|
07:05
lizmat left,
lizmat joined
07:34
patrickb joined
07:40
patrickz joined
07:41
patrickb left
19:01
patrickz left
|
|||
| lizmat | JRaspass: I just notice that on a page such as raku.land/cpan:JNTHN/Cro::HTTP | 20:54 | |
| the version that is activeis an old one (0.8.2.2) rather than the most recent (0.8.6) | 20:55 | ||
| looks like the versions with 4 parts are sorted before the ones with 3 parts ? | |||
| tonyo | it gets sorted that way in zef too | 21:06 | |
| lizmat | isn't that... wrong ? | 21:07 | |
| afk for the rest of the day& | 21:08 | ||
| JRaspass | ooo 4 part versions, that's a fun one | 21:09 | |
| I guess I ought to agree with raku: $ raku -e 'say max v0.8.2.2, v0.8.6' # v0.8.6 | 21:11 | ||
| oh i think it's a shocking bug in the sql | 21:22 | ||
| good spot | |||
| tonyo | the version comparison in raku is more than just parts comparison | 21:27 | |
| ~ ❱ raku -e 'say max Version.new("0.8.a"), v0.8.5' (base) | 21:28 | ||
| v0.8.5 | |||
| JRaspass | yeah raku's is smarter than mine, but my bug was silly, it's fixed now | 21:41 | |
| gitlab.com/raku-land/raku-land/-/c...t/f1ed9247 | |||
| i probably need to port more logic to mine but it also needs proper service tests with a spun up pg, this'll do for now | 21:42 | ||