Home liberachat/#haskell: Logs Calendar

Logs: liberachat/#haskell

←Prev  Next→
Page 1 .. 413 414 415 416 417 418 419 420 421 422 423 .. 17997
1,799,637 events total
2021-06-12 15:03:04 <kuribas> hmm, maybe using a lambda
2021-06-12 15:04:05 <kuribas> ok, it's ugly but it works.
2021-06-12 15:04:28 Crypt9a2e joins (~Crypt9a2e@2409:4072:301:9da3:81fa:805a:fd5b:12a4)
2021-06-12 15:05:00 × Crypt9a2e quits (~Crypt9a2e@2409:4072:301:9da3:81fa:805a:fd5b:12a4) (Read error: Connection reset by peer)
2021-06-12 15:05:20 Crypt9a2e joins (~Crypt9a2e@2409:4072:301:9da3:81fa:805a:fd5b:12a4)
2021-06-12 15:05:21 waleee joins (~waleee@h-98-128-228-119.NA.cust.bahnhof.se)
2021-06-12 15:05:54 bastelfreak joins (~bastelfre@basteles-bastelknecht.bastelfreak.org)
2021-06-12 15:05:57 <janus> kuribas: prefix all your record fields with the name of their record. If your record is called StudentGroupBinding, it's fields could be prefixed with "sgb"
2021-06-12 15:06:20 <kuribas> janus: that defeats the purpose for using them in the first place.
2021-06-12 15:07:42 <janus> kuribas: i don't understand. the point of a record is to group a bunch of data and get one value that refers to them. if the fields have a prefix, that is still possible
2021-06-12 15:08:33 <kuribas> I have types with a subset of fields. With recordWildCards I can do: toSmaller :: Bigger -> Smaller; toSmaller Bigger{..} = Smaller{..}
2021-06-12 15:08:43 <kuribas> But it depends on them having the same name.
2021-06-12 15:09:44 <kuribas> it saves a lot of boilerplate code.
2021-06-12 15:11:08 zebrag joins (~chris@user/zebrag)
2021-06-12 15:11:18 <janus> ok, i can see how it breaks that usecase. many of my records have smart constructors that i am using with applicative syntax, so i can't be doing that anyway
2021-06-12 15:11:44 <janus> i was told that the purescript records are much better, but i havn't tried them
2021-06-12 15:12:22 lavaman joins (~lavaman@98.38.249.169)
2021-06-12 15:12:41 <tomsmeding> maerwald: lol
2021-06-12 15:12:51 <maerwald> tomsmeding: not funny
2021-06-12 15:12:53 <tomsmeding> and it's not even windows
2021-06-12 15:13:21 <janus> kuribas: another approach is to have your data type be polymorphic and have "type Bigger = Polymorphic ExtraData" and "type Smaller = Polymorphic ()". then they can have the same fields ;)
2021-06-12 15:13:30 <maerwald> tomsmeding: yes... I sorta had this vision... if someone married windows and unix principles, would that be nix?
2021-06-12 15:13:43 <maerwald> powerful and bloated all the way
2021-06-12 15:13:46 × hueso quits (~root@152.170.216.40) (Read error: Connection reset by peer)
2021-06-12 15:14:03 <kuribas> janus: yeah, but bigger is autogenerated from a SQL schema :-)
2021-06-12 15:14:42 hueso joins (~root@152.170.216.40)
2021-06-12 15:16:35 nilof joins (~olofs@90-227-86-119-no542.tbcn.telia.com)
2021-06-12 15:17:09 × lavaman quits (~lavaman@98.38.249.169) (Ping timeout: 264 seconds)
2021-06-12 15:20:52 <DigitalKiwi> doesn't seem to be nix related
2021-06-12 15:20:54 jakalx joins (~jakalx@base.jakalx.net)
2021-06-12 15:21:55 wroathe joins (~wroathe@c-68-54-25-135.hsd1.mn.comcast.net)
2021-06-12 15:23:20 <DigitalKiwi> ls /usr/bin/
2021-06-12 15:23:21 <DigitalKiwi> env
2021-06-12 15:23:46 <maerwald> DigitalKiwi: I aborted and built it in 1/10th time with cabal
2021-06-12 15:24:14 <DigitalKiwi> kiwi@mvp-nixos ~ []$ welke mkdir
2021-06-12 15:24:18 <DigitalKiwi> /nix/store/wq2b68y4hi98zzm8p1jk7ybf23ly56m0-coreutils-8.32/bin/coreutils
2021-06-12 15:24:33 <tomsmeding> wtf is 'welke' and why is that dutch
2021-06-12 15:24:52 <DigitalKiwi> presumably it's which in dutch
2021-06-12 15:24:55 <tomsmeding> it is
2021-06-12 15:25:20 <DigitalKiwi> welke () { readlink -f "$(command -v "${1}")"}
2021-06-12 15:25:21 × dunkeln quits (~dunkeln@94.129.65.28) (Ping timeout: 252 seconds)
2021-06-12 15:26:01 <tomsmeding> when is that different from 'which'?
