| 14 Aug 2025 |
 emily | well, the large integers will still get clobbered precision-wise | 16:54:23 |
| emily | that's sort of inevitable though | 16:54:32 |
| emily | and why people who put large integers into JSON put them in strings generally | 16:54:42 |
 jade_ | yes, i just mean that they had it coming to a degree; having any json parsing error out is unfortunate | 16:54:52 |
| emily | Nixpkgs just does the funny thing because of Nix's lack of useful parsing functions | 16:54:53 |
| emily | right | 16:54:56 |
| emily | I will probably prepare a revert of that fromJSON change in the absence of strong objection / someone who wants to try and make it do the same for negative integers | 16:55:24 |
| emily | I wish someone had told Douglas Crockford what semantics are | 16:55:39 |
| emily | further context for this is https://github.com/NixOS/nixpkgs/pull/433710 because it turns out that fromTOML is a total mess (working on a stack locally) | 16:56:05 |
| jade_ | i would like consistent behaviour with pos/neg integers ideally and i would rather it be a total function. so ya | 16:56:08 |
| emily | btw what's the intended way to update functional2 snapshots with a just-built Lix? | 16:57:29 |
| emily | PATH=build/… hacks? | 16:57:33 |
| jade_ | "did we wire that in yet" -helle | 16:59:28 |
| emily | hm, seems that doesn't work | 16:59:29 |
| emily | btw, Nix recently started forbidding NUL bytes in string literals, fromJSON, etc.; how do we feel about backporting that? | 17:00:18 |
| emily | I know it's "language versioning" kind of territory but the behaviour is also super broken so if they're pulling off the bandaid already… | 17:00:36 |
| jade_ | I think it's the _NIX_TEST_ACCEPT=1 thing, and iirc you can just run functional2 with pytest in lix (assuming the fresh lix is in fact in PATH and you've rehashed in zsh etc) | 17:00:59 |
| emily | yes the problem is getting it in $PATH | 17:01:27 |
| emily | looks like you need to do an install first | 17:01:30 |
| emily | because the nix-* symlinks | 17:01:35 |
| jade_ | this seems like a good idea, yes. it will break some of puck 's eval-breaking art projects, but that's good actually. | 17:02:32 |
| emily | ok I'll pile those on the stack then :P | 17:02:49 |
| jade_ | that's correct, yeah, i don't know where that's supposed to be documented | 17:03:09 |
 puck | In reply to @jade_:matrix.org
this seems like a good idea, yes. it will break some of puck 's eval-breaking art projects, but that's good actually. i expected it to break some day, tbh | 17:03:10 |
| puck | and i think i no longer depend on it in any thing | 17:03:36 |
| emily | btw https://github.com/NixOS/nixpkgs/pull/433729 is also funny for CI | 17:03:57 |
| emily | turns out some Nixpkgs lib tests were not in fact testing the built interpreter | 17:04:11 |
| jade_ | unrelated question: why is lix not also codeowners on those files? i mean, i think we would have to convince someone to make us a group but ..... | 17:05:21 |
 Rutile (Commentator2.0) feel free to ping | In reply to @emilazy:matrix.org
btw what's the intended way to update functional2 snapshots with a just-built Lix? The env variable and just test-functional2 --accept-tests | 17:05:28 |
| emily | probably just because nobody did it? | 17:05:42 |
