ChanServ changed the topic of #crystal-lang to: The Crystal programming language | | Crystal 0.20.5 | Fund Crystal's development: | Paste > 3 lines of text to | GH: | Docs: | API: | Logs:
unshadow_ has joined #crystal-lang
unshadow has quit [Ping timeout: 258 seconds]
akwiatkowski has quit [Ping timeout: 240 seconds]
<shelvacu> I'm using emacs-crystal-mode, I haven't noticed any speed issues.
<Dreamer3> dyld: Library not loaded: /usr/local/opt/bdw-gc/lib/libgc.1.dylib
<Dreamer3> is there anything else i can pass to crystal build to make it look in the correct location?
<Dreamer3> i tried DYLD_FALLBACK_LIBRARY_PATH but either i'm doing it wrong or that's nto working
<Dreamer3> i'm trying to install via homebrew
sz0 has quit [Quit: Connection closed for inactivity]
<Dreamer3> is there a better channel?
rtl has quit [Remote host closed the connection]
<FromGitter> <MaxLap> Is there any doc explaining a little bit the inner details of the compiler?
sz0 has joined #crystal-lang
<FromGitter> <TheLonelyGhost> I'v spent far too much time messing with cross-compiling crystal to Alpine linux (short story, no luck so far), and a lot of the docs for things like libgc are in the wiki on github
<FromGitter> <TheLonelyGhost> libgc is actually BOEHM-GC
<FromGitter> <TheLonelyGhost> **is provided by
<FromGitter> <TheLonelyGhost> Does that answer anything?
<FromGitter> <MaxLap> Not really. I mean the steps that crystal does when compiling, how crystal interacts with llvm. I see it creates lots of ASTNodes, but i fail to see how it goes from that to llvm. That kind of inner details.
<FromGitter> <TheLonelyGhost> AFAIK, once upon a time the compiler was written in Ruby to compile to LLVM and into a binary. It eventually could use that previous compiler binary to compile a new version of itself. That in-between self-hosted magic is a big mystery to me too and I would definitely like some more clarification on it.
<FromGitter> <TheLonelyGhost> If you're feeling zesty, might try looking at *really* old copies of the compiler when it was written in Ruby...
<FromGitter> <MaxLap> Hum, that's not a bad idea, thnx
<FromGitter> <TheLonelyGhost> I'm not driven enough to do that research, but I'd read the heck out of a blog post if someone does
<FromGitter> <TheLonelyGhost> no prob
ome has joined #crystal-lang
mgarciaisaia has joined #crystal-lang
mgarciaisaia has quit [Ping timeout: 255 seconds]
mgarciaisaia has joined #crystal-lang
<Dreamer3> how long should it take to build crystal?
<Dreamer3> seems to be taking several minutes now
jinks7 has joined #crystal-lang
<jaitaiwan> Hey guys, just wondering about, why does one of its arguments: "args" not have a type associated with it? What sort of data structure does it accept
jinks7 has left #crystal-lang [#crystal-lang]
_jinks7 has joined #crystal-lang
_jinks7 is now known as jinks7
mgarciaisaia has quit [Ping timeout: 264 seconds]
<jinks7> jaitaiwan: i think its safe to assume its an array
<jaitaiwan> jinks7: bastard :|
<jaitaiwan> :P
<jinks7> jaitaiwan: just trying to help ;)
<jaitaiwan> when your colleagues get on IRC just to troll you
<wmoxam> Dreamer3: it shouldn't take much more than a couple of minutes
<wmoxam> It takes about 45 seconds on my desktop (relatively recent Intel I3 + SSD)
<Dreamer3> ok so it's broken somehow
<Dreamer3> it was complaining about the GC lib until i linked .brew/opt to /usr/local/opt
<Dreamer3> who knows what it's doing
* Dreamer3 sighs
<bmcginty> Dreamer3: hey. if you've got a symlink to crystal anywhere else on your $PATH, world-ending things happen when you try to compile... I used to keep a link to my copy of crystal git repo in ~/bin/crystal, but finally renamed to crystal2 so I could compile again.
