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

IRC log for #mojo, 2015-08-22

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

All times shown according to UTC.

Time Nick Message
00:20 meshl joined #mojo
00:26 jberger mst: without consensus on psgi extensions you could just jam async on top of it in any way that you happen to think of
00:27 jberger Hmmmmm sorry that was kinda snarky
00:28 * ZoffixWork didn't feel that
00:28 jberger (It is if you know what I'm talking about)
00:28 ZoffixWork :)
00:31 howitdo joined #mojo
00:44 jabberwok so if you switch from plain CSS to Less, then you have one more file, but once you use AssetPack, you have one less file, but it's not the Less file.
00:45 ZoffixWork lol jabberwok++
02:08 sri odd how the $c->log thing went nowhere, some folks seemed to be passionate about it
02:24 noganex_ joined #mojo
03:06 hesco joined #mojo
03:08 berov joined #mojo
03:10 SmokeMachine Hi again! Is there any "automatic" way to validate ->res->json? (like json-schema)
03:16 SmokeMachine the c->validator doesn't apply to json, doesn't it?
03:17 SmokeMachine *does it?
03:18 mst SmokeMachine: Mojolicious::Pluggin::Swagger2 ?
03:21 SmokeMachine swagger is to much for what I want... I just want validation...
03:22 preaction Test::Deep::NoTest then perhaps?
03:22 mst ok, so, JSON::Schema and Rx both exist
03:22 mst so there's two json schema systems on cpan
03:22 mst I'm not sure what you're missing
03:23 SmokeMachine mst: I am not sure either...
03:24 SmokeMachine I was thinking in something like the Mojolicious::Plugin::Validator but for son...
03:24 SmokeMachine *json
03:24 mst SmokeMachine: ok, so, I have no further opinions on solutions until you've had more of a think about what the actual problem is :)
03:26 SmokeMachine I think I'd just like to use Mojolicious::Plugin::Validator to validate my entering json...
03:26 mst there's a vast amount of complexity hidden behind that 'just'
03:27 SmokeMachine I suppose...
03:27 mst sri and I have both spent ten years trying to figure out a good API for that; if we had one, it'd be on CPAN by now
03:27 mst have you looked at JSON::Schema and Rx ?
03:30 SmokeMachine I already had seen JSON::Schema... and this is my problem with that: "This module offers good support for JSON Schema as described by draft specifications circa 2012"
03:30 mst and the differences between that and subsequent publications affect you how?
03:30 sri the one i liked most so far was joi.js https://github.com/hapijs/joi
03:33 SmokeMachine mst: I already have some 2013 son-schema files... I don't know how many problems I would have...
03:34 mst SmokeMachine: try it before assuming?
03:35 mst also, I know somebody's currently working on updating it, so if you think it would be useful bug reports would likely help especially if you can produce test cases for them
03:35 SmokeMachine https://www.irccloud.com/pastebin/9pD06VVy/
03:35 mst cpan works because people try things, and report bugs :)
03:36 mst sri: that last paste is an API idea, so I have no opinion
03:37 sri and how do you validate arrays?
03:38 sri hey, if you can make it fit well into Mojolicious::Validator, i'm totally interested
03:38 SmokeMachine $validator->pointer('/my/json/pointer')->required(1)->is_array->length(3, 20);  ?
03:38 sri api is not how i would do it, but that's not really important
03:39 sri $validation->pointer->required('/foo/bar/baz')->...
03:39 SmokeMachine sri: I don't know if I can do that... I am just trying to solve my problem... :P
03:40 sri maybe there's an easy way and i'm just not seeing it
03:42 SmokeMachine easy way to use it with a array?
03:42 sri no, handling structures in general
03:43 sri the simple array case should just work
03:43 sri you just can't require an array
03:43 sri it might be a string or an array of strings
03:44 SmokeMachine does that make sense on jsons? I can understand it on query string...
03:45 sri and i guess the you're not really dealing with json
03:45 sri just perl data structures
03:45 sri s/and i guess the//
03:45 SmokeMachine sure... but that came from a son...
03:45 SmokeMachine *json
03:45 sri which doesn't really matter
03:46 sri it's validation for arbitrary hash/array structures
03:46 sri and you're correct, the main focus of Mojolicious::Validator is query strings
03:46 SmokeMachine I think that it matter for the meaning of that data? doesn't it? I mean you could do different things for query string and for son...
