Home freenode/#haskell: Logs Calendar

Logs: freenode/#haskell

←Prev  Next→
Page 1 .. 861 862 863 864 865 866 867 868 869 870 871 .. 5022
— 502,152 events total
2020-10-26 02:23:39 <lambdabot> error:
2020-10-26 02:23:39 <lambdabot> • No instance for (Integral Rational) arising from a use of ‘%’
2020-10-26 02:23:39 <lambdabot> • In the expression:
2020-10-26 02:23:46 <mikevdg> toRational 1.2
2020-10-26 02:24:04 × tromp quits (~tromp@dhcp-077-249-230-040.chello.nl) (Remote host closed the connection)
2020-10-26 02:24:09 <mikevdg> > toRational 1.2
2020-10-26 02:24:10 <gattytto> dsal: deploying in kube sometimes gives more fine-grained expenditure, like if you containerize your apps in aws/gce it'll be cheaper than a vps because you pay as you use
2020-10-26 02:24:11 <lambdabot> 5404319552844595 % 4503599627370496
2020-10-26 02:24:24 <dsal> It's not clear what you're trying to accomplish. Why would you type `toRational 1.2` ?
2020-10-26 02:24:25 <Axman6> > decodeFloat 1.2
2020-10-26 02:24:27 <lambdabot> (5404319552844595,-52)
2020-10-26 02:25:04 <mikevdg> I'm just playing with the numbers and trying to work out how they fit together.
2020-10-26 02:25:47 <dsal> gattytto: But why would I want to do that? GoPro Plus already does all the cloud stuff for free. I wanted tools to free myself from that. I guess I did make a web interface as well, which maybe could use that, but the `backup` command runs locally and uses AWS Lambda to push all of the data to your own S3 bucket, tracking that state locally.
2020-10-26 02:26:07 <Axman6> > properFraction 1.2
2020-10-26 02:26:09 <lambdabot> (1,0.19999999999999996)
2020-10-26 02:26:20 <mikevdg> ew.
2020-10-26 02:26:49 <Axman6> that's why you get the weird result with 1.2, since it cannot be exactly represented in IEEE-754
2020-10-26 02:27:05 <Axman6> > iterate (*2) 1.2
2020-10-26 02:27:07 <lambdabot> [1.2,2.4,4.8,9.6,19.2,38.4,76.8,153.6,307.2,614.4,1228.8,2457.6,4915.2,9830....
2020-10-26 02:27:10 <dsal> My goal is: GoPro decides to shut down, I can copy all my data out in ~43 minutes. (I have a slow connection, lots of latency, and that was without much concurrency)
2020-10-26 02:27:22 <dsal> mikevdg: If you don't like IEEE standards, there are plenty of others to choose from.
2020-10-26 02:27:37 × dbmikus quits (~dbmikus@cpe-76-167-86-219.natsow.res.rr.com) (Ping timeout: 246 seconds)
2020-10-26 02:27:44 <Axman6> > properFraction (1.2 :: Rational)
2020-10-26 02:27:45 <lambdabot> (1,1 % 5)
2020-10-26 02:28:02 <Axman6> :t 1.2
2020-10-26 02:28:03 <lambdabot> Fractional p => p
2020-10-26 02:29:51 <Axman6> > properFraction (1.2 :: Float)
2020-10-26 02:29:53 <lambdabot> (1,0.20000005)
2020-10-26 02:30:10 <Axman6> > toRational (1.2 :: Float)
2020-10-26 02:30:12 <lambdabot> 5033165 % 4194304
2020-10-26 02:31:34 × Chi1thangoo quits (~Chi1thang@87.112.60.168) (Ping timeout: 258 seconds)
2020-10-26 02:31:57 hackage crdt-event-fold 1.0.0.1 - Garbage collected event folding CRDT. https://hackage.haskell.org/package/crdt-event-fold-1.0.0.1 (rickowens)
2020-10-26 02:34:25 × geowiesnot quits (~user@i15-les02-ix2-87-89-181-157.sfr.lns.abo.bbox.fr) (Ping timeout: 240 seconds)
2020-10-26 02:34:34 → danso joins (~dan@69-165-210-185.cable.teksavvy.com)
2020-10-26 02:35:21 × djellemah quits (~djellemah@2601:5c2:100:96c:e008:b638:39fe:6a54) (Quit: Leaving)
2020-10-26 02:35:47 → djellemah joins (~djellemah@2601:5c2:100:96c:e008:b638:39fe:6a54)
2020-10-26 02:35:54 <int-e> The thing is, I believe that you can get a majority of Haskell users to agree that the existing numerical classes are awful, but you cannot get them to agree on a better design for them. There's too many conflicting design goals.
2020-10-26 02:36:37 <Axman6> simple-reflect is the onlt reasonable number implementation, fight me
2020-10-26 02:37:01 <int-e> > scanl (+) x [1..]
2020-10-26 02:37:03 <lambdabot> [x,x + 1,x + 1 + 2,x + 1 + 2 + 3,x + 1 + 2 + 3 + 4,x + 1 + 2 + 3 + 4 + 5,x +...
2020-10-26 02:37:06 → dyeplexer joins (~lol@unaffiliated/terpin)
2020-10-26 02:37:18 <Axman6> keeping the AST and allowing the user to evaluate expressions if they need the answer is the proper thing to do
2020-10-26 02:37:32 <mikevdg> What is simple-reflect?
2020-10-26 02:37:40 <Axman6> > 1 + f x
2020-10-26 02:37:42 <lambdabot> error:
2020-10-26 02:37:42 <lambdabot> • Ambiguous type variable ‘a0’ arising from a use of ‘show_M820095627375...
2020-10-26 02:37:42 <lambdabot> prevents the constraint ‘(Show a0)’ from being solved.