<Dreamer3> ok well now my old crystal is gone
<Dreamer3> but that didn't help things any
<FromGitter> <martinium> how can I loop my program permanently
<FromGitter> <martinium> basically I want it to keep restarting
<FromGitter> <martinium> meh looping over and over until someone does a SIGINT
<Dreamer3> what would crystal getting stuck in a loop mean?
<FromGitter> <martinium> hmm top level namespace has a loop method
<FromGitter> <martinium> let me try that
<FromGitter> <martinium> got it
buggs has quit [Ping timeout: 256 seconds]
<FromGitter> <martinium> ```loop do ⏎ function ⏎ end``` []
<FromGitter> <martinium> that worked for me yay
<Dreamer3> grrrr
<Dreamer3> it compiles file outside of homebrew
<Dreamer3> ok
<Dreamer3> it's building a released
<Dreamer3> it gets 90% done and then spins
<Dreamer3> judging from the # of [warn] kq_init: detected broken kqueue; not using.: Undefined error: 0
<Dreamer3> is there a verbose compile?
<FromGitter> <mjago> @Dreamer3 have you plenty of ram? Maybe open an issue on Github
<Dreamer3> yes i have plenty of ram
buggs has joined #crystal-lang
<Dreamer3> some sort of linking issue
<Dreamer3> hmm
<Dreamer3> maybe
<Dreamer3> now i can get the non-release to build with homebrew also
<Dreamer3> so my issue is just with releas
<FromGitter> <mjago> which llvm version? I read about something like that under issues
<Dreamer3> "You will also need LLVM 3.5 or 3.6 present in the path. "
<Dreamer3> is this still an issue?
<Dreamer3> i have 3.9 and homebrew seems fine with that
<FromGitter> <mjago> @Dreamer3 have you seen this?
<FromGitter> <mjago> @dreamer3 maybe "brew install libevent --HEAD && brew install crystal-lang -v” (from home-brew issue)
akwiatkowski has joined #crystal-lang
<Dreamer3> looking
<Dreamer3> i can try that but i'm not getting a successful compile at all
<Dreamer3> and the kqueue errors don't seem to be the problem
<Dreamer3> trying it not with new libevent
<Dreamer3> now libevent just fails to be found by the dynamic linker
<Dreamer3> version numbers are diff
<FromGitter> <mjago> @dreamer3 Any progress?
<Dreamer3> FromGitter: no
<FromGitter> Dreamer3, I'm a bot, *bleep, bloop*. I relay messages between here and
<Dreamer3> mjago: uh
<Dreamer3> mjago: does this work
<FromGitter> <mjago> @Dreamer3 I had hell of a time getting Crystal set up on Mac 10.9.2 with homebrew
<Dreamer3> it used to be a pain, can't believe it sitll is
<FromGitter> <mjago> I got there in the end and can now compile the latest easily
<FromGitter> <mjago> Definitely open an issue - will probably get sorted much quicker - I don’t know enough about homebrew etc
<Dreamer3> i opened an issue on homebrew
<FromGitter> <mjago> Also I can’t find a record of what I did to fix in the end (I would usually make a note)
<Dreamer3> what is so special about compiling release vs not?
<FromGitter> <mjago> You have a working crystal compiler
<FromGitter> <mjago> crystal is self-hosted
<Dreamer3> so what do you get if ou don't compile release?
<Dreamer3> "release mode"
<FromGitter> <mjago> if you do make crystal it is release I believe ?
<Dreamer3> there is make and then make release=true
<Dreamer3> make works fine
akwiatkowski has quit [Ping timeout: 240 seconds]
<FromGitter> <mjago> debug info?
<FromGitter> <mjago> not sure
<FromGitter> <mjago> but I haven’t noticed issues with compilation speed
<FromGitter> <mjago> trying now make clean crystal release=true
<FromGitter> <mjago> taking much longer so I assume it is llvm optimisations of the compiler itself
<Dreamer3> 20.5+2 ?
