Home freenode/#haskell: Logs Calendar

Logs: freenode/#haskell

←Prev  Next→
Page 1 .. 697 698 699 700 701 702 703 704 705 706 707 .. 5022
— 502,152 events total
2020-10-17 13:51:32 <bifunc2> Is it appropriate to put an integration test into an HUnit testCase?
2020-10-17 13:52:13 × nyd quits (~lpy@unaffiliated/elysian) (Quit: nyd)
2020-10-17 13:53:03 × cdepillabout quits (~cdepillab@pl35600.ag1001.nttpc.ne.jp) (Quit: Leaving)
2020-10-17 13:55:06 → fendor joins (~fendor@046124068105.public.t-mobile.at)
2020-10-17 13:55:31 × drbean quits (~drbean@TC210-63-209-144.static.apol.com.tw) (Ping timeout: 256 seconds)
2020-10-17 13:55:49 × Nahra quits (~Nahra@unaffiliated/nahra) (Quit: leaving)
2020-10-17 13:57:14 × da39a3ee5e6b4b0d quits (~textual@n11211935170.netvigator.com) (Quit: My MacBook has gone to sleep. ZZZzzz…)
2020-10-17 13:57:36 × nbloomf quits (~nbloomf@2600:1700:83e0:1f40:a405:7897:267b:514b) (Quit: My MacBook has gone to sleep. ZZZzzz…)
2020-10-17 14:00:50 → petersen joins (~petersen@redhat/juhp)
2020-10-17 14:00:57 → da39a3ee5e6b4b0d joins (~textual@n11211935170.netvigator.com)
2020-10-17 14:01:52 → gnrp1 joins (~gnrp@84.39.117.57)
2020-10-17 14:06:46 × bifunc2 quits (bifunc2@gateway/vpn/protonvpn/bifunc2) (Ping timeout: 246 seconds)
2020-10-17 14:07:30 → Sheilong joins (uid293653@gateway/web/irccloud.com/x-mfevbueuacsxobra)
2020-10-17 14:15:20 → kotrcka joins (~kotrcka@ip-94-112-194-11.net.upcbroadband.cz)
2020-10-17 14:16:14 → howdoi joins (uid224@gateway/web/irccloud.com/x-ingbgflergmrjqkw)
2020-10-17 14:16:30 → djellemah joins (~djellemah@2601:5c2:100:96c:e008:b638:39fe:6a54)
2020-10-17 14:16:35 → nbloomf joins (~nbloomf@2600:1700:83e0:1f40:fc2b:b27d:96e3:a54e)
2020-10-17 14:17:50 × nbloomf quits (~nbloomf@2600:1700:83e0:1f40:fc2b:b27d:96e3:a54e) (Client Quit)
2020-10-17 14:21:06 × dwt quits (~dwt@c-98-200-58-177.hsd1.tx.comcast.net) (Ping timeout: 272 seconds)
2020-10-17 14:30:41 → wroathe joins (~wroathe@c-68-54-25-135.hsd1.mn.comcast.net)
2020-10-17 14:31:17 → GyroW joins (~GyroW@ptr-48ujrfd1ztq5fjywfw3.18120a2.ip6.access.telenet.be)
2020-10-17 14:31:17 × GyroW quits (~GyroW@ptr-48ujrfd1ztq5fjywfw3.18120a2.ip6.access.telenet.be) (Changing host)
2020-10-17 14:31:17 → GyroW joins (~GyroW@unaffiliated/gyrow)
2020-10-17 14:31:52 × GyroW_ quits (~GyroW@unaffiliated/gyrow) (Ping timeout: 272 seconds)
2020-10-17 14:32:25 × ericsagnes quits (~ericsagne@2405:6580:0:5100:1942:718:5977:178d) (Ping timeout: 240 seconds)
2020-10-17 14:33:47 × xff0x quits (~fox@2001:1a81:53c5:c200:800e:b588:fbcf:aa74) (Ping timeout: 260 seconds)
2020-10-17 14:33:59 <zincy__> Is it correct to refer to the parameter in Maybe's type constructor as a polymorphic type variable?
2020-10-17 14:34:38 → xff0x joins (~fox@2001:1a81:53c5:c200:3161:a29a:3186:4358)
2020-10-17 14:34:45 <L29Ah> s/polymorphic//
2020-10-17 14:35:15 <zincy__> Does that mean delete the word?
