess Mojo::Collection can stay
14:55 sugar joined #mojo
14:56 alnewkirk seems like a good addition
15:20 GitHub172 joined #mojo
15:20 GitHub172 [mojo] kraih pushed 1 new commit to master: http://bit.ly/oHfuDF
15:20 GitHub172 [mojo/master] fixed another small test portability bug - Sebastian Riedel
15:20 GitHub172 left #mojo
15:23 alnewkirk thats just f'ing sweet
15:24 alnewkirk url captures are available as template variables ... damn thats nice
15:27 alnewkirk why wouldn't url_for (if referring to a url under a bridge) render the url from its origin?
15:33 d4rkie joined #mojo
15:34 metaperl joined #mojo
15:45 xaka joined #mojo
15:46 xaka sri: are you still interested in %ENV dump?
15:47 sri xaka: yes
15:47 xaka ok, i'll attach it to pull request in few minutes
15:50 xaka https://github.com/kraih/mojo/​pull/203#issuecomment-1852262
15:51 xaka dump right before "Require ..." line
15:52 sri interesting
15:52 sri wonder why you don't have $ENV{USER}
15:52 xaka i think it's because of fakeroot which uses in debian when you build packages
15:53 xaka it creates fake env and so on
15:53 sri i see, the $ENV{HARNESS_ACTIVE} check should take care of that then
15:56 xaka let me know if you need to test something
15:59 sri well, it's already on github
15:59 sri so your problem should be solved
16:03 xaka when are you planning to make a new release? i'd like to upload stable version into repository
16:03 sri maybe later, we'll see
16:04 cosimo I found that $c->cookie(foo=>'bar', {expires=>'+1Y'}) doesn't work. Correct?
16:04 sri correct
16:04 cosimo sri: and, if correct,
16:04 sri we don't do the wonky +1Y style
16:04 sri just epoch
16:04 cosimo well, epoch makes sense ofc
16:05 cosimo however, people migrating from mod_perl / Apache::Cookie would really appreciate that I think
16:06 cosimo however epoch works for me
16:06 sri feel free to make a cookie helper plugin that understands +1Y ;)
16:06 xaka ok, all tests are passed
16:16 elb0w my future dog http://www.youtube.com/watch?v=87ASEG34rGI
16:16 sri argh… not available in your geographic region...
16:25 random left #mojo
16:46 tabbi joined #mojo
16:47 tabbi left #mojo
16:58 GitHub40 joined #mojo
16:58 GitHub40 [mojo] kraih pushed 1 new commit to master: http://bit.ly/oPrR2s
16:58 GitHub40 [mojo/master] added experimental sort method to Mojo::Collection - Sebastian Riedel
16:58 GitHub40 left #mojo
17:10 elb0w can yo guys upvote this? http://www.reddit.com/r/reddit.com/comments/jo0t2/​reddit_i_need_your_help_sweet_mother_in_law_being/
17:14 Eugene joined #mojo
17:35 SmokeMachine joined #mojo
17:36 GitHub190 joined #mojo
17:36 GitHub190 [mojo] kraih pushed 1 new commit to master: http://bit.ly/o4xAvP
17:36 GitHub190 [mojo/master] added experimental reverse and shuffle methods to Mojo::Collection - Sebastian Riedel
17:36 GitHub190 left #mojo
17:37 Eugene joined #mojo
17:38 sri ok, those are just for fun now :)
17:38 sri elb0w: upvoted
17:40 sri perl -Mojo -E'g("mojolicio.us")->dom("h1, h2, h3")->shuffle->join("\n")->say'
17:40 sri :D
17:53 elb0w thats all it takes now?
17:53 elb0w is that useragent?
17:56 GitHub42 joined #mojo
17:56 GitHub42 [mojo] kraih pushed 1 new commit to master: http://bit.ly/qlXtIv
17:56 GitHub42 [mojo/master] better documentation and tests - Sebastian Riedel
17:56 GitHub42 left #mojo
17:56 sri it's ojo, our oneliner module
17:57 sri at least the g() part
17:57 sri rest is in fact just useragent
18:01 sri but i think it is time to stop adding stuff to Mojo::Collection now and let it settle for a bit
18:05 GitHub37 joined #mojo
18:05 GitHub37 [mojo] kraih pushed 1 new commit to master: http://bit.ly/qfggsJ
18:05 GitHub37 [mojo/master] fixed typo - Sebastian Riedel
18:05 GitHub37 left #mojo
18:07 elb0w Whats that code editor thats on a website
18:07 elb0w ?