<FromGitter> <mjago> current HEAD
<Dreamer3> are you using homebrew?
<FromGitter> <mjago> make clean crystal release=true - so no
<Dreamer3> what compiler are you starting with to build
<Dreamer3> are you on mac?
<Dreamer3> you have the other dependencies in homebrew though, right?
<FromGitter> <mjago> yes - CC=clang++ (I added that to makefile to avoid using gcc)
<FromGitter> <mjago> @Dreamer3 I think most of my original problems came from llvm but I think I ended up using homebrew
<Dreamer3> i'm just trying to build it the "normal" way
<Dreamer3> is that not with llvm?
<Dreamer3> it uses that for it's compilerl i thought
<FromGitter> <mjago> back-end compiler yes - but make crystal is the normal way I believe
<FromGitter> <mjago> and crystal is written in crystal so you use the crystal compiler (which utilizes llvm as does clang)
<Dreamer3> homebrew is just calling make crystal
<Dreamer3> and trying to build with the prior version
<Dreamer3> though i tried to build with the current binary also, no diff
<FromGitter> <mjago> ahh ok
<Dreamer3> do yu have your homebrew stuff in /usr/local?
<FromGitter> <mjago> yes I believe so
<FromGitter> <mjago> I fight with homebrew since I’m on an old OS (Mavericks)
<Dreamer3> i can try head now
mark_66 has joined #crystal-lang
akwiatkowski has joined #crystal-lang
<Dreamer3> grrrr
<Dreamer3> i removed my symlinks and now i'm just back to the stupid dylib errors
<FromGitter> <mjago> which?
<Dreamer3> dyld: Library not loaded: /usr/local/opt/bdw-gc/lib/libgc.1.dylib
<Dreamer3> because my library is in .brew
<Dreamer3> yeah i install homebrew in .brew
<Dreamer3> i symlinked a few things earlier but then i just hit the other problems
<Dreamer3> so who knows if the root problem might be you can't symlink such things
<Dreamer3> hence my trying to figure out how to tell it where the correct location is
<FromGitter> <mjago> yes
<FromGitter> <mjago> I’ve seen this issue
<Dreamer3> and?
<Dreamer3> DYLD_LIBRARY_PATH should fix this but does not
<Dreamer3> i commented on the bottom
<Dreamer3> if you have a specific suggestion i'm all ears
<FromGitter> <mjago> just looking at the makefile! so you tried symlinking from /usr/local/opt/
<Dreamer3> yeah and that's when i get the other errors
<Dreamer3> so i'm not sure if i'm moving forward or not
<FromGitter> <mjago> I can’t believe building on OSX is such a minefield since some of the core developers use it
<Dreamer3> yeah
<Dreamer3> can anyone explain the diff between release=true and not?
bjz has joined #crystal-lang
<FromGitter> <mjago> is your libevent version 2.0.22 ?
<FromGitter> <mjago> Cellar/libevent/2.0.22
<Dreamer3> yes
<Dreamer3> runnings specs now
<Dreamer3> without release
<Dreamer3> damn takes long enough
<Dreamer3> ....
<FromGitter> <mjago> there’s quite a few - running with release=true
<FromGitter> <crisward> How should errors be returned from a method in crystal? I don't really want to wrap all my method calls in begin rescue. But it'd be good if I could set specific error messages at the method level to explain what went wrong. Any thoughts? (btw I'm used to js way of the first callback argument being the error).
<Dreamer3> is release suppsoed to take forever?
<FromGitter> <mjago> @dreamer3 still running here in release - loads of dwarf warnings - 2.4G memory usage
<Dreamer3> sepcs pass
<FromGitter> <mjago> @crisward you can `raise` exceptions
<Dreamer3> man, what is up with --release
<FromGitter> <mjago> release build is still running specs...
<FromGitter> <crisward> just looking at the raise, begin, rescue. Bit like throw, try, catch.