2020-10-26 02:37:48 <Axman6> > 1 + f x :: Expr
2020-10-26 02:37:50 <lambdabot> 1 + f x
2020-10-26 02:38:02 <Axman6> > 3 ^ 7 :: Expr
2020-10-26 02:38:04 <lambdabot> 3 * 3 * (3 * 3) * (3 * 3 * 3)
2020-10-26 02:38:18 <mikevdg> :t (^)
2020-10-26 02:38:19 <lambdabot> (Integral b, Num a) => a -> b -> a
2020-10-26 02:38:38 <Axman6> > 3 ** 7 :: Expr
2020-10-26 02:38:40 <lambdabot> 3**7
2020-10-26 02:38:50 <mikevdg> :t (**)
2020-10-26 02:38:51 <lambdabot> Floating a => a -> a -> a
2020-10-26 02:39:00 → dbmikus joins (~dbmikus@cpe-76-167-86-219.natsow.res.rr.com)
2020-10-26 02:39:07 <mikevdg> So (^) is exponent. What is (**) meant to be?
2020-10-26 02:39:13 <int-e> mikevdg: You're seeing it in action... it captures numeric expressions in an ADT that can be pretty-printed
2020-10-26 02:39:16 <Axman6> a different exponent...
2020-10-26 02:39:24 <Axman6> > 3 ** 3.5
2020-10-26 02:39:26 <lambdabot> 46.76537180435969
2020-10-26 02:39:29 <int-e> :t (^^)
2020-10-26 02:39:30 <lambdabot> (Fractional a, Integral b) => a -> b -> a
2020-10-26 02:39:46 <int-e> ^ is for natural number exponents, ^^ for integer exponents, ** for arbitrary exponents
2020-10-26 02:39:58 <mikevdg> :-o
2020-10-26 02:40:03 <Axman6> Haskell is an exponential language, we have at least three ways to compute exponentials
2020-10-26 02:40:25 <int-e> The operators live in different type classes.
2020-10-26 02:40:34 <Axman6> > 5 ^ (-2)
2020-10-26 02:40:37 <lambdabot> *Exception: Negative exponent
2020-10-26 02:40:38 <dsal> Learning Haskell by exploring the surface area is a daunting task.
2020-10-26 02:40:40 <Axman6> > 5 ^^ (-2)
2020-10-26 02:40:42 <lambdabot> 4.0e-2
2020-10-26 02:41:08 <int-e> The numeric type classes are one of the worse places to start.
2020-10-26 02:41:27 <int-e> (One other terrible place is the implementation of IO)
2020-10-26 02:41:39 <dsal> :t show. (^^ 2)
2020-10-26 02:41:40 <lambdabot> (Show a, Fractional a) => a -> String
2020-10-26 02:41:45 <mikevdg> Yea, I got hit in the face with IO and lazy evaluation.
2020-10-26 02:41:52 × jchia__ quits (~jchia@58.32.37.146) (Quit: Leaving.)
2020-10-26 02:43:01 × merijn quits (~merijn@83-160-49-249.ip.xs4all.nl) (Ping timeout: 260 seconds)
2020-10-26 02:43:27 <dsal> Lazy evaluation sometimes makes really hard things easy, but I rarely notice it in general. It can make some things work in unexpected ways when mixed with io, for sure.
2020-10-26 02:43:46 → T0pH4t joins (~josh@c-73-53-48-224.hsd1.wa.comcast.net)
2020-10-26 02:43:48 × dbmikus quits (~dbmikus@cpe-76-167-86-219.natsow.res.rr.com) (Ping timeout: 260 seconds)
2020-10-26 02:44:07 <monsterchrom> ** uses e^x and ln, and floating point approximations, it is inappropiate if you're doing natural numbers, or even your own number type for which ln doesn't make sense.
2020-10-26 02:44:36 × crestfallen quits (~jvw@135-180-15-188.fiber.dynamic.sonic.net) (Ping timeout: 256 seconds)
2020-10-26 02:44:37 <mikevdg> yea, like having a file be closed on you before you're done. Then you need to go read a chapter further ahead in the book that describes how `deepSeq` works.
2020-10-26 02:44:43 <monsterchrom> x^^y allows negative powers, therefore 1/x must make sense. Well this doesn't make sense for some number types either.
2020-10-26 02:45:22 <monsterchrom> Therefore, x^y is there for that.
2020-10-26 02:45:25 × machinedgod quits (~machinedg@24.105.81.50) (Ping timeout: 240 seconds)
2020-10-26 02:45:34 <dsal> mikevdg: that sounds... Wrong. I don't think I've ever used deepseq outside of benchmarks.
2020-10-26 02:46:06 <dsal> What book is this?
2020-10-26 02:46:09 <mikevdg> Well, I gave up on the IO stuff and just moved on.
2020-10-26 02:46:14 <Axman6> lazy IO is just a bad idea in general
2020-10-26 02:47:32 <T0pH4t> hello all, quick question. Is there a way to access the type declaration in a class so that I can use it in func signature? For example "class Foo a where type Bar " then "doStuff :: Foo a => a -> Bar a"
2020-10-26 02:47:47 <T0pH4t> Effectively I want access to the type declared for Bar in Foo
2020-10-26 02:48:17 <Axman6> (b ~ Bar a, Foo a) => ... ?
2020-10-26 02:48:34 <T0pH4t> oops that should be "type Bar a" in class def
2020-10-26 02:48:48 <Axman6> wait, does that not work the way you've written it?
2020-10-26 02:48:57 <T0pH4t> Well Bar is defined within class Foo, so Axman6 not sure that works
←Prev  Next→
Page 1 .. 861 862 863 864 865 866 867 868 869 870 871 .. 5022

All times are in UTC.