03:47 sri i have no idea if it fits in, that's your job to find out! ;p
03:47 SmokeMachine *json (my mac is auto-correcting me, sorry...)
03:48 mst autocotwreck is ... yeah
03:49 sri we've recently extended the scope actually with http://mojolicio.us/perldoc/Mojolicious/Validator#upload
03:50 sri it now validates a one-dimensional hash with string or array of strings/upload objects
03:50 SmokeMachine maybe the simplest way is $validator->json_schema("file.schema.json")
03:50 sri for core, json schema is definitely out of scope
03:51 sri some json pointer magic is a possibility though, if it fits in
03:51 SmokeMachine (I didn't mean that it should exists... but that would, probably, be the easiest way...)
03:53 SmokeMachine could it only accept one-dimensional pointer? it would be a big help... (it wouldn't be perfect... but could be a first step...)
03:54 davido_ joined #mojo
04:05 SmokeMachine https://www.irccloud.com/pastebin/W93KOlhR/
04:05 SmokeMachine sri: ^
04:06 SmokeMachine does that make sense?
04:09 PopeF joined #mojo
04:20 kaare joined #mojo
04:44 PopeFelix joined #mojo
04:45 sri SmokeMachine: i don't know, can it work?
06:40 sri \o\
06:40 sri /o/
06:47 bpmedley_ }o{
06:48 Grinnz tie fighter?
06:48 bpmedley_ I'm just tired.. No clue, really.. :-o
06:57 amon joined #mojo
07:09 batman ZoffixWork: I think you might be tricked by SASS_PATH
07:09 batman Zoffix ^
07:13 batman SmokeMachine: Swagger2::Validator doesn't care if it's swagger
07:13 batman You can use it to validate any schema.
07:14 * batman need to rename those modules! :(
07:14 batman Zoffix: glad you found out. Let me know if I could do something to improve the error messages.
07:43 ashimema SmokeMachine check out the Mojo Swagger stuff as batman said..
07:43 ashimema it;s bang up to date json chema validation.. batman++
07:43 * batman factor out the swagger stuff now, creating JSON::Validator
07:43 batman hope that name is something that will show up in search results
07:52 frogonia joined #mojo
08:05 dod joined #mojo
08:09 Vandal joined #mojo
08:10 dod joined #mojo
08:28 punter joined #mojo
08:28 punter left #mojo
08:29 KCL_ joined #mojo
09:08 dod joined #mojo
10:16 mattastrophe joined #mojo
10:27 melo joined #mojo
10:34 nigel joined #mojo
11:35 juikuen joined #mojo
11:36 juikuen joined #mojo
12:14 Zoffix batman, not sure how it can be improved without resulting in a confusing message for assets retrieved over HTTP... What tripped me up was that it was telling me it was "Fetching", it was using an Http handler, and the result was "Not Found", which is a 404 message, and the fact that the path had '/www/' in it made me think there was a bug that due to that 'www' the asset was a link to something.
12:14 Zoffix Older AssetPack had an even more interesting message: http://fpaste.scsys.co.uk/497441
13:14 gryphon joined #mojo
13:23 mattastrophe joined #mojo
13:23 meshl joined #mojo
13:30 * jberger yawns
14:07 odc joined #mojo
14:57 nic joined #mojo
14:59 nic crazy question..   How much work is it to have my.js.ep to render as my.js in a lite app?
15:00 nic joined #mojo
15:01 pink_mist sounds like it would be hardly any work at all
15:01 pink_mist just put it in the template directory
15:01 bpmedley_ nic: Are you trying to use a template for javascript?
15:01 nic y
15:02 bpmedley_ I agree with pink_mist.  Just letting you know you can use the DATA section, as well.
15:03 nic that's what I'm talking about
15:04 nic joined #mojo
15:04 pink_mist it's a bit hard to have a conversation with you if you quit every 2 minutes, nic
15:04 nic I guess I'll hit problems with escaping
15:04 pink_mist why?
15:04 nic HTML! =js
15:05 bpmedley_ nic: One option is base64 encoding if, if you need it.
15:05 pink_mist html makes use of <> ... js hardly makes use of those ... so HTML would have more issues with ep templates than js :P
15:05 nic anyway, giving it a spin now.