<FromGitter> <mjago> @Dreamer3 still running here so I guess IO dwarfs any release optimisations - still not finished
<FromGitter> <mjago> TCPServer#initialize error here! Gotta go
<Dreamer3> why is it trying to use c++ instead of clang++
<Dreamer3> just noticed that
<Dreamer3> ok CC and CXX, but no better
<Dreamer3> asme compile error in release mode
bjz has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<Dreamer3> mjago: any progress on those specs?
<Dreamer3> ok now --stats is helpful
<Dreamer3> some output
<Dreamer3> stuck in Codegen (bc+obj): now
<Dreamer3> memory use keeps going up
<Dreamer3> trying to use 20.5 to build HEAD
<Dreamer3> 2.5gb
<Dreamer3> surely that's not normal?
<Dreamer3> wish i just knew if it was working
<Dreamer3> mjago: can you build release with stats and share output?
<Dreamer3> Codegen (bc+obj): 00:00:20.5700970 ( 851.99MB)
<Dreamer3> non release
<Dreamer3> releases i gave up
<Dreamer3> maybe i should let it run overnight
ome has quit [Quit: Connection closed for inactivity]
Raimondii has joined #crystal-lang
Raimondi has quit [Ping timeout: 252 seconds]
Raimondii is now known as Raimondi
<FromGitter> <ezrast> Any way to force `crystal doc` to document private methods?
<jaitaiwan> Hey Guys, I have a variable that can be Nil or a class Foo. When I try to access a property of Foo after checking to see if it is nil or not, the compiler complains that undefined method "bar" for Nil (compile time type is Foo | Nil). How can I get around this?
<FromGitter> <ezrast> @jaitaiwan `var.try &.bar` or `if var;; end`; the latter only works if `var` is a local variable.
<Dreamer3> lets try llvm 3.8
<Dreamer3> this release thing is killing me
bjz has joined #crystal-lang
<jaitaiwan> FromGitter thanks but neither seem to work unless local var. How would you make it work to see if is nil? at the moment the compiler complains still about any code underneath that
<jaitaiwan> *outside of the class fyi
<FromGitter> <mjago> @Dreamer3 just got back - where are you?
<Dreamer3> where?
<Dreamer3> lousville
<Dreamer3> mjago: can you build release with stats?
<Dreamer3> i'm just curious
<Dreamer3> llvm 3.8 doesn't seem much diff so far
<Dreamer3> so weird
<jaitaiwan> Hey FromGitter, in this code: shouldn't be accessible without declaring a function?
<FromGitter> jaitaiwan, I'm a bot, *bleep, bloop*. I relay messages between here and
<Dreamer3> how it works great but not with --release
<FromGitter> <jaitaiwan> Hey ezrast, in this code: shouldn't be accessible without declaring a function?
<Dreamer3> Codegen (bc+obj): 00:00:01.8087280 ( 901.10MB)
<Dreamer3> damn so mine is def hung somewhere
<Dreamer3> what is bc + obj ?
<FromGitter> <ezrast> @jaitaiwan My bad, not sure why I thought `try` was different. I think `as` is what you want?
<Dreamer3> that's crazy fast though
<FromGitter> <ezrast> and no, you need `property name` inside `Foo`
<FromGitter> <jaitaiwan> Gotcha
<Dreamer3> mine takes 16 seconds in non-release
<FromGitter> <jaitaiwan> @ezrast just to confirm... that makes these docs wrong correct?
<FromGitter> <ezrast> @jaitaiwan I think it's supposed to be implied that they're building on the `Person` class from the previous section but that could probably stand to be made clearer.
<FromGitter> <mjago> @Dreamer3 bc == bitcode I believe
<FromGitter> <mjago> btw crystal used 2.4G to run my specs
<Dreamer3> i about to give up
<Dreamer3> no idea why release is broken
<FromGitter> <mjago> @Dreamer3 can you compile a simple in —release ?
<FromGitter> <mjago> *with release
<Dreamer3> yes
<Dreamer3> 2.5 seconds
<Dreamer3> how annoying
yopp has joined #crystal-lang
<FromGitter> <mjago> @Dreamer3 actually crystal just failed to compile for me in release but compiled in non-release so thats not good
<FromGitter> <mjago> trying again
<Dreamer3> what did you change?
