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

IRC log for #mojo, 2015-03-06

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

All times shown according to UTC.

Time Nick Message
00:23 mattastrophe joined #mojo
01:00 tempire sri: I've found that if you can highlight the code snippets within the block you're speaking about, when you speak about them, it helps comprehension a lot
01:05 asarch joined #mojo
01:09 riche tempire: i second that.  i use that technique throughout all my training / workshop decks.
01:11 cpan_mojo WWW-Lipsum 1.001009 by Zoffix Znet - http://metacpan.org/release/ZOFFIX/WWW-Lipsum-1.001009 (depends on Mojolicious)
01:14 sri well, i know that i want to do highlighting, the big question is *how* ;)
01:14 * sri is interested in pretty examples
01:17 riche when i use msft powerpoint I use 90% yellow boxes, no border, fade-in.
01:18 riche but the last few days I started converting to LaTeX, because I just have far too many slides to maintain with powerpoint... so I need to learn how to do that with LaTeX
01:19 riche well, beamer specifically.
01:19 sri this is more my style :) https://vimeo.com/phusionnl/review/98027409/03ba678684
01:21 riche it's ok, but i don't like borders because i feel it can obscure context
01:21 riche or distract, because it leaves the user wondering what's behind the border
01:22 riche ugh some of it's bad, the font is just way too small
01:25 riche lol 25:54 please use a sequence diagram before i lose my mind
01:28 zivester joined #mojo
01:43 Guest46 joined #mojo
01:44 Guest46 joined #mojo
01:45 mattastrophe joined #mojo
01:47 Guest46 joined #mojo
01:48 Guest46 joined #mojo
01:49 riche joined #mojo
01:52 tempire I think dark background and light text is better
01:53 tempire And I make the highlight colors either more vibrant, or much brighter than the surrounding text.
01:53 tempire The contrast between the two has to be high, though, because the projector bulb could be old
02:03 asarch joined #mojo
02:10 riche tempire: I agree it looks nice but I suggest you look deeper on the psychology of color and its effects on learning
02:13 sri lol
02:16 sri if you really want to play that game, you have to take it serious and switch between multiple color palettes to keep people engaged and stuff
02:23 mattastrophe joined #mojo
02:26 riche sri: yes, i do take my training development very seriously
02:30 klapperl_ joined #mojo
02:35 tempire that might be
02:36 tempire I just know what I respond to
02:36 tempire So I adjust accordingly.
02:36 tempire Like, for example, if it's ugly, I don't care. It's going to take a lot of effort for me to pay attention.
02:36 tempire And I'll probably miss stuff.
02:37 tempire And if you have an entire conference where nothing is pretty, then I don't return.
02:38 tempire Of which I won't name names.
02:38 * firnsy refunds is tickets to UglyCon 2015
02:38 firnsy *his
02:39 tempire There actually is an uglycon
02:39 tempire And even its marketing is pretty.
02:40 tempire http://plasticandplush.com/images/2013/03/06/550/Uglycon.jpg
02:40 hshong joined #mojo
02:40 firnsy hypocrites
02:40 purl hypocrites are such fun :)
02:41 sri tempire: bullshit!
02:42 sri tempire: don't tell me you wouldn't go back to http://dhtmlconf.com/
02:45 inokenty-w joined #mojo
02:45 tempire #ohsnap
02:45 tempire That is amazing.
02:45 tempire That is it's own kind of beauty
02:45 tempire That circles back around into gloriousness
02:46 tempire I can't stop looking at this site.
02:46 tempire It's wonderful.
02:46 * sri broke tempire
02:46 bpmedley the mouse pointer sparkle is hypnotic, a gate to the second dimension
02:48 tempire That song
02:48 purl That song is for sucks.
02:48 tempire I love it.
02:49 chansen joined #mojo
02:56 jberger hahahahahah
02:56 purl LOLCON 5.5 reached.
02:56 jberger sri++
03:10 gtodd joined #mojo
03:16 Flying_Squirrel joined #mojo
03:20 gtodd why does this question seem to get asked so frequently:  http://stackoverflow.com/questions/28880680/route-to-static-file-in-mojo  ... maybe because the answers are different depending on whether you are using Mojolicious::Lite and ./public/ versus writing a more complicated application   ....
03:23 bpmedley gtodd: Is it also one of the most common cases?  Wanting to get assets delivered to the browser when “/“ is accessed.
03:25 zivester joined #mojo
03:29 gtodd guess so I usually just put things under ./public so I didn't realize it was complicated :)
03:29 noganex_ joined #mojo
03:29 sri is that a common question?
03:31 bpmedley From a newbies perspective there’s a lot to learn.  I wonder if a lot of people get excited and do not read through or know about the tutorial.
03:38 good_news_everyon joined #mojo
03:38 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/xxbl
03:38 good_news_everyon mojo/master 2622ee0 Sebastian Riedel: show how to create an index.html alias
03:38 good_news_everyon left #mojo
04:05 sri bpmedley: any progress on the minion patch?
04:06 bpmedley sri: No, I’ll have open source hours tomorrow and saturday, is that cool?