15:18 cpan_mojo Mojolicious-Plugin-Qooxdoo-0.904 by OETIKER https://metacpan.org/release/OETIKER/Mojolicious-Plugin-Qooxdoo-0.904
15:20 * mst stares at GraphQL and Falcor
15:20 mst "REST is terrible because it isn't flexible enough for client side devs, let's reinvent everything"
15:20 mst (my first thought: how about you try and design a better REST API? :)
15:21 mattastrophe joined #mojo
15:26 Gedge joined #mojo
15:26 meshl joined #mojo
15:44 ZoffixWork joined #mojo
15:44 jberger for escaping javascript there is JavaScript::Value::Escape
15:47 ZoffixWork My local store is selling Mojo packages: http://i.imgur.com/u3S4XRG.jpg
15:47 ZoffixWork ^_^
15:54 pink_mist mojo on a bottle
16:12 ajr_ joined #mojo
16:36 mattastrophe joined #mojo
16:39 * sri wonders if anyone uses savepoints with Mojo::Pg https://github.com/kraih/mojo-pg/pull/10
16:47 lluad joined #mojo
16:50 melo1 joined #mojo
16:52 PryMar56 joined #mojo
17:05 lsm joined #mojo
17:55 dod joined #mojo
17:55 batman Zoffix: Ok. Thanks.
17:56 batman Let me know if you come up with anything better. Always +1 on improvements :)
17:58 batman I'm factoring out the JSON Schema part of swagger2 now
17:59 batman Please join #swagger if you have any feedback
17:59 batman https://github.com/jhthorsen/json-validator
17:59 batman SmokeMachine: ^^
18:24 PopeFelix joined #mojo
18:28 asm35 joined #mojo
19:00 dod joined #mojo
19:12 SmokeMachine batman: thank you very much! That can solve my problem!
19:17 batman SmokeMachine: does the api make sense to you?
19:17 SmokeMachine batman: it looks great!
19:17 batman haha. thanks :)
19:18 * batman laughing because it's probably not great, but time will tell...
19:26 marcus #evil_laugh
19:26 cpan_mojo JSON-Validator-0.48 by JHTHORSEN https://metacpan.org/release/JHTHORSEN/JSON-Validator-0.48
19:34 batman SmokeMachine: ^ go crazy
19:34 batman i will probably hate myself of going directly to version 0.48 :P
19:35 bpmedley_ batman: Be bold.  :)
19:35 batman i am. me and my paulaner.
19:35 batman hehe
19:37 good_news_everyon joined #mojo
19:37 good_news_everyon [mojo] kraih tagged v6.17 at 8f596cd: http://git.io/vsV6r
19:37 good_news_everyon left #mojo
19:38 good_news_everyon joined #mojo
19:38 good_news_everyon [mojo] kraih pushed 1 new commit to master: http://git.io/vsV6h
19:38 good_news_everyon mojo/master de1b1a5 Sebastian Riedel: bump version
19:38 good_news_everyon left #mojo
19:52 SmokeMachine batman: \o/
19:55 batman i wonder if it would make sense to mimic the Mojolicious::Validator::Validation api
20:01 thowe joined #mojo
20:05 SmokeMachine do you mean the user would call some methods to create a hidden son-schema?
20:05 SmokeMachine *json
20:07 batman i have no idea what you're asking, but i can try to explain: i was wondering if it would make sense to add methods such as input, output, failed, ...
20:07 batman but i guess it doesn't. i need to make a new module if that should happen.
20:09 SmokeMachine batman: I meant the API I was discussing yesterday...
20:12 batman do you mean $validator->pointer('/my/json/pointer')->required(1)->is_array->length(3, 20);  ?
20:12 batman that's not really json-schema validation imo. it's json validation alright, but that's not what i want to do.
20:14 batman the point with stuff like JSON::Validator is that you can say to an IOS developer (or whatever) here's my schema: make an application against it.
20:15 mst well, if you could then do $validator->write_schema('schema.json');
20:15 batman mst: that is true. but in what format?
20:16 mst well, swagger would be the obvious one, no?
20:16 batman no, it wouldn't. there's a bunch
20:16 batman or at least a couple of interesting one...