<Dreamer3> i thought it was working for you
<FromGitter> <mjago> It was but then it broke!
<Dreamer3> things don't just break :)
<Dreamer3> i haven't had it working since 0.7
<Dreamer3> so who knows what my issue is
<Dreamer3> but if you JUST had it working
<Dreamer3> what did you do
<FromGitter> <mjago> I did a make clean - then make crystal —release and it worked again
<FromGitter> <mjago> 10 minute compile
<FromGitter> <mjago> hello world with and without release:
<Dreamer3> what took 10 minutes?
<Dreamer3> crystal release?
<FromGitter> <mjago> yes
<FromGitter> <mjago> :)
<Dreamer3> ok let me give it 10 minutes
<Dreamer3> what CC and CXX are you using
<Dreamer3> i'm trying
<Dreamer3> CC=clang-3.8 CXX=clang++-3.8 PATH="../crystal-0.20.5-1/bin:$PATH" make crystal release=true stats=true
<FromGitter> <mjago> I added ` CXX=clang++` to top of Makefile clang version 3.9.1 (tags/RELEASE_391/final)
<Dreamer3> doing it command line does same thing, yes?
<FromGitter> <mjago> Used upto 2GB and 4 threads - all the time was in LLVM optimisation
<FromGitter> <mjago> using environment vars - I guess so
<FromGitter> <mjago> wait it out and see what happens. When it broke it exploded
<Dreamer3> what are you using to do the compile?
<Dreamer3> 0.20.5 ?
<FromGitter> <mjago> Crystal 0.20.4 (2017-01-17)
<Dreamer3> shouild i go back and try that?
<Dreamer3> i'm using 0.20.5
<FromGitter> <mjago> I don’t think it makes a difference - I think the extra time taken in —release is by llvm
<FromGitter> <mjago> …I assume llvm is optimising bytecode at that point
<Dreamer3> pretty sure i've gave it 10 minutes in the past
<Dreamer3> what hardware are you on
<FromGitter> <mjago> 1) 5GHz Intel Core i5 8GB 1600 MHz DDR3 MBP circa 2013
<FromGitter> <mjago> OSX 10.9.2
<Dreamer3> uh
<Dreamer3> they don't make em 5hz
<FromGitter> <mjago> huh?
<Dreamer3> they don't maybe 5ghz MBPs
<Dreamer3> anyways mine is 2016 so you'd think it'd compile faster
<FromGitter> <mjago> 2 point 5 GHz (perhaps got lost in IRC translation?)
<FromGitter> <mjago> you should be faster
<FromGitter> <mjago> I’m on Gitter
<FromGitter> <mjago> is it still crunching?
<Dreamer3> yes
<FromGitter> <mjago> is cpu maxed out ish
<Dreamer3> no
<FromGitter> <mjago> oh :(
<Dreamer3> threads 4 only seems to use 25%
<Dreamer3> was yours pegged
<FromGitter> <mjago> pretty much - not sure about individual threads (used all 4)
<FromGitter> <mjago> how much memory have you got ?
<Dreamer3> 16
<Dreamer3> what did you pass to threads?
<FromGitter> <mjago> ?
<Dreamer3> threads=
<Dreamer3> it never pegs more than 25% and passing threads=4 doesn't seem to change much
<Dreamer3> maybe i should try higher
<FromGitter> <mjago> ah I didn’t pass anything - activity monitor
<Dreamer3> it's pegging a single CPU
<Dreamer3> will it use more than 1?
<FromGitter> <mjago> it appeared to use 4 for me (so you may be looking at 40 minutes :)
<Dreamer3> well i dunno why it would do that
<Dreamer3> if it was going to peg all 4 4 threads should be enough
<Dreamer3> you were using the prebuilt x86 0.20.4 ?
<Dreamer3> or maybe it's some LLVM flag?
