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

IRC log for #mojo, 2014-06-16

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

All times shown according to UTC.

Time Nick Message
00:03 maze joined #mojo
00:19 jberger Thanks stephan48
00:21 mst sri: and I should probably write a DBIC plugin, just for the look of it :)
00:21 sri :D
00:27 maze joined #mojo
00:29 ccm joined #mojo
00:57 btyler joined #mojo
01:02 laouji joined #mojo
01:12 wingfold joined #mojo
01:25 klapperl joined #mojo
01:31 wingfold_ joined #mojo
01:34 basic6_ joined #mojo
02:08 jberger mst++
02:08 mst no guarantees I actually will
02:08 mst but it might be less annoying than helping people who've tried to do it themselves
02:08 jberger nope, there it is in writing!! :-P
02:11 jberger haha, yeah, I can imagine
02:13 wingfold joined #mojo
02:40 Zx3 joined #mojo
02:42 zivester joined #mojo
02:43 * sri wonders if it would be nice to have a way to completely reset an ioloop
02:44 sri like Mojo::IOLoop->reset
02:44 sri removes all acceptors, connections, timers, and low level event watchers
02:45 jberger sri: that would be really useful for ForkCall obviously
02:45 sri jberger: i think you asked for that before
02:45 sri heh :)
02:45 sri i just stumbled over it with minion
02:45 jberger I don't know so much if I asked for it, as much as I wondered if it was possible
02:45 jberger if it is possible, it would be seriously useful
02:45 sri i just wrote it, and was wondering if i should commit it
02:46 sri (gotta add more tests too...)
02:46 jberger got a patch handy? (not that I'm really qualified to review it)
02:55 sri jberger: sure http://pastie.org/9293614
02:55 sri it's rather trivial
02:59 sri just has to happen on both layers
03:00 crab patch looks fine
03:02 sri hahaha, the reactor test case already exists actually
03:08 tempire What's use case for resetting an IOLoop?
03:08 Godsey joined #mojo
03:12 jberger sri: interesting, I guess it would make sense that it would have to happen both places. Seems obvious now that you say it
03:13 jberger tempire: when you fork, the child process gets all of the ioloop stuff too, which can be scary since there isn't full communication between the processes
03:13 tempire oh forks
03:14 jberger https://metacpan.org/pod/release/JBERGER/Mojo-IOLoop-ForkCall-0.08/README.pod#WARNINGS
03:14 jberger I could remove that warning
03:14 good_news_everyon joined #mojo
03:14 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/8vVOzA
03:14 good_news_everyon mojo/master 80c5c6d Sebastian Riedel: added reset methods to a few classes
03:14 good_news_everyon left #mojo
03:15 jberger sri: no change to the EV Reactor?
03:15 sri not necessary
03:16 sri as you can see, it's also tested ;)
03:16 jberger I did see that, that's why I'm confused :-P
03:17 Godsey I'm new and stumbling my way around learning mojolicious so please forgive me for my super novice question.
03:17 Godsey Is there a better way to manage session data with MongoDB using Mango vs MojoX-Session-Store-MongoDB?
03:18 jberger Godsey: how much data do you need in the session?
03:19 jberger mostly we just store what we need in the session directly
03:19 tempire jberger: we do?
03:19 tempire I only store the session id
03:19 tempire *application session id
03:19 jberger correction, I do, I thought most of us did, but maybe I"m wrong
03:20 Godsey well maybe I'm barking up the wrong tree.  I want my app to run on multiple ec2 instances at same time and be able to float between them
03:20 Godsey that's why I was thinking about using the MongoDB backend for session/flash messages
03:20 jberger I guess that in an app proper you would do that, but in that case you already have essentially the equivalent of a session-store plugin
03:20 tempire I have an auth'd id, and that does lookups in some database when the relevant information is needed, keyed by the id
03:21 * tempire doesn't use a special module
03:21 jberger tempire, on that we agree, there really doesn't need to be a special module
03:21 Godsey how do flash messages work betwen server instances?
