| Time |
Nick |
Message |
| 03:00 |
|
SFENCE_arch joined #luanti-dev |
| 05:00 |
|
MTDiscord joined #luanti-dev |
| 07:27 |
|
fluxionary joined #luanti-dev |
| 11:07 |
sfan5 |
any more comments on #15768? |
| 11:07 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/15768 -- [no sq] Split blockpos into 3 different columns in sqlite3 map database by sfan5 |
| 11:49 |
[MatrxMT] |
<Zughy> I swear triaging with the GH UI update is so annoying. My phone lags when I look for a label, video uploader is broken on my browser and sometimes pages don't really load. For fuck sake |
| 11:49 |
celeron55 |
sfan5: my comment is: go ahead, this should have been done a decade ago |
| 11:50 |
celeron55 |
(but a decade ago we might not have had the resources to take advantage of this, so it's not entirely nonsensical) |
| 12:52 |
sfan5 |
planning to merge #15768 this evening then |
| 12:52 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/15768 -- [no sq] Split blockpos into 3 different columns in sqlite3 map database by sfan5 |
| 12:52 |
sfan5 |
along with whatever other PRs are ready |
| 12:55 |
pgimeno |
wouldn't it also help with making the world bigger? |
| 13:20 |
|
MTDiscord joined #luanti-dev |
| 14:44 |
|
SFENCE_arch joined #luanti-dev |
| 14:44 |
|
Thomas-S joined #luanti-dev |
| 15:12 |
|
hwpplayer1 joined #luanti-dev |
| 15:45 |
MTDiscord |
<herowl> pgimeno: pretty sure it's not an issue with that |
| 17:53 |
|
hwpplayer1 joined #luanti-dev |
| 18:18 |
sfan5 |
it is time |
| 18:18 |
sfan5 |
in 10m |
| 18:18 |
|
Sharpman joined #luanti-dev |
| 18:19 |
|
Sharpman joined #luanti-dev |
| 18:30 |
sfan5 |
done |
| 18:30 |
sfan5 |
in an incredible record minetestmapper is the first external tool to support the new database schema: in an incredible record minetestmapper is the first external tool to support |
| 18:30 |
sfan5 |
whoops |
| 18:30 |
sfan5 |
https://github.com/luanti-org/minetestmapper/commit/6947c5c4e4d22a3ae1a3f8dd9cfc2bfbf911baec meant to link this |
| 18:36 |
jcm |
Supported the new schema before the actual engine did |
| 18:59 |
[MatrxMT] |
<y5nw> Can we also merge #15754 and #14602? (IMO for the former PR we can always come up with a shorter name later - there is IMO no need to block the PR over naming) |
| 18:59 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/15754 -- Add `table.copy_with_metatables` by appgurueu |
| 18:59 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/14602 -- Add API to cancel async jobs by y5nw |
| 19:01 |
sfan5 |
14602 is ready now that you're a coredev |
| 19:02 |
sfan5 |
the other one two but I'd like to still see a reply for the nameing suggestion |
| 19:03 |
sfan5 |
s/two/too/ |
| 19:03 |
sfan5 |
since it would be better to add the function once with the final name, and not change it later |
| 19:04 |
[MatrxMT] |
<y5nw> Ok, I'll just merge 14602 then. In 15m |
| 19:22 |
[MatrxMT] |
<y5nw> Actually forget that. I think I noticed something weird with the async unittests |
| 19:36 |
|
Desour joined #luanti-dev |
| 19:40 |
[MatrxMT] |
<Zughy> metacopy sounds like it's only gonna copy the metatable |
| 19:42 |
MTDiscord |
<luatic> i think table.copy_with_metatables is verbose, but most clear. my long-term plan would be to make this the default for table.copy as well, hence the note in the docs. |
| 19:43 |
MTDiscord |
<luatic> quite frankly i think that if someone's code breaks due to table.copy not stripping metatables, that code is probably ass, but i still don't want to break it unnecessarily just yet when it's rather easy to avoid. |
| 19:43 |
MTDiscord |
<luatic> perhaps i should also add this to breakages.md, even though it's a rather minor breakage? |
| 19:46 |
[MatrxMT] |
<y5nw> Idea: table.mpcopy() (metatable-preserving copy)? |
| 19:50 |
Desour |
I'm fine with table.copy_with_metatables. the name makes it obvious, and it's not like we need to golf a short name here |
| 19:50 |
[MatrxMT] |
<y5nw> About 14602: apparently one thing I overlooked in #14369 was that I put a metatable registration code in an async unittest, so the registration fails when you try to run the test again - but the coroutine wrapping ignores that error so the second unittest run simply doesn't end. Moving the metatable registration out of said unittest works but then other unrelated tests fail so 🤷 |
| 19:50 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/14369 -- Try to preserve metatable when exchanging data with the async env by y5nw |
| 19:55 |
[MatrxMT] |
<Zughy> as a modder, +1 for copy_with_metatables |
| 20:45 |
sfan5 |
fwiw the reason I suggested "metacopy" is to have a short and memorable name |
| 20:46 |
sfan5 |
"copy" isn't totally unambigous either: deep or shallow? with or without metatables? |
| 20:46 |
sfan5 |
though that disagrees with the idea of making meta-copying the default |
| 20:46 |
sfan5 |
so it doesn't matter really |
| 22:29 |
|
AliasAlreadyTake joined #luanti-dev |
| 23:12 |
|
Fleckenstein joined #luanti-dev |
| 23:34 |
|
panwolfram joined #luanti-dev |
