_4d47 has quit [Quit: Connection closed for inactivity]
<FromGitter>
<codenoid> btw, idk, but my crystal wont work as frontend server :(
ruby_ has joined #crystal-lang
ruby_ has quit [Quit: Leaving]
codenoid has joined #crystal-lang
snsei has joined #crystal-lang
snsei has quit [Ping timeout: 246 seconds]
<FromGitter>
<picatz> Started working on porting Command Lion over to Crystal tonight.
<FromGitter>
<picatz> Mostly because I was really curious how hard it would be. Got some of the "Raw" ARGV argument parsing sorted out I think.
<FromGitter>
<picatz> What does it look like? Hopefully not too terrible. I'm open to feedback: https://carc.in/#/r/2o7v
<FromGitter>
<picatz> *The documentation wording it a bit weird, but #yolo.*
<FromGitter>
<ezrast> @picatz I'd do `def self.arguments_to(flag : String, flags : Array(String | Regex) = [/^-/])`, if that makes sense
<FromGitter>
<picatz> I think it makes sense.
<FromGitter>
<picatz> Not entirely, though. What would be the benefit of having the String | Regex union?
<FromGitter>
<ezrast> Default to assuming that any arg that begins with `-` is a flag
<FromGitter>
<picatz> In general, I'm not really enforcing any sort of flag format ( with or without the -'s ); and at least how I parsed command lion's apps.
<FromGitter>
<picatz> Like the crystal command for example. There's commands that don't have `-` in the flag.
<FromGitter>
<picatz> I'd like for command lion to have `-` flags totally optional.
<FromGitter>
<picatz> And, basically, I know all the flags ahead of time before the parsing of ARGV happens.
<FromGitter>
<ezrast> I'm probably misunderstanding the use case. I'm old-school about my flags I guess.
<FromGitter>
<picatz> No worries! I appreciate the feedback.
<FromGitter>
<ezrast> It just struck me as an opportunity for a reasonable default.
<FromGitter>
<picatz> For sure, I'm glad you brought it up.
<FromGitter>
<picatz> Maybe something I need to document in the future. The "flags philosophy" of command lion.
<FromGitter>
<picatz> There'd nothing stopping you from having the `-` flags if you ever end up using it :P
<FromGitter>
<picatz> Maybe I could do some sort of keyword/method to enforce the `-` flags. ✨
<FromGitter>
<picatz> Perhaps I should also investigate POSIX more and see if I could provide some sort of option for that. I dunno.
<FromGitter>
<ezrast> My assumption is that non-`-` flags are positional, so random access to them makes no sense. Being unopinionated is fine but I'd assume that's what most users will want to do.
sp4rrow has joined #crystal-lang
<FromGitter>
<picatz> That's also one thing I've sort of overlooked in command lion: positional arguments.
<FromGitter>
<picatz> Basically, right now, in the ruby version, there's no positional enforcing at all.
<FromGitter>
<picatz> Like, you couldn't if you wanted to.
<FromGitter>
<picatz> Probably something I should invest some time into working out.
sp4rrow has quit [Ping timeout: 246 seconds]
snsei has joined #crystal-lang
sp4rrow has joined #crystal-lang
rohitpaulk has joined #crystal-lang
sp4rrow has quit [Quit: The Internet needs a break and I need a cookie]
sp4rrow has joined #crystal-lang
sp4rrow_ has joined #crystal-lang
sp4rrow has quit [Ping timeout: 255 seconds]
sp4rrow_ has quit [Ping timeout: 255 seconds]
snsei has quit [Remote host closed the connection]
codenoid has quit [Quit: This computer has gone to sleep]
codenoid has joined #crystal-lang
codenoid has quit [Client Quit]
rohitpaulk has quit [Ping timeout: 260 seconds]
codenoid has joined #crystal-lang
sp4rrow has joined #crystal-lang
sp4rrow has quit [Ping timeout: 246 seconds]
snsei has joined #crystal-lang
rohitpaulk has joined #crystal-lang
<FromGitter>
<sdogruyol> Morning everyone!