20:17 mst ok, so you know more than me about what formats
20:17 mst I was just pointing out you could have a perl setup syntax
20:17 mst that generated whatever format you wanted
20:18 batman another thing is that i imagine (from the snippet) that $validator->pointer('/my/json/pointer')->required(1)->is_array would be inside an action method somewhere, not very easy to extract
20:18 batman works awesome in mojolicious world, but not very awesome in a collaborative world, with ios, js, whatever devs :(
20:24 mst well, yes, you'd abstract the validator building out somewhere so you can get the validator and call ->write_schema on it
20:25 SmokeMachine batman: i was talking about $validator->json_schema("file.schema.json")
20:25 sri a data structure seems more reasonable than a DSL, considering there's specs
20:26 batman sri, jberger, marcus, tempire, crab, Grinnz: any input? is it ok? http://act.yapc.eu/ye2015/wiki?node=Hackathons#Mojolicious
20:26 sri batman++
20:26 batman \o/
20:26 SmokeMachine mst: batman: or $validator->write_schema('schema.json' => "format");
20:27 batman too bad i was so slow ... planned to do it ten days ago, but... vacations :(
20:27 SmokeMachine sorry, i think now i understood
20:27 batman SmokeMachine: that's essentially what i made. @errors = JSON::Validator->new->schema("file.schema.json")->validate($data);
20:27 sri my $json = $validator->to_json_schema;
20:28 SmokeMachine batman: I meant it inside M::Validation
20:28 batman or even ->validate($c->req->json)
20:29 KCL joined #mojo
20:29 SmokeMachine like a plugin that includes that method
20:29 batman SmokeMachine: that won't happen, since JSON::Validator is tied to JSON Schema.
20:29 batman it's not a perl validator. it follows the JSON Schema spec.
20:29 marcus batman++
20:30 marcus I'll try to be available online during the yapc hackathon
20:30 sri marcus: play hearthstone, i need my free pack! ;p
20:30 batman marcus: thanks. how are your blog post coming along? :)
20:30 SmokeMachine file.schema.json is a file with the json-schema... the method would use it to validate ->res->json...
20:30 batman *is
20:30 marcus sri: Sorry, watching astronaut wives club with $wife
20:30 marcus Maybe in a bit.
20:30 sri (╯°□°)╯︵ ┻━┻
20:31 batman SmokeMachine: yeah. going to convert Mojolicious::Plugin::Swagger2 into Mojolicious::Plugin::JsonValidator (or something)
20:31 batman just need to find a clever name for the plugin
20:31 batman help me with that, and i'll make it
20:31 marcus batman: Got 509 words. Got to add links and examples.
20:32 batman marcus++ # shit! now i need to get blogging!
20:32 batman SORRY!
20:32 batman batman - - # language
20:32 sri -.-
20:32 batman i wonder why i can't give myself --... i get that i can't give myself ++, but...
20:34 SmokeMachine Mojolicious::Plugin::JsonValidator isnt a good name?
20:34 batman not sure. looks weird.
20:34 batman camelcase--
20:34 SmokeMachine Mojolicious::Plugin::Json::Validator?
20:35 batman SmokeMachine: so the api i'm considering is simply my @errors = $c->validate_json($data, $schema);
20:35 batman or maybe $errors, since everything else in mojo are refs.
20:36 batman or even just $c->validate_json($schema) # where $data defaults to $c->req->json
20:36 * sri wonders where we will draw the line for core validation in the future, now that json schema and the like keeps popping up
20:36 batman yeah. interesting question.
20:36 batman RULE ALL THE THINGS!
20:36 batman ...?
20:36 batman :)
20:37 sri form validation so far is a pretty natural fit, since it connects many different core features, filling out tags generated by helpers and the like
20:37 batman Lee: wanna attend? http://act.yapc.eu/ye2015/wiki?node=Hackathons#Mojolicious
20:38 sri guess json validation is not as essential
20:38 SmokeMachine maybe differentiate validate_param from validate_json could simplify the task...
20:39 batman sri: i think one of the issues is that json validation can be done in so many ways.
20:39 sri json schema is on its way to becoming a bunch of RFCs though
20:39 sri so it got that going for it
20:39 batman sri: something different while i remember... did you see my comment? https://github.com/kraih/mojo/issues/833#issuecomment-133517649
20:42 ZoffixWork joined #mojo
20:44 sri batman: no i did not
20:45 batman sorry for jumping in with that now in the middle of the conversation. just didn't want to forget about it again :/
20:46 ZoffixWork batman, Are you open to changing this line to $self->{sass_path} = $config->{sass_path} || [];  ? I can make a pull with docs and tests. https://github.com/jhthorsen/mojolicious-plugin-bootstrap3/blob/master/lib/Mojolicious/Plugin/Bootstrap3.pm#L335
20:46 sri not sure that's the right way to test it anyway
20:46 batman sri: i was thinking it was better than no test..?
