<    March 2017    >
Su Mo Tu We Th Fr Sa  
          1  2  3  4  
 5  6  7  8  9 10 11  
12 13 14 15 16 17 18  
19 20 21 22 23 _2_4 25  
26 27 28 29 30 31
00:02 markh joined
00:05 <bvcosta> what do you use for social login? custom code or any external library?
00:08 nighty joined
00:08 <notriddle> bvcosta: https://github.com/scrogson/oauth2
00:10 <bvcosta> is there anything out of the box?
00:10 gvaughn joined
00:10 <scrogson> bvcosta: look at ueberauth
00:10 <scrogson> https://github.com/ueberauth
00:11 <bvcosta> thanks!
00:16 r00tl3ss_ joined
00:17 sumi joined
00:17 sumi joined
00:18 Quintasan joined
00:18 Quintasan joined
00:19 io_bora joined
00:20 cmn joined
00:24 talklittle joined
00:25 fckfreenode joined
00:25 <fckfreenode> will there be new versions of any of the elixir books soon? (programming elixir, phoenix, meta)?
00:25 <fckfreenode> am i safe to buy the collection?
00:25 <carterparks> i think there's supposed to be a new version of the phoenix book soon
00:26 <carterparks> for 1.3
00:26 <carterparks> amazon has it set for a 5/25 release date
00:26 <fckfreenode> carterparks: ah ok, thanks. any idea when it's released?
00:26 <carterparks> ^
00:26 <fckfreenode> ah ok, thank you
00:26 <fckfreenode> guess i'll buy all but the programming phoenix book and wait until then
00:27 <carterparks> programming elixir was released in november so i wouldn't expect a new version anytime soon
00:27 <carterparks> i started reading the phoenix book and wish i had red the elixir book first
00:27 <fckfreenode> yeah that's my plan, go from prog elxir, then either phoenix or meta
00:32 PaReeOhNos joined
00:33 <carterparks> *read
00:33 willyk joined
00:35 chriswarren joined
00:36 rschmukler joined
00:39 sillyotter joined
00:39 sumi joined
00:43 chriswarren joined
00:52 chriswarren joined
00:53 montanonic joined
00:55 fowlduck joined
00:59 talklittle joined
00:59 rschmukler joined
01:01 chriswarren joined
01:06 sfbw joined
01:08 FIFOd joined
01:11 chriswarren joined
01:12 oetjenj joined
01:15 talklittle joined
01:17 skohz joined
01:19 fhoffmann joined
01:20 chriswarren joined
01:22 alex88 joined
01:29 greyhound93 joined
01:32 fckfreenode joined
01:34 <greyhound93> "Hello" |> IO.puts
01:35 <ivan> Hello
01:37 allanmac_ joined
01:44 w-t-h_ joined
01:44 TarVanim_ joined
01:46 jdvs joined
01:55 fowlduck joined
01:57 hipertracker joined
01:58 fowlduck_ joined
02:00 mikeyhew joined
02:01 PaReeOhNos joined
02:04 allanmacgregor joined
02:07 refriedchicken joined
02:13 chrismccord joined
02:19 oetjenj joined
02:23 jerel joined
02:24 hipertracker joined
02:33 wtetzner joined
02:34 sfaxon joined
02:41 tomterl joined
02:49 MasterNayru joined
02:51 sumi joined
02:58 raycoll joined
02:58 jdvs joined
02:59 fowlduck joined
02:59 allanmac_ joined
03:00 refriedchicken66 joined
03:04 cdg joined
03:04 hakunin joined
03:14 sfbw joined
03:17 refriedchicken joined
03:21 hakunin joined
03:21 mikeyhew joined
03:21 hakunin joined
03:24 raycoll joined
03:27 fowlduck joined
03:28 marcdel joined
03:31 PaReeOhNos joined
03:47 proteusguy joined
03:50 haennar joined
03:51 hipertracker joined
04:01 racycle_ joined
04:05 rkazak joined
04:06 r00tl3ss joined
04:06 gvaughn joined
04:11 jdvs joined
04:13 icecreamcohen joined
04:15 gvaughn joined
04:25 craigger joined
04:28 fowlduck joined
04:30 lizzin joined
04:33 glennular joined
04:37 skohz joined
04:39 chriswarren joined
04:48 chriswarren joined
04:53 alex88 joined
04:58 systemfault joined
04:58 chriswarren joined
05:02 PaReeOhNos joined
05:07 chriswarren joined
05:08 willyk joined
05:09 zabirauf joined
05:10 zabirauf joined
05:14 raycoll joined
05:16 chriswarren joined
05:22 zabirauf joined
05:27 zabirauf_ joined
05:29 fowlduck joined
05:34 rschmukler joined
05:37 hipertracker joined
05:47 rschmukler joined
06:01 Tica2 joined
06:03 dec0n joined
06:04 craigp_ joined
06:05 alexarnon joined
06:09 systemfault joined
06:16 refriedchicken joined
06:17 hbraun joined
06:27 icanhazbroccoli joined
06:28 rschmukler joined
06:30 fowlduck joined
06:33 PaReeOhNos joined
06:36 voltone joined
06:47 Tica2_ joined
06:48 markh joined
06:49 zabirauf joined
06:53 <ivan> is it OK to align -> in case expressions in elixir-lang/elixir or not?
