Logs: liberachat/#haskell
| 2021-05-28 14:29:42 | → | tromp joins (~textual@dhcp-077-249-230-040.chello.nl) |
| 2021-05-28 14:31:50 | × | ddellacosta quits (~ddellacos@86.106.121.62) (Ping timeout: 244 seconds) |
| 2021-05-28 14:32:37 | → | safinaskar joins (~user@109-252-90-89.nat.spd-mgts.ru) |
| 2021-05-28 14:33:33 | → | sciencentistguy joins (~sciencent@212.102.63.133) |
| 2021-05-28 14:34:43 | <safinaskar> | hi! some time ago i asked on #haskell on freenode, why "length []" works. i got answer from geekosaur that answer is ExtendedDefaultRules. now i found that this is not true! look to ghc sources to https://github.com/ghc/ghc/blob/master/compiler/GHC/Builtin/Types.hs#L419 ! |
| 2021-05-28 14:34:47 | <safinaskar> | geekosaur: ping |
| 2021-05-28 14:36:08 | <safinaskar> | length has type "Foldable t => t a -> Int" |
| 2021-05-28 14:36:08 | <merijn> | What? Why would "length []" need ExtendedDefaultrules? |
| 2021-05-28 14:36:29 | × | sedeki quits (~textual@user/sedeki) (Ping timeout: 272 seconds) |
| 2021-05-28 14:36:35 | <safinaskar> | merijn: of course, it doesn't need ExtendedDefaultrules, as i learned right now |
| 2021-05-28 14:36:39 | <merijn> | [] isn't polymorphic unless you enable extensions |
| 2021-05-28 14:36:53 | <merijn> | So it's straightforward typechecking |
| 2021-05-28 14:36:56 | enikar | thinks it was with ghci |
| 2021-05-28 14:36:57 | <safinaskar> | merijn: [] has polymorphic type "[a]" |
| 2021-05-28 14:36:59 | × | ukari quits (~ukari@user/ukari) (Remote host closed the connection) |
| 2021-05-28 14:37:24 | <safinaskar> | ghc picks type "Any" for "a" in "Foldable t => t a -> Int", not "()", not anything else |
| 2021-05-28 14:37:28 | <merijn> | safinaskar: But Foldable doesn't care about 'a', so it's not relevant |
| 2021-05-28 14:37:30 | → | ukari joins (~ukari@user/ukari) |
| 2021-05-28 14:37:33 | <safinaskar> | read https://github.com/ghc/ghc/blob/master/compiler/GHC/Builtin/Types.hs#L419 |
| 2021-05-28 14:37:45 | <merijn> | GHC doesn't have to pick anything for 'a' |
| 2021-05-28 14:38:01 | → | ddellacosta joins (~ddellacos@89.46.62.116) |
| 2021-05-28 14:38:07 | <safinaskar> | merijn: but according to https://github.com/ghc/ghc/blob/master/compiler/GHC/Builtin/Types.hs#L419 , ghc still needs to pick something for "a", and ghc picks "Any" |
| 2021-05-28 14:42:17 | <cdsmith> | It's polymorphic in the element type. |
| 2021-05-28 14:42:19 | → | nate1 joins (~nate@108-233-125-227.lightspeed.sntcca.sbcglobal.net) |
| 2021-05-28 14:42:37 | → | monadlight joins (~chris@bras-vprn-nwmkon8540w-lp130-19-184-147-249-234.dsl.bell.ca) |
| 2021-05-28 14:42:50 | <merijn> | Yeah, but that's not relevant in the surface language |
| 2021-05-28 14:43:06 | <merijn> | The fact that GHC needs to pick something for it's Core IR is an implementation detail of GHC |
| 2021-05-28 14:43:44 | <geekosaur> | interesting |
| 2021-05-28 14:44:22 | <monadlight> | Hi haskellers! Newbie here. I'm trying to set up Emacs. Is haskell-language-server the way to go now? Looks like ghc-mod and dante are not updated for a while and possibly deprecated? |
| 2021-05-28 14:44:30 | × | z0k quits (~z0k@101.50.108.132) (Ping timeout: 248 seconds) |
| 2021-05-28 14:44:39 | × | ddellacosta quits (~ddellacos@89.46.62.116) (Remote host closed the connection) |