03:21 Godsey oops between
03:21 jberger flash messages are stored in the session
03:22 crab flash messages are stored in the session
03:22 jberger which is a cookie
03:22 tempire It sounds like you're using flash for something unintended
03:22 Godsey I didn't realize it was a cookie
03:22 crab no cookie for you?
03:22 jberger the cookie is signed to prevent tampering
03:22 jberger but of course it then comes with a size limitation
03:22 Godsey flash as "Thank you for registering" type messages
03:22 jberger yep, that works perfectly
03:23 tempire oh, you're load balancing.
03:23 Godsey yes load balancing
03:23 good_news_everyon joined #mojo
03:23 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/R9LI0g
03:23 good_news_everyon mojo/master 64c1e1a Sebastian Riedel: better timer tests
03:23 good_news_everyon left #mojo
03:23 jberger (self-promotion alert: see also Mojolicious::Plugin::Humane)
03:23 jberger if not for the actual module, at least for the concepts
03:24 jberger (but the module is good too :-) )
03:25 jberger sri: so is this all I would need to do then: http://pastie.org/9293641
03:27 sri jberger: yes
03:27 jberger oh man, that would make two-way communication sooo easy!
03:27 sri forgot to stop the ioloop though... gotta add that
03:28 jberger the child does
03:28 jberger or both
03:29 Godsey thank you.  I was over thinking it and missed that the data is just stored in a cookie
03:30 sri actually... i guess without anything to watch all reactors stop automatically :)
03:30 crab sri: somewhere in the documentation i saw that "some" reactors stop if they have nothing to do
03:30 crab i wondered about that
03:30 crab (as in, which ones don't)
03:32 jberger sri: oh wait, I see the problem
03:32 jberger if I have to call this in the parent too, then I would lose anything else going on in the parent's loop
03:33 jberger making this much more useful for say Minion than say ForkCall
03:33 jberger in ForkCall I have no guarantee that I am the only thing being done in the parent's loop
03:33 crab why do you have to call it in the parent?
03:34 sri yea, why?
03:35 tempire yeah!
03:36 crab we demand answers!
03:36 good_news_everyon joined #mojo
03:36 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/AhQICw
03:36 good_news_everyon mojo/master 8a587ae Sebastian Riedel: make sure reset stops the event loop
03:36 good_news_everyon left #mojo
03:36 sri WHAT DO WE WANT?!
03:36 purl BOOOOOOOOBIES
03:36 sri oi!
03:36 nicomen WHEN DO WE WANT IT?!
03:36 purl BRAAAIIIINS
03:36 nicomen purl: now you are just making a fool of yourself
03:36 purl ...but now purl is accusing me of taking over the channel?...
03:41 sri this is what support in minion looks like https://github.com/kraih/minion/commit/c5a1c2e5e535af2f1774603e0b8f87605672d6a0
03:41 Godsey I had not come across humane.js before  double thanks for your module and pointing me in this direction.
03:41 sri i guess we have our first bigger 5.x feature \o/
03:46 crab what, resetting the ioloop?
03:49 Adura It's odd how Cygwin's perl is fine with the fork-in' and the lock-in'.
03:49 crab what does that mean?
03:49 purl You're a nut! You're crazy in the coconut!
03:50 Adura Minion's file backend tests on Strawberry failing, but not Cygwin.