04:56 jb360_ joined #mojo
05:09 preaction and now i have a strange urge to make a DHTML theme for Statocles, and bundle it with the app
05:09 preaction ... and maybe on April 1, force the daemon to use that theme 0.0001% of the time...
05:21 ashimema joined #mojo
05:21 ashimema hi chaps
05:22 preaction hello
05:22 ashimema I'm just experimenting with websockets and have reached a minor confusion point..
05:23 ashimema basically, I'm trying to get a shared editor type app working, and so far it's working well for a single editor instance as such..
05:23 ashimema I'm now wondering what the general thoughts are for expanding it to allow for multiple editor instances..
05:24 preaction two people editing the same thing at the same time?
05:24 ashimema i.e. user a, b and c are working on document 1 in one 'editor' attached to one socket..
05:24 ashimema indeed preaction..
05:25 preaction i would say "no". then i would say "three-way merge" and "good luck"
05:25 ashimema if I then get users d, and e come along and want to work on document 2 collaboratively in their own editor..
05:25 ashimema what makes you say that preaction?
05:26 preaction because if two people edit the same character, word, line, sentence, paragraph, who wins?
05:26 preaction and if they're doing it in real time, when do you stop them to merge?
05:26 preaction it sounds, to me, like a ux nightmare
05:26 ashimema coolios.. that's a fair question.
05:27 ashimema it's somewhat like etherpad.. ever used it?
05:27 preaction no, i avoid these things for this reason
05:27 firnsy it's a pretty well solved google docs and the plethora of other collaborative editors
05:28 firnsy uggh ... "pretty well solved problem by ..."
05:28 preaction okay, so as i said, since the answer isn't "no", then the answer is "three-way merge"
05:28 ashimema fiar enough..
05:28 preaction and "good luck"
05:28 purl "good luck" is not plural.
05:28 firnsy maybe draw inspiration from owncloud's collaborative editor
05:29 ashimema indeed.. it's a pretty well solved problem.. so I'm using those as inspiration.. the documents we're editing are their own strictly structured format in my app.. hense not just going ahead and using google docs.etherpad off the shelf ;)
05:30 firnsy right, then hopefully your wheel is rounder ;)
05:30 ashimema my real question.. as I've not spotted it in the code out there yet, is whether I should be using 'one socket per editor instance', or using a shared socket and identifying the messages for individual editors using an id
05:30 preaction i mean, if i were to say "no" was unnacceptable, i'd probably break the document into paragraphs, or some atomic section where entering and leaving the section caused a merge, with conflicts queued up and pausing all merging until resolved (saving still to the server, but not as the golden copy)
05:30 preaction so, basically, a git workflow, at a rapid pace and automatic
05:31 firnsy ashimema: 50/50, both are sane choices
05:31 ashimema preaction, funily enough that's exactly what I'm doing ;)
05:31 bpmedley ashimema: WebRTC editor
05:31 firnsy your document id could be part of the websocket endpoint or you could choose to pass the document id with every message at a single endpoint
05:33 ashimema interesting.. hadn't found WebRTC editor.. will have a read of that.. thanks bpmedley
05:34 ashimema firnsy, glad I'm not going mad... I had a feeling it was pretty much a 50/50 thing
05:34 bpmedley ashimema: WebRTC is more of browser to browser communication; however, it can be used with other technologies to realize something like you’re talking about.
05:34 bpmedley ashimema: You would also be able to add voice, video, and maybe screen sharing.
05:35 ashimema haha.. that's a tad out of scope for my project ;)
05:35 bpmedley The ability to “fire up” an editing session that was as easy as a gist or pastie would be sweet.
05:35 ashimema my gut tells me I'd prefer to go the one socket per editor instance route.. though I'm not sure how to achieve this at the minute
05:37 bpmedley You might consider using a git repo for every set of files.  Then, the user can branch and merge other people’s edits, perhaps?
05:37 bpmedley https://www.vhx.tv/ <— I’d pay to watch mojoconf
05:38 ashimema that's a nice idea.. though again probably out of scope for this particular experiment..
05:38 ashimema I'm loving the idea's though :)
05:39 bpmedley So, you just want an app that 2-3 people can edit a structed text file?
05:41 ashimema sorta... it's actually a 'list' and I want them to be able to re-order, add/remove and update items on the list
05:41 firnsy ashimema: websocket('/document/:id') => sub {}
05:41 firnsy apart from bad syntax that'll give you an specific endpoint per document :id
05:42 ashimema firnsy.. does that not still open one socket, but pass id in per message?
05:42 ashimema hang on.. brain melty feeling
05:42 ashimema of course.. i'm an idiot.. that's exactly what I've done!
05:42 ashimema I just thought it was working the other way like that
05:43 irq joined #mojo
05:46 sri bpmedley: if i say no you'll make a patch faster? :)
05:46 bpmedley Gaw!
05:48 * ashimema disappears off to play with this further
05:48 ashimema cheers chaps.. great sounding board.. as usual :)
05:49 bpmedley would you stay a few minutes?  maybe we can work on a proof of concept?
