| 12 Jan 2026 |
 raitobezarius | in practice, breaking backward compat of it is not a good idea | 18:44:31 |
 Robert Hensing (roberth) | they're already part of the sediment record | 18:44:40 |
| raitobezarius | flakerefs as a term should die | 18:44:46 |
| raitobezarius | and we should all pretend they are actually URLs | 18:44:52 |
| raitobezarius | there, they're not about flakes anymore | 18:45:00 |
| Robert Hensing (roberth) | I've called them sourcerefs occasionally when talking about fetchTree | 18:45:33 |
| Robert Hensing (roberth) | flakref = sourceref + ?dir | 18:45:49 |
 Sergei Zimmerman (xokdvium) | And also git's URL (but I think nowadays nix doesn't actually support some of these cuz fetchTree) is a huge mess too. Just ran into:
GIT_TRACE=1 git clone "xokdvium@[localhost]hahahahahahhahaha:/~/work/code/nix"
| 18:46:09 |
| Sergei Zimmerman (xokdvium) | Any bets if this works or not? | 18:46:16 |
| Sergei Zimmerman (xokdvium) | * And also git's URL (scp-like paths they call these) (but I think nowadays nix doesn't actually support some of these cuz fetchTree) is a huge mess too. Just ran into:
GIT_TRACE=1 git clone "xokdvium@[localhost]hahahahahahhahaha:/~/work/code/nix"
| 18:47:04 |
 Ilan Joselevich (Kranzes) | is -I supposed to behave differently of NIX_PATH? | 18:47:21 |
| raitobezarius | yes | 18:47:26 |
| raitobezarius | because -I uses the setting system | 18:47:46 |
| Ilan Joselevich (Kranzes) | Is this documented somewhere? what is the intended behavior? | 18:47:59 |
| raitobezarius | in many reasonable software, an environment variable serves to seed an initial default | 18:48:32 |
| raitobezarius | there's 3 provenances for the setting system: NIX_CONF, the default settings file locations, -I | 18:49:07 |
| raitobezarius | and 1 provenance via env var: NIX_PATH | 18:49:12 |
|  isabel changed their profile picture. | 18:59:01 |
 John Ericson | snix should just not implement NIX_PATH I think | 19:55:37 |
| John Ericson | at least not like it currently exists | 19:55:41 |
| John Ericson | I do not wish that pain upon you all! :) | 19:56:09 |
 flokli | We don't have the luxury, at least the common / sensible usages should work | 19:56:17 |
| John Ericson | I thought snix was deprioritzing the evaluator as a whole? | 19:56:38 |
| flokli | Who said that? 😄 | 19:57:22 |
| John Ericson | a conservative NIX_PATH implementation would split on :, and require every entry to start with a / | 19:57:50 |
| John Ericson | just don't support URLs at all | 19:58:05 |
| John Ericson | and don't accidentally treat a URL as two valid relative paths | 19:58:27 |
| John Ericson | * and don't accidentally treat a URL as two valid relative path entries | 19:58:31 |
| John Ericson | (no foobar://x becomes foobar and ///x) | 19:59:12 |
| John Ericson | * (no foobar://x becomes foobar and //x) | 19:59:20 |
