<    April 2017    >
Su Mo Tu We Th Fr Sa  
 2  3  4  5  6  7  8  
 9 10 11 12 13 14 15  
16 17 18 19 20 21 22  
23 24 25 26 27 _2_8 29  
00:01 Ayey_ joined
00:01 rschmukler joined
00:12 nahtnam joined
00:15 icecreamcohen joined
00:16 NeverDie joined
00:18 zetashift joined
00:23 <* benwilson512> ericmj: hey, is there documentation for the hexpm web API?
00:23 zeta joined
00:24 <benwilson512> other than the code obviously
00:25 <benwilson512> ah nvm found it https://github.com/hexpm/specifications/blob/master/apiary.apib
00:26 <benwilson512> apologies
00:30 nolan330 joined
00:32 nolan330 joined
00:34 <nolan330> hey everyone, wondering if there is a idiomatic preference between composing async_stream calls (e.g. Task.async_stream(Mod1, :func1) |> Task.async_stream(Mod2, :func2)) or passing a function to run as an argument to a single async_stream (e.g. Task.async_stream(Mod1, :func1, [&Mod2.func2/N]) and then calling the function parameter within :func1)
00:35 <nolan330> :func1 is a very common function, and sometimes i want to immediately follow it up with a :func2 transformation
00:36 <nolan330> or, more generally, sometimes i want :func1 alone, and sometimes i want to process the results of :func1 concurrently
00:40 gvaughn joined
00:43 <nolan330> it also might matter that both :func1 and :func2 are (or could be) IO heavy
00:47 rschmukler joined
00:49 rpip joined
00:49 codestorm joined
00:55 codestorm joined
00:57 nighty-- joined
01:10 steffkes joined
01:10 steffkes joined
01:12 <nolan330> if anyone gets a chance to answer a silly question about Task.async_stream, I just posted a question over here http://stackoverflow.com/questions/43599963/composing-task-async-stream-vs-continuation-passing
01:30 jkreeftmeijer joined
01:30 bambanx joined
01:30 willyk joined
01:35 felideon joined
01:39 cschneid_ joined
01:39 tomterl joined
01:41 adgtl joined
01:42 rschmukler joined
01:44 codestorm joined
01:54 duane joined
01:59 rschmukler joined
02:02 Ayey_ joined
02:10 codestorm joined
02:14 proteus-guy joined
02:14 ianleeclark joined
02:31 jkreeftmeijer joined
02:31 kau joined
02:31 jschneck joined
02:44 willyk joined
02:47 mwbrown joined
03:08 mika__ joined
03:11 steffkes joined
03:11 steffkes joined
03:14 rschmukler joined
03:15 mwbrown joined
03:31 RexM_ joined
03:34 adgtl joined
03:41 skace joined
03:43 r00tl3ss joined
03:51 larshesel joined
03:55 isaiah joined
03:57 mwbrown joined
03:58 cschneid_ joined
03:59 iFire joined
04:08 mwbrown joined
04:22 codestorm joined
04:24 Tica2 joined
04:32 jkreeftmeijer joined
04:33 mwbrown joined
04:41 mwbrown joined
04:43 mika__ joined
04:54 refriedchicken joined
05:01 refriedchicken joined
05:02 refriedchicken66 joined
05:02 rschmukler joined
05:03 refriedchicken66 joined
05:05 refriedchicken joined
05:06 refriedchicken66 joined
05:07 refriedchicken66 joined
05:12 duane joined
05:12 steffkes joined
05:16 refriedchicken33 joined
05:16 gokr joined
05:18 gvaughn joined
05:28 milad joined
05:30 TwentyChung joined
05:32 refriedchicken joined
05:40 jschneck joined
05:43 Shados joined
05:46 codestorm joined
05:52 justelex joined
05:55 milad joined
05:56 justelex joined
05:59 rschmukler joined
06:10 wyzzard_ joined
06:13 gvaughn joined
06:18 proteus-guy joined
06:21 nepiktaz joined
06:22 Guest67 joined
06:25 refriedchicken joined
06:28 tuacker joined
06:31 Ayey_ joined
06:34 vaultin joined
06:34 jkreeftmeijer joined
06:39 cschneid_ joined
06:40 willyk joined
06:41 voltone joined
06:41 codestorm joined
06:44 tomaz_b joined
06:44 johanw joined
06:45 milad joined
06:46 rc1140 joined
06:48 m_m joined
06:54 robinsjdotcom joined
06:58 m_m joined