2021-06-12 15:27:04 <geekosaur> when you start a new process to see it, you see what the next shell to open would see, not what your current shell sees
2021-06-12 15:27:19 <geekosaur> this is why posix specifies "type" as a shell built-in
2021-06-12 15:27:19 × zebrag quits (~chris@user/zebrag) (Remote host closed the connection)
2021-06-12 15:27:23 <DigitalKiwi> kiwi@mvp-nixos ~ []$ which mkdir
2021-06-12 15:27:25 <DigitalKiwi> /run/current-system/sw/bin/mkdir
2021-06-12 15:27:30 yoctocell_ joins (~yoctocell@h87-96-130-155.cust.a3fiber.se)
2021-06-12 15:28:25 × yoctocell_ quits (~yoctocell@h87-96-130-155.cust.a3fiber.se) (Client Quit)
2021-06-12 15:28:33 × haltux quits (~haltux@a89-154-181-47.cpe.netcabo.pt) (Ping timeout: 264 seconds)
2021-06-12 15:28:56 yoctocell joins (~yoctocell@h87-96-130-155.cust.a3fiber.se)
2021-06-12 15:29:05 <DigitalKiwi> https://mostlyabsurd.com/files/2021-06-12-152839_2202x1023_scrot.png
2021-06-12 15:30:06 <DigitalKiwi> the next message from them was a gist to welke lol
2021-06-12 15:30:22 wallymathieu joins (~wallymath@81-234-151-21-no94.tbcn.telia.com)
2021-06-12 15:30:41 <DigitalKiwi> any more questions
2021-06-12 15:30:48 fizbin joins (~fizbin@c-73-33-197-160.hsd1.nj.comcast.net)
2021-06-12 15:33:05 dunkeln joins (~dunkeln@94.129.65.28)
2021-06-12 15:33:21 sedeki joins (~textual@user/sedeki)
2021-06-12 15:34:23 <DigitalKiwi> anyway i think maerwald is probably having PATH issues since...it's not actually using nix
2021-06-12 15:34:47 zebrag joins (~chris@user/zebrag)
2021-06-12 15:35:31 zmt00 joins (~zmt00@user/zmt00)
2021-06-12 15:37:32 × sedeki quits (~textual@user/sedeki) (Client Quit)
2021-06-12 15:39:15 × da39a3ee5e6b4b0d quits (~textual@2403:6200:8876:2955:f1c5:68ee:212f:e12f) (Quit: My MacBook has gone to sleep. ZZZzzz…)
2021-06-12 15:39:24 jakalx parts (~jakalx@base.jakalx.net) (Error from remote client)
2021-06-12 15:40:24 <maerwald> well, nix is in PATH
2021-06-12 15:40:33 <maerwald> The rest is up to nix, not me
2021-06-12 15:40:34 jakalx joins (~jakalx@base.jakalx.net)
2021-06-12 15:40:51 <DigitalKiwi> https://www.youtube.com/watch?v=0XH3oMNKApI this is really about nix
2021-06-12 15:41:06 × jao quits (~jao@cpc103048-sgyl39-2-0-cust502.18-2.cable.virginm.net) (Ping timeout: 268 seconds)
2021-06-12 15:45:11 <DigitalKiwi> i don't know who's responsibility it is to make the PATH right
2021-06-12 15:46:15 <tomsmeding> geekosaur: and... how is that different exactly? it's not like a new process would suddenly get a different environment than the current process magically
2021-06-12 15:46:39 <DigitalKiwi> you very clearly do not want it to work though and i don't know enough to help nor do you want helped so *shrug*
2021-06-12 15:47:22 <geekosaur> tomsmeding, PATH doesn't have to be exported from a shell to be used, plus there's caching of seen commands so changes aren't necessarily visible immediately
2021-06-12 15:47:49 <tomsmeding> geekosaur: ah right
2021-06-12 15:47:53 × dunkeln quits (~dunkeln@94.129.65.28) (Ping timeout: 268 seconds)
2021-06-12 15:48:07 <tomsmeding> having an unexported modified PATH sounds like a recipe for trouble though
2021-06-12 15:48:08 <DigitalKiwi> i can nixos-rebuild switch and that /run/current-system/sw/bin/mkdir will no longer point to /nix/store/wq2b68y4hi98zzm8p1jk7ybf23ly56m0-coreutils-8.32/bin/coreutils
2021-06-12 15:48:38 × lbseale quits (~lbseale@user/ep1ctetus) (Read error: Connection reset by peer)
2021-06-12 15:48:45 <DigitalKiwi> or even just nix-shell nix-env etc...