05:52 ashimema i'll certainly be around
05:52 ashimema right at the minute the proof of concept is so much more on paper and less in code that there's very little to see ;)
05:53 ashimema i'm converting from a locking angular editor to hopefully this more fluid editor
05:56 arpadszasz joined #mojo
05:57 melo joined #mojo
06:02 arpadszasz joined #mojo
06:06 Grinnz joined #mojo
06:07 arpadszasz joined #mojo
06:10 bpmedley ashimema: https://gist.github.com/brianmed/1b33a2273986a12d792f <— Thoughts?
06:10 bpmedley Line 34 will have issues, eventually
06:13 reneeb joined #mojo
06:14 ashimema interesting
06:14 ashimema not an approach I thought about before..
06:15 ashimema lots of patch events per say
06:15 ashimema cool
06:16 bpmedley Yes, and think (however, don’t have the command) that it will be possible to check for a rejected patch, and then allow the user to merge.
06:23 arpadszasz joined #mojo
06:25 ashimema that's certainly pretty nice in terms of a completely flexible file and patching..
06:25 ashimema I may have to adopt some of that ;)
06:25 bpmedley Cool.  Are you going to work on it in the next few minutes?  Just asking, because I’d be willing to stay up a little longer to help.
06:26 ashimema I've got some other avenues to continue on before going back to that i'm afraid
06:26 ashimema (bloomin customers do like to call aye ;) )
06:26 ashimema thanks for the support though mate.. you have a good night
06:26 bpmedley Awright, no issues.  If you can, please share your results when you get them.
06:34 ashimema will do
06:38 marmez joined #mojo
06:42 marcus Good morning
06:42 purl Lies!
06:48 bpmedley o/
06:49 batman good morning marcus :)
06:50 bpmedley batman: Mojolicious::Plugin::Bootstrap3 is sweet.  Makes for some rapid development.
06:50 batman bpmedley: nice to hear :)
06:50 batman oh! i've totally forgot to improve that module :/
06:50 batman want to make it easier to customize
06:51 bpmedley In what way?
06:51 batman bpmedley: right now, you have to do a lot of manual copying if you want to compile your own bootstrap with sass
06:51 batman copy+configuring
06:52 bpmedley Such as themeing bootstrap?  What reasons are there to compile and distribute your own bootstrap?
06:52 batman bpmedley: the default is rather big. most people don't actually need all of bs
06:53 batman bpmedley: https://metacpan.org/source/JHTHORSEN/Mojolicious-Plugin-Bootstrap3-3.2005/lib/Mojolicious/Plugin/Bootstrap3/sass/bootstrap.scss
06:53 batman the default includes quite a lot of functionality
06:53 batman at least i mostly never use jumbotron or modals
06:53 batman and there's a lot more i comment out
07:11 jb360 joined #mojo
07:18 bramirez joined #mojo
07:34 polettix joined #mojo
07:38 Lee jberger: yes, that's me
07:40 batman bpmedley: themeing comes on top of that again.
07:43 dod joined #mojo
07:49 dod joined #mojo
08:03 eseyman joined #mojo
08:05 oetiker_ joined #mojo
08:06 oetiker joined #mojo
08:12 Vandal joined #mojo
08:20 Lee jberger: what's the state of https://gist.github.com/jberger/72527a9702c2ed8b5e2f ?
08:25 trone joined #mojo
08:26 AndrewIsh joined #mojo
08:34 salv0 joined #mojo
08:58 dod joined #mojo
09:45 fhelmber_ joined #mojo
10:05 jontaylor joined #mojo
10:40 bramirez joined #mojo
10:54 irq_ joined #mojo
10:59 ZadYree joined #mojo
11:09 dvinciguerra joined #mojo
11:13 berov joined #mojo
11:36 trone is it a good practice after a redirect_to inside a controller? (a return or something)
11:37 trone rewrite: is it a good practice return after a redirect_to inside a controller ? (or something)
11:37 nicomen I would do that at least
11:37 nicomen to not risk unneccesary code running
11:38 bpmedley trone: I believe it’s up to how you want to structure your logic.
11:39 marcus Anyone tried writing an xmpp bot with mojo?
11:55 xdg joined #mojo
11:57 meshl joined #mojo
11:58 ver joined #mojo
12:06 meshl joined #mojo
12:30 Andreas2 joined #mojo
12:32 crab yes sort of
12:35 jberger Lee: those work, but see also https://github.com/avkhozov/Test-Mojo-Session/pull/2
12:35 niczero joined #mojo
12:37 jontaylor with the change to param, is there a way to get a full list of parameters now?
12:37 jontaylor names() isn’t exposed at the controller level, and using params through request seems to miss params that come in through routing
12:37 Lee jberger: yep, saw that - i'm trying to figure if the test failures i'm seeing with my app is related to v6, or Test::Mojo::More not doing the right thing or my attempts to patch it being wrong
12:38 Lee i've pushed a release of Test::Mojo::Most that patches Test::Mojo::More since the author seems to be MIA, but might rethink it having seen the Roles stuff
12:40 jberger my reason for the rules thing is that I have a test module in the works and I don't want to see ::MuchMoreMost
12:40 Lee agreed
12:40 jberger *roles (obviously)
12:41 jberger I'd love to see a whole test ecosystem if possible
12:42 Lee mostly i'm just frustrated that the author of Test::Mojo::More would relase a module in the Test:: namespace that doesn't have any tests itself!