06:58 jeffweiss joined
07:01 tomaz_b joined
07:02 refriedchicken joined
07:02 akeating joined
07:05 jkreeftmeijer joined
07:08 milad joined
07:08 hbraun joined
07:09 craigp joined
07:09 flaviodesousa joined
07:10 akeating joined
07:13 steffkes joined
07:13 steffkes joined
07:14 duane joined
07:14 ZippoWeb3 joined
07:15 nd___ joined
07:22 m_m joined
07:24 elgenie joined
07:26 asabil joined
07:29 mark_66 joined
07:32 marr joined
07:34 cemilowski joined
07:35 codestorm joined
07:36 isHavvy joined
07:37 craigp joined
07:38 vendethiel- joined
07:39 mattyw joined
07:40 ultralazer joined
07:41 ashp_ joined
07:42 steffkes joined
07:45 dastagg joined
07:54 jkreeftmeijer joined
07:56 jkreeftmeijer joined
07:56 refriedchicken joined
07:58 akeating joined
07:58 codestorm joined
07:59 stephen_m joined
08:00 akeating_ joined
08:00 [0__0] joined
08:01 gvaughn joined
08:02 seggy joined
08:05 akeating joined
08:12 robinsjdotcom joined
08:17 robinsjdotcom joined
08:22 mattyw joined
08:32 codestorm joined
08:37 cschneid_ joined
08:42 josevalim joined
08:45 lexmag joined
08:52 tomaz_b joined
08:55 imack joined
08:57 ariejan joined
08:59 Ayey_ joined
09:00 siruf__ joined
09:04 mattyw joined
09:07 Dev0n joined
09:08 cemilowski joined
09:13 vmoravec joined
09:13 miwa joined
09:16 duane joined
09:22 justelex joined
09:26 codestorm joined
09:29 akeating joined
09:30 mika__ joined
09:32 elgenie joined
09:38 rschmukler joined
09:39 cemilowski joined
09:40 akeating joined
09:49 gvaughn joined
09:52 jkreeftmeijer joined
09:57 sumi joined
10:02 nighty-- joined
10:03 JoelMcCracken joined
10:05 akeating joined
10:05 yourname_ joined
10:14 PaReeOhNos joined
10:17 duane joined
10:17 mr_bmt joined
10:18 mr_bmt joined
10:20 codestorm joined
10:22 <ciawal> it seems that specifying a `mix test file:line` referencing a describe block doesn't run the contents of the describe block as I would expect
10:22 <ciawal> which is a shame
10:24 <fishcakez> you'd need to use a describe tag
10:26 akeating joined
10:34 jkreeftmeijer joined
10:41 asabil joined
10:43 cemilowski joined
10:43 mr_bmt joined
10:44 mr_bmt joined
10:45 mr_bmt joined
10:46 mr_bmt joined
10:47 mr_bmt joined
10:47 mr_bmt joined
10:49 mr_bmt joined
10:49 gmcabrita joined
10:49 mr_bmt joined
10:50 mr_bmt joined
10:51 mr_bmt joined
10:52 mr_bmt joined
10:53 mr_bmt joined
10:53 robinsjdotcom joined
10:54 cemilowski joined
10:57 jkreeftmeijer joined
11:03 milad joined
11:06 gem joined
11:06 cevado joined
11:12 will_ joined
11:12 milad joined
11:13 <will_> hey can anyone help with an Ecto query question?
11:14 codestorm joined
11:14 <Nicd-> don't ask to ask, just ask :)
11:14 <will_> im trying to do a nearest neighbour search with PostGis, as shown here
11:14 <will_> https://boundlessgeo.com/2011/09/indexed-nearest-neighbour-search-in-postgis/
11:15 <will_> is there a way to do `<#>`
11:15 <will_> trying to do this `order_by: fragment("? <#> ST_SetSRID(ST_PointFromText('POINT(? 123)', -1), 4362)", shop.geom, ^x)`
11:15 <will_> but keep getting the error parameters must be of length 0 for query %Postgrex
11:16 <will_> is there a way to accomplish this better?
11:16 nepiktaz joined
11:16 <micmus> will_: you have ? inside a string - postgres will see it only as a string with the ? character, not a place for a parameter
11:16 <ciawal> yeah, construct that whole string and pass that in instead
11:17 <will_> into the fragment?
11:17 <micmus> or use concatenation in postgres: 'POINT(' || ? || ' 123)'
11:17 <will_> I tried constructing the string above like "? <#> ST_SetSRID(ST_PointFromText('POINT(#{x} #{y})', -1), 4362)"
11:18 duane joined
11:19 <ciawal> but?