<FromGitter>
<drujensen> Good Morning!
<FromGitter>
<drujensen> 11pm for me
<FromGitter>
<drujensen> lol
<FromGitter>
<sdogruyol> :D
<FromGitter>
<sdogruyol> How are you @drujensen
<FromGitter>
<drujensen> doing great. I’m writing an issue for TechEmpower regarding the results for Kemal and Amber
<FromGitter>
<sdogruyol> awesome 👍
<FromGitter>
<sdogruyol> Have you found the root cause yet?
<FromGitter>
<drujensen> We have ran the benchmarks several times and we are seeing way better results on smaller systems
<FromGitter>
<sdogruyol> I've also done small runs and can say the same
<FromGitter>
<sdogruyol> Not sure about what's wrong with TE setup
<FromGitter>
<sdogruyol> I'm getting around 100K RPS / core
<FromGitter>
<drujensen> I followed their setup and made sure the environments are to their specs
<FromGitter>
<sdogruyol> great
<FromGitter>
<drujensen> on a 64 core, Amber is getting around 1m rps
<FromGitter>
<drujensen> but they were 1.2ghz
<FromGitter>
<sdogruyol> on an i7 or so that'd be double
<FromGitter>
<drujensen> but that is still 10 times better than what they are showing
<FromGitter>
<drujensen> right
<FromGitter>
<sdogruyol> yeah
<FromGitter>
<drujensen> we should be around 1.8 to 2m
<FromGitter>
<sdogruyol> I get that result on a single core i7
<FromGitter>
<drujensen> hoping they can figure out what the discrepency is and provide us an update
<FromGitter>
<drujensen> I will include you on the PR
snsei has quit [Remote host closed the connection]
<FromGitter>
<drujensen> ok, @sdogruyol . I included you on the ticket. I’m hoping we can determine the cause. It’s as if the reuseport was not being used by all 3 frameworks.
<FromGitter>
<sdogruyol> Yeah
<FromGitter>
<sdogruyol> This is just gut feeling but
<FromGitter>
<sdogruyol> What if crystal-raw reuse_port somehow runs first and then the others can't use it later?
<FromGitter>
<sdogruyol> sounds silly but..
<FromGitter>
<drujensen> I’ve asked if he could rerun the plaintext for the 3 projects. I don’t know if they will do that for us.
<FromGitter>
<drujensen> I think I provided enough evidence that something went wrong
<FromGitter>
<sdogruyol> Have you tried running crystal-raw first and then amber later?
<FromGitter>
<drujensen> yeah
<FromGitter>
<sdogruyol> I'm out of clues then :/
<FromGitter>
<drujensen> well, no. I think it runs them alphabeticallly
<FromGitter>
<sdogruyol> yes
<FromGitter>
<drujensen> but i’ve ran them together
<FromGitter>
<sdogruyol> hmm
<FromGitter>
<drujensen> we are seeing about a 25% reduction in speed between crystal-raw and amber
<FromGitter>
<drujensen> which is to be expected
<FromGitter>
<drujensen> btw, i am removing kemalyst from the tests since amber is now available
<FromGitter>
<sdogruyol> makes sense 👍
<FromGitter>
<drujensen> so we should have raw, micro and full
<FromGitter>
<drujensen> time for bed. have a great day!
<FromGitter>
<sdogruyol> thanks, good night @drujensen 😴
<FromGitter>
<krypton97> Morning all, lot of good news
<FromGitter>
<krypton97> Good night:)
<FromGitter>
<sdogruyol> morning @krypton97
<FromGitter>
<sdogruyol> What news?
<FromGitter>
<krypton97> Well, the result of the amber's bench :)
Tuxified has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
claudiuinberlin has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
claudiuinberlin has joined #crystal-lang
Tuxified has joined #crystal-lang
codenoid has quit [Quit: This computer has gone to sleep]
sp4rrow has joined #crystal-lang
codenoid has joined #crystal-lang
sp4rrow has quit [Ping timeout: 246 seconds]
codenoid has quit [Client Quit]
claudiuinberlin has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
Tuxified has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
Tuxified has joined #crystal-lang
Groogy has quit [Disconnected by services]
Groogy2 is now known as Groogy
<Groogy>
Lol that was interesting, got a notification on Github on rubinus
<Groogy>
apparently I still count as a contributor? wut?