12:42 Lee but anyway...
12:42 niczero joined #mojo
12:42 pink_mist jontaylor: $c->req->params()?
12:42 batman berov: why did you close the pr?
12:42 batman ...for assetpack
12:42 jberger Lee: the new Test::Simple is awesome for testing your tester
12:43 pink_mist jontaylor: oh, missed your second line
12:43 berov batman: because I am stupid
12:43 berov i messed up something
12:43 batman ok.
12:43 berov not sure what
12:43 berov I did git pull --rebase upstream master in my clone
12:44 berov now my new changes are not seena nad the old ones reappere in the new diff I ma not sure what to do
12:44 jberger Lee: https://github.com/jberger/Test-Mojo-Phantom/blob/master/t/die.t
12:44 berov nad->and
12:45 jontaylor pink_mist: I’m basically trying to update a validation plugin to support Mojolicious 6, but i’m struggling as we’ve lost the capability of getting all parameter names
12:45 Lee jberger: cool, ta. going to spend this aft sorting this out
12:46 berov batman: can you compare the two repos and advise?
12:46 batman berov: git reset --hard; git remote update; git merge origin/master; + resolve conflicts
12:46 batman i think your branch is "messed up" because there's a conflict in ./Changes
12:46 Lee i might just deprecate the module and point at the roles stuff
12:47 batman i could be wrong, but i'm not cleaning up (at least not now)
12:47 berov yes changes conflicted
12:47 batman then resolve the conflict.
12:47 berov but I accepted what is in your repo
12:47 berov ok, let me try and see how it goes
12:48 jberger Lee: well that's kinda what I was beating about the bush after, though an alternative might be just to keep it as a Task::Kenscho of mojo testing
12:48 Lee yep
12:49 Lee it has no rev deps so i'm happy to do that
12:51 bpmedley https://github.com/kraih/minion/compare/master...brianmed:master <— How is this; although, I probably need to remove the debug output in _work: sri, jberger?
12:53 bpmedley jontaylor: http://mojolicio.us/perldoc/Mojo/Parameters#names <— Will this work?
12:53 meshl joined #mojo
12:53 pink_mist bpmedley: that's what he gets from req->params() ... which doesn't work, no
12:53 jontaylor bpmedley: the only way I can find to get the parameters object is through the request object, which will miss any paramters added from the routing engine
12:55 jontaylor It can be hacked around my manually poking inside $self->stash->{'mojo.captures’}, but that feels like something that will come to bite me later on
12:56 pink_mist yeah, that doesn't sounds like a good strategy
12:56 pink_mist -s
12:57 jberger hmmmmm did that change? I never use matches from the stash as parameters
12:57 jberger always felt icky to me
12:57 jberger I always get them out of the stash
12:57 jontaylor ideally I think the solution is for a controller to has_a Mojolicious::Parameters object itself, which calls merge with $self->stash->{'mojo.captures’} and $self->req->params
12:59 jontaylor jberger: I could do that on a case by case basis but i’m trying to update Mojolicious::Plugin::ValidateTiny, and that needs a list of all params you try and validate against
13:01 reneeb Lee, jberger: I added the error_logs to the matrix, see the "not ok"s at http://matrix.perl-services.de/mojolicious?perl=5.10.1&amp;framework=6.0
13:01 jberger but path matches have to match your conditions or it doesn't match the route
13:01 reneeb And now it's possible to get the matrix as JSON, see http://matrix.perl-services.de/mojolicious.json?perl=5.10.1&amp;framework=6.0
13:01 Lee reneeb: excellent
13:02 jontaylor you might have something like /object/:some_id. and do validation that checks that :some_id is accessible by the current user for example.
13:02 reneeb It will take some time, to re-run all the not-oks as we have now 9 perl versions and 9 mojolicious versions...
13:02 dvinciguerra joined #mojo
13:02 jberger reneeb++
13:04 reneeb jberger: Galileo is skipped from the matrix for the moment as it takes some time to install it. Will add it again when the current run is over...
13:04 cpan_mojo Test-Mojo-Most 0.07 by Lee Johnson - http://metacpan.org/release/LEEJO/Test-Mojo-Most-0.07
13:04 jontaylor biab, getitng lunch, i’ll raise a github issue, or submit a pull request depending on what I come up with
13:05 jberger reneeb: I blame dbicdh
13:05 jberger :-P
13:06 reneeb jberger: could be. I really like dbicdh, it makes db deployment a no-op ;-)
13:07 reneeb Currently I use dbicd, MySQL::Workbench::DBIC, Mojolicious::Plugin::WebAPI to create APIs in just a few minutes...
13:08 jberger reneeb: I had to customize mine quite a bit
13:08 jberger mst is opposed
13:08 ribasushi reneeb: there's RapidApp that targets the same problem space, in case you haven't looked at it
13:08 ribasushi jberger: opposed to...?
