00:15 <FromGitter> <akzhan> looks ike we need need hashed from @funyy-falcon and will what em all… lustly joke

00:34 <RX14> well

00:34 <RX14> looks like the crystal arch package is on 0.24.0

00:35 <RX14> which isn't even a real release

00:35 <RX14> such a mess

01:07 <FromGitter> <jwoertink> Is there a way to call one macro from another?

01:11 <Papierkorb> just use the macro in a macro?

01:11 <FromGitter> <jwoertink> I get `can't nest macro expressions`

01:12 <Papierkorb> "No code no help"

01:14 <FromGitter> <jwoertink> ```code paste, see link``` [https://gitter.im/crystal-lang/crystal?at=5a18c40571ad3f873628a46f]

01:15 <FromGitter> <jwoertink> `undefined local variable or method 'a`

01:15 <FromGitter> <jwoertink> which would make sense here, and I understand why

01:15 <FromGitter> <jwoertink> but this is the most simple way I can show "call one macro from another"

01:16 <Papierkorb> If you understand why, why don't you post working code?

01:16 <FromGitter> <jwoertink> lol. because If I had working code, I wouldn't be asking the question

01:17 <Papierkorb> `::Thing.a` in b.

01:17 <FromGitter> <jwoertink> ah!

01:17 <FromGitter> <jwoertink> awesome! Thanks.

01:17 <Papierkorb> macros are *expanded*. They work on the AST. They're expanded locally where you "call" them (hence we say macros are expanded, not called or invoked).

01:18 <Papierkorb> Thus, in a macro body, you have to take into account that your code is actually living in the scope of the users code, and not in your class

01:18 <FromGitter> <jwoertink> oh, ok. Yeah, that makes sense.

03:13 <FromGitter> <codenoid> morning crystaller

04:01 <bmcginty> I was just going through some of the crystal code, trying to modernize txe's crystal-win branch to get Crystal to run with the current commit. However, I see there are all these crystal::system changes. I'd like to assist with the windows porting, but don't want to do work that's going to just be scrapped as others have partiall done it already. Anyone have suggestions?

11:11 <FromGitter> <extremety1989> https://github.com/will/crystal-pg/issues/119

11:11 <FromGitter> <extremety1989> hope someone will help me =)

14:41 <maxpert> hmmm

14:45 <FromGitter> <lwakefield> Hey, I am sure I am going about this the wrong way, but what is the easiest way to create a `JSON::Any` from a hash literal: ex. `JSON::Any.new({"foo": "bar"})`

14:46 <FromGitter> <extremety1989> @lwakefield have you tried it on postgres ?

14:48 <FromGitter> <lwakefield> No, but unless I am missing something @extremety1989, I don't see how that is related?

14:51 <FromGitter> <extremety1989> @lwakefield well

14:51 <FromGitter> <extremety1989> im trying to insert json data to postgres

14:51 <FromGitter> <extremety1989> using token field as jsnb

14:51 <FromGitter> <extremety1989> jsonb

14:51 <FromGitter> <extremety1989> ```code paste, see link``` [https://gitter.im/crystal-lang/crystal?at=5a19837c8b3a9e2c0c182830]

14:52 <FromGitter> <extremety1989> in my database tokens table, the field token is jsonb

14:53 <FromGitter> <lwakefield> I think we may be talking at cross purposes @extremety1989, I am asking a separate question :)

14:54 <FromGitter> <extremety1989> @lwakefield aa ))

14:59 <FromGitter> <bew> @lwakefield don't think there's another way, but why would you want a JSON::Any object?

15:11 <FromGitter> <unreadable> some fun with shaders and sine function

15:11 <FromGitter> <unreadable> http://thebookofshaders.com/edit.php?log=171125151933

15:16 <FromGitter> <lwakefield> @bew bummer. Ideally it would be a convenience thing, for testing some framework code.

15:17 <FromGitter> <lwakefield> I suppose serializing & deserializing isn't that big of a deal if it is just testing code.