20:47 sri dunno, are there no portability issues?
20:47 sri like windows stuff
20:47 ZoffixWork I'm basically trying to have several paths where sass would pull the files from. I have foo.scss; and I want it to pull it from location A, if it's present; if not, pull from locations B
20:47 sri had you said it like 30 minutes earlier i would have tried it with the 6.17 release... but now... i'd rather not have it sit in the repo forever ;p
20:49 batman sri: i don't think so. i believe the unlink() will fail on windows since the file is open, but we're not testing that :-)
20:49 batman believe = pretty sure
20:49 sri so, the test will die?
20:50 sri anyway, too much uncertainty... -1 from me
20:50 batman not unless you're using autodie. no core perl functions die.
20:51 batman anyone on windows here that can test it?
20:51 ZoffixWork I am
20:51 ZoffixWork What are we testing?
20:52 ZoffixWork unlink on an open file?
20:52 batman ZoffixWork: i can't think about bs3 right now. i'll be back to you soon-ish.
20:52 batman ZoffixWork: yes.
20:52 ZoffixWork 1 sec
20:52 batman ZoffixWork: or even better.. do you have mojo cloned with test files? if so, could you apply this? https://github.com/kraih/mojo/issues/833#issuecomment-133517649
20:56 batman ZoffixWork: so... back to your bs3 issue: just set $ENV{SASS_PATH}
20:56 ZoffixWork batman, ah, that works :)
20:56 batman it works like PATH="here:or:here:or:there"
20:56 ZoffixWork As for the unlink. No, it doesn't unlink: http://fpaste.scsys.co.uk/497456
20:56 * ZoffixWork tries the patch in the comments
20:57 batman awesome!
20:58 sri is that correct, not a single one of the perl core functions will ever die?
20:58 marcus osx keeps running out of open files for me lately, can't figure out what does it :(
20:58 marcus this last time I wasn't even able to fork a shell to try to figure it ou
20:58 marcus +T
20:59 sri fd limit is like super low by default, 128 or so
21:00 ZoffixWork batman, passes: http://fpaste.scsys.co.uk/497457  This is the file I tested; I manually applied your patch, 'cause I don't know how to use 'patch' to do it :) http://fpaste.scsys.co.uk/497458
21:00 batman sri: i can't name one that will die. but a lot will die with autodie :)
21:01 sri batman: so, you're just guessing?
21:01 batman sweet! thanks ZoffixWork. no you can run off trying out $ENV{SASS_PATH} :)
21:01 ZoffixWork :)
21:03 batman sri: i'm not guessing that unlink() can die. it can probably coredump, but not die
21:03 batman ...unless you load autodie;
21:03 batman oh! i know one core function that dies......... ;) die()
21:03 sri batman: you said "none", and i found that very interesting
21:04 sri so, is that true, or a guess?
21:04 batman it's neither. i haven't used all the core functions, but noen of the functions i've used can die.
21:05 sri ok
21:05 batman mst: can you tell me if any of the core functions can die? (except the obvious one)
21:06 * jberger surfaces after an all day Cities::Skylines bender
21:07 batman sri: i'm dead wrong.
21:07 mst batman: certainly the I/O type ones like unlink don't by default
21:07 batman perl -le'$foo=123;shift $foo;print "dead wrong :("'
21:08 ZoffixWork perl -le'$foo=123;shift @$foo;print "dead wrong :("'
21:08 batman mst: yeah. i was thinking about i/o functions, but then when i looked at "perlfunc" i saw things like each, keys, ....
21:08 ZoffixWork Original was a compilation error, I believe.
21:09 batman ZoffixWork: nope. perl -le'$foo=123;shift $foo;BEGIN{print "dead wrong"}'
21:10 batman sri: so... let's rewind: noen of the i/o functions i've used can die (without autodie) if you give them the right arguments.
21:11 ZoffixWork Hm.
21:11 batman they always return false and tells you to check $!
21:11 ZoffixWork That's interesting.
21:11 marcus sri: Ima play now.
21:19 sri \o/
21:20 mst batman: actually, that's the @$foo deref blowing up
21:21 batman but "blow up" == die...?
21:21 mst yes. but my point is that, mooostly, core *functions* don't
21:21 tempire die is a core perl function
21:21 mst but derefs can
21:22 mst tempire: yes, but that was excluded right at the start of the conversation
21:22 * tempire scampers away
21:22 mst of course the hopefully-to-be-deleted 'push $array_ref' thing
21:22 mst means that now, effectively, push can die
21:24 batman i see. thanks.