13:08 reneeb which customizations?
13:09 reneeb ribasushi: I didn't know it, will have a look, thanks...
13:09 jberger I regenerate the ddl on the client side as needed
13:10 jberger that way it supports all the dbs that sql abstract does
13:10 amon joined #mojo
13:11 ribasushi reneeb: for a quick-feel: https://metacpan.org/pod/distribution/RapidApp/script/rdbic.pl#SYNOPSIS
13:11 jberger the queries are very simple CRUD, if every db doesn't support them I'd be surprised
13:12 ribasushi jberger: ddl != CRUD (which is dml), also sql-abstract is explicitly dml-only, sql-translator does the ddl
13:12 ribasushi so your sentence is confusing ;)
13:15 mattastrophe joined #mojo
13:16 pink_mist <reneeb> Lee, jberger: I added the error_logs to the matrix, see the "not ok"s at http://matrix.perl-services.de/mojolicious?perl=5.10.1&amp;framework=6.0 <-- I don't see a single "not ok" there ... in fact, I don't see a single result there at all :P
13:16 danejx joined #mojo
13:17 Lee pink_mist: results are rebuilding
13:17 marcusr joined #mojo
13:17 Lee change it to 5.20.2 and 6.0
13:17 jberger ribasushi: true, I should say that, I feel comfortable attempting to support the whole ranges of dbs, since the app's db need are so simple
13:17 pink_mist ah yes, with 5.20.2 I see things =)
13:17 pink_mist reneeb++
13:19 pink_mist reneeb: the description for Mojolicious-Plugin-DirectoryQueue 1.002 looks a little borked though =)
13:19 pink_mist same with a few others
13:20 Andreas2 joined #mojo
13:33 denny-- joined #mojo
13:38 Andreas2 joined #mojo
13:41 berov batman: I resolved the conflicts but now the build fails because of "Bareword "CSS::Sass::SASS_STYLE_COMPRESSED" not allowed..."
13:42 berov batman: can you look at it?
13:44 denny-- joined #mojo
13:44 berov it is not defined anywhere
13:45 batman berov: that's really weird that at least _that_ didn't fail here :(
13:46 berov I changed only POD
13:46 batman yeah, what i mean is that it _should_ have failed on my side as well
13:46 berov seems we have $ENV{ENABLE_LIBSASS_BINDINGS}=1
13:47 batman who's "we" ?
13:47 berov ok you :)
13:47 batman i've pushed by the way.
13:47 berov on Travis CI
13:47 batman i don't understand what you're saying :(
13:48 berov please look at https://travis-ci.org/jhthorsen/mojolicious-plugin-assetpack/jobs/53337400 to see what is going on
13:49 batman berov: https://travis-ci.org/jhthorsen/mojolicious-plugin-assetpack/jobs/53340654
13:50 berov batman: ok :)
13:51 dod joined #mojo
13:51 berov batman: then I think you can merge my pull request
13:51 jberger reneeb: I'm not seeing any failures, is that right?
13:52 jberger seem ... implausible
13:52 jberger oh nm
13:52 jberger I refreshed and see some
13:53 Lee jberger: https://gist.github.com/jberger/72527a9702c2ed8b5e2f # i think it makes sense to prefix $p with / if this isn't passed in
13:53 jberger hey hey, it looks like 5.21.9 fixed autobox!
13:54 Lee (otherwise Mojo::JSON::Pointer will always return something)
13:54 jberger Lee: '' is a valid json pointer
13:55 Lee so that means you have to call session_hasnt( '/foo' ) ?
13:55 jberger yes, just like in other mojo tests
13:56 jberger has(nt) doesn't make sense without a pointer though?
13:56 Lee right, that's not the way Test::Mojo::Session was doing it, but it makes sense
13:56 jberger you want to be able to say session_hasnt('foo') ?
13:57 Lee that's what my tests were doing, yes (with Test::Mojo::Session)
13:57 jberger I'm not sure I like it, if these are supposed to be json pointers
13:57 Lee i'm agnostic, but it's a trivial change on my end to fix the calls to session_*
13:58 jberger Lee: but json_has doesn't support that for example
13:58 batman berov: thanks :) *merged*
13:59 Lee yes, but it's explicit in the call that you're examining the json there whereas with the session tests you need to know something about the internals? either way i don't mind
13:59 batman berov: are the test suite ok on your side now?