06:53 nighty joined
07:02 sfbw joined
07:03 oetjenj joined
07:08 jamenmarz joined
07:20 chaterrero joined
07:20 hipertracker joined
07:21 <chaterrero> hi, I'm using Phoenix framework and it seems I've got a problem with automatic asset reloading,
07:21 oetjenj joined
07:22 zabirauf joined
07:22 oetjenj joined
07:23 mounibec joined
07:23 <chaterrero> almost every time I make a change to one of my .css files, of course, they are getting "compiled" into one app.css file, but there's a problem - it's not rendered to the full,
07:23 <chaterrero> it's actually cut, more-or-less at half,
07:24 oetjenj joined
07:25 <chaterrero> and because of that every time I make a change (I'm using SFTP plugin to automatically send changes to the server) to .css file, I need to manualy build assets using "brunch build"
07:25 <chaterrero> does anyone have a similar problem?
07:25 bigkevmcd joined
07:30 refriedchicken joined
07:32 fowlduck joined
07:33 tuacker joined
07:37 harfangk joined
07:40 montanonic joined
07:45 cemilowski joined
07:50 _0xAX joined
07:53 acscherp joined
07:55 mounibec joined
07:57 Cohedrin joined
08:01 jdvs joined
08:01 refriedchicken joined
08:04 inoas joined
08:04 PaReeOhNos joined
08:10 mounibec joined
08:14 ZippoWeb joined
08:16 mattyw joined
08:17 rschmukler joined
08:18 vmoravec joined
08:19 PaReeOhNos joined
08:19 Guest67__ joined
08:22 celyr joined
08:23 fowlduck joined
08:23 mark_66 joined
08:30 thiagoalessio joined
08:40 flaviodesousa joined
08:45 tomaz_b joined
08:49 proteusguy joined
09:03 refriedchicken joined
09:06 acscherp joined
09:07 <OliverMT> is there a way to "fake" system env variables?
09:07 <OliverMT> I actually *do* want it in my config
09:07 <OliverMT> but the lib only support system envs
09:08 <OliverMT> aha! System.put_env
09:12 rschmukler joined
09:12 marr joined
09:16 hbraun joined
09:22 lexmag joined
09:23 fowlduck joined
09:24 nd__ joined
09:25 Uniaika joined
09:31 <whodidthis> 1.3 happening :O when someone gets to it link some gists how you are supposed to use action_fallback
09:34 <gazler> whodidthis: https://github.com/phoenixframework/phoenix/blob/2f475135b8ee5e4cbbd7b6e616ad6eb91e04c3fa/test/phoenix/controller/pipeline_test.exs#L47
09:35 <gazler> The argument is a plug that is called when the controller action returns anything other than a Plug.Comm struct
09:35 sumi joined
09:36 vendethiel joined
09:36 <gazler> s/Comm/Conn
09:37 haennar joined
09:42 apotry joined
09:50 BeerLover joined
09:50 BeerLover joined
09:53 sumi joined
09:59 skohz joined
09:59 josevalim joined
10:07 josevalim_ joined
10:22 josevalim joined
10:24 fowlduck joined
10:34 nonninz joined
10:36 cemilowski joined
11:00 rschmukler joined
11:07 refriedchicken joined
11:09 nighty joined
11:10 <OliverMT> http://erlang.org/doc/man/zip.html#create-3
11:10 <OliverMT> when "options" is simply a single atom
11:10 <OliverMT> [:verbose] is what you would send in, no?
11:12 cdg joined
11:15 <ciawal> looks that way
11:20 craigp_ joined
11:21 gmcabrita joined
11:22 Guedes joined
11:22 Guedes joined
11:25 fowlduck joined
11:33 haennar joined
11:34 MaSven joined
11:34 <OliverMT> I wonder why path.join does not make locale specific separators
11:34 <OliverMT> now I get paths of \\ first and then with / later
11:37 justicefries left
11:40 <micmus> OliverMT: because the VM converts / to \\ if it's needed on file operations
11:41 haennar joined
11:47 <OliverMT> oh
11:47 cristianR_ joined
11:47 <micmus> ericmj: I can't publish a package
11:47 <micmus> {:failed_connect, [{:to_address, {'hex.pm', 443}}, {:inet, [:inet], :nxdomain}]}
11:47 <OliverMT> my zip.create just returns einval
11:47 <OliverMT> wondering if its because of paths
11:47 skohz joined
11:47 <micmus> ericmj: also - if publishing fails the task tries to continue with publishing the docs. It should probably immediately abort
11:48 watersoul joined
11:48 <micmus> It makes little sense to publish docs for a package that wasn't published
11:48 <ciawal> are you using charlist OliverMT
11:48 <ericmj> micmus: probably a local error
11:49 <ericmj> Can you show output when it tries to publish docs after package fails?