03:51 crab ah
03:56 jberger sri: ok reading back I saw you said "just has to happen on both layers" which I took to mean parent and child
03:56 jberger which I was able to talk myself into believing, you know, shared resources etc
03:56 jberger but now that I look back, you meant loop and reactor
03:56 sri i meant ioloop and reactor
03:57 jberger right
03:57 jberger so then all i need to do is to reset in the child
03:58 jberger good, so then it is what I need after all
03:58 * jberger wipes brow
03:58 jberger perhaps I have done too much driving this weekend, my brain has left me
03:59 jberger time to wind down for bed with an episode of GoT
03:59 jberger night all
03:59 jberger sri++
04:00 rem_lex| joined #mojo
04:05 purl joined #mojo
04:18 good_news_everyon joined #mojo
04:18 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/Dk1dDw
04:18 good_news_everyon mojo/master b7969aa Sebastian Riedel: reuse more code
04:18 good_news_everyon left #mojo
04:46 ryanc joined #mojo
05:14 chankey joined #mojo
05:23 laouji joined #mojo
05:25 laouji_ joined #mojo
05:53 disputin joined #mojo
06:01 Lee joined #mojo
06:01 Eke- joined #mojo
06:21 igorsutton joined #mojo
06:23 dpetrov_ joined #mojo
06:23 ryanc joined #mojo
06:46 dod joined #mojo
06:50 basiliscos joined #mojo
06:57 laouji joined #mojo
07:05 eseyman joined #mojo
07:16 Vandal joined #mojo
07:21 arthas joined #mojo
07:22 trone joined #mojo
07:31 denis_boyun joined #mojo
07:38 jberger joined #mojo
07:40 Eke|| joined #mojo
07:44 chankey joined #mojo
07:58 laouji_ joined #mojo
08:00 chankeypathak joined #mojo
08:03 chankey joined #mojo
08:12 babl joined #mojo
08:31 edestler joined #mojo
08:35 edestler joined #mojo
08:55 ua joined #mojo
09:03 dod joined #mojo
09:08 chankeypathak joined #mojo
09:23 oetiker joined #mojo
09:25 crab hi oetiker.
09:26 dexteruk joined #mojo
09:30 fhelmber_ joined #mojo
09:49 rem_lex| joined #mojo
10:07 laouji joined #mojo
10:10 chankey joined #mojo
10:11 laouji joined #mojo
10:31 d4rkie joined #mojo
10:33 alnewkirk joined #mojo
10:33 mire joined #mojo
10:48 d4rkie joined #mojo
10:51 meshl joined #mojo
10:53 nic sri,crab,tempire,marcus,jberger: Feature request at https://gist.github.com/niczero/12e8e07bc0b231324716
10:54 nic This is for supporting dynamic layouts in combination with extending templates
10:54 nic use cases, feature branch (including tests), example code
10:58 nic I vaguely remember abandoning 'extends' last year cos I didn't have time to investigate why it broke my app
10:58 nic I'm guessing I'm not the only one, so this FR would improve take-up and make template re-use more awesome
11:04 * crab reads
11:09 crab nic: it's awesome that you've presented it in this way
11:12 nic it does lack colour though
11:12 nic and animated gifs
11:12 nic crab: thanks
11:12 crab i can see black. and white.
11:13 basiliscos joined #mojo
11:17 batman nic: why didn't you make an issue?