13:59 berov batman: thanks
13:59 jberger Lee: hmmm, fair point
14:00 jberger notice also that Mojo::JSON::Pointer doesn't act on JSON
14:00 jberger it acts on a data structure
14:00 jberger which the session clearly is, you can have a deep structure in the session
14:00 Lee indeed
14:01 jberger my vote would be for consistency with Test::Mojo, but I do see your point
14:01 Lee i just had to jump into the debugger to figure what was going on here, but given this isn't an actual module yet something perhaps to point out in any docs for it if/when if does become a module
14:02 Lee s/if does/it does/
14:02 jberger I'm kinda hoping that avkhozov_ just takes the code wholesale and makes it the next release of Test::Mojo::Session
14:02 Lee yeah :D
14:02 jberger then I don't have to worry about it :-)
14:02 Lee i'm using this with Test::Mojo::WithRoles so i'm happy :)
14:02 jberger :-)
14:03 jberger yeah, I'm liking ::WithRoles
14:03 jberger seems to work nicely
14:03 jberger its the first time that the hints hash was actually what I was looking for
14:03 jberger I've almost used it about a dozen times
14:05 zivester joined #mojo
14:06 reneeb jberger: it's correct that you can't see many failures as I had to delete the not ok's from the database as the script tests only those combinations that aren't already in the database...
14:06 jberger also its looking like there might not actually be a Test::Mojo::Phantom, it might just be Mojo::Phantom and Test::Mojo::Role::Phantom
14:06 jberger reneeb: ah
14:07 reneeb pink_mist: please try again. You should see something now...
14:10 pink_mist no, still the same gibberish in the description
14:12 dod joined #mojo
14:14 reneeb pink_mist: I wanted to reply to you 'I don't see a single result at all...
14:14 pink_mist oh, well you already replied to that earlier :P
14:17 doby joined #mojo
14:19 Ptolemarch joined #mojo
14:30 mishantil Lets say you have '<div>foo <a href="/bar">baz</a> fez</div>' , what's the easiest way to replace the '<a ... /a>'-part with 'baz(/bar)' using Mojo?
14:31 mishantil So the end-result is '<div>foo baz(/bar) fez</div>'.
14:32 mishantil Doh! ->replace() *facepalm*
14:32 mishantil _as usual_ what I need is staring me in the face.
14:36 Ptolemarch joined #mojo
14:58 sh4 joined #mojo
15:09 dod joined #mojo
15:10 asarch joined #mojo
15:27 denny joined #mojo
15:36 sri bpmedley: as far as i can see it does exactly what i didn't want it to do ;p
15:36 sri which is completely separate code paths for the old ->perform and new ->start/->finished
15:38 sri your new methods have no documentation at all
15:39 jontaylor sri: With $c->param() no longer returning a list, there doesn’t seem to be a reliable way of getting a list of all parameters any more, and going through $c->req doesn’t seem to include params defined in the route
15:39 sri in fact, you've duplicated stuff like "$self->fail('Non-zero exit status')", i don't like that at all
15:40 sri jontaylor: you're too late, that was up for discussion a few weeks ago
15:40 jontaylor oh, what was the outcome?
15:41 sri nobody could come up with a use case, so we just removed it instead of bringing back a new solution
15:41 jontaylor my use case was to make a validation plugin work again, I’ve just sent this pull request for a plugin, but I hate the way i’ve done it https://github.com/koorchik/Mojolicious-Plugin-ValidateTiny/pull/6
15:42 sri bpmedley: why do you compare "$waited == $pid"?
15:43 sri bpmedley: is "my $waited = waitpid($pid, WNOHANG)" unreliable?
15:47 sri bpmedley: why are you mixing select() and usleep()?
15:49 sri those tests with a hard timeout seem very fragile
15:49 punter joined #mojo
15:49 sri fork can be pretty slow on some test boxes
15:51 sri is there a reason you use "== scalar("?
15:57 riche sri: did you see http://blogs.perl.org/users/byterock/2015/03/mojo.html
15:57 cpan_mojo Mojolicious-Plugin-JSLoader 0.05 by Renee Baecker - http://metacpan.org/release/RENEEB/Mojolicious-Plugin-JSLoader-0.05
15:58 riche not sure why s/he doesn
15:58 riche t use carton
15:58 sri riche: what about it?
15:59 riche it basically reads: mojolicious doesnt respect backward compatibility
15:59 sri he literally skipped *two* major releases
15:59 jberger and even then it only took him an hour
15:59 sri as inllustrated by the render_to_string change mentioned
16:00 sri if anything, that's a compliment for mojolicious
16:00 riche i agree ... but it's a strange post
16:01 sri just 30 minutes to get apps from 4.x to 6.x
16:01 sri try that with rails!
16:02 riche maybe he was on here to get the answers
16:03 genio That seems like a rather bitchy post about something simple to fix
16:04 sri bpmedley: anyway, what bugs me most is that it's just tacked on, and not integrated into the rest of the api
16:05 sri i'm ok with lower level methods being exposed in the Minion::Job api, but don't want duplicated functionality
16:05 cpan_mojo Mojolicious-Plugin-CSSLoader 0.07 by Renee Baecker - http://metacpan.org/release/RENEEB/Mojolicious-Plugin-CSSLoader-0.07
16:07 sri riche: you could point out the false information though
16:07 btyler yeah, I imagine it'd be tricky to get people to pay for mojo LTS when most of the changes are pretty quick renames. https://railslts.com/ for comparison
16:07 sri he went from 4.x to 6.x, not 5.x to 6.x as the article claims
16:08 cpan_mojo Mojolicious-Plugin-AssetPack 0.39 by Jan Henning Thorsen - http://metacpan.org/release/JHTHORSEN/Mojolicious-Plugin-AssetPack-0.39
16:10 jberger sri: notice he mentions that the last time he went from 3.0 - **~** 5.0
16:10 riche yah it's strange
16:10 jberger sounds like he knew he didn't make it all the way to 5.0
16:11 riche yah that's why i said that about carton
16:11 riche he may have a mismatch across many machines
16:11 riche anyways
16:14 Lee byte_rock still hasn't responded to my PR from 6 months ago so i'm not sure he's a great example to take as being on top of changes in Mojo
16:16 genio The "About byterock" section at the top-right...