Groogy_ has joined #crystal-lang
<Groogy>
oh Rubinus has +400 contributors, I guess once you do a PR it never goes away
<livcd>
rubinus or rubinius ?
<Groogy>
ah yeah Rubinius
<Groogy>
It's like 5 years ago? Or something like that, did some stuff for the C api
claudiuinberlin has joined #crystal-lang
<oprypin>
Groogy, this contributor thing means nothing for normal github notifications, it's literally "do i have a commit in it"
<oprypin>
unless someone manually acted using that list
claudiuinberlin has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
sz0 has joined #crystal-lang
claudiuinberlin has joined #crystal-lang
rohitpaulk has joined #crystal-lang
sp4rrow has joined #crystal-lang
sp4rrow has quit [Ping timeout: 246 seconds]
<FromGitter>
<crisward> Got a really odd bug with sockets. It only happens every few days and seems impossible to recreate. I have an array of connected sockets which I send messages to. However occasionally one seems to close without triggering a close. I have send wrapped in a begin rescue which removed the socket from the array on error, but it's not clear if this is running. I probably need to chuck some debug messages in to find
<FromGitter>
... out what's happening. If anyone has ideas on this issues, and thoughts on causes / recreation that'd really help.
<FromGitter>
<crisward> I also have a `SIGNAL::PIPE.trap` as without this the app crashes.
<oprypin>
crisward, what do you mean by "close without triggering a close"? what is the symptom?
<oprypin>
anyhow, that seems like completely normal socket behavior. detecting a socket close was never reliable and never will be
<oprypin>
to make your protocol reliable, you add periodical "heartbeat" messages, and detect timeouts
<oprypin>
when you receive 0-length data from a socket, you know it's closed for sure
<oprypin>
but not all closes are possible to find out
<oprypin>
of course, I get the heartbeat for free as part of the irc protocol
<Papierkorb_>
TCP itself also has a heartbeat functionality you can enable if it's not, or shorten the interval to get around close-happy routers
claudiuinberlin has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<oprypin>
i think i would've heard more about this functionality if it was a widely accepted practice to rely on it. something must be wrong with using it.
<Papierkorb_>
Though I never relied on it, as in, either the protocol had a heartbeat message, or the connection was shortlived in any case
<Papierkorb_>
oprypin: To be fair, there are many perfectly working features nobody uses for no reason
<Papierkorb_>
SCTP <3
<crystal-gh>
[crystal] akzhan opened pull request #4931: [WIP] try to reproduce bug in 32bit (master...example-of-bad-code-for-x86-bigint) https://git.io/v5afi
claudiuinberlin has joined #crystal-lang
claudiuinberlin has quit [Client Quit]
bmcginty has quit [Ping timeout: 240 seconds]
bmcginty has joined #crystal-lang
sp4rrow has joined #crystal-lang
sp4rrow has quit [Ping timeout: 246 seconds]
ShalokShalom_ has joined #crystal-lang
ShalokShalom has quit [Ping timeout: 260 seconds]
claudiuinberlin has joined #crystal-lang
snsei has joined #crystal-lang
claudiuinberlin has quit [Client Quit]
<crystal-gh>
[crystal] akzhan closed pull request #4931: [WIP] try to reproduce bug in 32bit (master...example-of-bad-code-for-x86-bigint) https://git.io/v5afi
snsei has quit [Ping timeout: 246 seconds]
<crystal-gh>
[crystal] RX14 closed pull request #4925: Process markdown in block quotes (master...master) https://git.io/v5uU3
<FromGitter>
<Qwerp-Derp> How hard would it be to make a binding for Nuklear?
rohitpaulk has quit [Ping timeout: 240 seconds]
<FromGitter>
<sdogruyol> what's that?