11:49 <micmus> ericmj: it looks like it was some intermittent error... it worked now
11:50 <micmus> ericmj: https://gist.github.com/michalmuskala/d62b1be1a16a459e4577b09b8f533177
11:53 glennular joined
11:53 Uniaika joined
11:54 rschmukler joined
11:55 <OliverMT> ciw
11:55 <OliverMT> ciawal: yes
11:56 <OliverMT> ciawal: https://gist.github.com/olivermt/a3a0e4716d27eb50eaf40d30ec2b2259
11:58 lexmag joined
12:01 <OliverMT> I wonder if i should be using cwd to that folder instead
12:04 <ciawal> does it work without :memory perhaps?
12:05 <ciawal> you have mixed forward and backslashes there
12:05 <ciawal> maybe an issue
12:05 <OliverMT> no, not accoirding to micmus
12:05 <OliverMT> and no, doesnt work without memory
12:06 <ciawal> where does the module/function info come from in the log call?
12:07 <ciawal> are you sure the path is correct also
12:08 kommen joined
12:10 Ven joined
12:10 CapNemo joined
12:16 kommen_ joined
12:21 <OliverMT> Yes and yes
12:21 <OliverMT> The path is made and filled with files further up with same dest_dir
12:22 <OliverMT> Im downloading from s3 to make bundle and re-upload
12:25 skohz joined
12:26 fowlduck joined
12:27 pbrooko joined
12:29 gvaughn joined
12:30 pbrooko_ joined
12:35 sfbw joined
12:40 Ven joined
12:40 chriswarren joined
12:41 dimitarvp joined
12:42 elixir5234 joined
12:43 <elixir5234> Can anyone clarify what `embedded_schema` does in the context of Ecto schemas? I tried reading the docs but it's not clear to me.
12:43 haennar joined
12:48 CapNemo joined
12:50 <PaReeOhNos> elixir5234: Where `schema` defines the structure of a model backed by a table, the `embedded_schema` defines the structure of a "model" that will be stored embedded in another record as JSON
12:50 chriswarren joined
12:51 <PaReeOhNos> so you'd have a model which has something like `embeds_one :other_model, MyApp.OtherModel`, then in the `MyApp.OtherModel` module, you'd define your `embedded_schema`
12:53 <elixir5234> PaReeOhNos: Hmm, what would be the use case for an embedded_schema?
12:53 mattyw joined
12:54 <PaReeOhNos> elixir5234: Unstructured data, or maybe meta-data? There's loads of use-cases that I'm sure others can give examples of, I don't generally use them. If you've got structured data then it probably makes more sense to use a normal schema
12:55 <elixir5234> PaReeOhNos: Thanks for your help! :)
12:57 <dimitarvp> I've also seen `embedded_schema` used for virtual attributes, including composite (record / struct) ones.
12:59 chriswarren joined
13:00 pbrooko joined
13:01 simonoram joined
13:04 cemilowski joined
13:07 laurocaetano joined
13:08 r00tl3ss joined
13:08 chriswarren joined
13:09 mattyw joined
13:09 <hansihe> would anyone familiar with how enumerables work internally mind answering some questions? https://gist.github.com/hansihe/605652b55cbf8a4fc776df5c326fa707
13:13 Uniaika joined
13:13 refriedchicken joined
13:14 haennar joined
13:17 acscherp joined
13:17 <dimitarvp> hansihe: after a quick look, maybe it's `Stream.resource/3` what you're looking for.
13:18 <dimitarvp> I successfully made use of it for a server side of an API in order to be able to emit paginated results
13:18 <dimitarvp> not 100% your use case I know -- but I believe it's applicable.
13:22 <hansihe> dimitarvp: yeah, I think I can make that work
13:22 <hansihe> thanks
13:24 <dimitarvp> hansihe: Took me a while (2 halves of free evenings). The philosophy of FP languages isn't to give you a rich toolset with which you can assemble your programs with almost no effort; their philosophy is to give you less but more powerful tools with which you can build anything you like =)
13:25 <dimitarvp> `Stream.resource/3` is one such tool -- it doesn't look like much but you can actually build *anything* custom that requires streaming with it.
13:27 fowlduck joined
13:28 laut joined
13:28 <hansihe> the question then is, since I want to take a stream as an input, how to I take one item off the stream and keep going later? Enum.take doesn't enable you to keep going from where you left off
13:28 <hansihe> Stream.take seems to be the exact same thing except lazy
13:29 <hansihe> I suppose what I want is conceptually more like a flat_map_reduce operation, except with the option of emitting additional items after you know the input stream has ended
13:29 b14d3 left
13:30 <dimitarvp> hansihe: that's where `Stream.resource/3` comes in unless I am misunderstanding you. You can emit `{:halt, final_value}` at any time. But you CANNOT resume such a stream IMO.
