Home freenode/#haskell: Logs Calendar

Logs: freenode/#haskell

←Prev  Next→
Page 1 .. 453 454 455 456 457 458 459 460 461 462 463 .. 5022
— 502,152 events total
2020-10-06 04:36:31 <hololeap> :t \x y -> _ (x + y)
2020-10-06 04:36:36 <lambdabot> error:
2020-10-06 04:36:36 <lambdabot> • Found hole: _ :: t -> t1
2020-10-06 04:36:36 <lambdabot> Where: ‘t’, ‘t1’ are rigid type variables bound by
2020-10-06 04:37:13 → haritz joins (~hrtz@62.3.70.206)
2020-10-06 04:37:14 × haritz quits (~hrtz@62.3.70.206) (Changing host)
2020-10-06 04:37:14 → haritz joins (~hrtz@unaffiliated/haritz)
2020-10-06 04:38:39 → day_ joins (~Unknown@unaffiliated/day)
2020-10-06 04:39:29 <gnumonik> I'm not even sure how to break this down. I've been at this for more than a week... This seems like it should work: myFunc adt val adtLens fieldPrism i = set' (adtLens . ix i . fieldPrism) val adt. But it doesn't for the list-of-sums-inside-a-product. The sketchy part, as far as I can tell, is that the product type doesn't have an instance of a class that it can't possibly have. Oh well.
2020-10-06 04:39:53 → takuan joins (~takuan@178-116-218-225.access.telenet.be)
2020-10-06 04:40:29 → wroathe joins (~wroathe@c-68-54-25-135.hsd1.mn.comcast.net)
2020-10-06 04:41:02 × zoom84 quits (~zoom55aa@068-190-177-201.res.spectrum.com) (Quit: WeeChat 2.9)
2020-10-06 04:41:18 → zoom84 joins (~user@068-190-177-201.res.spectrum.com)
2020-10-06 04:41:56 × day quits (~Unknown@unaffiliated/day) (Ping timeout: 272 seconds)
2020-10-06 04:41:56 day_ is now known as day
2020-10-06 04:42:21 <Axman6> gnumonik: can you share a more concrete example?
2020-10-06 04:44:39 <Axman6> gnumonik: given that code it's difficult to tell what the types are, I can sort of infer them, but some type signatures and definitions of the sums and products would be good
2020-10-06 04:45:44 × wroathe quits (~wroathe@c-68-54-25-135.hsd1.mn.comcast.net) (Ping timeout: 272 seconds)
2020-10-06 04:46:09 <Axman6> :t set'
2020-10-06 04:46:11 <lambdabot> ASetter' s a -> a -> s -> s
2020-10-06 04:46:16 <Axman6> :t set
2020-10-06 04:46:18 <lambdabot> ASetter s t a b -> b -> s -> t
2020-10-06 04:47:32 <hololeap> doesn't lens composition get reversed from the usual order, e.g. (fieldPrism . ix i . adtLens), assuming that the sum type is the "topmost" structure
2020-10-06 04:47:56 × solonarv quits (~solonarv@adijon-655-1-70-207.w90-13.abo.wanadoo.fr) (Ping timeout: 256 seconds)
2020-10-06 04:47:58 × zacts quits (~zacts@dragora/developer/zacts) (Quit: leaving)
2020-10-06 04:48:11 × brandly quits (~brandly@c-73-68-15-46.hsd1.ma.comcast.net) (Ping timeout: 240 seconds)
2020-10-06 04:48:32 <Axman6> yeah if it's a product containing a list containing a sum then it would be fieldLens . ix i . sumPrism
2020-10-06 04:48:43 <hololeap> "topmost" isn't very descriptive. i mean the first structure you navigate
2020-10-06 04:48:54 <hololeap> right
2020-10-06 04:49:07 <Axman6> it looks to me like it is doing the right thing, assuming it is what I said it is
2020-10-06 04:50:43 <hololeap> oh, so it would be the normal order of composition then (same as functions)? i remembered composing lens' to be the opposite of what you usually do
2020-10-06 04:51:54 <gnumonik> It's actually a sum containing a product containing a list of sums. The actual context of this is pretty substantial. I'm writing a protocol analyzer that wraps different message types in one big sum type. The 'adtLens' thing is actually a composition of lenses and prisms but I dunno that that matters
2020-10-06 04:52:28 <hololeap> sorry, i misread
2020-10-06 04:52:35 hololeap edges away from the conversation
2020-10-06 04:54:08 <dsal> > Right (1, [Right 2, Left 3, Right 4]) ^.. _Right . _2 . folded . _Right
2020-10-06 04:54:11 <lambdabot> [2,4]
2020-10-06 04:56:23 <dsal> > Right (1, [Right 2, Left 3, Right 4]) & _Right . _2 . traversed . _Right %~ (^2)
2020-10-06 04:56:25 <lambdabot> Right (1,[Right 4,Left 3,Right 16])
2020-10-06 04:56:32 <Axman6> :t (folded, each)
2020-10-06 04:56:34 <lambdabot> (Indexable Int p, Contravariant f1, Each s t a1 b, Foldable f2, Applicative f1, Applicative f3) => (p a2 (f1 a2) -> f2 a2 -> f1 (f2 a2), (a1 -> f3 b) -> s -> f3 t)
2020-10-06 04:56:44 × zoom84 quits (~user@068-190-177-201.res.spectrum.com) (Quit: WeeChat 2.9)
2020-10-06 04:57:03 <dsal> Oh, I never think of each like that. That's pretty neat.