21:26 PopeFelix joined #mojo
21:29 mst but that never stopped being experimental, only hopefully will go away again
21:31 batman :)
21:40 jberger mst: that's gone isn't it?
21:40 pink_mist https://metacpan.org/pod/release/WOLFSAGE/perl-5.23.1/pod/perldelta.pod#The-autoderef-feature-has-been-removed
21:41 pink_mist jberger: if you're on blead
21:46 ZoffixWork joined #mojo
21:46 ZoffixWork batman, simply setting SASS_PATH appears to introduce the same issue we were having the other day with custom _variables.scss file.
21:47 batman ZoffixWork: that's weird. are you sure you're setting it to the right values?
21:48 batman also try running with MOJO_ASSETPACK_DEBUG=1 too see the complete value
21:48 batman *to
21:50 ZoffixWork Pretty sure, because it finds the file. Essentially, I have   foo/style.scss and bar/style.scss   and "foo:bar" in SASS_PATH. If I make a change in foo/style.scss it won't propagate until I modify bar/style.scss in any way or nuke packed/
21:54 batman yeah. that's not quite the same problem...
21:54 ZoffixWork MOJO_ASSETPACK_DEBUG=1 says "Using existing asset at foo/packed/" ... doesn't seem to even mention "bar/", which is fine
21:56 batman i hope the files you are referring to is @included...?
21:56 batman *are
21:56 sri marcus: (offtopic) http://www.hearthpwn.com/decks/250941-rank-1-legend-legend-anti-meta-deathlord-hunter
21:56 mst jberger: oh, good, it got culled already
21:57 ZoffixWork batman, imported.... @import "bootstrap/company-variables";   in bootstrap.scss file that's located in "bar/"
21:57 ZoffixWork foo/ doesn't have a bootstrap.scss in it
21:57 batman ZoffixWork: the issue is checksum() https://github.com/jhthorsen/mojolicious-plugin-assetpack/blob/master/lib/Mojolicious/Plugin/AssetPack/Preprocessor/Scss.pm#L94
21:57 batman it's not clever enough to look in SASS_PATH.
21:57 DaniBunny joined #mojo
21:58 batman please file a bug report to assetpack with this conversation and "...scss preprocessor need to look in SASS_PATH, and not just dirname($path);
21:58 ZoffixWork k
21:59 batman not sure why you need so complicated setup though :/
21:59 ZoffixWork http://xtatik.org :)
22:00 batman but either way, it's a bug. but i'm not going to start working on that when it's 00:00, sunday morning :/
22:00 ZoffixWork I have "site" stuff that overrides "company" stuff that overrides "core" stuff. The end result is I can launch a new eCommerce site with like 6 lines of config :)
22:02 batman right. that was the only thing i could think of, haha
22:03 batman ZoffixWork: i will try to fix it tomorrow.
22:03 batman but you need to be around to test before i make a release
22:04 Adura joined #mojo
22:48 batman sri: i'm going to bed now, but... are you still -1 to the "unlink" test, after ZoffixWork ran the test on windows?
22:53 cpan_mojo JSON-Validator-0.49 by JHTHORSEN https://metacpan.org/release/JHTHORSEN/JSON-Validator-0.49
22:53 PryMar56 joined #mojo
23:04 pink_mist it sounded to me that once it was confirmed that that kind of thing couldn't cause it to die() that he was all for it .. but that may have just been wishful thinking on my part
23:19 ZoffixWork batman, sent you a pull. It fixes the issue I was having: https://github.com/jhthorsen/mojolicious-plugin-assetpack/pull/60
23:19 ZoffixWork batman, I also stumbled upon this area and the way it's handling SASS_PATH looks suspicious: https://github.com/jhthorsen/mojolicious-plugin-assetpack/blob/master/lib/Mojolicious/Plugin/AssetPack/Preprocessor/Scss.pm#L141
23:20 ZoffixWork Well... that particular line doesn't include SASS_PATH (does it need to?), and the line above that does include SASS_PATH puts dirname() first instead of last (should it be last?)
23:20 ZoffixWork But nothing is broken for me, so I kinda didn't investigate any of that further :)
23:20 ZoffixWork Time to go out and try out that Mojo drink :) \o/

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