13:31 simonoram_ joined
13:32 <hansihe> right, I understand how `Stream.resource/3` can be used to emit values in the way I want
13:33 <hansihe> what I don't know how to do is consume an input stream one item at a time in a way that is compatible with it
13:33 <dimitarvp> that's your job sadly -- you must encode that in the `start_fun` and `next_fun` functions
13:33 <benwilson512> hansihe: what is the source of data?
13:33 <dimitarvp> your best bet is probably to buffer a good chunk of the input data
13:33 <hansihe> a Stream
13:34 <benwilson512> hansihe: you want to wrap a stream w/ stream.resource?
13:34 stephen_m joined
13:34 <hansihe> well, not specifically
13:35 <hansihe> I guess what I really want is flat_map_reduce on a stream with the option of emitting a final value when the input stream ends
13:35 <benwilson512> ah
13:35 <dimitarvp> benwilson512: more like, after getting a rough idea of what he needs, I advised him to do so. Might not have been the best advice though, since I don't know his *exact* needs.
13:35 <hansihe> Stream.resource might be a way to accomplish that, but i'm not sure
13:36 <hansihe> benwilson512: https://gist.github.com/hansihe/605652b55cbf8a4fc776df5c326fa707
13:36 <hansihe> there is some more information on exactly what I want to do in there
13:37 juanjoc joined
13:38 <benwilson512> ah
13:38 <benwilson512> so
13:38 <benwilson512> if I were you
13:38 <benwilson512> I'd implement enumerable from scratch
13:38 <benwilson512> instead of relying on one of the stream functions
13:38 <benwilson512> it'll give you the most control
13:38 <benwilson512> and the least performance overhead
13:39 wpcarro_ joined
13:39 <hansihe> yeah, I kinda suspected I would want to do that
13:39 <OliverMT> benwilson512: that bug yesterday? hackney 1.6.6 too :)
13:39 <OliverMT> just a different error msg on the upload than on the other commands
13:39 <benwilson512> OliverMT: yeah I know
13:39 <benwilson512> so
13:40 <benwilson512> so I used to do hackney "1.6.3" exactly
13:40 <hansihe> I looked at it a bit, but there doesn't seem to be all that much documentation on how Enumerables work internally and how to write them correctly
13:40 <hansihe> I guess I will be reading code
13:40 <benwilson512> and then everyone complained that that was overly strict
13:40 <benwilson512> I'm going back to doing that
13:40 gvaughn joined
13:41 <benwilson512> AWS is super picky about stuff, and once a version works I'm gonna just peg the ExAws dependency to that version. I'll support multiple versions by whitelisting
13:41 <benwilson512> not gonna go back to ~>, been burned twice by that now
13:42 rschmukler joined
13:44 <gazler> Shouldn't running mix deps.get ignore retired packages? and 1.6.6 is retured
13:44 <gazler> *retired
13:44 <benwilson512> it is now
13:45 <micmus> AFAIK it will warn, but not skip the retired versions
13:45 <benwilson512> micmus: orly?
13:45 <benwilson512> OliverMT: can you confirm that 1.7.1 works?
13:45 <gazler> Just checked, it only warns.
13:45 <OliverMT> I went back to 1.6.5
13:46 <gazler> hackney 1.6.6 RETIRED! (invalid)
13:46 <benwilson512> OliverMT: ah ok. feel like checking out 1.7.1 ?
13:47 <micmus> benwilson512: it seems the problem is with hackney violating semver (accidentally)
13:47 <benwilson512> micmus: yup
13:48 <benwilson512> it's a perfectly reasonable set of mistakes, it's just resulting in a lot of github issues for ExAws, so I'm gonna take the conservative route and just whitelist known good versions
13:48 <OliverMT> I can do later when on that pc again benwilson512
13:48 <benwilson512> k
13:48 <OliverMT> on laptop babysitting and doing some other stuff now :p
13:51 alex88 joined
13:52 skohz joined
13:55 gvaughn joined
14:01 r00tl3ss joined
14:02 <OliverMT> ciawal: the issue is that erlang cant handle / slashes
14:02 Tallys__ joined
14:03 cemilowski joined
14:03 jerel joined
14:08 <OliverMT> oh lol, FileList is a list of files, with paths relative to the current directory,
14:08 simonoram joined
14:08 <OliverMT> so an absolute path doesnt work it seems?