11:18 * batman is just curious
11:20 nic If an issue is wanted, it'll be easy to create one
11:21 nic Initially I'm interested in whether it'll be integrated 'soon', 'v6', never
11:21 batman it's not easy to ignore an issue, the gist can however be "lost"
11:21 nic I'm guessing an issue is only useful if it'll be parked till v6
11:21 * batman need to look into "extends"
11:21 batman thanks
11:22 batman i would say an issue is useful if you want it to be closed :)
11:22 * batman lays off
11:24 nic tbh I couldn't remember how FRs are handled; I guess it should have an issue, but I'd like to check if sri says that's the next step
11:26 wingfold joined #mojo
11:28 nic http://mojolicio.us/perldoc/Mojolicious::Guides::Contributing#Feature_requests
11:28 nic phew! No issue should be raised
11:33 batman nic++
11:49 igorsutton joined #mojo
11:53 rem_lex|pivo joined #mojo
12:25 edestler joined #mojo
12:40 crab i do not use template inheritance much
12:41 crab so maybe i'm missing something when i say the patch looks pretty good to me
12:43 crab (but i have to admit that i'm strongly biased in favour of accepting simply because of the presentation :-)
13:08 mire joined #mojo
13:10 neilhwatson joined #mojo
13:24 zivester joined #mojo
13:25 dexteruk joined #mojo
13:36 bluescreen joined #mojo
13:52 sri nic: i'm -1 since i don't understand any of the use cases
13:53 sri crab: if you do understand the use cases please explain them
13:54 sri to make matters worse, the original inheritance recipe in the rendering guide makes sense to me, the changed one from your patch doesn't
13:58 btyler joined #mojo
14:01 sri as far as i can see you don't actually explain the value of using them together, just "now you can"
14:02 sri you're ignoring the fact that layout and extends are currently the exact same thing, with the small difference that layouts have a "layouts/" prefix
14:05 sri to sum it up, inheritance is already a complicated topic, if you patch makes the guide section more complicated i'm -1
14:05 oetiker joined #mojo
14:05 sri *+r
14:07 amarnus joined #mojo
14:07 oetiker crab: hi
14:10 sh4 joined #mojo
14:18 hesperaux joined #mojo
14:22 hesperaux_ joined #mojo
14:25 amarnus left #mojo
14:26 nic sri: under the covers, templates and layouts are exactly the same thing, but conceptually they are different
14:27 nic If layouts aren't thought of differently, why put them in a separate location and use a separate name?
14:28 nic Currently the guide is missing the section that explains to the user why they can't use the two features together
14:29 nic Currently there is no warning that if the reader tries to use everything in the guide they will enter a world of mystery and pain :D
14:30 sri "Inheritance takes the layout concept above one step further..." is pretty close, making that a little more specific is *a lot* easier than your test breaking patch
14:32 nic The tests that broke were testing undocumented behaviour, and the fix was trivial
14:33 sri not the point really, it's all about good use cases
14:33 nic What's wrong with Use Case1?
14:34 sri that's not a use case at all, you're just saying "now you can use them together"
14:35 sri i want something our users would like to do that gets easier conceptually after the patch
14:35 sri your new guide example is actually more complicated
14:35 sri to me it makes no sense at all
14:36 sri there's two layouts, one doesn't appear to even be used for anything
14:37 nic I'm surprised you don't understand the two use cases, but I'm afraid I don't have more time to devote to this
14:37 sri a "now you can do it" is not of muh value to me without a "and this is why you would want to"
14:37 nic I gave it a lot of what should have been family time at the weekend
14:38 nic It'll be a shame if nobody else gets to use this awesomeness
14:38 sri i'm indifferent on the feature atm. and strongly -1 on the changes to the rendering guide
14:40 sri if it's that awesome you really should be able to explain why
14:41 maze joined #mojo
14:41 sri crab, marcus, jberger, tempire: if you feel differently about it please chime in
14:43 nic "less horrible than the alternatives" is what my marketing team are going with
14:43 sri that doesn't tell me anything, i don't understand the feature
14:44 dotan nic: you talk about your application in the feature request, can you create some minimal example based on your own code?
14:45 rem_lex joined #mojo
14:45 sri i usually explain new features by comparing how something used to be, and how much smaller the code has become after the patch
14:46 nic sri: You understand the features "default(layout => 'L')" and "->render('B')"
14:47 nic Why should that break when B is refactored to "% extends 'A'" ?
14:48 sri because extends is just another form of layout
14:48 sri in both cases you start with an html skeleton and fill in the blanks
14:49 kyshtynbai Guys, is it possible for a web application to tell browser not to cache it's content?