2021-06-12 15:53:19 × fizbin quits (~fizbin@c-73-33-197-160.hsd1.nj.comcast.net) (Remote host closed the connection)
2021-06-12 15:53:48 <geekosaur> tomsmeding, in general it is. csh auto-exported changes to $path for that reason, and you can still set that behavior in zsh iirc
2021-06-12 15:53:55 <DigitalKiwi> kiwi@mvp-nixos ~ []$ echo $NIX_PATH
2021-06-12 15:54:00 <DigitalKiwi> /home/kiwi/.nix-defexpr/channels:nixpkgs=/nix/var/nix/profiles/per-user/root/channels/nixos:nixos-config=/etc/nixos/configuration.nix:/nix/var/nix/profiles/per-user/root/channelskiwi@mvp-nixos ~ []$ echo $PATH/home/kiwi/bin:/run/wrappers/bin:/home/kiwi/.nix-profile/bin:/etc/profiles/per-user/kiwi/bin:/nix/var/nix/profiles/default/bin:/run/current-system/sw/bin
2021-06-12 15:57:06 <tomsmeding> geekosaur: if a shell variable is already set to exported, then the variable isn't any different from the exported one, right?
2021-06-12 15:57:24 <tomsmeding> so "auto-exporting changes to $path" sounds like a redundant statement to me assuming that PATH is exported anyway
2021-06-12 15:57:39 <geekosaur> correct, but shell variables don't inherit that from being imported
2021-06-12 15:57:47 <geekosaur> which is weird and counterintuitive
2021-06-12 15:58:18 ubert joins (~Thunderbi@p200300ecdf259ddf6d4493bcfbeaba19.dip0.t-ipconnect.de)
2021-06-12 15:58:24 <geekosaur> so while a shell's $PATH comes in from the environment, it is not marked as exported; the shell has to do an explicit `export PATH` for its changes to get exported
2021-06-12 15:59:20 <tomsmeding> ...no? $ bash -c 'echo $PATH' ; PATH="abc:$PATH" ; bash -c 'echo $PATH'
2021-06-12 15:59:34 <tomsmeding> the second time it shows an abc-prefixed path for me
2021-06-12 16:00:14 <geekosaur> are you sure something didn't do an export earlier, maybe in your dotfiles?
2021-06-12 16:00:19 <tomsmeding> but perhaps this is shell-dependent (this example is in bash)
2021-06-12 16:00:25 <geekosaur> it doesn't have to be repeated on every change.
2021-06-12 16:00:26 <tomsmeding> oh right
2021-06-12 16:00:33 <tomsmeding> oh that's dumb
2021-06-12 16:00:37 <geekosaur> yeh
2021-06-12 16:00:46 <geekosaur> weird and nonintuitive, like I said
2021-06-12 16:01:25 <geekosaur> and rather than fix it, POSIX specified ways to see what's really going on
2021-06-12 16:02:12 lavaman joins (~lavaman@98.38.249.169)

All times are in UTC.