11:19 <will_> ** (Ecto.Query.CompileError) fragment(...) expects the first argument to be a string for SQL fragments, a keyword list, or an interpolated value, got: `query`
11:19 <micmus> will_: that would open you to SQL injection, so ecto won't allow it - query part of fragment has to be a literal
11:19 <will_> yeah I thought so
11:19 <ciawal> you should do as micmus said above instead
11:20 <will_> ok let me try
11:20 <will_> thanks
11:21 <micmus> or interpolate outside of query:
11:21 <micmus> fragment("? <#> ... ST_PointFromTest(?, ...", shop.geom, "POINT(#{x}, #{y})")
11:21 <micmus> sorry, should be ^"POINT...
11:21 <will_> is this what you means? `order_by: fragment("? <#> ST_SetSRID(ST_PointFromText('POINT(" || x || " " || y || ")', -1), 4362)", shop.geom),`
11:22 <ciawal> it should be ?s
11:22 <ciawal> instead of x/y
11:22 <ciawal> you're passing them into fragment
11:23 <ciawal> and not closing the query string
11:23 <ciawal> 'POINT(' || ? || ' ' || ? || ')'
11:23 <ciawal> if you want to pass both in
11:24 dimitarvp joined
11:24 <will_> quick one, why doesn't this work? order_by: fragment("? <#> ST_SetSRID(ST_PointFromText('POINT(? ?)', -1), 4362)", shop.geom, ^x, ^y),
11:25 <fredsir> Is there a good guide for setting up a new phoenix project using 1.3 RC, and things to be aware of an so on?
11:25 <will_> I thought the fragment should just interpolate x and y
11:25 <micmus> will_: the ? are inside a postgres string ''
11:25 <micmus> you can't do that
11:25 <will_> ah ok
11:25 PaReeOhNos joined
11:26 rschmukler joined
11:27 <will_> woohoo got it workind :D
11:27 <will_> constructed point = "POINT(#{x} #{y})"
11:27 <will_> then used that in the fragment
11:27 <will_> awesome
11:28 <fishcakez> is it simplier to use ST_Point(?, ?), ^x, ^y ?
11:29 <will_> ah yeah guess so, left over from my prev query
11:29 <fishcakez> ecto/postgrex is trying to avoid casting to text and back again
11:29 jkreeftmeijer joined
11:29 <will_> would there be a way to accomplish this without using fragments?
11:33 <micmus> will_: what does the <#> operator do? I can't find it
11:34 <ciawal> https://postgis.net/docs/geometry_distance_box.html
11:34 <nox> micmus: a <#> b tweets "b #a".
11:35 <micmus> the geo library has all the functions encoded as macros, but not operators. It seems there's no function to replace <#>
11:36 <micmus> if there were (you can write it yourself), it could be written as:
11:36 robinsjdotcom joined
11:38 gvaughn joined
11:39 <micmus> oh, hmm... you probably could use the Geo.Point struct directly
11:40 <ciawal> does it actually matter? should we be trying to avoid using fragment?
11:40 <will_> that's my question
11:41 <micmus> fragments are fine, wrapping them in macros if you use something a lot, can make the code a lot cleaner, though
11:41 <will_> also another, if I wanted to calculate the distance between the x,y passed in and the geom of the shop, would I do that as part of the postgres query, or in elixir itself?
11:41 chrismccord joined
11:43 akeating joined
11:49 will___ joined
11:50 nepiktaz joined
11:54 will___ joined
11:55 robinsjdotcom joined
11:57 cristianR_ joined
11:58 jschneck joined
12:00 <jschneck> Good morning (for me) everyone
12:00 <Nicd-> good evening jschneck
12:01 mattyw joined
12:01 <jschneck> I've slumped into a discussion yesterday that I could use some community opinion on.
12:01 mattyw joined
12:01 <jschneck> Its regarding a named gen server process that multiple dependencies are interested in running
12:02 <jschneck> Lets say we have apps A, B, and C. A depends on B and A depends on C. In addition, A, B and C each depend on D.
12:03 <jschneck> A, B, and C are all interested to make sure that lets say a specific named Registry is started in D. Therefore, it seams reasonable that they should all declare in their supervision trees that they want Registry [MyScope]
12:04 <josevalim> so they need to declare a registry that is going to be used by D?
12:04 <jschneck> well, I am using Registry as an example here.