14:09 harfangk joined
14:16 cdg joined
14:20 sfbw joined
14:21 alex88 joined
14:25 spinningtopsofdo joined
14:26 acscherp joined
14:26 cdg joined
14:27 fowlduck joined
14:37 rschmukler joined
14:39 simonoram joined
14:39 dec0n joined
14:43 jschoolcraft joined
14:44 simonoram joined
14:50 jerel joined
14:55 kommen joined
14:57 Stratus3D joined
14:58 <gazler> Anyone on a mac using Elixir 1.4.x able to help me confirm an issue?
14:58 <drewolson> gazler: sure
14:59 <gazler> drewolson: would you mind running the following for me? git clone git@github.com:VoiceLayer/dispatch.git && cd dispatch && mix deps.get && mix test
14:59 <drewolson> gazler: doing so right now, hold on a sec
14:59 <gazler> drewolson: You can see the Repo at https://github.com/VoiceLayer/dispatch if you don't want to blindly execute my task. :)
15:00 kommen joined
15:00 <gazler> I am getting ` ** (EXIT) {:error, {:error, {:open_error, -10}}}`
15:00 allanmacgregor joined
15:00 <gazler> https://gist.github.com/Gazler/6ff814f03e6e8970f78b3ccc09725009 is my trace. Just want to confirm it isn't an OS specific issue.
15:01 <drewolson> gazler: https://gist.github.com/drewolson/b6ca10afa0f896777cd2e3eab4631295
15:01 <drewolson> seems to reproduce the issue
15:01 <gazler> drewolson: Awesome, thanks!
15:01 <drewolson> np
15:03 <gazler> drewolson: Thanks for releasing 1.1.4 of scrivener_ecto too
15:03 <drewolson> gazler: thank for the PR! sorry it took a day or two longer than expected. hope your issue is fixed now.
15:04 <gazler> drewolson: Yeah, I was calculating a relevance in the db and using it for sorting.
15:04 <drewolson> gazler: yes, that makes a lot of sense.
15:06 <Ankhers> I'm trying to use https://gist.github.com/chrismccord/71ab10d433c98b714b75c886eff17357 to upgrade an app from phoenix 1.2.x to 1.3.0-rc.0. I just did the first step, bumping the phoenix version. When I try to compile my application, I'm getting an error in lib/my_app/endpoint.ex saying module MyApp.Router is not available. Would anyone have any idea where to look for why this isn't working?
15:07 <Ankhers> Not that it should make a difference, but it worked 2 minutes ago when it was still using phoenix 1.2.x
15:07 rpihlak joined
15:07 <gazler> Ankhers: Did you rename it to MyApp.Web.Router?
15:07 <Ankhers> I haven't done any of the renaming yet. I just bumped the version.
15:08 <Ankhers> Which, according to the gist, says should still be enough to have everything compile.
15:09 potatosalad joined
15:09 cdg_ joined
15:09 refriedchicken joined
15:13 racycle_ joined
15:14 <Ankhers> Alright, disregard that. I did remove "web" from the elixirc_paths/1 already... Which of course makes it fail...
15:14 Or1on joined
15:14 kommen_ joined
15:14 <gazler> Ankhers: That'd do it. :)
15:15 <Ankhers> gazler: Indeed. I am dumb...
15:15 Guest76598 joined
15:17 kommen__ joined
15:18 milmazz joined
15:19 kommen joined
15:20 sumi joined
15:22 nhu joined
15:23 kommen_ joined
15:23 nanohest joined
15:24 kommen joined
15:26 Gringill joined
15:26 meh` joined
15:28 Stratus3D joined
15:28 fowlduck joined
15:32 maxbeizer joined
15:41 Guest67__ joined
15:41 <Gringill> ping bitwalker - question about distillery config
15:46 fckfreenode joined
15:49 samdvr joined
15:50 laurocaetano joined
15:54 MaSven joined
15:54 jdvs joined
15:59 rhnonose joined
16:00 Guedes joined
16:02 icecreamcohen joined
16:03 rschmukler joined
16:07 hbraun joined
16:16 chrismccord joined
16:18 tuacker joined
16:20 raycoll joined
16:22 <OliverMT> can I borrow an exstra pair of eyes:
16:22 <OliverMT> https://gist.github.com/olivermt/399796f2aa2b751fcc97b35ffd828104
16:23 gvaughn joined
16:26 <Fenne> OliverMT: line 12, shouldn't be zip_files instead of files?
16:26 <OliverMT> hoooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo my god
16:26 <OliverMT> hahaha
16:26 <OliverMT> I need to hire an intern solely to be my pair programmer for when I get code blindness
16:26 <OliverMT> thanks Fenne
16:27 chriswarren joined
16:27 <Fenne> np, OliverMT ;)
16:27 <Fenne> and about the intern, good idea, I need to consider it myself
16:29 fowlduck joined
16:34 chriswar_ joined
16:35 ekmartin joined
16:39 racycle_ joined
16:40 rschmukler joined
16:42 nanohest joined
16:45 kommen_ joined
16:47 <Gringill> Trying to migrate an umbrella app from exrm to distillery. It has two apps in it (:core, and :net). :net uses :core as an in_umbrella dep. I have a release defined with `set applications: [net: :permanent]` in it. When I attempt to start the release :core and :net are loaded, but not started. It seems like all their dependencies get started though -- any
16:47 <Gringill> advice here?