| 2021-05-28 14:44:50 | <merijn> | ghc-mod is deprecated and has been for multiple years by now (which it says in the readme) |
| 2021-05-28 14:45:08 | → | ddellacosta joins (~ddellacos@89.46.62.116) |
| 2021-05-28 14:45:08 | <merijn> | dante is stack only, afaik, and no clue how maintained it is |
| 2021-05-28 14:45:30 | <monadlight> | BTW, I haven't used IRC for a loooooog time... pardon my lack of proper etiquette if I do anything newbie-ish. |
| 2021-05-28 14:45:34 | <geekosaur> | I thought dante was the doesn't-require-stack version of intero? |
| 2021-05-28 14:45:42 | <merijn> | Oh, maybe? |
| 2021-05-28 14:45:44 | <merijn> | I dunno |
| 2021-05-28 14:46:16 | <monadlight> | merijn: you use lsp in emacs? |
| 2021-05-28 14:46:31 | <merijn> | I don't use emacs :p |
| 2021-05-28 14:46:38 | <geekosaur> | but in any case hls has been getting all the work over the past several months |
| 2021-05-28 14:46:50 | <geekosaur> | ather than have N different projects all going their own directions |
| 2021-05-28 14:46:52 | <merijn> | s/months/year(s) |
| 2021-05-28 14:47:15 | × | involans quits (~alex@cpc92718-cmbg20-2-0-cust157.5-4.cable.virginm.net) (Ping timeout: 272 seconds) |
| 2021-05-28 14:47:24 | <monadlight> | merijn: vim / vscode? |
| 2021-05-28 14:48:01 | <merijn> | vim |
| 2021-05-28 14:48:10 | <merijn> | My brain is to vim addicted to use anything else |
| 2021-05-28 14:48:22 | <monadlight> | N diffrent projects all going their own directions... Is that a good thing or bad thing? LOL |
| 2021-05-28 14:48:39 | → | dunham joins (~dunham@97-113-35-16.tukw.qwest.net) |
| 2021-05-28 14:48:43 | <merijn> | Any editors other than vim literally feels like missing a limb |
| 2021-05-28 14:48:49 | <geekosaur> | it means lots of duplicated or wasted effort |
| 2021-05-28 14:48:51 | <Hafydd> | My brain is, to vim, addicted. |
| 2021-05-28 14:48:52 | <monadlight> | merijn: I use Doom Emacs with vim (evil) binding |
| 2021-05-28 14:49:24 | → | zava joins (~zava@ip5f5bdf0f.dynamic.kabel-deutschland.de) |
| 2021-05-28 14:50:11 | <monadlight> | Might switch to Onivim when it becomes stable and more usable. |
| 2021-05-28 14:50:13 | × | ddellacosta quits (~ddellacos@89.46.62.116) (Ping timeout: 272 seconds) |
| 2021-05-28 14:50:14 | <merijn> | monadlight: I considered it, but migrating my current setup to emacs will probably take me weeks/months |
| 2021-05-28 14:50:17 | → | ddellaco_ joins (~ddellacos@86.106.121.23) |
| 2021-05-28 14:50:38 | <enikar> | merijn: I use dante, it's not stack only. |
| 2021-05-28 14:50:46 | <monadlight> | merijn: I can't live without magit. It's still the best git client |
| 2021-05-28 14:50:47 | ← | safinaskar parts (~user@109-252-90-89.nat.spd-mgts.ru) () |
| 2021-05-28 14:50:57 | <merijn> | monadlight: Oh, I have a simple life-hack for that |
| 2021-05-28 14:50:58 | <maerwald> | yeah, git in vim is just bad |
| 2021-05-28 14:51:00 | <maerwald> | I use tig |
| 2021-05-28 14:51:08 | <merijn> | I just don't use git #Winning |
| 2021-05-28 14:51:13 | <monadlight> | LOL |
| 2021-05-28 14:51:17 | <merijn> | Mercurial is my git interface |
| 2021-05-28 14:51:17 | <sciencentistguy> | > /cursor |
| 2021-05-28 14:51:19 | <lambdabot> | <hint>:1:1: error: parse error on input ‘/’ |
| 2021-05-28 14:51:24 | <sciencentistguy> | ah shit whoops |
| 2021-05-28 14:51:35 | → | slowButPresent joins (~slowButPr@user/slowbutpresent) |