12:04 <josevalim> why doesn't D declares the registry? and maybe you can use tuples in the registry to avoid conflicts betweens apps when necessary?
12:05 <jschneck> if I understand that correctly, D would start one big registry and they would use the fist part of a tuple to declare a scope
12:05 <jschneck> *the rest would use...
12:06 <jschneck> This was a choice I was leaning towards, but it got me thinking about apps which share interest in a named process running
12:07 <jschneck> it got me curious if worker/3 could take something like :ensure_started as an arg for situations where :already_started were returned
12:07 akeating joined
12:08 jkreeftmeijer joined
12:08 <jschneck> the pratical example of this above is that there are 3 apps which make up Nerves wifi and all of them want to push bits of info to a scope Nerves.Networking.
12:09 codestorm joined
12:10 <jschneck> so, in that case someone needs to make sure that registry is started. Its starting to feel that one big Nerves registry is more appropiate and that the sub namespacing happens under it.
12:11 akeating_ joined
12:12 PaReeOhNos joined
12:13 duane joined
12:16 <fishcakez> jschneck: unfortunately an ensure_started would likely cause circular application dependence
12:16 <fishcakez> because if a crash happens then the supervision trees are all going to rush to try to start that supervisor
12:16 <fishcakez> and then C and depend on A
12:17 <fishcakez> s/start the supervisor/start the worker/
12:17 <fishcakez> but you can implement this feature by doing case start_link do ... {:error, {:already_started, _}} -> :ignore end
12:18 <jschneck> yeah in that case there will still be a rush to start it, but the others should get :ok anyways since someone would've had to get there first
12:20 <fishcakez> it also increase the error kernel for the registry because if A, B or C manage to claim the registry, it possible that a crash in other processes in their tree can propagate up to terminate the regstry, this will cause processes in all the other apps to crash
12:20 <fishcakez> so you have very poor fault tolerance
12:21 rschmukler joined
12:22 <fishcakez> shutdown of the node also becomes messy I think because the app with the registry might get terminated first
12:22 <jschneck> thats a solid argument.
12:23 <fishcakez> apps are shutdown in the reverse order they are started iirc
12:26 mattyw joined
12:35 mr_bmt joined
12:41 akeating joined
12:43 aedigix joined
12:46 steffkes joined
12:47 akeating joined
12:50 inoas joined
12:51 rodolfojcj joined
12:54 cschneid_ joined
12:55 DeadTrickster joined
12:56 akeating joined
12:57 TinkerTyper joined
13:03 ramortegui joined
13:03 codestorm joined
13:05 sfbw joined
13:11 gazler joined
13:15 jimmyrcom joined
13:23 bpmcd joined
13:26 gvaughn joined
13:27 jmiven joined
13:31 <flupke> hi, I'm getting an error when trying to compile Elixir on a mac with erlang 19.3 (also tried 19.1): https://www.pastiebin.com/58ff4f7a99745
13:32 duane joined
13:33 <fishcakez> flupke: likely need to update to newer elixir
13:33 <fishcakez> looks like that source only works on =< OTP 18
13:36 <flupke> fishcakez: I'm compiling an up-to-date elixir master
13:36 <flupke> oh wait, my branch is not up-to-date, so sorry
13:36 <flupke> s/branch/fork/
13:38 lexmag joined
13:40 cschneid_ joined
13:41 gvaughn joined
13:41 r00tl3ss joined
13:44 hashpuppy joined
13:49 tomaz_b joined
13:49 jschneck joined
13:50 elixir-lang708 joined
13:53 duane_ joined
13:56 robinsjdotcom joined
13:56 cristianR_ joined
13:57 codestorm joined
14:00 cdg joined
14:04 douglascorrea joined
14:04 noark9 joined
14:05 vnz joined
14:05 vnz joined
14:10 gvaughn joined
14:10 rschmukler joined
14:10 jkreeftm_ joined
14:10 clem_ joined
14:11 gvaughn_ joined
14:16 brguy joined
14:20 willyk joined
14:20 zetashift joined
14:26 MarioBranco joined
14:32 gem joined
14:34 wyzzard joined
14:35 lexmag joined
14:35 papachan joined
14:39 cevado joined
14:43 proteusguy joined
14:44 gvaughn joined
14:45 gvaughn_ joined
14:47 dtusk joined
14:49 akeating joined
14:51 codestorm joined
14:54 justelex_ joined
14:56 julienfayad joined
14:58 cemilowski joined
14:59 <julienfayad> Hello! Wondering has anyone worked with SSHex ? I’m facing an issue in using public ssh keys to connect
14:59 <ciawal> I have julienfayad
15:00 <julienfayad> Hi ciawal maybe you can give me a hand
15:00 <julienfayad> as I can’t see what I’m doing wrong here and the ssh debug/verbose is quite limited (or at least I don’t know how to get more output).