14:50 nic that's way too complicated -- your code broke cos you used two features that happen to be implemented the same under the covers
14:50 rem_lex joined #mojo
14:51 rem_lex joined #mojo
14:55 genio kyshtynbai: the Cache-Control header allows   Public, Private, No-Cache, No-Store
14:56 good_news_everyon joined #mojo
14:56 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/XmJLFA
14:56 good_news_everyon mojo/master 59061a6 Sebastian Riedel: added another layer to the template inheritance example
14:56 good_news_everyon left #mojo
14:56 sri maybe that makes the problem more clear
14:56 sri a default template would be applied on the first layer, and totally fuck up the inheritance chain
14:56 sri umm
14:56 sri default layout
14:56 purl default layout is ridiculously ugly
14:57 sri nic: thinking some more about it, now i'm -1 on the feature as a whole
14:57 sri it just doesn't work
14:59 nic Those three levels of inheritance -- the same as what the guys here are using very happily as of this morning
15:01 kyshtynbai genio: Thanks.
15:01 nic Trialling the feature branch, been a long time since I've seen them so enthusiastic; it's revolutionising our templates
15:01 nic and drastically reducing our overall volume of templates
15:02 sri ok, i give up, maybe try explaining it to the other core devs first
15:02 dotan kyshtynbai: http://stackoverflow.com/questions/49547/making-sure-a-web-page-is-not-cached-across-all-browsers
15:02 sri looks like i just don't get it
15:02 sri something is missing from your explanations
15:04 kyshtynbai dotan: thanks!
15:04 dotan kyshtynbai: and that question's from the time of firefox 3.0, IE 8... who knows if it works in modern browsers. :(
15:05 kyshtynbai dotan: why do we at all need caching ng 21 century...
15:06 kyshtynbai *in
15:07 sri i suspect crab didn't understand the feature either though
15:07 suy kyshtynbai: I'd say because latency is still quite bound to the speed of light, and that hasn't changed in some centuries :)
15:07 kyshtynbai I cant' figure out how to turn off caching in firefox without dancing with plugins bu the way, the settings menu is weird in the latest version of ff.
15:10 kyshtynbai suy: well, with all this 1mbit channels in every microwave owen...
15:10 kyshtynbai s/1mbit/100mbit
15:12 dotan I hear that mobile ushers in a dark new age of sucky connectivity
15:12 suy Dunno, I've always read that latency hasn't improved much in those years, though. Our frontend developer has lots of issues with Chrome's caching. He has to clear it completely sometimes, because it refuses to fetch new versions of his templates/JS files.
15:18 nic sri: Have a think about why layouts are a good thing.  'Separation of concerns' is probably in there.  Now, all of those reasons also apply even if some of your templates are 'extending' templates
15:21 sri some kind of information is still missing, because all i see is extends and layouts interfering with each other
15:24 nic You don't think the user deserves a warning that combining them at the moment leads to strange breakage?
15:25 rem_lex joined #mojo
15:28 kyshtynbai render method sends headers, does it?
15:35 beyondcreed joined #mojo
15:44 sri nic: the point is you can't combine them, i think that is the part of your expanation that confuses me
15:44 sri looking at the code, what you really want is to *separate* them
15:44 sri your code makes the renderer not consume layout and extends together
15:45 sri instead the layout always comes after extends
15:46 sri quoting from your feature request "Benefits: Now two important rendering features can be combined."
15:46 sri when actually they can now be used separately
15:46 dotan kyshtynbai: not quite sure it does, but you can set the headers explicitly on the response object ($self->res on your controller).
15:46 dotan http://mojolicio.us/perldoc/Mojolicious/Controller#res
15:48 nic the two features can now be used in combination -- ie you don't have to abandon one if you use the other
15:48 nic the user doesn't care how it's implemented
15:50 sri nic: not sure it works the way you imagine, considering how much trouble i had understanding
15:50 sri (for our users)
15:51 sri i really want to see you explain it to marcus, jberger, tempire or crab
15:52 sri i had to read the code to get it
15:52 nic consider that nobody has explained the current behaviour
15:53 sri but that's the current behavior
15:53 nic I had to read the code to understand why following the guide leads to breakage
15:53 sri you want to make a change, so the burden of proof is on you
15:54 nic and my goodness what a burden
15:54 sri actually, i'd like to see anyone here understand it
15:55 nic The guide makes a good case for separating out html scaffolding to layouts
15:55 sri doesn't even have to be core devs
15:55 sri well, your new guide example is terrible
15:55 nic but then both the guide and the unit tests switch to using templates as layouts whenever 'extends' is in play
15:56 sri that will always get a -1 from me
15:56 nic I'm guessing I'm not the only reader who found that confusing
15:56 sri i've not seen any complaints
15:56 nic Oh, have I misunderstood what layouts are for?  But why didn't he use a layout here?