15:23 <Papierkorb> lwakefield, there's a convenient thing to do it: Simply `JSON.parse` a json document. For testing I this may be fine.

15:24 <Papierkorb> Note that you can use `%<...>` instead of double-quotes, so you don't have to unescape them in the document when writing it manually.

15:25 <FromGitter> <lwakefield> Oh nice, that makes much more sense than having the overhead of serializing.

15:25 <FromGitter> <lwakefield> Thanks @Papierkorb!

16:44 <hightower3> Is it expected that 0.24.0 prerelease reports "Crystal 0.23.0+313 [ea4187c57] (2017-10-27)" in crystal -v ?

16:55 <FromGitter> <straight-shoota> hightower3, yes

16:57 <FromGitter> <straight-shoota> but I think this could be improved

20:21 <FromGitter> <MrSorcus> https://github.com/crystal-community/leveldb - why here PR doesn't accepted? This shard failed when building, but in PR available fix. (unexpected token: as) change line 9 to `@err_ptr = pointerof(@err_address).as(Pointer(UInt64))`

20:22 <oprypin> I don't understand what you're saying

20:26 <oprypin> MrSorcus, perhaps your version of Crystal is outdated?

20:27 <FromGitter> <MrSorcus> > **<oprypin>** @MrSorcus, perhaps your version of Crystal is outdated? ⏎ ⏎ ```code paste, see link``` [https://gitter.im/crystal-lang/crystal?at=5a19d24b540c78242d4a58aa]

20:28 <oprypin> MrSorcus, it's likely that your download of the shard doesnt have the latest code

20:28 <oprypin> because the released version is old

20:28 <oprypin> do you use shard.yml ? maybe specify branch: master for the dependency

20:29 <bmcginty> I'm going through the windows code that txe and others have written, and just planning on working on everything until specs pass. Is that a good plan, or is someone else working on any part of it currently?

20:30 <oprypin> bmcginty, well there's the opinion that this should be planned and done gradually and that the fork is messy,

20:30 <oprypin> but nobody's actually doing the work

20:31 <oprypin> i'd love to merge any such fork but i dont have a say

20:34 <bmcginty> oprypin: Okay. I couldn't eve figure out which way the crystal::system thing is headed, or I'd pull all my changes to the latest commit.

20:34 <FromGitter> <MrSorcus> > **<oprypin>** do you use shard.yml ? maybe specify branch: master for the dependency ⏎ ⏎ Oh, you right. Thank you.

20:35 <oprypin> well it's staying in, unfortunately, and not really moving anywhere, like everything else

20:37 <bmcginty> I'll try to get all this working as quick as I can and maybe that'll be enough to push things forward a bit. Though contributers to the original windows PR have done the difficult bits with stacks and the like.

20:38 <oprypin> bmcginty, well you can keep doing the work, just don't expect it to be merged ¯\_(ツ)_/¯

20:48 <FromGitter> <MrSorcus> https://crystal-lang.org/api/0.23.1/OpenSSL/Digest.html#new%28name%2Cctx%3ALibCrypto%3A%3AEVP_MD_CTX%29-class-method ⏎ Broken link.

20:50 <oprypin> MrSorcus, it's really just broken API. lib things are not supposed to show up in docs

20:57 <FromGitter> <jwoertink> Hey @oprypin, can you explain how this works? https://github.com/oprypin/crsfml-examples/blob/master/2048/2048.cr#L325 Mainly just the fact that you're doing addition with a tuple. Is there an override somewhere that adds a `def +(nums : Tuple(Int32, Int32))`?

20:58 <oprypin> jwoertink, https://github.com/oprypin/crsfml/blob/sources/src/system/system.cr#L74

20:58 <FromGitter> <jwoertink> Oohh! Ok. Vectors!

20:58 <FromGitter> <jwoertink> Thanks :D

20:59 <oprypin> it's kind of a hack, really should be `SF.vector2(0, -75)` but too neat to pass up

20:59 <FromGitter> <jwoertink> haha yeah. That makes sense. This whole thing is pretty wild.