Time |
Nick |
Message |
04:00 |
|
MTDiscord joined #luanti-dev |
07:04 |
|
TenPlus1 joined #luanti-dev |
07:04 |
TenPlus1 |
Hi folks |
07:04 |
TenPlus1 |
Question, are we moving away from using the minetest. commands entirely and replacing them with core. ??? |
07:10 |
Krock |
TenPlus1: for historical reasons this can only be enforced in a major release, such as 6.0 |
07:10 |
Krock |
realistically I don't think it will ever be enforced. Maybe there will be recommendation. Who knows. |
07:11 |
TenPlus1 |
Hi Krock, I'm hoping minetest. will be around as an alias so older mods wont break |
07:12 |
TenPlus1 |
*around for a long time |
07:13 |
* TenPlus1 |
awaits an official forum post to state their intentions :P |
07:15 |
Krock |
I think every core developer and modder is aware of that. No need to panic. |
07:22 |
TenPlus1 |
I just saw some mods in ContentDB being updated to use core. and it had me worried for a sec there :P |
07:27 |
* TenPlus1 |
wonders if mod.conf will have similar changes eventually e.g. min_minetest_version |
07:28 |
Krock |
TenPlus1: the more difficult question is likely on how to migrate the .conf files and ~/.minetest directory. |
07:31 |
TenPlus1 |
good question, I suppose you could check for .minetest first of all and .luanti second if it isnt found |
07:32 |
TenPlus1 |
same with minetest.conf, if found copy to luanti.conf and rename original |
07:35 |
Krock |
renaming breaks scripts that rely on exact paths :3 |
07:36 |
Krock |
and you cannot symlink the files because Windows and filesystems that do not support symlinks. |
07:36 |
TenPlus1 |
yeah, unless you alias the folders but dunno if windows handles thast well |
07:36 |
TenPlus1 |
lol |
07:37 |
Krock |
and a --migrate CLI option would only be used by a handful of users. So it can only be applied to new installs, hence just as you said. |
07:37 |
TenPlus1 |
would have been easier sticking to Minetest as the name, lolol |
07:38 |
sfan5 |
I think pro users can live with needing to adjust their scripts once in 10 years after a name change |
07:39 |
TenPlus1 |
it's an easy change in the end, just gotta post something on the forums about it so everyone knows |
07:39 |
TenPlus1 |
also Hi sfan5 |
07:41 |
TenPlus1 |
I suppose Luanti 6.0 would make for a good turning point when it comes to file/folder names and settings |
07:47 |
TenPlus1 |
breakfast time, laters o// |
08:22 |
|
Warr1024 joined #luanti-dev |
08:46 |
|
Warr1024 joined #luanti-dev |
09:15 |
[MatrxMT] |
<Zughy> Reminder: meeting today |
09:16 |
[MatrxMT] |
<Zughy> I think inventory key and GUIDs should be the pivotal points, since we are/were supposed to release in a week, but I frankly don't see it happening |
10:06 |
|
YuGiOhJCJ joined #luanti-dev |
14:19 |
|
MTDiscord joined #luanti-dev |
14:20 |
MTDiscord |
<luatic> merging #16109, #16111, #16083 in 10m |
14:20 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/16109 -- Minor code improvements to image algorithms by sfan5 |
14:20 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/16111 -- Purge some dead code (mostly Irrlicht) by appgurueu |
14:20 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/16083 -- [no squash] Revamp `dump` by appgurueu |
14:22 |
|
Warr1024 joined #luanti-dev |
14:33 |
MTDiscord |
<luatic> done |
15:18 |
|
MTDiscord1 joined #luanti-dev |
15:19 |
sfan5 |
when is the meeting again |
15:19 |
sfan5 |
or did we miss it |
15:21 |
MTDiscord1 |
<luatic> idk i'm just waiting for it to happen |
15:22 |
sfan5 |
well I have to hang up my laundry now |
15:27 |
MTDiscord |
<nathan4220776> Excellent. |
15:39 |
sfan5 |
ok done |
15:40 |
Krock |
> Time 18:00 UTC |
15:41 |
Krock |
that's in 2 hours and 20 minutes |
16:53 |
|
Desour joined #luanti-dev |
16:57 |
Desour |