<Dreamer3> wish it had a bit more verbosity
<Dreamer3> geez
<Dreamer3> Codegen (bc+obj): 00:27:06.7952120 ( 859.99MB)
<Dreamer3> well shit
<FromGitter> <mjago> just `make clean crystal stats=true release=true` with `CXX=clang++` in Makefile which I guess would use 0.20.4 (which I built some time ago from 0.20.1)
<FromGitter> <mjago> good to go?
<Dreamer3> doing a compile of HEAD with HEAD now to see if that's any faster :)
<Dreamer3> yeah after all this i'll try with 3.9 again
<Dreamer3> vs 3.8
<Dreamer3> now that i know it just fucking takes forever
<FromGitter> <mjago> ;)
<FromGitter> <mjago> It also seems to vary
akwiatkowski has quit [Ping timeout: 255 seconds]
<Dreamer3> well i'm watching amazon tv too so not expecting it to be exact science
<Dreamer3> just "about same" or not
<Dreamer3> although i did 8 threads this time
<FromGitter> <mjago> Ah well - probably sub 10 on a quiet computer
<Dreamer3> cpu seems same
<Dreamer3> ha
<Dreamer3> i have 3 cpus doing nothing
<Dreamer3> so i don't think it should matter that much
bjz has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
soveran has joined #crystal-lang
icezimm has joined #crystal-lang
akwiatkowski has joined #crystal-lang
Liothen has quit [Quit: The Dogmatic Law of Shadowsong]
Liothen has joined #crystal-lang
Liothen has quit [Changing host]
Liothen has joined #crystal-lang
unshadow_ has quit [Ping timeout: 240 seconds]
unshadow has joined #crystal-lang
akwiatkowski has quit [Ping timeout: 276 seconds]
jinks7 has left #crystal-lang [#crystal-lang]
akwiatkowski has joined #crystal-lang
<rkeene> Anyone want to try my crystal bootstrapping build ?
<Dreamer3> seems the distributed bootstraps could be a lot better
<Dreamer3> or maybe they have to be blah cause of CPU optimizations
kostya_ has joined #crystal-lang
<splitty_> How do I link a Crystal library with another Crystal library?
<splitty_> I'm getting linking errors because both libraries define a main_module:main
<rkeene> Dreamer3, Distributed ?
<Dreamer3> the binaries github publishes
<Dreamer3> crystal/releases
<Dreamer3> the x86 builds
<Dreamer3> the one you're using to compile the compiler
<rkeene> Well, in this case I'm compiling the compiler to bitcode which can be translated to any platform's object code
<rkeene> There may still be differences in calling convientions and maybe object name mangling, unsure
<Dreamer3> no idea
<Dreamer3> i've just been compiling from source a lot
<rkeene> This should help with being able to compile from source
<Dreamer3> so what step do you need to covert the bitcode back into compiler?
kostya_ has quit [Quit: Leaving]
<rkeene> Running ./configure then make on that source should do it
<rkeene> The script modifies the Makefile to compile the bitcode into object code and then link that object code into an executable
<Dreamer3> interesting
unshadow has quit [Ping timeout: 240 seconds]
<rkeene> So it's definitely possible to compile Crystal (from bitcode) without Crystal... but I need some testing on other platforms to see where the results fall down
unshadow has joined #crystal-lang
<Dreamer3> hmmm
<FromGitter> <jaitaiwan> Hey guys, any example of PrettyPrint being used somewhere?
<Papierkorb> jaitaiwan: you mean `pp`?
<FromGitter> <jaitaiwan> No sorry. PrettyPrint class
mark_66 has quit [Remote host closed the connection]
nomethoderror has joined #crystal-lang
nomethoderror_ has joined #crystal-lang
nomethoderror has quit [Read error: Connection reset by peer]
nomethoderror_ has quit [Client Quit]
unshadow has quit [Ping timeout: 258 seconds]
unshadow has joined #crystal-lang
unshadow has quit [Ping timeout: 240 seconds]
unshadow has joined #crystal-lang
soveran has quit [Remote host closed the connection]
jokke has quit [Quit: WeeChat 1.6]
mgarciaisaia has joined #crystal-lang
jokke has joined #crystal-lang
dhk has joined #crystal-lang
soveran has joined #crystal-lang
soveran has joined #crystal-lang
soveran has quit [Changing host]
snsei has joined #crystal-lang
snsei has quit [Client Quit]
mgarciaisaia1 has joined #crystal-lang
mgarciaisaia has quit [Read error: Connection reset by peer]
soveran has quit [Remote host closed the connection]
<TheGillies> Any rough eta for parallelism>
<TheGillies> ?
