2024/04/29

Newest at the top

2024-04-29 06:28:16 +0200 <jackdk> You'll need a 9.8-series `wasm32-wasi-ghc`, from wherever you got the 9.10-series one from. I'm not really sure, sorry, I mostly work with native code
2024-04-29 06:27:17 +0200 <andrewboltachev> i.e. which has the wasm32-wasi-ghc
2024-04-29 06:27:02 +0200 <andrewboltachev> the whole `wasm32-wasi` stack?
2024-04-29 06:26:30 +0200 <andrewboltachev> okay. but, what should I reinstall?
2024-04-29 06:23:47 +0200 <jackdk> You could try that, and if it has enough WASM support for what you're trying I'd recommend it
2024-04-29 06:23:22 +0200 <andrewboltachev> jackdk: maybe I can instead shift down to 4.19?
2024-04-29 06:23:21 +0200 <jackdk> `constraints: bytestring ^>= 0.12.1.0, unordered-containers ^>= 0.2.20`, I guess. Welcome to life on the bleeding edge
2024-04-29 06:22:44 +0200 <jackdk> The re-export of `foldl'` from `Prelude` is a new change in `base-4.20.0.0`: https://gitlab.haskell.org/ghc/ghc/-/blob/master/libraries/base/changelog.md?ref_type=heads . Add another constraint to your `cabal.project` to select `unordered-containers ^>=0.2.0`, and you may need to add more `allow-newer`: https://hackage.haskell.org/package/unordered-containers-0.2.20/changelog
2024-04-29 06:21:40 +0200tri(~tri@ool-18bc2e74.dyn.optonline.net)
2024-04-29 06:19:46 +0200 <andrewboltachev> I guess earliest one that doesn't clash with foldl'
2024-04-29 06:19:25 +0200 <andrewboltachev> how do I define which ver. to pick for unordered-containers (and other library)?
2024-04-29 06:18:59 +0200 <andrewboltachev> similar foldl' error though: https://paste.tomsmeding.com/ao8zbeZA
2024-04-29 06:18:34 +0200 <jackdk> I am delighted to hear this - does the PR have an output sample?
2024-04-29 06:17:34 +0200 <sclv> may not help enough, but its something
2024-04-29 06:16:59 +0200 <sclv> we have a big pr landing on it
2024-04-29 06:16:36 +0200 <jackdk> Someone should do something about cabal's solver output, because while it has the information most people need, it fails to communicate it clearly.
2024-04-29 06:15:47 +0200michalz(~michalz@185.246.207.203)
2024-04-29 06:15:38 +0200 <andrewboltachev> jackdk: thanks. I'm still a noob in Haskell, even though I created this project myself: https://github.com/andrewboltachev/matcher/blob/master/src/Logicore/Matcher/Core.hs
2024-04-29 06:12:39 +0200 <jackdk> That is a three-year-old version of `bytestring`, and it had a correct upper bound which we just told GHC to ignore. Try adding `constraints: bytestring ^>= 0.12.1.0` to your `cabal.project` to see if it gives you a better shot at building.
2024-04-29 06:11:45 +0200hgolden(~hgolden@2603:8000:9d00:3ed1:2678:8497:aa5c:7fa9)
2024-04-29 06:11:40 +0200 <andrewboltachev> will try cabal.project file with this now (didn't work 'cause of "base" error as well): https://github.com/tweag/ormolu/tree/master/ormolu-live
2024-04-29 06:10:27 +0200 <andrewboltachev> wow: https://paste.tomsmeding.com/AKjcvAqu
2024-04-29 06:09:30 +0200ski(~ski@ext-1-033.eduroam.chalmers.se)
2024-04-29 06:09:08 +0200ski(~ski@ext-1-033.eduroam.chalmers.se) (Ping timeout: 252 seconds)
2024-04-29 06:08:19 +0200hgolden(~hgolden@2603:8000:9d00:3ed1:f849:272c:fda5:33c9) (Remote host closed the connection)
2024-04-29 06:07:34 +0200 <jackdk> Create a `cabal.project` file next to your `.cabal` file, telling cabal that it's allowed to relax the `base` bound for all pacakges: https://www.irccloud.com/pastebin/anN72RgX/cabal.project
2024-04-29 06:07:34 +0200 <andrewboltachev> I think I've launched an exported function from the wasm file successfully
2024-04-29 06:06:33 +0200 <andrewboltachev> The Glorious Glasgow Haskell Compilation System, version 9.11.20240420
2024-04-29 06:06:32 +0200 <andrewboltachev> ➜ h1 git:(master) ✗ wasm32-wasi-ghc --version
2024-04-29 06:05:56 +0200 <andrewboltachev> yes, I think I need that
2024-04-29 06:05:22 +0200 <jackdk> Also, if you're playing with WASM, only 9.10 has the WASM->JS FFI
2024-04-29 06:05:20 +0200 <andrewboltachev> as deps that are deep in the tree ask for 4.19. 4.20 is too new (and an answer to all of the universe or sth)
2024-04-29 06:05:07 +0200 <jackdk> It is possible to override those bounds, one sec
2024-04-29 06:04:30 +0200 <andrewboltachev> so I think I should get compiler that will provide 4.19
2024-04-29 06:04:09 +0200 <andrewboltachev> well it's this then: https://paste.tomsmeding.com/fT1guof4
2024-04-29 06:03:59 +0200_ht(~Thunderbi@28-52-174-82.ftth.glasoperator.nl)
2024-04-29 06:02:32 +0200 <andrewboltachev> jackdk: aha, trying
2024-04-29 05:56:02 +0200 <jackdk> (`s/I would get/I would guess/`)
2024-04-29 05:55:34 +0200 <jackdk> andrewboltachev: whatever your `h1` package is (is it the name of your test program?), it's claiming a dependency on `base-4.19.1.0`, but the compiler you're using provides `base-4.20.0.0`. https://wiki.haskell.org/Base_package has not updated to list `4.20.0.0`, but I would get it's the GHC 9.10 RC? I would try adjusting the `base` bound in `h1.cabal` to `== 4.20.0.0` to see if that gets you moving.
2024-04-29 05:53:16 +0200cashew(~cashewsta@65.17.175.150) (Remote host closed the connection)
2024-04-29 05:52:41 +0200 <andrewboltachev> base==4.19.1.0)
2024-04-29 05:52:41 +0200 <andrewboltachev> [__1] rejecting: base-4.20.0.0/installed-inplace (conflict: h1 =>
2024-04-29 05:52:40 +0200 <andrewboltachev> [__1] next goal: base (dependency of h1)
2024-04-29 05:52:40 +0200 <andrewboltachev> [__0] trying: h1-0.1.0.0 (user goal)
2024-04-29 05:52:39 +0200 <andrewboltachev> Error: cabal: Could not resolve dependencies:
2024-04-29 05:52:38 +0200 <andrewboltachev> I have a simple WebAssembly project and Cabal is saying me:
2024-04-29 05:52:09 +0200xigua(~xigua@user/xigua)
2024-04-29 05:51:55 +0200xigua(~xigua@user/xigua) (Read error: Connection reset by peer)
2024-04-29 05:51:47 +0200 <andrewboltachev> Hello! Does anyone has example of successful wasm32-wasi-cabal builds? Aka cabal in non-standard environments
2024-04-29 05:51:01 +0200andrewboltachev(~andrewbol@178.141.238.156)