15:00 <julienfayad> my problem boils down to being asked a password
15:01 <julienfayad> where as if a I ssh into the server from the terminal (same user that is running the elixir app) the connection establishes right away
15:03 icecreamcohen joined
15:04 tuacker joined
15:05 cschneid_ joined
15:05 <ciawal> can you show your code?
15:05 <ciawal> and the ssh command you run
15:06 <ciawal> with anything necessary blanked out ofc
15:06 <julienfayad> SSHEx.connect ip: to_charlist(Application.get_env(:sb_api_elixir, :server_ip)), user: to_charlist(Application.get_env(:sb_api_elixir, :server_user))
15:06 <julienfayad> that’s the connect command I use
15:07 ZippoWeb3 joined
15:07 <icecreamcohen> ‘morning peeps. In a monorepo, is it standard practice to include intra-repo deps like this {:dep_name, path: “../other_project”, env: Mix.env, override: true} ?
15:07 <Ankhers> I don't believe you need to call to_charlist/1 yourself.
15:07 <ciawal> you can try specifying user_dir: '/path/to/your/.ssh' julienfayad
15:08 <julienfayad> Hi Ankhers, I wasn’t initially but diggin into the SSHex I noticed there is an example where it’s being done
15:08 <julienfayad> so I gave it a shot
15:08 <gazler> icecreamcohen: I've seen it before. I prefer using an umbrella or a separate git repo.
15:08 <julienfayad> as for the user_dir: option I tried
15:08 <julienfayad> and same result, it asks for a password
15:08 <icecreamcohen> gazler: I don’t have that option ;-/
15:08 <Ankhers> julienfayad: I believe you need to specify the key file you want to use.
15:08 <Ankhers> It just passes the opts down the erlangs ssh module.
15:09 <Ankhers> (by default)
15:09 <gazler> icecreamcohen: Why not?
15:09 <icecreamcohen> because I just started at a new place and this is how they’ve organized their repo. It’s a single monorepo, and it’s been like that for a while.
15:09 <Ankhers> Though, reading the documentation, it doesn't sound like you need to do that.
15:10 beatpanic joined
15:10 <julienfayad> ok Ankhers I can specify the dir with user_dir and tried that, but that didn’t work
15:10 <julienfayad> by the way ciawal Ankhers I also tried to use SSHEx.ConfigurableClientKeys.get_cb_module
15:10 <ciawal> julienfayad: have you tried specifying the IP/user as literals
15:10 <gazler> icecreamcohen: That's no excuse! What you said is probably the best option then.
15:10 <ciawal> to check it's not a config issue
15:10 <icecreamcohen> lol, gazler
15:10 <ciawal> julienfayad: i.e. copy what you use in your terminal ssh command
15:11 <Ankhers> julienfayad: Do you by chance have a .ssh/config file that specifies a differnt key to use to for the host you are attempting to connect to?
15:11 <julienfayad> yes ciawal, I confirmed it’s not a config issue by putting the values as literals
15:11 <julienfayad> nop Ankhers, I checked that there is no config file for this user
15:14 <benwilson512> icecreamcohen: by new place do you mean no longer pinterest?
15:14 hexkey[m] joined
15:14 <icecreamcohen> benwilson512: Yep
15:14 TheGillies joined
15:14 logos[m] joined
15:14 krigare[m] joined
15:14 M107262[m] joined
15:14 mindflayer[m] joined
15:14 M04n0[m] joined
15:14 <benwilson512> icecreamcohen: very cool, where you at now?
15:14 dp[m] joined
15:14 m3tti[m] joined
15:14 OAK0[m] joined
15:14 malteof[m] joined
15:14 seequ_ joined
15:14 bin7me[m] joined
15:14 wilo[m] joined
15:14 nolan_d joined
15:14 notriddle joined
15:14 technikhil[m] joined
15:14 <icecreamcohen> Discord
15:14 uhoreg joined
15:14 spawnthink[m] joined
15:14 M-Quora joined
15:14 M-nickgal joined
15:14 <benwilson512> NICE
15:14 <icecreamcohen> Yeah; they use elixir a lot1
15:15 <benwilson512> that's what I hear
15:15 <icecreamcohen> :-)
15:15 <gazler> Thanks for reminding me, I haven't logged in in a while!