2020-10-17 14:35:28 → urodna joins (~urodna@unaffiliated/urodna)
2020-10-17 14:35:33 × lnlsn quits (~lnlsn@2804:14c:2b:891d::1007) (Ping timeout: 272 seconds)
2020-10-17 14:35:35 × kotrcka quits (~kotrcka@ip-94-112-194-11.net.upcbroadband.cz) (Quit: kotrcka)
2020-10-17 14:35:49 <L29Ah> yes
2020-10-17 14:36:14 → conal joins (~conal@64.71.133.70)
2020-10-17 14:36:18 × invaser quits (~Thunderbi@31.148.23.125) (Quit: invaser)
2020-10-17 14:36:44 → invaser joins (~Thunderbi@31.148.23.125)
2020-10-17 14:36:51 <zincy__> Ok thanks
2020-10-17 14:37:54 → dwt joins (~dwt@c-98-200-58-177.hsd1.tx.comcast.net)
2020-10-17 14:37:57 → bitmapper joins (uid464869@gateway/web/irccloud.com/x-nisutrzlzcwlcitv)
2020-10-17 14:38:48 <ski> zincy__ : not polymorphic
2020-10-17 14:39:03 × dhil quits (~dhil@195.213.192.122) (Ping timeout: 260 seconds)
2020-10-17 14:39:33 → Tops2 joins (~Tobias@dyndsl-095-033-019-139.ewe-ip-backbone.de)
2020-10-17 14:39:57 <zincy__> ski: Maybe I dont understand what polymorphism is but when I see a type variable I just think oh look polymorphic
2020-10-17 14:40:02 <zincy__> Is that misguided?
2020-10-17 14:40:03 <ski> (parametric types begets polymorphic operations on such types. so there is a relation)
2020-10-17 14:40:45 <ski> `length' is polymorphic. its type `[a] -> Int' is not polymorphic (nor is the type variable `a', in that type, polymorphic)
2020-10-17 14:41:15 <zincy__> Oh so polymorphism refers to operations on values of different types
2020-10-17 14:41:41 <zincy__> Whereas type variable is just about representation
2020-10-17 14:41:48 <ski> the explicit type of `length' is `forall a. [a] -> Int'. Haskell allows you to leave out the `forall' in source, and it'll be inserted implicitly by the language. but it's always, conceptually, there
2020-10-17 14:42:12 <ski> a value is polymorphic if and only if it has a type of general shape `forall a. ..a..'
2020-10-17 14:42:24 <zincy__> Ah thanks!
2020-10-17 14:42:47 <ski> just like a value is a list if and only if it has a type of general shape `[...]'. or is a function if and only if it has a type of general shape `... -> ...'
2020-10-17 14:42:54 <zincy__> So just functions can be polymorphic?
2020-10-17 14:43:10 × Tops2 quits (~Tobias@dyndsl-095-033-019-139.ewe-ip-backbone.de) (Client Quit)
2020-10-17 14:43:15 <ski> no. e.g. `Nothing' is not a function, but is still polymorphic. has type `forall a. Maybe a'
2020-10-17 14:43:56 <zincy__> Is Nothing not a data constructor and data constructors are functions?
2020-10-17 14:44:07 <ski> not all deata constructors are functions, no
2020-10-17 14:44:25 <ski> `Nothing' doesn't have a type that looks like `... -> ...'. hence it's not a function
2020-10-17 14:45:06 <zincy__> So only data constructors which are parameterised by at least one other value are functions?
2020-10-17 14:45:10 × jpcooper quits (~user@unaffiliated/jpcooper) (Ping timeout: 272 seconds)
2020-10-17 14:45:13 <ski> in fact, strictly speaking, `Just' isn't a function, either. it has type `forall a. a -> Maybe a'. it's a "polymorphic value (that when specialized, will become a function)"
2020-10-17 14:45:14 → ericsagnes joins (~ericsagne@2405:6580:0:5100:54da:6b21:e514:b8a2)
2020-10-17 14:45:18 × fragamus_ quits (~michaelgo@73.93.152.141) (Read error: Connection reset by peer)
2020-10-17 14:45:38 <ski> yes, data constructors which take arguments, which "pack data fields", are functions
2020-10-17 14:46:07 <ski> (or, in this case, "polymorphic functions", meaning "polymorphic value, that when specialized, will become a function")
2020-10-17 14:46:24 <zincy__> So forall a. id :: a -> a isn't a function?
2020-10-17 14:46:33 <zincy__> Until you parameterise the `a`?