| 2021-05-28 14:52:03 | <merijn> | I'm not joking, I use actual git probably less than twice a month on average |
| 2021-05-28 14:52:57 | × | j4ck quits (~jack@net-93-151-148-49.cust.dsl.teletu.it) (Quit: Leaving) |
| 2021-05-28 14:54:21 | <maerwald> | as if mercurial was any better... it's like Go is to C... sure, maybe better, but not to the point that anyone cares |
| 2021-05-28 14:54:37 | <monadlight> | Well, anything you can elimnate in life is a good thing. If I don't have to use git, I would be happier too. |
| 2021-05-28 14:54:59 | <maerwald> | just remember: git rebase is the one magic command that includes all others |
| 2021-05-28 14:55:07 | <monadlight> | maerwald: Agree... I never care for Go. Waste of my ever shortening life... ;-) |
| 2021-05-28 14:55:07 | <maerwald> | you basically just need that one |
| 2021-05-28 14:55:45 | <cdsmith> | I always wanted darcs to win, but that hope has died... |
| 2021-05-28 14:56:06 | <merijn> | maerwald: It's better from a UI perspective |
| 2021-05-28 14:56:07 | <maerwald> | well, because GHC dropped it |
| 2021-05-28 14:56:17 | <sciencentistguy> | not a fan of Go either. Rust is by far my favourite low-level language |
| 2021-05-28 14:56:22 | → | wroathe joins (~wroathe@c-68-54-25-135.hsd1.mn.comcast.net) |
| 2021-05-28 14:56:23 | <maerwald> | and they must have had goot reasons |
| 2021-05-28 14:56:49 | <merijn> | maerwald: I hope for something like darc/pijul, but neither of those are currently winning and I need to somehow interoperate with a damage world that prefers git |
| 2021-05-28 14:57:22 | <cdsmith> | Yeah, being the only user of a version control system is not useful, no matter how great the system! |
| 2021-05-28 14:57:38 | <nsilv> | the only reason i don't strongy dislike git right now is because i felt the pain of working with _Team Foundation Server_... still recovering from that |
| 2021-05-28 14:57:38 | <maerwald> | neither darcs nor pijul are simple. I don't believe in sophisticated stuff when it comes to everyday tools |
| 2021-05-28 14:57:44 | <jrm> | I just wish Rust didn't take ridiculous amounts of memory and time to build itself. |
| 2021-05-28 14:57:57 | <maerwald> | Implementation details will leak through... all the time, so you want the simplest and dumbest concept |
| 2021-05-28 14:57:59 | <maerwald> | and that is git |
| 2021-05-28 14:58:22 | × | _73 quits (~user@pool-96-252-123-136.bstnma.fios.verizon.net) (Ping timeout: 248 seconds) |
| 2021-05-28 14:58:24 | <maerwald> | the CLI interface is a different story |
| 2021-05-28 14:58:27 | <sciencentistguy> | jrm: To build rustc? yeah that is the downside of using an LLVM-based compiler |
| 2021-05-28 14:58:50 | <cdsmith> | maerwald: Not sure what you mean there. git looks far more complex than darcs ever was, to me. |
| 2021-05-28 14:58:57 | <Hafydd> | Does it take longer/more memory than GHC? |
| 2021-05-28 14:59:03 | <monadlight> | I've just gone through most of "Get Programming with Haskell" book. |
| 2021-05-28 14:59:08 | <maerwald> | cdsmith: the object model... a commit is a STATE. Period. |
| 2021-05-28 14:59:21 | <jrm> | sciencentistguy: Yes. When testing package builds, it's the only one that can take down my system if I'm not careful. |
| 2021-05-28 14:59:23 | <maerwald> | it's not a patch, because patch... what patch, compared to what? |
All times are in UTC.