16:51 Stratus3D joined
16:52 davidw joined
16:54 drincruz joined
16:55 kommen__ joined
16:57 amclain joined
16:58 CapNemo joined
17:01 gvaughn joined
17:02 wsieroci joined
17:11 <OliverMT> can someone tell me if this is possible, to save me some trouble of typing out the whole prototype
17:12 <OliverMT> I am processing a bunch of files async, downloading them, making zip, then reuploading the zip
17:12 <OliverMT> I'd like to send incremental updates of progress down the conn, which should work with send_chunk
17:12 <bitwalker> Gringill: sup
17:12 montanonic joined
17:12 <OliverMT> could I send self() down into the service and loop for status updates?
17:12 <OliverMT> a conn lives in its own process doesnt it?
17:13 cemilowski joined
17:13 wsieroci joined
17:14 <bitwalker> Gringill: If I understand your setup, you need to have both :core and :net in the `set applications` line
17:15 kommen joined
17:15 <Gringill> bitwalker: Yeah I believe I tried that as well. Retesting to see where that lands me, one moment.
17:16 Guedes joined
17:17 <Gringill> `set applications: [net: :permanent, core: :permanent]`. They both are loaded, neither are started. Starting with `_build/prod/rel/my_umbrella_name/bin/my_umbrella_name console` in these tests.
17:19 <Gringill> bitwalker: If I remove the :core dep from :net, then they both are loaded, but only :net is started.
17:23 rhnonose joined
17:23 <OliverMT> benwilson512: ping
17:23 <bitwalker> Gringill: use `set applications [:net, :core]`
17:24 <bitwalker> s/use/try
17:24 PaReeOhNos joined
17:25 <OliverMT> is there a way to make a stream from a large binary?
17:26 jeffweiss joined
17:29 <Gringill> bitwalker: :core and :net are loaded but not started.
17:29 nanohest joined
17:29 <bitwalker> are they started when you run `mix run --no-halt`?
17:29 rschmukler joined
17:30 fowlduck joined
17:30 <Gringill> Yes
17:31 <Gringill> bitwalker: To be clear about all "connections" between those two apps, :net declares :core as a dep, and there is a place in :core that aliases a module of :net and uses a function in there. Idk if that last part is bad practice / impacting this.
17:33 <bitwalker> Is that function called during :core's startup?
17:34 kommen_ joined
17:35 chriswa__ joined
17:35 <bitwalker> Gringill: I just tested with distillery/conform master using my umbrella test project (https://github.com/bitwalker/distillery-umbrella-test) and it works as expected, there must be something else going on
17:36 nahtnam joined
17:36 <Gringill> bitwalker: Using the combined release?
17:37 Scramblejams joined
17:37 <Gringill> I am not using a master_app. Is that the issue?
17:38 wilo joined
17:40 <bitwalker> No, you can remove it from the combined release and it makes no difference, app_one and app_two are both still loaded/started
17:40 ekmartin joined
17:40 fowlduck joined
17:41 Stratus3D joined
17:41 CapNemo joined
17:42 vmoravec joined
17:43 <Gringill> Yeah I think the issue here is the dependency of my apps on eachother, :core and :net. I am just unsure in what manner it is unhappy.
17:43 <Gringill> The two apps in your example repo do not depend on eachother.
17:44 Cohedrin joined
17:45 <Gringill> I am going to pull out that function I referenced above and see if that has any impact here.
17:54 squall joined
17:55 CapNemo joined
18:00 achauvin joined
18:00 <Gringill> y
18:01 <Gringill> mistell.
18:01 stephen_m joined
18:02 wsieroci joined
18:02 wsieroci joined
18:08 inoas joined
18:08 markh joined
18:11 Stratus3D joined
18:17 Scramblejams joined
18:19 rhnonose joined
18:20 kommen joined
18:22 Scramblejams joined
18:25 ZippoWeb joined
18:34 samdvr joined
18:38 cemilowski joined
18:38 nanohest joined
18:39 rschmukler joined
18:39 willyk joined
18:40 drincruz joined
18:40 Ven joined
18:42 elixir4134 joined
18:42 <elixir4134> Hey, playing around with Phoenix 1.3.0rc now. What's the difference between phx.new and phx.new.web?
18:43 <elixir4134> There doesn't seem to be any description for the phx.new.web mix task.
18:43 <chrismccord> elixir4134 https://github.com/phoenixframework/phoenix/blob/master/installer/lib/mix/tasks/phx.new.web.ex#L2-L26
18:45 Ilyes512 joined
18:45 <elixir4134> chrismccord: thx!