15:56 sri nobody complained yet
15:57 sri maybe nobody bothered to learn it, or it works for everyone who tried, we can't know
15:57 nic It's only after a weekend of debugging that I understand it's cos you'd have to use hard-coded layouts
15:58 sri i want to hear from others
15:58 nic hence me calling the FR "dynamic layout selection"
15:59 sri get someone else here to support your patch
16:00 nic tempire,marcus,jberger: When you adopt this patch, even your coffee tastes better!
16:00 sri anyone will do, doesn't have to be a core dev
16:00 sri i just want to see that they get your version better than what we have now
16:01 sri you can also open a pull request or post to the list if you think it's easier to reach people that way
16:01 sri (of course the pull request would ultimately get rejected for the documentation thing)
16:02 nic I haven't persuaded you so far => I will never persuade you
16:03 sri honestly, i'm indifferent on how it should work, i'm -1 on the patch for the documentation thing
16:03 dotan I understood layouts easily enough, found "extends" confusing and potentially scary and didn't dig further. I'm guessing most users did the same.
16:03 sri and since i'm indifferent, i want to see evidence that there is a broader interest in changing what we have now
16:03 nic tempire,marcus,jberger might see my arguments and the awesomeness of template re-use, but if they have sense they won't read all of that log
16:04 nic life is too short
16:04 beyondcreed_ joined #mojo
16:04 sri you can always make a pull request with a tldr description
16:06 nic My director is now saying that the patch could save us enough development time that we could maintain only patched versions of Mojolicious
16:06 nic I just think it would be a shame to keep that functionality private
16:07 sri that kind of statement only makes me more sceptical, i want to hear from people directly
16:14 denis_boyun joined #mojo
16:15 disputin joined #mojo
16:26 denis_boyun_ joined #mojo
16:38 nic Next week I'm supposed to do some benchmarking
16:41 nic I had been expecting that template inheritance might make the app slower, but I guess there's a chance the smaller templates will make it faster; it'll be interesting
16:42 disputin joined #mojo
16:42 basiliscos joined #mojo
16:47 sri dotan: does the pull request make extends any easier for you to understand?
16:47 sri s/pull request/feature request/
16:49 sri here we have the inheritance example edited to use layout and extends... if anything it got worse imo http://pastie.org/9295599
16:51 meshl joined #mojo
16:52 sri nic: make this section easier to understand for the people here http://pastie.org/9295606
16:56 nic sri: My patch has no impact on that pastie
16:57 nic And as for the guide example, my patch doesn't require any change (though I personally don't find the current text helpful)
16:57 sri nic: of course it does https://github.com/niczero/mojo/compare/dynamic_layout_selection#diff-65cf575515a0ad467194433b39cd30d3L641
16:57 sri you even add a useless layout
16:58 dotan sri: I think you're looking at it from the inside while nic is looking at it from the outside. For you, layout is a synonym for extend. For nic, those are two different, complementary ways to modularize the design of his templates.