>is there a reason we don't allow waving for mesh nodes? |
16:57 |
Desour |
I wish waving was always just allowed drawtype specific |
16:57 |
Desour |
liquid-waving nodeboxes are horrible |
16:58 |
Desour |
for example |
17:02 |
Krock |
wobbling liquids would actually be funny to have to build a jelly factory |
17:04 |
MTDiscord |
<nathan4220776> And a physics simulation nightmare. |
17:05 |
MTDiscord |
<nathan4220776> Unless it's something simple like the "mad milk" in tf2. |
17:05 |
Desour |
that wouldn't be our liquid waving then though |
17:47 |
Desour |
is there a reason why flowingliquid uses special_tiles? |
17:50 |
Krock |
Wasn't there a setting to disable liquid animations? I think Minetest (0.4.7 or whichever version) would then just use the static images |
17:50 |
Krock |
but don't quote me on that. It's mostly just interpretation |
17:51 |
Krock |
Meeting in 9 minutes |
17:51 |
Desour |
but source liquid uses usual tiles, so that doesn't explain it |
17:53 |
|
SFENCE joined #luanti-dev |
17:59 |
Krock |
Pinging those who reacted to the meeting post: Zughy sfan5 y5ne @y5ne (me) |
18:00 |
SFENCE |
I am here, but in the train on the way. |
18:00 |
Krock |
thankfully we don't need to be at the same location to participate in this meeting :3 |
18:01 |
sfan5 |
I am sill composing a github comment |
18:01 |
sfan5 |
still* |
18:01 |
SFENCE |
Yes, but connection in train can be unstable sometimes. :D |
18:02 |
Krock |
Let's wait a few minutes. Today's goal is to enter feature freeze --> https://github.com/luanti-org/luanti/milestone/26 |
18:02 |
MTDiscord |
<luatic> I am here too |
18:02 |
Krock |
on a side note, we might as well discuss https://github.com/luanti-org/docs.luanti.org/pull/228 (release schedule) after processing the milestone |
18:05 |
Krock |
Starting with the PRs first. |
18:05 |
Krock |
> #16053 |
18:05 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/16053 -- Change default inventory key to E, default Aux1 key to F by minlemon |
18:06 |
Krock |
I already left a comment there. Is it urgent to address? |
18:06 |
sfan5 |
quoting myself: |
18:06 |
sfan5 |
That is true, but this isn't my reason for proposing to do this change in 5.12. |
18:06 |
sfan5 |
The reasoning is that for many people this release will cause at least one noticeable key change, so it would be a good idea to merge it with another change (change fatigue). |
18:06 |
|
Desour joined #luanti-dev |
18:07 |
Krock |
oh that's the comment you were referring to |
18:08 |
sfan5 |
IMO if we decide not to do it now then it's also no problem to delay it a few more releases |
18:08 |
sfan5 |
(though I still think the engine default bindings should switch away from I) |
18:09 |
sfan5 |
so what we need to discuss is: should any change to the default bindings be done now, or not? |
18:10 |
[MatrxMT] |
<y5nw> IMO it would be nice to have it now, but if we want to start a poll or don't have a consensus yet then we can also delay it |
18:11 |
Krock |
I do see that it's a great opportunity for changes. Question is whether that's generally accepted. I actually wouldn't mind too much. |
18:13 |
Krock |
counter-proposal: merge it now for an rc1 build and collect user feedback this way |
18:13 |
sfan5 |
ideally we would also have the dialog ready for rc1 then |
18:14 |
sfan5 |
though windows users having run_in_place builds mostly breaking any migration ideas (we should fix that some day) |
18:14 |
sfan5 |
mostly breaks* |
18:14 |
SFENCE |
Potencially... are we able to add/show some windows on first lunati start and show controls to user? So, it should fix confusion problem for users and existing tutorials. |
18:14 |
Krock |
I'm sure they'll carry over the minetest.conf file once they customized Luanti enough |
18:15 |
Krock |