<crystal-gh>
[crystal] akzhan reopened pull request #4931: [WIP] try to reproduce bug in 32bit (master...example-of-bad-code-for-x86-bigint) https://git.io/v5afi
claudiuinberlin has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<Papierkorb_>
<3
<FromGitter>
<Qwerp-Derp> I've run into a brick wall with Nuklear... :(
<FromGitter>
<Qwerp-Derp> oprypin, qt5.cr doesn't work for me yet (because it hasn't been tested on Mac OS X, which is what I'm using right now)
<Papierkorb_>
oprypin: I moved on friday, so I wasn't able to do much the last days. Will start today, maybe with a commit or tomorrow on bindgen. Gotta work on the architecture more. Will also enable tons of more possibilities :3
<Papierkorb_>
Qwerp-Derp, as I outlined, I don't own (or plan to own) a mac computer, so I can't port nor test it on that platform
<FromGitter>
<Qwerp-Derp> Papierkorb_ yeah, IK
<Papierkorb_>
Hopefully someone else will be willing to maintain the OS X side at some point
<Papierkorb_>
At which point one could work on getting rid of the numerous pathing assumptions. Should be the majority of the work™.
<FromGitter>
<Qwerp-Derp> It's strange though, since Qt is cross platform you would expect this to work
<Papierkorb_>
oprypin: Does Crystal work on Windows outside the linux subsystem? Or is the scheduler still broken?
<FromGitter>
<Qwerp-Derp> How do you create a binding to a local `.h` file in Crystal?
<FromGitter>
<Qwerp-Derp> For example, say I have this structure:
<vegai>
I believe we'd be seeing a blog post if Crystal on Windows was a thing :)
<FromGitter>
<Qwerp-Derp> There doesn't seem to be any good options available for Crystal GUI at the moment for Mac OS X, except for `libui.cr` which lacks documentation and features IMO
<FromGitter>
<Qwerp-Derp> Primarily because `libui` itself lacks documentation and features
<crystal-gh>
[crystal] asterite opened pull request #4932: Don't check call of non-self instance when computing read instance vars (master...bug/4830-fix-instance-var-access-other-instance) https://git.io/v5a7a
claudiuinberlin has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
claudiuinberlin has joined #crystal-lang
greenbigfrog has quit [Remote host closed the connection]
Tuxified has joined #crystal-lang
mark_66 has quit [Remote host closed the connection]
greenbigfrog has joined #crystal-lang
Tuxified has quit [Ping timeout: 240 seconds]
claudiuinberlin has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
claudiuinberlin has joined #crystal-lang
claudiuinberlin has quit [Client Quit]
claudiuinberlin has joined #crystal-lang
_4d47 has quit [Quit: Connection closed for inactivity]
<FromGitter>
<asterite> Where's the AMA? The dev.to page is so... empty
<Yxhuvud>
it starts in an hour?
<Yxhuvud>
thanks for the reminder though :)
<ragmaanir>
do i have to create an acc on dev.to to ask questions?
<FromGitter>
<mgarciaisaia> The post will be published during this next hour
<FromGitter>
<mgarciaisaia> We were copying the blog over to dev.to, but then found some issues, and then forgot about that :P
<FromGitter>
<drujensen> @asterite good to see you here (even if it was a mistake) ;-)
<FromGitter>
<drujensen> oh, wait. @rumenzu make the mistake, nm. still good to see you.
<FromGitter>
<asterite> lol, no mistake :)
<FromGitter>
<asterite> I just have ups and downs with Crystal :-P
<FromGitter>
<drujensen> haha, gotta get that bi-polar’ness under control. you scared us.
rumenzu_ has joined #crystal-lang
<FromGitter>
<sdogruyol> @asterite Crystal is great ❤️ Keep up the great work
rumenzu has quit [Ping timeout: 240 seconds]
rumenzu_ has quit [Ping timeout: 248 seconds]
rohitpaulk has quit [Ping timeout: 260 seconds]
<FromGitter>
<elorest> @asterite Thanks so much for your work on crystal. I know how the ups and downs go. I’ve only been working for 6 months on the amber team I let things get to me too much sometimes. In order to do good work you have to care, but sometimes its easy to care too much or get burned out.