2020-10-06 04:57:15 <dsal> Lens doesn't fit in my tiny headcache
2020-10-06 04:58:55 × ixian quits (~mgold@2002:4a74:ba78:1701:0:ff:fe78:6269) (Ping timeout: 240 seconds)
2020-10-06 05:00:37 → ixian joins (~mgold@terra.bitplane.org)
2020-10-06 05:01:26 <gnumonik> Ugh I can't even make this more concrete. If there's nothing wrong with that function then I think I just can't do what I want with lens. That's literally the function I use.
2020-10-06 05:03:28 × dansho quits (~dansho@ip68-108-167-185.lv.lv.cox.net) (Remote host closed the connection)
2020-10-06 05:04:12 <dsal> The above examples match your description. It's hard to guess what you're doing differently.
2020-10-06 05:06:28 hackage reanimate-svg 0.12.1.0 - SVG file loader and serializer https://hackage.haskell.org/package/reanimate-svg-0.12.1.0 (DavidHimmelstrup)
2020-10-06 05:09:44 × mathlover2 quits (~mathlover@2604:6000:1013:129e:755e:eccf:8735:da37) (Quit: Leaving)
2020-10-06 05:09:55 × falafel quits (~falafel@2605:e000:1527:d491:99fe:5613:f0a7:56f0) (Ping timeout: 240 seconds)
2020-10-06 05:10:35 → xff0x joins (~fox@2001:1a81:5339:8000:bdc7:ef33:4985:5e4f)
2020-10-06 05:10:44 → GyroW_ joins (~GyroW@ptr-48ujrfd1ztq5fjywfw3.18120a2.ip6.access.telenet.be)
2020-10-06 05:10:45 × GyroW_ quits (~GyroW@ptr-48ujrfd1ztq5fjywfw3.18120a2.ip6.access.telenet.be) (Changing host)
2020-10-06 05:10:45 → GyroW_ joins (~GyroW@unaffiliated/gyrow)
2020-10-06 05:11:07 × GyroW quits (~GyroW@unaffiliated/gyrow) (Ping timeout: 260 seconds)
2020-10-06 05:11:53 → wroathe joins (~wroathe@c-68-54-25-135.hsd1.mn.comcast.net)
2020-10-06 05:13:29 <gnumonik> Well, if I'm going to start over I have a question: My goal here is to find a way to expose lens functionality to the user without writing a bunch of boilerplate for every record field/constructor. I was trying to use TH to generate parsers for every valid combination of nested records/type constructors (i.e. the string "foo.bar" parses to the composition of lenses foo and bar). Is there an... obvious easier way to do that?
2020-10-06 05:13:54 → coot joins (~coot@37.30.49.218.nat.umts.dynamic.t-mobile.pl)
2020-10-06 05:14:42 <dsal> the lens library does that for you.
2020-10-06 05:16:42 <gnumonik> huh? how?