SFENCE: that's exactly what sfan5 just mentioned. PR #16049 |
18:15 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/16049 -- Migrate existing keycode-based keybindings; add warning dialog about migration by y5nw |
18:16 |
Krock |
I'll review that soon so that we can provide an rc1 soon after |
18:16 |
SFENCE |
Krock: Nice. |
18:16 |
sfan5 |
can we ping the entire engine team on the PR to get feedback? |
18:17 |
Krock |
sounds good. by counting reactions. |
18:19 |
Krock |
'@luanti-org/engine' should do the trick. Would you like to do that, sfan5 ? |
18:19 |
sfan5 |
sure |
18:19 |
Krock |
Thank you :) |
18:20 |
Krock |
> continuing. #16017 |
18:20 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/16017 -- Complete rename to Luanti in package metadata by appgurueu |
18:20 |
Krock |
skipping. to be merged before the 5.12.0 release |
18:21 |
Krock |
> #15898 |
18:21 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/15898 -- Node specular needs to be disabled *or* server-configurable and off by default |
18:21 |
sfan5 |
disabling it is a relatively easy fix |
18:21 |
|
behalebabo joined #luanti-dev |
18:21 |
sfan5 |
s/relatively // |
18:21 |
Krock |
no blocker, thus strictly speaking not necessary to fix 8) |
18:22 |
Desour |
I see a blocker label |
18:22 |
Krock |
I'm blind. I can propose a PR for that. Removing 'enable_node_specular' for now, default to false. |
18:23 |
Krock |
s/I can/I will/ |
18:23 |
Krock |
> #15883 |
18:23 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/15883 -- Font texture turns black after app is in background (Android) |
18:23 |
Krock |
> It would be helpful if someone could reproduce this on a debug build ~ sfan5 |
18:23 |
Krock |
are there any debug builds available? |
18:24 |
Krock |
or "bake it yourself"? |
18:24 |
sfan5 |
the CI artifact is a debug one IIRC |
18:25 |
Krock |
the APK is 15 MB small, thus I don't think so. |
18:25 |
[MatrxMT] |
<Zughy> RE Inventory key (sorry but meetings are at dinner time): of course E will prevail, 99% of games we currently have are Minecraft-related and/or with an inventory. That makes me the minority but again, I don't see why I should be punished for not sticking to a Minecraft/inventory-featuring game |
18:26 |
sfan5 |
I believe the debug symbols are gone but it will be a build with debug code paths active |
18:26 |
Krock |
Zughy: on the other hand - you can still it back easily. it's not like we're hardcoding those :) |
18:26 |
|
Guest56 joined #luanti-dev |
18:27 |
Krock |
sfan5: hmm interesting. In that case I'll try to reproduce it again. It's rather difficult. It seems to only happen once after a reboot so far. Who knows what's going wrong there.... |
18:27 |
[MatrxMT] |
<Zughy> People playing both my server and other games will clearly stick to inventory = E, so they'll have to get used to F |
18:27 |
Krock |
> skipping 15030 (unconfirmed) |
18:27 |
[MatrxMT] |
<Zughy> I don't see how people will switch every time |
18:28 |
Krock |
> skipping 15027 (the physics calculations are more accurate now, and this isn't a regression) |
18:28 |
Krock |
Zughy: the switch only happens once, when you derive from the defaults. after that it's preserved in minetest.conf, which you need to re-use anyway to keep the graphical settings |
18:30 |
Krock |
> skipping 16049 (unconfirmed issue) |
18:30 |
[MatrxMT] |
<Zughy> What I mean is, if people play games with inventories, they'll use E for the inventory. As a consequence, they'll use F on my server, because they don't want to switch every time AUX1 between E and F |
18:31 |
Krock |
Zughy: these settings are client-side. I currently do not understand what kind of role the server plays here. The "open inventory key" will always be on the key that the player configured (or the default one) |