18:09 perlrocks Twitter: "Mojolicious 1.84 by SRI - http://t.co/fIauBTe" (ms) --cpan_new http://twitter.com/cpan_new​/status/104615967040086016
18:10 perlrocks Twitter: "Released #mojolicious 1.84 with a few fun new Mojo::DOM collection features, enjoy! :) http://t.co/65qYLS3 #perl" --kraih http://twitter.com/kraih/status/104616213686128640
18:11 perlrocks Twitter: "Released #mojolicious 1.84 with a few fun new Mojo::DOM collection features, enjoy! :) http://t.co/NES1Cid #perl" --kraih http://twitter.com/kraih/status/104616513885052928
18:12 sri xaka: test fix is included too
18:18 perlrocks Twitter: "#mojolicious hack of the day: "perl -Mojo -E'g("mojolicio.us")->dom("h1, h2, h3")->map(sub { $_->text })->shuffle->join("\n")->say'" #perl" --kraih http://twitter.com/kraih/status/104618342144421888
18:25 xaka what about json/xml/... in couple with "say"? :)
18:33 xaka when i run application, can i be sure that exactly Mojo::IOLoop->singleton instance is used for io/loop?
18:34 sri no
18:34 sri but you can check with Mojo::IOLoop->is_running
18:36 sri blocking backends like cgi/fastcgi/psgi don't have a running ioloop
18:36 xaka i'm only interested in morbo and development mode
18:37 perlrocks Twitter: "“@kraih: #mojolicious hack of the day: "perl -Mojo -E'g("mojolicio.us")->dom("h1, h2, h3")->map(sub { $_->text })->shuffle->join("\n")->say'" --sharifulin http://twitter.com/sharifuli​n/status/104623129267154944
18:39 SmokeMachine_ joined #mojo
18:40 Firkin left #mojo
18:52 xaka last few weeks i'm writting so much for (1..N) { iowatcher->one_tick(...) } so i'd like see it as a part of mojo :) we could pass timeout value to "start" and use timer inside it, so when timer happens - call stop and we done
18:53 sri that does not sound like a very smart implementation
18:54 xaka for me it looks like: start for N seconds
18:54 sri yes, not smart
18:55 sri ideally Mojo::IOLoop would notice that there is nothing more to do
18:56 xaka nothing more to do? what do you mean? in my example i have recurring so i'd like that it has been run (fired) 2 times, so i'm using one_tick twice
18:57 sri it would mean that if you clean up after yourself Mojo::IOLoop would stop on its own
18:57 xaka understand, but in case of many handlers (timers, recurring, streams, ...) it wont work
18:58 sri only if you don't clean up after yourself
18:58 sri anyway, i'm not very keen on adding bandaid fixes
18:59 xaka btw, net::dbus::reactor works in this way, nothing to do - run finishes
18:59 ispy_ joined #mojo
19:26 sugar joined #mojo
19:31 xaka sri: i'm looking at UserAgent/IOLoop code and can't find right place to extend "premature connection close" with a timeout information, drop/close methods/hooks have a lack of "reason" argument, and to add it need to do more than 1 line of changes
19:33 sri xaka: don't know how to solve that
19:35 sri quite sure i challenged you to solve it for exactly that reason a few days ago
19:41 Eugene joined #mojo
19:47 j-v-e joined #mojo
20:00 noganex_ joined #mojo
20:06 tabbi joined #mojo
20:06 tabbi left #mojo
21:25 alnewkirk hello? anyone here?
21:25 tempire echo echo echo
21:35 sri *crickets*
21:56 alnewkirk so sri, about the bridge issue?
22:11 sri alnewkirk: i don't see a question i could answer
22:12 alnewkirk sri, why wouldn't url_for (if referring to a url under a bridge) render the url from its origin?
22:12 alnewkirk i asked that 2-3 times since last night
22:12 sri if i knew what that question meant i would prolly have answered
22:12 sri or someone else
22:12 purl someone else is, like, the compiler
22:14 xxtjaxx sounds to me like an interesting wordpun...
22:15 tempire alnewkirk: I don't understand the question either.  test with expectation?
22:15 * tempire throws the url under a bridge
22:15 alnewkirk If I create a route under a bridge and name it, then refer to it by name, I'd expect it to give me the full route from its origin which would include the bridge part
22:15 alnewkirk .. is that wrong way to look at it?
22:15 alnewkirk lol
22:15 alnewkirk brb, pasting
22:18 tempire Behold, I have now learned Lisp.
22:18 alnewkirk here we go
22:18 tempire Bow before my pretentiousness.