15:15 <icecreamcohen> There’s an elixir discord
15:16 <ciawal> oh awesome
15:16 <icecreamcohen> and, get this, it’ll keep its history indefinitely.
15:16 <ciawal> I love discord, congrats
15:16 <ciawal> I think all their backend is elixir right?
15:16 <icecreamcohen> https://discord.gg/Shk6E
15:16 <icecreamcohen> Yes!
15:16 <icecreamcohen> large portions of it are.
15:16 <ciawal> that's not all then :D
15:17 <icecreamcohen> hmm… not all; most. The lion’s share.
15:17 <ciawal> what's the remainder in?
15:17 <gazler> C++
15:17 <icecreamcohen> There’s a couple python services and a couple go ones.
15:18 <ciawal> I imagine they're the biggest elixir user atm
15:18 <icecreamcohen> Well, it depends on how you define "biggest"
15:19 <icecreamcohen> They’re probably company with the most users who are most dependent on elixir
15:19 <ciawal> sure, I mean company using elixir most, rather than biggest company using elixir
15:19 <icecreamcohen> I’d wager that Pinterest puts the most QPS through elixir.
15:19 whharris joined
15:23 meh` joined
15:28 cevado joined
15:33 codestorm joined
15:34 vmoravec joined
15:35 PaReeOhNos joined
15:47 rschmukler joined
15:53 asabil joined
15:54 davidw joined
15:54 <julienfayad> ciawal, Ankhers, no idea ?
15:55 <ciawal> not sure, sorry
15:55 <ciawal> does sshex provide any debug logging?
15:57 <Ankhers> julienfayad: Sorry, I looked at the ssh module a little bit, but had to get back to work :(
15:58 amclain joined
15:58 <Ankhers> I was having issues connecting to one of my servers because of timeout.
16:01 drincruz joined
16:02 MarioBranco joined
16:02 gem joined
16:02 <julienfayad> ciawal, I was looking for that but reading the connect method I can’t see anything related to debug
16:03 <julienfayad> Ankhers, thanks :-) my issue is different though, as I can ssh from the terminal, but when it’s the elixir script that tries to connect it prompts for a password (running the elixir script with the same user)
16:06 <Ankhers> julienfayad: What if you just tried passing your arguments to the ssh module, since that is what is being used?
16:07 <julienfayad> what do you mean Ankhers ?
16:07 <Ankhers> :shh.connect('hostname', 22, ssh_options)
16:07 <Ankhers> That is all SSHEx is doing.
16:07 asabil joined
16:10 gokr joined
16:10 rozap joined
16:12 meandi_2 joined
16:14 milad joined
16:17 jeffweiss joined
16:19 OtherAllan joined
16:21 baweaver left
16:31 imack joined
16:31 imack joined
16:32 PaReeOhNos joined
16:41 asabil joined
16:45 cdg_ joined
16:52 cdg joined
16:53 wsieroci joined
16:55 smeevil joined
16:56 jkreeftmeijer joined
16:57 gnatt joined
16:58 jkreeftm_ joined
16:59 codestorm joined
17:02 refriedchicken joined
17:03 rschmukler joined
17:04 nd___ joined
17:05 rschmukler joined
17:09 cevado joined
17:10 steffkes joined
17:10 mika__ joined
17:10 chrismccord joined
17:11 cdg joined
17:11 Ven joined
17:11 m_m joined
17:13 MarioBranco joined
17:16 PaReeOhNos joined
17:18 jkreeftmeijer joined
17:20 jkreeftm_ joined
17:20 rozap joined
17:23 gem joined
17:26 <bitmod_> what's the deal with the send() function, and why does send(self(), :ping) display in the terminal, even if you don't receive/flush self()?
17:27 steffkes joined
17:28 <Ankhers> bitmod_: Because `send/2` returns the message that was sent. If you were to send to a different process, it would still return the message.
17:28 <Ankhers> If you receive or flush now, you will still see the message.
17:28 <Ankhers> https://hexdocs.pm/elixir/Kernel.html#send/2
17:29 <bitmod_> Ankhers: i see, thanks
17:29 Gasher joined
17:32 jkreeftmeijer joined
17:34 will_ joined
17:35 jkreeftm_ joined
17:36 baweaver joined
17:38 gem joined
17:55 codestorm7 joined
17:57 <codestorm7> benwilson512: hi, we're deciding on camel vs snake case for REST and also GraphQL. I noticed that Absinthe does a translation back and forth. Is that a standard for GraphQL in general?
