11:01 <FromGitter> <jmoriau> I randomly get "#<Errno:0xa59c40 @message="SSL_shutdown: Operation now in progress", @cause=nil, @callstack=CallStack(@callstack=[Pointer(Void)@0x414fbe, Pointer(Void)@0x449ac7, Pointer(Void)@0x48ddc5, Pointer(Void)@0x4f33b0, Pointer(Void)@0x505895, Pointer(Void)@0x417072, Pointer(Void).null], @backtrace=nil), @errno=115>" every once in a while is it worth mentioning in an issue or is it due to my setup?

11:01 <FromGitter> <jmoriau> (I have a long running process handling the lights in my apartment and every few days the process stops with this error)

11:06 <RX14> huh

11:07 <FromGitter> <jmoriau> (it's most likely while making a POST call to LIFX's servers)

11:07 <RX14> @jmoriau can you print that backtrace properly?

11:07 <FromGitter> <jmoriau> well in few days sure...

11:07 <RX14> *sigh*

11:07 <FromGitter> <jmoriau> just have to wait for it to happen again

11:08 <FromGitter> <jmoriau> is `puts e.backtrace` enough for next time ?

11:09 <RX14> seems like a crystal bug

11:09 <RX14> http://stackoverflow.com/questions/10204134/tcp-connect-error-115-operation-in-progress-what-is-the-cause

11:09 <RX14> well there's a method on exception which prints it properly

11:10 <FromGitter> <jmoriau> Uh had no idea which one is it

11:10 <RX14> inspect_with_backtrace(STDOUT)

11:10 <FromGitter> <jmoriau> ok cool thanks well I'll see in a few days then

11:11 <RX14> no make a bug report for it

11:11 <FromGitter> <jmoriau> ok

11:11 <RX14> it seems to me like an error in crystal's interction between nonblocking sockets/evented io and openssl

11:12 <RX14> interaction*

11:13 <FromGitter> <jmoriau> Well I'll open an issue and include a proper backtrace when I get one then

11:14 <FromGitter> <ssaatya> How do you use base level socket

11:14 <jhass> ssaatya: I don't get your question, sorry

11:15 <FromGitter> <ssaatya> How can you use TCPSocket first i wrote to socket next i read from socket i didnt get anything

11:16 <jhass> share what you got?

11:17 <FromGitter> <ssaatya> It is at https://carc.in/#/r/17at

11:20 <jhass> you need two \r\n after the headers

11:20 <FromGitter> <ssaatya> Ok thanks for your help once again

11:21 <RX14> also you should phobably use HTTP::Client

11:21 <RX14> instead of writing your own

11:47 <FromGitter> <ssaatya> Ok it doesnt redirect . It passes url directory with protocol and doesnt redirect

11:47 <FromGitter> <ssaatya> directly

15:21 <Papierkorb> Does the compiler already do automatic inlining?

15:23 <tilpner> Even if the crystal frontend itself doesn't, LLVM does a lot more than that.

15:25 <crystal-gh> [crystal] ysbaddaden opened pull request #3170: [WIP] Add file:line information to backtraces (master...core-callstack-refactor) https://git.io/v6oca

15:32 <jhass> as said we let LLVM do the inlining, there's the @[AlwaysInline] attribute to force it. yield statements are always inlined

15:36 <Criskov> Hello Crystal community!, Does Crystal have BigMath.E or BigMath.PI?

15:38 <jhass> >> Math::PI

15:38 <DeBot> jhass: # => 3.1415926535897931 - https://carc.in/#/r/17bu

15:38 <jhass> >> Math::E

15:38 <DeBot> jhass: # => 2.7182818284590451 - https://carc.in/#/r/17bv

15:39 <jhass> that's all I see

15:44 <Criskov> Ok, thanks jhass, Crystal rules! :-D

17:25 <FromGitter> <binary132> jhass: I saw that discussion, but it doesn't seem like native threading is more than a twinkle in someone's eye at the moment..... that conversation was Christmastime

17:25 <FromGitter> <binary132> *should* I have any expectations about timeframe to see it land?

17:26 <jhass> I guess better not tbh

17:26 <jhass> I expect something to happen this year still personally, but that's just my personal gut feelings

17:27 <FromGitter> <binary132> ok, that clarifies. thanks for the honesty :) I'll keep an eye out!

17:34 <crystal-gh> [crystal] marciogm opened pull request #3172: returns nil if xpath_nodes is empty (master...fix-xpath_node) https://git.io/v6oXx

23:20 <jeromegn> anybody can help with this macro error? https://gist.github.com/jeromegn/a68ceac28afb6c0c8e9c24eea4240f97 (at the bottom)

23:25 <jeromegn> says it was used, but it wasn’t. just trying to initialize it.

23:56 <willl> jeromegn: it can't know that you'll actually have a case 1, so it's nillable

23:56 <jeromegn> yea, I just figured out I should follow more closely the JSON.mapping example :)

23:56 <jeromegn> seems to work for my use case