2025/01/10

Newest at the top

2025-01-10 03:18:44 +0100vanishing(~vanishing@user/vanishingideal) vanishingideal
2025-01-10 03:17:39 +0100merijn(~merijn@128-137-045-062.dynamic.caiway.nl) (Ping timeout: 265 seconds)
2025-01-10 03:17:00 +0100orangeFlu(orangeFlu@gateway/vpn/protonvpn/orangeflu) orangeFlu
2025-01-10 03:16:56 +0100 <haskellbridge> <sm> stack is well designed, but haskell tooling is a bit complex, it takes more than a few minutes to grok
2025-01-10 03:16:25 +0100 <haskellbridge> <sm> oh yes, unless you specify it on command line it will probably look at the package.yaml generated from the template, and try to pick the snapshot that is likely to work with that
2025-01-10 03:16:02 +0100 <euouae> sm, you must be right. this is so difficult lol.
2025-01-10 03:15:14 +0100 <haskellbridge> <sm> eg {{name}} and {{category}} in https://github.com/commercialhaskell/stack-templates/blob/master/new-template.hsfiles. The stack.yaml snapshot is not one of those
2025-01-10 03:14:35 +0100 <haskellbridge> <sm> euouae: the parameters you can set, might be just the parameters defined in the template you are using ? not sure
2025-01-10 03:13:56 +0100 <euouae> It seems like it's deciding based on some heuristic instead of my setting
2025-01-10 03:13:48 +0100 <euouae> When I read the message under `stack new` I see: "Selecting the best among 13 snapshots..."
2025-01-10 03:13:43 +0100 <haskellbridge> <sm> if I hear more, I'll cc it here
2025-01-10 03:13:25 +0100 <haskellbridge> <sm> I can't test right now because of the usual transient network failures with some back end server :(
2025-01-10 03:13:10 +0100 <euouae> I've tried both resolver: lts-21.25, resolver: compiler: ghc-9.4.8, resolver: ghc-9.4.8 and snapshot: lts-21.25
2025-01-10 03:13:03 +0100merijn(~merijn@128-137-045-062.dynamic.caiway.nl) merijn
2025-01-10 03:12:50 +0100 <euouae> although `--resolver 21.25` works from the command line for `stack new`, and although `stack new` informs me that I can set parameters under ~/.stack/config.yaml uner templates: params: it does seem to get ignored
2025-01-10 03:11:53 +0100xff0x(~xff0x@fsb6a9491c.tkyc517.ap.nuro.jp)
2025-01-10 03:11:22 +0100itscaleb(~itscaleb@user/itscaleb) (Ping timeout: 265 seconds)
2025-01-10 03:10:58 +0100rdcdr(~rdcdr@user/rdcdr) (Ping timeout: 252 seconds)
2025-01-10 03:10:25 +0100 <haskellbridge> <sm> thank you. I don't use it for privacy currently
2025-01-10 03:09:24 +0100 <euouae> sm: if you care, <https://gitlab.com/libremonde-org/papers/research/privacy-matrix.org>, it's criticism of the privacy "features" of matrix
2025-01-10 03:04:15 +0100vanishingideal(~vanishing@user/vanishingideal) vanishingideal
2025-01-10 03:02:58 +0100 <euouae> I'm wondering if that was me being an idiot back then. Well I've made up my mind, I'm switching to cabal before I go nuts
2025-01-10 03:02:23 +0100 <euouae> it also calls it copyright: ... when it's a license. o_O
2025-01-10 03:02:10 +0100merijn(~merijn@128-137-045-062.dynamic.caiway.nl) (Ping timeout: 252 seconds)
2025-01-10 03:01:53 +0100 <haskellbridge> <sm> I think you're right that "stack new" will use whatever snapshot you have in your global user config
2025-01-10 03:01:44 +0100 <euouae> interesting that I've specified GPL-3.0 in my ~/.stack/config.yaml template but it still gave me BSD too
2025-01-10 03:00:29 +0100 <euouae> no, I meant to say that I dislike vector.im and their phony privacy thing
2025-01-10 03:00:05 +0100 <haskellbridge> <sm> ah. Well these are all FOSS rooms, unencrypted
2025-01-10 02:59:33 +0100 <euouae> I don't use matrix though, I'm bothered by their encryption
2025-01-10 02:58:55 +0100 <haskellbridge> <sm> #haskell-stack:matrix.org (https://matrix.to/#/#haskell-stack:matrix.org) will know for sure
2025-01-10 02:58:48 +0100 <geekosaur> it might not be mentioned because recent versions of stack renamed it to "snapshot"
2025-01-10 02:58:33 +0100 <haskellbridge> <sm> if I'm reading it right, https://github.com/commercialhaskell/stack-templates/blob/master/new-template.hsfiles seems to be the default template. But it doesn't seem to provide the stack.yaml file itself.
2025-01-10 02:57:41 +0100merijn(~merijn@128-137-045-062.dynamic.caiway.nl) merijn
2025-01-10 02:57:37 +0100 <euouae> apparently the value in ~/.stack/config.yaml must be indented to take effect
2025-01-10 02:56:55 +0100 <haskellbridge> <sm> https://docs.haskellstack.org/en/stable/commands/new_command/
2025-01-10 02:56:32 +0100 <haskellbridge> <sm> "stack new" is a command I think not many people use, do its docs say how to configure it ?
2025-01-10 02:56:27 +0100 <euouae> it does make me a little sekptical because I can't figure out in the docs where 'resolver' is mentioned
2025-01-10 02:56:09 +0100 <geekosaur> (I'm mostly a cabal user and only use stack when debuggig an xmonad user's config that's stack-based)
2025-01-10 02:55:48 +0100 <geekosaur> no, it just means I'm not actually a stack expert and don't know the exact place to edit off the top of my head 🙂
2025-01-10 02:54:46 +0100 <euouae> Maybe I shouldn't use stack?
2025-01-10 02:54:23 +0100 <euouae> I've tried to edit ~/.stack/config.yaml and ~/.stack/global-projects/stack.yaml but neither works with `stack new`
2025-01-10 02:54:06 +0100 <geekosaur> oh right, that's what you got at first, sorry
2025-01-10 02:54:00 +0100housemate(~housemate@pa49-185-174-252.pa.vic.optusnet.com.au) housemate
2025-01-10 02:53:37 +0100 <geekosaur> and nightlies are not especially stable
2025-01-10 02:53:24 +0100 <geekosaur> you probably want to do that anyway as that's a relatively old nightly; if you want ghc 9.8 you want LTS 23.3
2025-01-10 02:52:05 +0100 <geekosaur> edit that file in ~/.stack to specify the snapshot you want by default
2025-01-10 02:51:10 +0100 <euouae> sigh this is difficult to figure out... I guess I can use stack new --resolver
2025-01-10 02:51:02 +0100 <euouae> how can I make this the default behavior?
2025-01-10 02:48:03 +0100hawer(~newyear@2.219.56.221)
2025-01-10 02:47:52 +0100_73(~user@pool-173-76-100-193.bstnma.fios.verizon.net) _73