18:01 rschmukler joined
18:01 julienfayad joined
18:02 julienfayad joined
18:02 Ven joined
18:03 julienfayad joined
18:03 squallstter joined
18:04 julienfayad joined
18:05 julienfayad joined
18:11 icecreamcohen joined
18:11 MarioBranco joined
18:13 <notriddle> graphql-ruby has it, too.
18:14 <fredsir> When running a phoenix project with mix phx.server, I'm getting this repeatedly: `warning: mtime (modified time) for "lib/umbrella_web/templates/page/index.html.eex" was set to the future, resetting to now", what is up with that?
18:15 <notriddle> Actually, graphql-ruby camelization is still WIP: https://github.com/rmosolgo/graphql-ruby/pull/555
18:16 <ivan> fredsir: are your files on a remote computer?
18:16 <fredsir> ivan: inside a docker container
18:16 <benwilson512> codestorm7: I'm not sure, I think ruby is the other major implementation that would consider snake case language normal
18:16 <benwilson512> javascript and scala would both use camelcase internally
18:17 <benwilson512> codestorm7: it's actually super low overhead for us
18:17 <benwilson512> since we just re-use the camelcase strings that are given to us
18:17 <benwilson512> instead of calling a camelization function on every output key
18:17 JEG2 joined
18:18 <benwilson512> we do have to snake case incoming document field names but that isn't particularly expensive, and scales with merely the size of the document, vs the size of the results
18:18 <benwilson512> we
18:18 <benwilson512> *we've considered an optimization where we precompute camelized field names on all the schema types at compile time
18:18 <benwilson512> and then do lookups off of those, but I don't think that would be a huge win
18:20 <benwilson512> codestorm7: what's cool actually is that you could have different endpoints have different conventions, but with the same schema
18:20 <benwilson512> so you could have like /api/web that used the camelcase stuff and your schema for JS clients, and then /api/whatever that didn't
18:20 <benwilson512> and used snake case for snake case clients
18:20 <benwilson512> hell, I'm pretty sure you could make it header based if you wanted to, we're pretty flexible
18:21 <bitmod_> what were the primary factors in making brunch the official asset management tool, as opposed to webpack?
18:22 <benwilson512> bitmod_: it might be more accurate to call it the default, phoenix itself is happy to work with any asset management tool
18:22 <benwilson512> IIRC it was seen as simpler
18:25 chrismccord joined
18:30 cdg_ joined
18:32 icanhazbroccoli joined
18:32 codestorm7 joined
18:33 <codestorm7> sorry, benwilson512, I disconnected and lost the scroll,
18:33 <codestorm7> I'm not sure I follow you as far as what would be a win, so let me rephrase a little. When I In Elixir, I'm using all snake case. When I make GQL requests, the field names (even from autocomplete) are camel case
18:33 <codestorm7> so something is doing a translation now
18:35 gokr joined
18:37 Ven_ joined
18:39 PaReeOhNos joined
18:44 ZippoWeb3 joined
18:48 rodolfojcj joined
18:50 wsieroci joined
18:50 NeverDie_ joined
18:50 d10n-work joined
18:53 <bitmod_> benwilson512: ah ok, thanks
18:59 meandi joined
19:00 nath joined
19:07 kronicdeth joined
19:09 <kronicdeth> With ecto, if I want to do `where: c.expires_on + 1 > type(^iso8601_formatted_date_time, :date)`, is the only way to +1 the date is to drop into a fragment? I'm getting `(Ecto.Query.CompileError) `c.expires_on() + 1` is not a valid query expression.`
19:09 <kronicdeth> wait, nvm, fround date_dat
19:09 <ivan> kronicdeth: you could do math on the date instead?
19:10 <kronicdeth> *date_add
19:10 <ivan> ah
19:15 steffkes joined
19:15 whharris joined
19:20 Cohedrin joined
19:20 phansch joined
19:20 lattenwald joined
19:21 <bitmod_> what is Application.Endpoint?
