2024/09/19

Newest at the top

2024-09-20 00:25:51 +0200tromp(~textual@92-110-219-57.cable.dynamic.v4.ziggo.nl) (Quit: My iMac has gone to sleep. ZZZzzz…)
2024-09-20 00:21:27 +0200chromo(~chromo@210.185.108.225)
2024-09-20 00:11:15 +0200Natch(~natch@c-9e07225c.038-60-73746f7.bbcust.telenor.se)
2024-09-20 00:10:16 +0200chromo(~chromo@210.185.108.225) (Ping timeout: 252 seconds)
2024-09-20 00:09:10 +0200 <geekosaur> oh, you could also put that in cabal.project.local, which is not intended to be in CI
2024-09-20 00:08:41 +0200 <geekosaur> not that that will help now
2024-09-20 00:08:32 +0200 <geekosaur> fwiw I just filed a cabal feature request
2024-09-20 00:06:47 +0200__monty__(~toonn@user/toonn) (Quit: leaving)
2024-09-20 00:05:57 +0200Natch(~natch@c-9e07225c.038-60-73746f7.bbcust.telenor.se) (Read error: Connection reset by peer)
2024-09-20 00:05:53 +0200chromo(~chromo@210.185.108.225)
2024-09-20 00:03:27 +0200target_i(~target_i@user/target-i/x-6023099) (Quit: leaving)
2024-09-20 00:03:23 +0200 <tritlo> Ah, it's in a CI so can't change the `cabal.project`. Thanks anyway, I'll use the `--prefer-oldest` until the issue is resolved
2024-09-20 00:03:06 +0200pavonia(~user@user/siracusa)
2024-09-20 00:02:03 +0200 <geekosaur> unfortunately, it doesn't look like '--build-tool-depends' is supported on the command line. if you can use a cabal.project, try https://paste.tomsmeding.com/W3BRWzp1
2024-09-20 00:00:51 +0200 <geekosaur> that constraint only applies to libraries, not executables.
2024-09-19 23:49:00 +0200peterbecich(~Thunderbi@syn-047-229-123-186.res.spectrum.com) (Ping timeout: 252 seconds)
2024-09-19 23:48:07 +0200 <tritlo> Any way I can force cabal to select `happy` 1.20.1.1? I can get a build plan with `happy` 1.19 if I do `--prefer-oldest`, but I'd rather select only the one. `--constraint "happy == 1.20.1.1"` doesn't seem to cut it
2024-09-19 23:46:50 +0200takuan(~takuan@178-116-218-225.access.telenet.be) (Remote host closed the connection)
2024-09-19 23:44:48 +0200chromo(~chromo@210.185.108.225) (Ping timeout: 246 seconds)
2024-09-19 23:40:29 +0200chromo(~chromo@210.185.108.225)
2024-09-19 23:36:06 +0200 <mauke> iamiam cedant tristia
2024-09-19 23:35:46 +0200tromp(~textual@92-110-219-57.cable.dynamic.v4.ziggo.nl)
2024-09-19 23:33:46 +0200ystael(~ystael@user/ystael) (Ping timeout: 265 seconds)
2024-09-19 23:28:41 +0200 <glguy> already already
2024-09-19 23:28:19 +0200 <geekosaur> already commented on 😛
2024-09-19 23:28:05 +0200 <tritlo> Already done so :)
2024-09-19 23:26:21 +0200 <glguy> You can set bounds on your build-tools
2024-09-19 23:24:04 +0200chromo(~chromo@210.185.108.225) (Ping timeout: 252 seconds)
2024-09-19 23:22:29 +0200misterfish(~misterfis@87.215.131.102)
2024-09-19 23:20:49 +0200peterbecich(~Thunderbi@syn-047-229-123-186.res.spectrum.com)
2024-09-19 23:19:49 +0200chromo(~chromo@210.185.108.225)
2024-09-19 23:18:38 +0200 <geekosaur> and sure enough, with happy-2.0 I get that error. go file a bug against happy, and tell the haskell-lib-parser folks (or maybe a hackage trustee) that haskell-lib-parser needs a constraint added (I think to build-tool-depends)
2024-09-19 23:18:26 +0200 <tritlo> Yeah seems like there's no version bound on `happy` in the executable
2024-09-19 23:16:43 +0200misterfish(~misterfis@87.215.131.102) (Ping timeout: 252 seconds)
2024-09-19 23:13:45 +0200 <mauke> https://www.youtube.com/watch?v=AzVPelmFKPk
2024-09-19 23:13:43 +0200 <mauke> happy 2: manic
2024-09-19 23:12:51 +0200 <geekosaur> executable version handling's a hack in the Haskell ecosystem
2024-09-19 23:12:26 +0200 <geekosaur> but actually I don't think either one of those will help because both are for libraries and happy's an executable
2024-09-19 23:11:35 +0200 <haskellbridge> <sm> 😂
2024-09-19 23:11:23 +0200 <mauke> my happiness knows no bounds
2024-09-19 23:11:04 +0200weary-traveler(~user@user/user363627) (Remote host closed the connection)
2024-09-19 23:10:58 +0200 <mauke> discovered
2024-09-19 23:10:55 +0200 <geekosaur> people are supposed to set bounds
2024-09-19 23:10:45 +0200 <haskellbridge> <sm> * created
2024-09-19 23:10:31 +0200 <haskellbridge> <sm> This is why stackage was invented
2024-09-19 23:10:11 +0200 <geekosaur> my local build just succeeded, with `happy-1.20.1.1`
2024-09-19 23:10:10 +0200 <tritlo> hmm, yes, the CI does run cabal update
2024-09-19 23:09:20 +0200 <geekosaur> like, earlier today
2024-09-19 23:09:02 +0200 <geekosaur> did you happen to do a `cabal update`? that looks like `happy`-generated code, and I note that `happy-2.0` was just released
2024-09-19 23:08:54 +0200 <tritlo> I think this file is autogenerated from the parser, so there might be something going on there. But I don't see how that would have changed now