16:58 nic sri: http://pastie.org/9295606 has no layouts, so my patch doesn't touch it
16:59 nic My patch is only of interest to people who want to use both layout and extends
16:59 sri nic: THIS IS YOUR VERSION https://github.com/niczero/mojo/blob/f01e4171a9dbccaf05db8d6bc1927fbf6c45decc/lib/Mojolicious/Guides/Rendering.pod#template-inheritance
16:59 nic and since all it does is remove a nasty surprise, it doesn't strictly need any documentation change
17:00 sri don't tell me there are no changes
17:00 * sri is starting to get a little upset
17:01 nic Now, for the unit tests, I really would like your input, cos I'm not confident mine are as good as they could be
17:01 nic One of the reasons it's a feature branch and not a pull request is cos this is not an all-or-nothing situation
17:01 nic I was hoping it would be the basis of some fruitful discussion
17:02 nic like suggestions how to make the unit tests better, or tests that are missing
17:02 sri i don't care about the tests, my concern is the documentation
17:02 dotan layout is for wrapping all you site's pages in standard web site chrome - home button, nav bar, footer, etc, while extends is for saying something like "these pages all use two column layout and have the profile navigation widget". You don't want the 2nd page you create that extends the 2-col layout template to suddenly lose all the site's chrome.
17:02 sri like i said before, the patch has a strong -1 vote from me for that alone
17:03 nic The only impact the 2-line patch has on documentation is: now nobody needs to add warning/explanation of why things break
17:05 sri the example is terrible, those layouts add nothing but distractions
17:05 nic sri: If you don't like the change to Rendering.pod, please ignore it, it's not required
17:06 sri i do want change, good change
17:06 sri actually, i *expect* change if the behavior changes
17:07 sri if the documentation doesn't reflect the improvement, the whole feature is not worth it imo
17:07 nic The behaviour causing breakage wasn't documented
17:07 sri more reason to document the new behavior
17:08 sri if it's better the documentation should get better, that's how i see it
17:08 nic Ignore the feature and add a warning in Rendering.pod about the breakage that a 2-line patch could have avoided
17:08 sri i don't see this discussion going anywhere
17:09 * sri sticks with "indifferent on the feature", and -1 on the documentation patch
17:09 sri if you want the feature to happen go through marcus, jberger, tempire and/or crab
17:10 nic If they have any sense, they've already added me to their 'ignore' list :)
17:11 nic I need to leave now, but maybe tomorrow you can give me some constructive pointers about what you want added to Rendering.pod
17:12 sri i don't know what i want, for me the current documentation works... you have to make it better and prove that it is better (by getting support from others around here)
17:14 nic One of my points was, the documentation is fine if the behaviour matches; my patch achieves that
17:14 nic user reads guide... doesn't get stung
17:14 sri does anyone here support that opinion?
17:14 nic everyone goes home for tea happy
17:15 denis_boyun joined #mojo
17:16 sri btw. pushing me to add a note o the current documentation is really not a good idea... i might do it and release a new version later today... killing any chance your patch had for getting in before 6.0
17:17 nic that would harm Mojolicious, which would make us all sad
17:25 sri i suspect the overly dramatic tone this discussion took every now and then might have sealed the fate of the feature request
17:46 denis_boyun joined #mojo
17:56 basic6 joined #mojo
18:01 mport joined #mojo
18:23 * tempire sprinkles glitter on everyone
18:28 bodgix joined #mojo
18:32 kyshtynbai joined #mojo
18:34 * marty_ twiggles
18:43 denis_boyun joined #mojo
18:46 * sri coughs glitter
18:48 genio you should have that looked at
18:53 sri at least i'm not spewing rainbows yet
18:58 genio was it multi-color glitter?
19:01 dod joined #mojo
19:03 sri yea :S
19:04 * sri wonders if miyagawa is right about google making dart the next android language... the number of dart sessions at google i/o is suspicious
19:05 denis_boyun_ joined #mojo
19:05 basiliscos joined #mojo
19:06 dod joined #mojo
19:07 Eke- joined #mojo
19:14 marty joined #mojo
19:25 marty joined #mojo
19:28 mad_hatter joined #mojo
19:32 kyshtynbai I've done my gallery with authentication at last. Thanks everyone for advices, but I'm afraid new questions are near.