<FromGitter> <denysvitali> Does anyone know why `Kemal` (or, generally speaking, HTTP Server) raises an `Error writing file: Broken pipe (Errno)` when something too big is written to env.response ?
<FromGitter> <denysvitali> I'm trying to implement `RFC7233` to `Kemal`, but I'm blocked with this problem. I can't understand why the connection gets reset by peer
<FromGitter> <denysvitali> This is the stacktrace
akwiatkowski has quit [Ping timeout: 256 seconds]
<FromGitter> <denysvitali> And this is the .cr file
<FromGitter> <denysvitali> The error comes from line 80
jokke is now known as jokkepi
jokkepi is now known as jokke
dhk has quit [Quit: Leaving]
soveran has joined #crystal-lang
soveran has joined #crystal-lang
soveran has quit [Changing host]
DeBot has quit [Remote host closed the connection]
DeBot has joined #crystal-lang
p0p0pr37_ has joined #crystal-lang
p0p0pr37 has quit [Ping timeout: 276 seconds]
p0p0pr37_ is now known as p0p0pr37
unshadow_ has joined #crystal-lang
soveran has quit [Remote host closed the connection]
unshadow has quit [Ping timeout: 248 seconds]
soveran has joined #crystal-lang
soveran has joined #crystal-lang
soveran has quit [Changing host]
soveran has quit [Remote host closed the connection]
bjz has joined #crystal-lang
<FromGitter> <denysvitali> Nevermind, fixed. Now Kemal supports multipart :D
<FromGitter> <sdogruyol> @denysvitali it's been like that for a long time :)
soveran has joined #crystal-lang
soveran has quit [Changing host]
soveran has joined #crystal-lang
unshadow_ has quit [Ping timeout: 256 seconds]
unshadow has joined #crystal-lang
dhk has joined #crystal-lang
sz0 has quit [Quit: Connection closed for inactivity]
dhk has quit [Quit: Leaving]
<FromGitter> <drosehn> .. dreamer3: for what it's worth, the homebrew guys just replied to an issue I had entered, and they said that homebrew is just not going to work unless the homebrew prefix is `/usr/local`. I know it's a lot of work to implement a good package manager, so I can understand why they're giving up on trying to keep that option working.
<FromGitter> <drosehn> It has worked for me in the past, with me using a prefix of `/opt/homebrew`. But I started to run into odd problems in the last month or so, and couldn't get around them.
bjz has quit [Ping timeout: 255 seconds]
bjz_ has joined #crystal-lang
<FromGitter> <drosehn> I did a quick test where I did build `brew` in `/usr/local`, and that fixed (avoided?) all the odd problems I have been seeing.
<FromGitter> <drosehn> In the case of crystal-lang, I've been thinking of building it from the github repo anyway, so this is a good time to try that out.
<FromGitter> <drosehn> Well, not right this minute. but sometime soon!
unshadow has quit [Ping timeout: 264 seconds]
akwiatkowski has joined #crystal-lang
<wmoxam> speaking of building Crystal, I updated my OpenBSD port to compile 0.20.5, and fixed some issues with it:
<FromGitter> <drosehn> It's nice to see the port to openbsd, btw.
soveran has quit [Remote host closed the connection]
<FromGitter> <Sija> what ya’ll think about capitalizing exception messages throughout Crystal codebase? @asterite @ysbaddaden @bcardiff?
soveran has joined #crystal-lang
soveran has quit [Ping timeout: 245 seconds]