16:20 riche sri: you must feel super strong about signatures
16:21 sri i do
16:21 riche hell, they didn't even want to consider making filetest a compile option
16:22 riche some things i guess just want to stay in 1994
16:22 sri of course... today i'm not in a great mood, only seen complaints in the mail so far... so...
16:22 riche well, your target for the energy displacement is a fair choice
16:23 mst also, that blog post does specifically say "I am not complaining"
16:24 riche mst: of course it's a complaint, and then a pat on the back
16:24 riche that's why i said it was strange
16:25 dod joined #mojo
16:25 * jberger pats sri on the back
16:25 mst I don't find it strange at all, it's somebody who's clearly accepted that fixing stuff up is the price of progress, but still wants to note what things he had to fix
16:26 mst I could probably write a similar post about dist-upgrading debian or whatever, and still wouldn't feel like I was mad at debian
16:28 riche sri: note it's not just 30-60 min.  it's " 30~60 mins per app."
16:28 sri mst: fuck you, that comparison to reini went too far
16:28 mst was kicked by sri: mst
16:29 sri it's not a good day, i'll be taking a break from open source i suppose (close to a burnout)
16:29 riche maybe a weekend of afk
16:30 riche you did after all release 6.0
16:31 riche wow p5p is wack
16:32 jberger I think it time for jberger to state the obligatory "lets keep it calm in #mojo at least"
16:32 riche now i feel sorry i brought up that blog post
16:32 riche gheesus
16:34 disputin joined #mojo
16:35 Adura "Having the courage & commitment to venture into the ‘unknown’ is required 4 #mojo"
16:36 jberger Adura: I hope you mean that in the positive sense
16:37 Adura Lightening the mood with the help of Debbie.
16:37 avenj left #mojo
16:38 jberger I guess I don't know the quote
16:38 Adura https://twitter.com/deborahkagan/status/573740041362673664 her, heh.
16:39 Oleg joined #mojo
16:39 jberger Adura: AHHAHAHAHAHAH
16:39 gtodd riche: the mailing list or the channel? :-)
16:40 jberger I think we should all retweet that one :-)
16:41 jberger I've done my part ;D
16:41 riche gtodd: both.  Since 1997 I have extremely limited success speaking with them.  i feel that's the case because i don't dedicate every waking moment to perl
16:41 Adura With the amount of people asking for help with Perl here along with Mojo, it's a pretty accurate quote.
16:42 gtodd sigh
16:42 riche that twitter link is just absurd
16:43 jberger riche: she does this all the time
16:43 jberger she has no idea who we are
16:43 jberger completely unrelate
16:43 jberger d
16:43 gtodd mojolicious is a personal healing thingie too
16:43 riche gtodd: gross
16:43 gtodd I think they have camps and retreats etc
16:44 gtodd do NOT mix up these events
16:44 jberger did mst leave?
16:44 jberger :(
16:44 riche he got kicked
16:44 Adura He was kicked, ror.
16:44 jberger :o
16:44 gtodd with perl/mojolicious
16:45 riche we should hijack their shit with mojoconf marketing
16:45 jberger convos doesn't give a lot of status on things like that
16:45 gryphon joined #mojo
16:45 tudorconstantin joined #mojo
16:46 jberger ok, well I guess we should let the channel cool down a bit
17:04 sri it's been a shitty day and the p5p thing pushed me over the edge
17:05 * jberger hands sri a beer
17:05 jberger just think, mojoconf is coming
17:05 jberger purl: mojoconf!
17:05 jberger purl?
17:05 jberger oh NO!
17:06 jberger now it really is a shitty day
17:06 * jberger weeps
17:06 batman sri <3
17:07 Adura Something unfortunate was decided about signatures?