22:18 alnewkirk http://paste.scsys.co.uk/134763
22:18 * sri bows
22:18 alnewkirk $self->url_for('account-login') produces /login
22:18 sri alnewkirk: now if you reduce it to maybe 2-3 lines i'll take a look
22:18 alnewkirk lol
22:18 alnewkirk ok
22:20 alnewkirk sri, http://paste.scsys.co.uk/134764
22:20 alnewkirk but i may not be looking at it correctly
22:21 sri ok, your expectation is correct
22:22 sri or wait
22:22 sri the url_for call looks wrong
22:23 sri it's either $self->url_for or $self->url_for('account-login', account => 'something')
22:26 alnewkirk sri, well actually its <%= url_for 'account-login' %> in code in writting now producing /login, i didnt think i had to explicitly add account => '...' because account still exists in the current captures
22:27 alnewkirk so im at /sri/login, login and go to /sri/dashboard, in the dashboard template i have <%= url_for 'account-login' %> to go back to the login page
22:27 sri it's a very complicated problem
22:28 sri we only allow captures to be used if we are rendering the current url
22:28 alnewkirk ah
22:28 sri i don't know the actual reasons, but back when we made the decision it was because of a huge problem
22:29 alnewkirk <%= url_for 'account-login', account => $self->param('account') %> seems redundant
22:29 alnewkirk .. but i dont know all the use cases and I could see it being an issue psosibly
22:29 sri i think it might have something to do with optional placeholders
22:30 sri best way to find out would possibly to allow placeholders to be used and see what tests break :)
22:30 sri *+be
22:31 alnewkirk sri, sounds like a plan to me :}
22:34 sri you actually only have to comment out a single line to change it :)
22:34 sri captures are manually cleared
22:35 alnewkirk nice, can we get a beta branch I can pull and test against
22:35 sri interesting
22:35 sri we have 2 tests breaking
22:35 alnewkirk nevermind :)
22:35 sri gonna investigate
22:36 sri i would have expected a lot more if it was a serious issue
22:36 sri i'm not sloppy with tests for serious stuff, maybe something else made the problem less problematic
22:38 sri we might be on to something here
22:39 sri the two tests are for cases that make no sense in the real world and could be considered broken
22:40 sri specifically /foo/:bar/baz returning /foo/bar when rendered from another endpoint
22:40 sri it makes no sense in this context
22:40 sri and that's the only broken test
22:41 sri i'm willing to give this a try
22:42 sri looking over the code, i think the decision might have been made because of formats, which are handled differently now
22:43 sri heck, i'm always for making less code do more :)
22:46 alnewkirk sounds good
22:46 sri old behavior was useless, so we only have to watch out for new untested side effects
22:49 GitHub28 joined #mojo
22:49 GitHub28 [mojo] kraih pushed 1 new commit to master: http://bit.ly/os6voD
22:49 GitHub28 [mojo/master] fixed a url_for bug where route captures would be ignored - Sebastian Riedel
22:49 GitHub28 left #mojo
22:51 sri almost certain it was format related
22:52 xaka ohhh no, one more verssion, ahh! i just uploaded 1.84 to repo :)
22:52 sri oh noes!
22:53 xaka ok, the fix isn't critical for me so i can wait for the next one
22:54 xaka what will happen when 1.99 be present? 2.xx?
22:54 sri ye
22:54 sri or if we have to 1.999999
22:55 xaka cool, looks like you're faster than google chrome now ;)
22:55 sri chrome is at 14.0 or so
22:55 xaka but it's about few more weeks for you, isn't? :-P
22:56 xaka last week was really productive, from 1.7x to 1.8x
22:57 xaka and, hell, i'm so happy that i did upgrade from 1.6x to 1.8x and it still works!
22:58 sri well, we've gotten really good with backcompat
22:59 sri just crazy stuff we don't officially supports breaks now ;p
22:59 xaka except some stuffs like premature close and monkey one_tick callings, still believe it can be improved (whispers)
23:00 sri well, i still think you're using it wrong or so… but i really don't want taht discussion again
23:05 xaka hohoho, discussion is fun!
23:08 sri not if the alternative is to hack on something really fun
23:09 sri think i used to enjoy discussions more when i was wrong more often :/
23:09 sri takes more work to win then :)
23:10 xaka how much beer help you in that? :)
23:10 sri good point, i need increase my beer consumption
23:11 * sri goes to get some
23:12 * xaka haven't drunk a beer for a 3 or 4 weeks and switched to the wine
23:35 sri oh, skitch is now free
23:41 xaka skitch>
23:41 xaka ?

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