19:34 Ptolemarch joined #mojo
19:37 denis_boyun joined #mojo
19:40 bodgix hi there
19:43 bodgix I'm wondering if it makes sense to use Mango in asynchronous mode from an already asynchornous request handler. How can I return a value back to the request handler if I do an asynchronous find_one
19:54 kyshtynbai If a file is rendered via render_static and is located outside public directory, is it possible to put it inside html tag "environet"?
19:57 meshl joined #mojo
19:58 marty_ joined #mojo
20:02 * sri wonders what happened to jbergers patch for $self/$c
20:06 mad_hatter I'm having trouble understanding what the difference between %, %=, %== and similarly <%, <%=, <%==
20:09 lb % is perl, %= is print escaped, %= is print
20:09 preaction % has to be the first non-whitespace character, <% can be anywhere.
20:09 lb sorry, %== is print
20:10 * sri is still so happy we made the change to allow whitepsace before % a few years back :)
20:10 lb less succinctly explained in the section Embedded Perl in Mojolicious::Guides::Rendering
20:11 lb oh wait, i should be releasing
20:30 marty joined #mojo
20:37 marty joined #mojo
20:50 lb has not released anything to cpan in 7 years, fucks it up on first try
20:57 maze joined #mojo
20:58 meshl joined #mojo
21:06 torbjorn who doesnt
21:09 lb it was a mild fuck up, and i blame github fully
21:12 basic6_ joined #mojo
21:25 marty joined #mojo
21:25 sri i guess we have our 6.0 unicode character
21:25 sri REVERSED HAND WITH MIDDLE FINGER EXTENDED
21:26 sri 1F595
21:26 marty joined #mojo
21:27 sri i do like this one too though
21:27 sri THUMBS UP SIGN
21:27 sri 1F44D
21:32 lb what about INDEX FINGERS EXTENDED BUT REALLY MEANS MIDDLE FINGER
21:32 lb http://i.4cdn.org/sp/1402938165299.jpg
21:32 tempire no!
21:32 tempire ALIEN MONSTER (U+1F47E)
21:32 tempire purple space invader!
21:34 torbjorn thats awesome, how do i get fonts for those?
21:39 lb <+cpan_mojo> Mojolicious-Plugin-PetalTinyRenderer - https://metacpan.org/release/Mojolicious-Plugin-PetalTinyRenderer
21:39 lb whee
21:40 lb stupid cpan_mojo
21:46 basiliscos joined #mojo
21:53 meshl joined #mojo
21:54 good_news_everyon joined #mojo
21:54 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/QhLNiw
21:54 good_news_everyon mojo/master 27d7831 Sebastian Riedel: mention stash values
21:54 good_news_everyon left #mojo
22:05 basiliscos joined #mojo
22:06 bobkare joined #mojo
22:08 KindOne joined #mojo
22:09 meshl joined #mojo
22:25 marty joined #mojo
23:07 d4rkie joined #mojo
23:13 good_news_everyon joined #mojo
23:13 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/IQLy5Q
23:13 good_news_everyon mojo/master a9ab209 Sebastian Riedel: do not end error messages with a dot
23:13 good_news_everyon left #mojo
23:16 alnewkirk joined #mojo
23:43 good_news_everyon joined #mojo
23:43 good_news_everyon [mojo] kraih tagged v5.08 at 67ef024: http://git.io/lisd3Q
23:43 good_news_everyon left #mojo
23:44 good_news_everyon joined #mojo
23:44 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/3eL0uA
23:44 good_news_everyon mojo/master d40456b Sebastian Riedel: bump version
23:44 good_news_everyon left #mojo
23:46 rem_lex|pivo joined #mojo
23:52 mst sri: I'm sort of tempted to change /topic to "hear the roar of a speeding ferret" now :)
23:53 sri ;p
23:59 jberger lb: did cpan_mojo miss one?

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