18:49 kommen joined
18:50 Stratus3D joined
18:53 josevalim joined
18:53 wsieroci joined
18:55 <elixir4134> chrismccord: I've been playing around with the new generators, and contexts seem like a great idea, I'm just having a little trouble figuring out the best way to pick out good contexts. Let's say I'm creating a blog engine with posts--would it be appropriate to have a context and resource with the same name? (e.g. Posts)
18:58 <elixir4134> chrismccord: still a beginner, so I'm just figuring this stuff out. sorry if it seems like a stupid question. :)
19:00 PaReeOhNos joined
19:01 bvcosta joined
19:01 kommen joined
19:06 DeadZen joined
19:07 rhnonose joined
19:08 Guedes joined
19:08 Guedes joined
19:12 icanhazbroccoli joined
19:14 <chrismccord> elixir4134 : it's a blance, the "little bit of trouble" is because you have to think about the design :)
19:15 <chrismccord> elixir4134 Posts.Post is not what you want. Think about what you are building. What if this was a separate application, what would you name it?
19:15 <chrismccord> what does it accomplish? What bits of your domain does it take care?
19:16 <chrismccord> If you are creating a blog engine, then naturally you are creating a blog, and the context could be Blog
19:18 hbraun joined
19:19 nanohest joined
19:21 achauvin joined
19:22 juanjoc joined
19:27 Ven joined
19:27 <Gringill> bitwalker: Holy rabbit hole. Ok so the issue ended up being that one of my applications, :core, had a dependency :red, that in turn had a dependency :blue, that, on init did an application:stop(lager), followed by an application:start(lager). Ripping that out caused everything to work correctly. I noticed this because even when I tried to release :core
19:27 <Gringill> alone, it would load itself but not start. So I just started ripping things out until it worked. Still unsure why stopping lager causes this issue though, it didn't seem to matter when I was using exrm, if you have any thoughts on that I would appreciate it.
19:29 <bitwalker> Gringill: interesting, you'd need to explicitly mark lager as an included_application if you are doing that, is that how you had it set up?
19:30 FIFOd joined
19:31 sfbw joined
19:31 kommen joined
19:33 <Gringill> bitwalker: Indeed, my umbrella's root mix.exs has `[extra_applications: [:lager]]` in it. It is unclear immediately why that was put there, would have to investigate a little. I would imagine because of some warning at some point. Removing it from there (while keeping in the app:stop/start of lager) does not fix the issue.
19:34 <bitwalker> extra_applications is different than included_applications though
19:36 <Gringill> Indeed, it is not set anywhere as an included_application, I was just calling out a similar setting that mentioned it :)
19:37 <Gringill> You would expect having it as an included_app to cause this it seems?
19:38 <bitwalker> Other way around, you would need to have it in included_applications if you are taking over it's lifecycle, *or*, use `set applications [lager: :load]` in rel/config.exs, at which point it will be loaded but you'll need to manually start it, the latter is preferable, but either approach will work
19:39 <alisdair> re: ecto & dynamic configuration of repos (at runtime, not compile time). short of the discussed solution where static (modules + macros) and dynamic repos have completely disjoint interfaces and code paths, i think the only way to get dynamic repos in a reasonable way is to make them an interface with two implementations, one module based and one struct
19:39 <alisdair> based
19:40 <alisdair> i'm not really comfortable taking on that level of work myself tho without reasonable assurance it'll be accepted
19:41 <alisdair> i'd still like to use ecto ultimately, so i'd love to know the best way to proceed from here
19:41 <alisdair> minimally functional prototype? more discussion?
19:43 codestorm joined
19:43 cdg joined
19:46 hashpuppy joined
19:48 markholmes joined
19:50 TarVanimelde joined
19:50 drincruz joined
19:53 acscherp joined
19:54 wsieroci joined
19:55 craiglyons joined
19:56 markh joined
19:56 <micmus> alisdair: I think josevalim had the idea to expose the functions accepting an explicit repo argument
19:57 <micmus> we could use the name to name the top level supervisor of the repo
19:57 <micmus> and associate the pool config to that using the Registry
19:57 <micmus> we mostly already have those functions spread across the Ecto.Repo.* modules
19:59 <Gringill> bitwalker: Ok with the stop/start commented out, no extra or included apps, and just :net and :core in the set_applications. It works. From that point if I add :lager to included_applications, it still works. If from there I remove :lager from included_applications and add him to my set applications in rel/config.exs as lager: :load it actually causes :net
19:59 <Gringill> and :core to not start again (keep in mind the "fix" is still in place, commenting out the stop/start of lager in that dep).
20:02 <Gringill> No errors or anything...
20:03 rschmukler joined
20:03 <OliverMT> micmus: could I pilfer your asdf setup for elixir for circle again?
20:05 <targaf> With ecto/phoenix, if I have a model tools with a has_many relationship to a model attributes is there a way I can show n attribute fields on the new tools page?