2020-10-17 14:46:45 × Katarushisu quits (~Katarushi@82.30.254.24) (Quit: The Lounge - https://thelounge.chat)
2020-10-17 14:47:18 <ski> strictly speaking, it's not a function. but if we take `id :: forall a. a -> a', and specialize this, replacing `a' by `Bool' say, we get `id :: Bool -> Bool', which is a function
2020-10-17 14:47:51 <zincy__> Thanks, I have upgraded my thinking
2020-10-17 14:47:55 × unlink2 quits (~unlink2@p200300ebcf17c500f1cd4c5efbdddd1e.dip0.t-ipconnect.de) (Ping timeout: 240 seconds)
2020-10-17 14:48:02 <ski> one perhaps confusing part here is that the specializing of a polymorphic value is written as nothing, in the syntax. we still write `id'. but conceptually, it's an operation
2020-10-17 14:48:07 <zincy__> And to think I understood data constructors and Maybe :D
2020-10-17 14:48:21 <ski> (with a language extension, you can actually write `id @Bool' for this)
2020-10-17 14:49:00 <zincy__> Yeah so the polymorphic value becomes a function at some point during run time
2020-10-17 14:49:44 <ski> you can think of it as being a kind of expression node, in the abstract syntax tree in the implementation. it's just that it's (usually) written as nothing, in the source code
2020-10-17 14:50:00 → cole-h_ joins (~cole-h@c-73-48-197-220.hsd1.ca.comcast.net)
2020-10-17 14:50:25 → fragamus joins (~michaelgo@73.93.152.141)
2020-10-17 14:50:26 <zincy__> Ah thanks
2020-10-17 14:51:28 hackage path-io 1.6.1 - Interface to ‘directory’ package for users of ‘path’ https://hackage.haskell.org/package/path-io-1.6.1 (mrkkrp)
2020-10-17 14:51:47 × isBEKaml quits (~~.~@unaffiliated/isbekaml) (Quit: Leaving)
2020-10-17 14:52:00 → jneira_ joins (~jneira@111.red-176-83-68.dynamicip.rima-tde.net)
2020-10-17 14:52:11 <ski> something like `data Exp | Var Ident | Con Ident | App Exp Exp | Spec Exp Typ | ...' with `data Typ = TyVar Ident | TyCon Ident | Fun Typ Typ | Forall Ident Typ | ...'
2020-10-17 14:52:14 <zincy__> So is the general shape the same for parametric and adhoc polymorphism?
2020-10-17 14:52:29 <zincy__> Or is the difference there just the behaviour of the operation
2020-10-17 14:53:25 <ski> you could represent an expression like `not (id False)', using this AST, as `App (Var "not") (App (Spec (Var "id") (TyCon "Bool")) (Con "False"))'
2020-10-17 14:53:54 <ski> zincy__ : the difference is that the latter also involve type class constraints
2020-10-17 14:54:08 <zincy__> Ok gotcha
2020-10-17 14:54:34 <ski> `sort :: forall a. Ord a => [a] -> [a]' means `sort :: forall a. ((Ord a) => ([a] -> [a]))'
2020-10-17 14:54:45 <zincy__> So will that AST be reqritten when the `a` is specialised?
2020-10-17 14:54:59 <zincy__> And when would the rewriting occur?
2020-10-17 14:55:01 <ski> `... => ...' is a separate operation on types. you could call it the type of overloaded operations, i suppose
2020-10-17 14:55:56 <zincy__> Interesting
2020-10-17 14:57:08 <ski> well, in that example, `Var "id"' would be classified as having the type (namely `forall a. a -> a') expressed by `Forall "a" (Fun (TyVar "a") (TyVar "a"))'
2020-10-17 14:57:40 → knupfer joins (~Thunderbi@200116b82cc5f5003098c64aed4ab9ee.dip.versatel-1u1.de)
2020-10-17 14:58:02 → alp_ joins (~alp@2a01:e0a:58b:4920:417:348e:da6e:d990)
2020-10-17 14:58:37 <ski> specializing that to `Bool', we have that `Spec (Var "id")' is classified by `Fun (TyCon "Bool") (TyCon "Bool")'
2020-10-17 14:59:32 × hiroaki quits (~hiroaki@2a02:908:4b18:e20::a362) (Ping timeout: 246 seconds)
2020-10-17 15:00:01 × gnrp1 quits (~gnrp@84.39.117.57) ()
←Prev  Next→
Page 1 .. 697 698 699 700 701 702 703 704 705 706 707 .. 5022

All times are in UTC.