Newest at the top
| 2025-02-03 17:01:48 +0100 | <dminuoso> | Athas: Try haskell.nix |
| 2025-02-03 17:01:45 +0100 | <tomsmeding> | I see |
| 2025-02-03 17:01:25 +0100 | <Athas> | Not system libraries, necessarily - other libraries. gplk is one I needed recently. |
| 2025-02-03 17:01:21 +0100 | <tomsmeding> | the haskell stuff would just come from ghcup |
| 2025-02-03 17:01:05 +0100 | <tomsmeding> | what system libraries do you need? |
| 2025-02-03 17:00:53 +0100 | <Athas> | Then I need to learn how to use alpine, and perhaps I need a library that is not in alpine (or not statically linked), etc. It doesn't look like an easy path. |
| 2025-02-03 17:00:43 +0100 | <tomsmeding> | but I guess my threshold for bwrap is a lot smaller because I already know how it works from the playground :) |
| 2025-02-03 17:00:26 +0100 | <tomsmeding> | you need alpine I guess, but no need for docker, bwrap works fine enough |
| 2025-02-03 17:00:07 +0100 | <Athas> | That adds yet another piece of infrastructure (docker, alpine). |
| 2025-02-03 16:59:44 +0100 | <tomsmeding> | then see maerwald's post that I linked |
| 2025-02-03 16:59:23 +0100 | <Athas> | tomsmeding: the latter, sorry. |
| 2025-02-03 16:59:07 +0100 | <tomsmeding> | (and while you _can_ chain ViewPatterns, if you brain allows it, they would chain to the right, because the right-hand side of the ViewPatterns `->` is again a pattern, which could be a ViewPattern if you so desire) |
| 2025-02-03 16:58:33 +0100 | <tomsmeding> | but I didn't because it looks funny without |
| 2025-02-03 16:58:27 +0100 | <tomsmeding> | I recommend you put parentheses around the lambda |
| 2025-02-03 16:58:19 +0100 | <carbolymer> | right |
| 2025-02-03 16:58:18 +0100 | <carbolymer> | oh wait |
| 2025-02-03 16:58:17 +0100 | <tomsmeding> | the second is a ViewPattern arrow |
| 2025-02-03 16:58:14 +0100 | <tomsmeding> | the first arrow is a regular lambda arrow |
| 2025-02-03 16:58:08 +0100 | <carbolymer> | i.e. -> thing -> other thing |
| 2025-02-03 16:58:06 +0100 | <tomsmeding> | no this is not a chain |
| 2025-02-03 16:58:02 +0100 | <carbolymer> | ah I can chain them |
| 2025-02-03 16:57:58 +0100 | <tomsmeding> | the next step is just doing what you can always do in haskell: if you have `f = ...` then you can replace uses of `f` with `...` |
| 2025-02-03 16:57:39 +0100 | <tomsmeding> | you're already using viewpatterns |
| 2025-02-03 16:57:36 +0100 | <tomsmeding> | pattern Foo' s i <- (\(Foo s i) -> (s, i + length s) -> (s, i)) |
| 2025-02-03 16:57:31 +0100 | <carbolymer> | tomsmeding: with viewpatterns? |
| 2025-02-03 16:57:14 +0100 | <tomsmeding> | haskell is composable |
| 2025-02-03 16:57:12 +0100 | <tomsmeding> | carbolymer: you can make f an inline lambda function :p |
| 2025-02-03 16:54:21 +0100 | <tomsmeding> | for the latter: https://hasufell.github.io/posts/2024-04-21-static-linking.html (disclaimer: I haven't tried this, but it seems a lot more convenient than being forced to use nix) |
| 2025-02-03 16:54:00 +0100 | <tomsmeding> | Athas: statically link ghc, or statically link your own program with ghc? |
| 2025-02-03 16:53:32 +0100 | <carbolymer> | I understood that you're trying to build your program |
| 2025-02-03 16:53:22 +0100 | <carbolymer> | ah, link ghc |
| 2025-02-03 16:52:53 +0100 | <carbolymer> | tbf it's quite easy with haskell.nix |
| 2025-02-03 16:51:49 +0100 | <Athas> | I am getting quite depressed at how annoying it is becoming to statically link GHC with Nixpkgs. It is teethering on the bring of being so complicated that I cannot justify learning how to do it. |
| 2025-02-03 16:46:26 +0100 | fp1 | (~Thunderbi@2001:708:20:1406::10c5) fp |
| 2025-02-03 16:37:21 +0100 | alfiee | (~alfiee@user/alfiee) (Ping timeout: 265 seconds) |
| 2025-02-03 16:33:00 +0100 | alfiee | (~alfiee@user/alfiee) alfiee |
| 2025-02-03 16:29:26 +0100 | <ski> | mm, but yes, you could either put the context as another parameter of the judgement relation. or push that to the (linear) meta-logic |
| 2025-02-03 16:26:20 +0100 | eL_Bart0 | (eL_Bart0@dietunichtguten.org) |
| 2025-02-03 16:23:24 +0100 | <ski> | tomsmeding : well, you could use Lolli (linear logic programming language), for that resource splitting |
| 2025-02-03 16:23:23 +0100 | <ski> | kuribas : doing generalization can be a problem |
| 2025-02-03 16:21:20 +0100 | merijn | (~merijn@77.242.116.146) merijn |
| 2025-02-03 16:16:36 +0100 | merijn | (~merijn@77.242.116.146) (Ping timeout: 246 seconds) |
| 2025-02-03 16:08:10 +0100 | euleritian | (~euleritia@dynamic-176-006-146-039.176.6.pool.telefonica.de) |
| 2025-02-03 16:07:33 +0100 | euleritian | (~euleritia@ip4d17fae8.dynamic.kabel-deutschland.de) (Ping timeout: 248 seconds) |
| 2025-02-03 16:05:31 +0100 | forell | (~forell@user/forell) forell |
| 2025-02-03 16:04:06 +0100 | forell_ | (~forell@host-178-216-90-220.sta.tvknaszapraca.pl) (Quit: ZNC - https://znc.in) |
| 2025-02-03 15:58:33 +0100 | <carbolymer> | I mean, is it possible :D |
| 2025-02-03 15:58:23 +0100 | <carbolymer> | can I simplify this pattern: https://play.haskell.org/saved/nAhyanfL |
| 2025-02-03 15:57:49 +0100 | lxsameer | (~lxsameer@Serene/lxsameer) (Ping timeout: 252 seconds) |
| 2025-02-03 15:50:57 +0100 | sawilagar | (~sawilagar@user/sawilagar) (Quit: Leaving) |