19:32 dch_ joined
19:32 josevalim_ joined
19:36 AndChat-282900 joined
19:38 cevado joined
19:41 icecreamcohen joined
19:42 nath__ joined
19:45 miwa joined
19:45 miwa joined
19:47 lexmag joined
19:47 nath__ joined
19:48 nath__ left
19:50 sfbw joined
19:52 sfbw_ joined
19:53 PaReeOhNos joined
19:53 sfbw joined
19:54 meh`_ joined
19:54 sfbw joined
19:55 sfbw_ joined
19:57 sfbw joined
19:58 sfbw_ joined
19:59 sfbw joined
20:00 sfbw_ joined
20:01 sfbw joined
20:02 sfbw_ joined
20:03 sfbw joined
20:03 asabil joined
20:04 jdqx joined
20:04 sfbw joined
20:05 sfbw joined
20:06 sfbw_ joined
20:07 sfbw joined
20:08 sfbw_ joined
20:09 sfbw joined
20:10 sfbw_ joined
20:11 sfbw joined
20:15 devbug joined
20:16 <devbug> Mix won't let me override a dependency?
20:16 <devbug> https://gist.github.com/anonymous/23909de54c8c56142d6815392da6e2d4
20:17 dastagg joined
20:27 mbwe joined
20:29 mbwe joined
20:29 cdg joined
20:34 zetashift joined
20:35 PaReeOhNos joined
20:38 <fredsir> Does `phx.gen in 1.3 have an understanding of umbrella apps, so say I run `phx.gen.html`, will it put the views and templates and so into the xxx_web app and the "rest" into the "core" app?
20:38 milad joined
20:38 <fredsir> And when I say umbrella apps, I mean umbrellas created with `mix phx.new --umbrella`
20:44 <fredsir> seems not so
20:46 cevado joined
20:47 wsieroci joined
20:47 refriedchicken joined
20:49 strykerkkd joined
20:50 MarioBranco joined
20:53 Ven joined
20:53 PaReeOhNos joined
20:56 ikcerog joined
21:04 NeverDie joined
21:04 kronicdeth joined
21:10 dch_ joined
21:15 justelex joined
21:19 steffkes joined
21:27 wpcarro joined
21:28 asabil joined
21:28 <wpcarro> Hey all. Is there a preferred way to override a project's dependency's config variable?
21:29 <wpcarro> I'm using a dependency that reads a credentials file when starting up the application. I'd like to instead have that variable be defined by my project instead of leaking my dependency's variable through to the end consumer... is this possible?
21:29 jlouis joined
21:29 cevado joined
21:31 dch_ joined
21:33 <wpcarro> I think I could resolve the issue if I could prevent one of my dependencies, which is an application, from starting automatically. Does this seem reasonable to anyone?
21:40 gem joined
21:48 cevado joined
21:54 kronicdeth joined
21:58 myronmarston joined
22:00 <myronmarston> bitwalker: I’m upgrading our app from exrm to distillery and have run into an issue I haven’t been able to solve. Do you have a few minutes to chat?
22:01 tuacker joined
22:07 rschmukler joined
22:07 <benwilson512> wpcarro: the dependency hard codes the file?
22:07 <benwilson512> what exactly does it do with it, can you link to the code?
22:23 rschmukler joined
22:38 wpcarro joined
22:50 rschmukler joined
22:52 wpcarro joined
22:55 PaReeOhNos joined
23:05 kronicdeth joined
23:11 Cohedrin joined
23:12 smt_ joined
23:13 <bitmod> benwilson512: is something to do with models chaning in phoenix 1.3?
23:13 <bitmod> changing*
23:14 elgenie joined
23:15 cschneid_ joined
23:20 steffkes joined
23:21 <benwilson512> bitmod: the brunch stuff?
23:21 rschmukler joined
23:22 <bitmod> benwilson512: ecto models
23:22 <benwilson512> bitmod: what's the context?
23:22 <benwilson512> oh
23:22 <benwilson512> sorry I get what you're saying
23:22 <benwilson512> yes
23:23 <benwilson512> https://www.youtube.com/watch?v=tMO28ar0lW8
23:23 <bitmod> benwilson512: that's the one, thanks
23:32 wpcarro joined
23:33 myronmarston left
23:34 <Radar> >> DateAttr = Struct.new(:day, :foo); attrs = [DateAttr.new(1,2), DateAttr.new(3,4)]; attrs.detect { |a| a.foo == 4 }
23:34 <Radar> Whoops.
23:39 wpcarro joined
23:43 akeating joined
23:44 wpcarro joined
23:47 miwa_ joined
23:49 icecreamcohen joined
23:52 jhack joined
23:54 cevado joined
23:54 refriedchicken joined
23:56 cschneid_ joined
23:58 wpcarro joined
23:58 mika__ joined