<FromGitter>
<asterite> Definitely! Thanks for the kind words :)
<Papierkorb>
asterite, Great to see you around! Hope the time off was refreshing
<oprypin>
wonder how many questions will be something that wasnt asked before. expect around 15%
<oprypin>
then again, even if it was, maybe things have changed... like parallelism
<FromGitter>
<mgarciaisaia> He also happens to be a really nice human being ^_^
<FromGitter>
<faustinoaq> OMG, Thanks you so much 👏
<FromGitter>
<krypton97> Any good openid package or I should stick with the stdlib oauth2
<FromGitter>
<krypton97> In doubts
trojan[m] has joined #crystal-lang
<FromGitter>
<asterite> i think oauth2 should work
<FromGitter>
<sdogruyol> AMA was great!
<FromGitter>
<sdogruyol> Thanks everone
<FromGitter>
<drujensen> Yes it was. Thanks to Manas.Tech as well.
<FromGitter>
<mgarciaisaia> Really fun experience - thanks to everyone for joining!
<FromGitter>
<mgarciaisaia> The comments are open, so you can still send questions whenever you want to :)
<trojan[m]>
I'm Looking for a good customer to buy Cvv and dumps everyday and long-Term i will discount or give you bonus if you order bulk I got High balance fullz my icq 2336262
<FromGitter>
<fridgerator> wut
<Papierkorb>
Here we go
trojan[m] was kicked from #crystal-lang by RX14 [trojan[m]]
<FromGitter>
<sdogruyol> wtf
<RX14>
spammers
<Papierkorb>
-o
<RX14>
i was wondering if they'd come back
<RX14>
because it was only a kick
<FromGitter>
<asterite> but... icq?
<Papierkorb>
And a low number at that
<Papierkorb>
Comparably
<FromGitter>
<sdogruyol> that icq must be at least 15 years old or so lol
<FromGitter>
<asterite> maybe someone with a time machine, and we kicked him...
<FromGitter>
<sdogruyol> :D
<Papierkorb>
"Throwing out John Titor" is a once in a lifetime achievement
<FromGitter>
<bararchy> When those will get fixed, I'll re-run it again, right now it produces too many crashes to test
<FromGitter>
<bararchy> hahah
<FromGitter>
<bararchy> @crisward BTW, if your project is written in Crystal and you want to try out our new Tool, I can give you a Crystal-User only deal ;) free tests, for credits and referances
Tuxified has joined #crystal-lang
Tuxified has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<RX14>
bartgrantham, macos has a weird arrangement with openssl
<RX14>
i don't use osx so i don't know how to fix it
<FromGitter>
<opensas> Have fun, any feedback is welcome
<bartgrantham>
I installed openssl and crystal with homebrew. GG, I'll look into it and see if I find a solution. If so I'll come back and post it.
bartgrantham has quit [Quit: Page closed]
<FromGitter>
<ziprandom> hey there was a talk today?
<FromGitter>
<ziprandom> can't find anything on the blog ..
<FromGitter>
<krypton97> @ziprandom hello dude, love that graphql implementation:)
<FromGitter>
<krypton97> I'm interested in that talk as well, had to go out that time..
<FromGitter>
<ezrast> @ziprandom Not a talk, but an AMA at https://dev.to/
<FromGitter>
<ziprandom> @ezrast yeah found it!
<FromGitter>
<ziprandom> @krypton97 thanks! I'm excited to see it being used and fix bugs
<FromGitter>
<ziprandom> aren't we in for the parser rewrite and a benchmark? :)
<FromGitter>
<ziprandom> btw. today I started with smth. that could become a gitlab clone using the graphql lib and https://github.com/mosop/git. it's amazing how fast api development cant be with graphql
Tuxified has joined #crystal-lang
Tuxified has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]