2020-10-06 05:16:43 × wroathe quits (~wroathe@c-68-54-25-135.hsd1.mn.comcast.net) (Ping timeout: 260 seconds)
2020-10-06 05:18:27 → snakemasterflex joins (~snakemast@213.100.206.23)
2020-10-06 05:18:33 × Tritlo quits (sid58727@gateway/web/irccloud.com/x-vuffzyrxnybrguux) (Ping timeout: 272 seconds)
2020-10-06 05:18:50 × howdoi quits (uid224@gateway/web/irccloud.com/x-bfkecngtkcukjezu) (Quit: Connection closed for inactivity)
2020-10-06 05:19:55 → ishutin_ joins (~Ishutin@193-110-63-54.cable-modem.hdsnet.hu)
2020-10-06 05:20:10 → Tritlo joins (sid58727@gateway/web/irccloud.com/x-gcmzpwtjgmfrhnky)
2020-10-06 05:20:13 <gnumonik> Er maybe that wasn't clear. What I really need is a function :: String -> Lens, since I was hoping to use the functionality in a small DSL. When I say I'm trying to generate parsers, I mean functions that take strings and spit out lenses/prisms/traversals. (I'm sorry if this is stupid, Haskell's my first programming language and I've been teaching myself without much guidance)
2020-10-06 05:22:45 × snakemasterflex quits (~snakemast@213.100.206.23) (Ping timeout: 240 seconds)
2020-10-06 05:23:18 × ishutin quits (~Ishutin@92-249-150-146.static.digikabel.hu) (Ping timeout: 260 seconds)
2020-10-06 05:24:41 → petersen joins (~petersen@redhat/juhp)
2020-10-06 05:26:45 → snakemasterflex joins (~snakemast@213.100.206.23)
2020-10-06 05:27:01 × kori quits (~kori@arrowheads/kori) (Ping timeout: 244 seconds)
2020-10-06 05:27:27 × spew quits (uid195861@gateway/web/irccloud.com/x-zoqmilcrntqznxpj) (Quit: Connection closed for inactivity)
2020-10-06 05:36:10 → idhugo joins (~idhugo@563472ae.rev.stofanet.dk)
2020-10-06 05:37:18 × snakemasterflex quits (~snakemast@213.100.206.23) (Ping timeout: 260 seconds)
2020-10-06 05:37:43 → falafel joins (~falafel@2605:e000:1527:d491:99fe:5613:f0a7:56f0)
2020-10-06 05:37:47 × shailangsa quits (~shailangs@host165-120-169-97.range165-120.btcentralplus.com) (Ping timeout: 240 seconds)
2020-10-06 05:37:48 → snakemasterflex joins (~snakemast@213.100.206.23)
2020-10-06 05:41:42 <jackdk> I think dsal's saying that the point of lens is that lenses themselves compose with `.`, so you shouldn't feel like you have to generate all the compositions yourself
2020-10-06 05:43:50 → kori joins (~kori@arrowheads/kori)
2020-10-06 05:45:31 <dsal> Well, I mean you don't have to generate the boilerplate for every field. That's what makeLenses does. But doing this at runtime is a bit weird.
2020-10-06 05:46:47 <[exa]> kindof guess that gnumonik only wants an autogenerated dictionary that does f "foo" = foo
2020-10-06 05:47:10 <[exa]> the rest (like parsing out the dot and joining the results) can be easily done with the parser
2020-10-06 05:47:57 hackage hoauth2 1.16.0 - Haskell OAuth2 authentication client https://hackage.haskell.org/package/hoauth2-1.16.0 (HaishengWu)
2020-10-06 05:48:28 <gnumonik> yeah, ideally i'd not have to pre-generate the entire string, but that seemed to work (up to a point anyway) whereas storing a variety of differently-typed lenses in any kind of data structure did not
2020-10-06 05:48:58 × idhugo quits (~idhugo@563472ae.rev.stofanet.dk) (Ping timeout: 260 seconds)
2020-10-06 05:49:06 <dsal> How would a user use this mechanism?
2020-10-06 05:49:43 <dsal> Is it just about making runtime errors out of lens expressions?
2020-10-06 05:51:09 → albert_91 joins (~Albert@p200300e5ff0b5b4248a33bded2872db1.dip0.t-ipconnect.de)
2020-10-06 05:52:51 <gnumonik> Not really. I wanted to somehow leverage lens to allow a user to select fields in data types that represent network packets and filter or replace those fields. But given that there are... i dunno, a few thousand network protocols out there, I wanted some way of automating the construction of functions for doing those things.
2020-10-06 05:53:01 → bahamas joins (~lucian@unaffiliated/bahamas)
2020-10-06 05:54:26 <dsal> I think that makes sense. The string is throwing me off, though. I think it's a confusing detail. If you didn't have the string, and someone were using your library, what would they have to do now in one example?
2020-10-06 05:55:25 × albert_91 quits (~Albert@p200300e5ff0b5b4248a33bded2872db1.dip0.t-ipconnect.de) (Ping timeout: 240 seconds)
2020-10-06 05:55:47 → shatriff joins (~vitaliish@217.27.153.240)
2020-10-06 05:56:55 → thir joins (~thir@p200300f27f0fc60038c1b16891cbfa03.dip0.t-ipconnect.de)
2020-10-06 05:57:23 <dsal> because, as jackdk says, you don't need to generate all the combinations, but you might be able to write some useful helpers that skip a few common steps.
2020-10-06 06:00:01 × MasterGruntR75 quits (~MasterGru@185.244.214.216) ()
2020-10-06 06:03:07 → aplainze1akind joins (~johndoe@captainludd.powered.by.lunarbnc.net)
2020-10-06 06:03:15 → jcarpenter2 joins (~rofl@96.78.87.197)
←Prev  Next→
Page 1 .. 453 454 455 456 457 458 459 460 461 462 463 .. 5022

All times are in UTC.