17:07 jberger Adura: just indecision basically
17:07 sri it's the apathy gets to me
17:07 sri *+that
17:08 jberger should the implemenation be very specific and fast, less specific and less fast but more reusable, or "the way it is" which is basically "rewrite the code to something you could have written by hand"
17:09 jberger I don't think anyone really wants the latter, but it was all that could get through on the first round of bikeshedding
17:09 sri certain p5p people like DaveM could be doing great things for perl, if only someone cared enough to steer them in the right direction
17:09 sri instead there's no clear rules and ambiguity everywhere leading to endless bikeshedding
17:10 jberger it does seem like rjbs could make a ruling OP_SIGNATURE is either ok or too specific and at least get the rest of the team on the same page
17:10 sri rjbs needs to decide if ops are public api or not
17:10 jberger true, but it is a big question
17:11 sri but yea, it's much easier to call me a rambling maniac ("reini") and shrug it off
17:17 riche someone who knows what they're doing and actually has the capability and capacity to mentor this DaveM guy should help
17:17 sri jberger: actually no, that's not the case, if ops are an implementation detail it doesn't matter how specific they are
17:17 gtodd left #mojo
17:17 riche it sounds like someone said to this DaveM guy 'i like the concept, but your approach sucks'
17:19 sri one side wants a faster perl and considers ops an implementation detail, the other side wants a more hackable perl and considers ops to be part of the public api
17:19 riche it reminds me of those horrible movies during the 50's about the peak of the roman empire and arguments between senators
17:21 btyler riche: DaveM is probably the one who would be doing any mentoring. he's one of the more active and prolific current contributors to the perl core
17:22 btyler it would be a much simpler situation, politically, if this were a new committer with some flaws in a patch
17:24 riche and now he lost his chance to play that card
17:25 riche that could be a new trend: posting patches to core under fake aliases
17:26 btyler I tend to agree that this would be a good moment for a pumpking to decide on whether ops are part of the public API, but I don't envy having to make that call and then defend it
17:27 btyler risk of alienating talented people either way
17:28 sri just saying ops are public api wouldn't work anyway
17:29 sri ops have always been in flux, and you have different trees before and after peephole optimization
17:31 btyler good point, I didn't appreciate the additional distinction wrt the different phases
17:31 sri of course, perl always allowed changing of the public api in major releases
17:36 jb360 joined #mojo
17:40 sri anyway, maybe better if i take a break from irc for a bit
17:46 disputin joined #mojo
17:54 jberger I can say one thing, you certainly started some debate
17:54 jberger I'm not sure I would say that it is a tactic that should be used regularly :-)
17:55 mattastrophe joined #mojo
17:58 sh4 joined #mojo
17:58 juikuen joined #mojo
18:06 marmez left #mojo
18:25 frogaincia joined #mojo
18:46 mattastrophe joined #mojo
18:47 harryl joined #mojo
19:03 dod joined #mojo
19:10 trone joined #mojo
19:34 mattastrophe joined #mojo
19:36 mattastrophe joined #mojo
19:38 disputin joined #mojo
19:40 mattastrophe joined #mojo
19:40 Flying_Squirrel joined #mojo
19:42 meshl joined #mojo
19:47 marmez joined #mojo
19:49 disputin joined #mojo
19:54 Trelane joined #mojo
20:17 foristh joined #mojo
20:18 foristh Hi. I developed some site with partial rendering ($self->render(partial => 1)) with mojolicious <5 and I see that in version >5 this no longer works. Is there any alternative for this feature?
20:19 jberger foristh: its now its own method, ->render_to_string
20:20 foristh oh, this one. thanks!
20:49 juikuen joined #mojo
21:02 meshl joined #mojo
21:14 mattastrophe joined #mojo
21:18 cpan_mojo SVG-Fill 0.05 by Jens Gassmann - http://metacpan.org/release/JEGADE/SVG-Fill-0.05 (depends on Mojolicious)
21:38 ispyhumanfly joined #mojo
21:50 dvinciguerra joined #mojo
21:51 quibbit joined #mojo
21:57 Flying_Squirrel joined #mojo
22:34 mishantil joined #mojo
22:50 melo joined #mojo
23:02 zykz joined #mojo
23:03 zykz https://github.com/kraih/mojo/issues/756 is bug?
23:06 cpan_mojo SVG-Fill 0.06 by Jens Gassmann - http://metacpan.org/release/JEGADE/SVG-Fill-0.06 (depends on Mojolicious)
23:08 riche zykz: friday night ... i think usual players are not around
23:10 bpmedley wow, didn’t wait very long for an answer
23:10 riche lol
23:10 riche well it's been pretty quiet since the 'episode' earlier
23:11 bpmedley It’s the lounging around after some BDSM
23:11 riche speak for yourself
23:11 bpmedley :)
23:31 bpmedley For posterify: http://pastie.org/10006233 <— Is that correct usage?
23:32 jberger bpmedley: why build a controller in there?
23:34 bpmedley https://github.com/kraih/mojo/blob/master/lib/Mojolicious.pm#L58
23:35 bpmedley From the original, the app->dumper calls monkey_patch that invokes AUTOLOAD for “dumper”.
23:35 bpmedley $tx ||= $self->build_tx; <— Calls the after_hook_tx hook..
23:36 bpmedley Did I peruse the code correctly?
23:37 jberger bpmedley: OOOHHHH, you are saying your works
23:37 bpmedley It seems to.
23:37 jberger yes I think that it should (haven't tested)
23:37 jberger yes, please reply to the OP
23:38 bpmedley I think they left.. :)
23:38 jberger no, on here: https://github.com/kraih/mojo/issues/756
23:38 bpmedley Will do
23:38 jberger thanks
23:38 jberger I'll close it once you have
23:42 bpmedley Updated
23:44 howitdo joined #mojo
23:47 jberger bpmedley++

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