00:54 <FromGitter> <Blacksmoke16> hey @vladfaust, is there a way to provide the actual ID of a FK relation?

00:54 <FromGitter> <Blacksmoke16> ```code paste, see link``` [https://gitter.im/crystal-lang/crystal?at=5d019ec2c7a8fc263a136f5a]

00:55 <FromGitter> <Blacksmoke16> `Group.new(id: 1, name: "Ore", category_id: 1)`, or do i have to provide it a `Category` type im assume?

00:56 <FromGitter> <Blacksmoke16> currently im just defining the raw column, and using a method to return the actual category

01:51 <FromGitter> <watzon> Would a sigmoid function be classified as calculus or linear algebra?

01:51 <FromGitter> <watzon> I'm leaning more towards the latter

04:23 <FromGitter> <Nbotz> How would i send Time.local to all clients in a websocket?

04:23 <FromGitter> <Nbotz> through ECR and JS?

07:33 <FromGitter> <girng> what is ecr?

07:34 <FromGitter> <girng> should be able to just send a server timestamp to all the clients

07:47 <FromGitter> <bew> ECR is for Embedded CRystal, it's a compile time templating system.. @Nbotz I don't see how that would help though, if you a have a websocket on the backend, you could simply send a message from Crystal with the data you want

07:55 <FromGitter> <alex-lairan> I see a lot of Crystal matrix shards, but no one is realy more used than other. ⏎ ⏎ Maybe we could have one shard on `crystal-community` that handle Matrix, and all matrix repo owners join they forces ? ⏎ ⏎ :) [https://gitter.im/crystal-lang/crystal?at=5d020172c12fd52664fc10b3]

08:03 <FromGitter> <yxhuvud> there is a bunch of matrix shards based on integrating with external math libraries. Perhaps it is time someone built a native matrix shard, focusing on nice api, while keeping the integration against external libraries separate (but still reasonably feasible to implement)

09:05 <FromGitter> <akrejczinger_gitlab> How can I create a spec for testing the print output of a program? I tried to find a way to reroute output to a file but I don't see a way to do it.

09:58 <FromGitter> <vladfaust> @Blacksmoke16 there is an issue for that -- https://github.com/onyxframework/sql/issues/92#issuecomment-488242220

09:59 <FromGitter> <vladfaust> Also `Group.new(id: 1, name: "Ore", category: Category.new(id: 1))`, which doesn't harm performance a lot

10:55 <livcd> ararchy: did you check the Cyber 2.0 ? It sounds like bullshit...

11:12 <FromGitter> <MrSorcus> Hi, why https://github.com/crystal-lang/crystal/issues/6754 can't be done?

13:03 <FromGitter> <straight-shoota> @MrSorcus It *can* be done, but supporting LLVM 7 is not a high priority right now. You can simply use an older version.

13:04 <FromGitter> <MrSorcus> @straight-shoota oh, ok. Thanks for answer.

16:28 <FromGitter> <girng> wtf

16:29 <FromGitter> <girng> why would yuou need ECR to send a message to a client from a server

16:37 <FromGitter> <watzon> @alex-lairan I'm trying to write an all-encompassing linear algebra library at watzon/apatite (https://github.com/watzon/apatite). I would love to see a Vector and Matrix implementation in the standard library though.

18:43 <FromGitter> <wontruefree> is there a reason you are monkey patching parts of the stdlib?

18:44 <FromGitter> <wontruefree> I just found cadmium. It looks really cool

18:47 <FromGitter> <watzon> Adding methods that should be there and aren't. I'll probably refactor that stuff out though. Monkey patching always feels dirty.

20:54 <pracabor> searching the https://crystal-lang.org/api/0.29.0 docs I don't see the equivalent of https://docs.ruby-lang.org/en/2.6.0/String.html#method-i-unpack . The semantics of pack/unpack are admittedly opaque and difficult to grok -- is this functionality implemented elsewhere under some other method name?

20:57 <FromGitter> <watzon> There is no unpack in Crystal unfortunately

20:58 <FromGitter> <watzon> There is a shard I believe that replicates some of the functionality that `String#unpack` has though

21:01 <FromGitter> <straight-shoota> In Crystal most use cases for String#unpack can be implemented using IO#read_* methods which is more flexible and doesn't require to load all data into a string

21:05 <pracabor> good to know. Perhaps I'm overthinking this... Given that I'm trying to take a string and operate on it in a bitwise way, do I even need thi