20:06 PaReeOhNos joined
20:08 <alisdair> micmus: the problem with that is that ecto treating `Repo` as a module is endemic throughout the code base, and you'd have to thread the pool name through a ton of functions to achieve that. i think it's probably less work to make Repo an interface
20:09 robinsjdotcom joined
20:12 leandroalemao joined
20:15 alex88 joined
20:18 haennar joined
20:19 <micmus> alisdair: I'm not sure what you mean by "an interface"
20:20 rschmukler joined
20:20 <micmus> OliverMT: https://github.com/codetogether-editor/codetogether-backend/blob/master/circle.yml and https://github.com/codetogether-editor/codetogether-backend/blob/master/scripts/ci-setup.sh for example
20:21 <OliverMT> micmus: thanks
20:24 <alisdair> micmus: sorry, a protocol
20:24 Scramblejams joined
20:25 <micmus> how would that work? A protocol needs different data structutres
20:27 Ven joined
20:27 nd__ joined
20:30 MotherFlojo joined
20:34 <alisdair> right, so the protocol would have implementations for module (actually atom) and for struct
20:34 davidw joined
20:34 <alisdair> the module implementation would be identical to the current implementation
20:37 simonoram joined
20:40 w-t-h_ joined
20:53 travis-ci joined
20:53 <travis-ci> elixir-lang/elixir#13694 (master - 824b268 : Wojtek Mach): The build passed.
20:53 <travis-ci> Change view : https://github.com/elixir-lang/elixir/compare/c1ca6fb8990d...824b26842052
20:53 <travis-ci> Build details : https://travis-ci.org/elixir-lang/elixir/builds/207142896
20:53 travis-ci left
20:54 PaReeOhNos joined
20:56 meh` joined
20:57 chrismccord joined
20:58 sumi joined
21:00 <aaronjensen> chrismccord: should we add mix phoenix.digest.clean to the changelog?
21:00 <chrismccord> aaronjensen yes indeed!
21:01 <chrismccord> aaronjensen can you send a PR so you get some love for the contrib?
21:01 <aaronjensen> Sure, though it was mostly my team, not me :)
21:02 <aaronjensen> chrismccord: did things change to phx?
21:03 <chrismccord> aaronjensen : yes, but I just realized that the phoenix.digest.clean was unreleased so there is no need to keep the only name around with a deprecation warning
21:03 <chrismccord> keep the old* name
21:03 <aaronjensen> good call
21:03 <aaronjensen> https://github.com/phoenixframework/phoenix/pull/2140
21:06 rschmukler joined
21:07 <aaronjensen> and https://github.com/phoenixframework/phoenix/pull/2141
21:08 <chrismccord> <3
21:08 <aaronjensen> congrats on rc0 release, looking forward to upgrading
21:16 rschmukler joined
21:17 rschmukl_ joined
21:17 allanmacgregor joined
21:20 codestorm joined
21:22 jeregrine joined
21:29 oetjenj joined
21:31 rschmukler joined
21:33 Schalla joined
21:36 wess joined
21:36 <wess> hello all
21:39 Scramblejams joined
21:39 <OliverMT> congrats on rc0 release
21:39 <OliverMT> do not look forward AT ALL to upgrading
21:39 <OliverMT> :D
21:39 <OliverMT> will be doing 1.3 and 1.4.* cleanup at the same time on our oldest project
21:39 <OliverMT> I did a test compile on 1.4 and the console basically just scrolled warnings at me for 2-3 minutes
21:40 wess left
21:40 cdg joined
21:42 montanonic joined
21:44 jamenmarz joined
21:47 jeregrine joined
21:49 codestorm joined
22:24 irclogger_com joined
22:24 Topic for
22:26 Guest1200 joined
22:29 marcdel joined
22:29 naratcis joined
22:30 <naratcis> hello
22:30 <Guest1200> hi
22:33 rschmukler joined
22:35 spinningtopsofdo left
22:42 pbogut joined
22:49 Ilyes512 joined
22:59 chriswarren joined
23:00 chriswarren joined
23:01 chriswarren joined
23:01 chriswarren joined
23:02 chriswarren joined
23:03 chriswarren joined
23:03 squall joined
23:04 chriswarren joined
23:04 Aeyris joined
23:05 chriswarren joined
23:05 Aeyris left
23:05 codestorm joined
23:12 chriswar_ joined
23:14 jeffweiss joined
23:14 nd__ joined
23:15 apotry joined
23:16 markh joined
23:21 chriswarren joined
23:22 vadviktor joined
23:24 aedigix joined
23:30 chriswarren joined
23:34 <drincruz> quit
23:36 PaReeOhNos joined
23:37 PaReeOhNos joined
23:39 mozzarella joined
23:39 chriswarren joined
23:45 codestorm joined
23:46 jeregrine joined
23:50 willyk joined
23:58 io_bora joined