18:31 |
MTDiscord |
<luatic> we will need per-server / per-game keybinds. |
18:31 |
MTDiscord |
<luatic> (not now, but when we do proper custom controls) |
18:32 |
Krock |
> #14135 |
18:32 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/14135 -- Add persistent unique identifiers for objects by sfence |
18:33 |
Krock |
Two weeks ago we tried to get this done before feature freeze. Shall we still wait? |
18:33 |
Desour |
s/16049/15010/ ? |
18:33 |
[MatrxMT] |
<Zughy> My server doesn't use inventories, but it also uses AUX a lot. Which is currently on E because it's faster than F, and if we move it on F that will worsen the experience. Please read what I wrote in the PR, I feel like I'm repeating myself over and over |
18:34 |
Krock |
Desour: yes. thank you. |
18:34 |
sfan5 |
didn't we way that we don't want to delay releases for feature PRs anymore? |
18:34 |
SFENCE |
I will be able to look at this one mayby tomorrow. If we want to start feature freeze today, should be postponed. |
18:34 |
sfan5 |
we say* |
18:34 |
MTDiscord |
<luatic> i'm fine with postponing. better not to rush. |
18:35 |
Krock |
moving to 5.13.0 |
18:35 |
Krock |
The remaining issues are regressions and unlikely to be fixed soon. |
18:36 |
Krock |
(#13961 and #14545) |
18:36 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/13961 -- Strange behavior when dragging&dropping from MTG creative inventory (since #13146) |
18:36 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/14545 -- SDL: Some keybinds broken due to missing character lookup |
18:37 |
Krock |
okay that's it for the milestone. Any questions, remarks for those points? |
18:38 |
SFENCE |
14545 problem was discussed before new keybingind PR was merged. And we merge it with knowing, that this problem will appear. So, shoudl nod be blocker. |
18:40 |
MTDiscord |
<luatic> #14545 |
18:40 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/14545 -- SDL: Some keybinds broken due to missing character lookup |
18:46 |
[MatrxMT] |
<Zughy> So FF time? |
18:46 |
sfan5 |
this only applies to /, doesn't it? |
18:46 |
Krock |
sfan5: to any printable characters produced with Shift or Alt |
18:47 |
SFENCE |
I think so. I do not know about other key than /. |
18:47 |
Krock |
in practice, / is generally affected on non-US layouts, yes. |
18:48 |
sfan5 |
or rather |
18:48 |
sfan5 |
does this issue even make sense anymore? we do not bind keys to "characters" anymore |
18:49 |
Krock |
good point. Perhaps the request would then correctly be 'allow binding to characters' as a feature, but that's impossible to implement with SDL2 |
18:49 |
[MatrxMT] |
<y5nw> IMO it depends on whether we still want to have keybindings with modifier keys. It is doable (even without SDL3) but I would not prioritize it at the moment |
18:49 |
[MatrxMT] |
<y5nw> Although showing the correct character (e.g. / instead of Shift-7) would involve SDL3 |
18:50 |
Krock |
I see. |
18:50 |
SFENCE |
Multiple-key binging for one actions is the best solution for this problem, I believe. So should be better that binging to printable characters. |
18:51 |
sfan5 |
in fact on my german keyboard I can press - to get "open chat console and type a /" |
18:51 |
sfan5 |
it's just scancodes working as intended |
18:51 |
sfan5 |
(had to check here to know which key to press https://www.farah.cl/Keyboardery/A-Visual-Comparison-of-Different-National-Layouts/#enUS) |
18:51 |
Krock |
what does it show in the settings? |
18:52 |
Krock |
`-` ? |
18:52 |
SFENCE |
From my side: lets go with SDL, add multiple keybingins in some future release... |
18:52 |
sfan5 |
Krock: yep |
18:52 |
Krock |
sfan5: so that's technically correct. Nice. |
18:53 |
Desour |
imo, if we get the feature to bind key combinations to actions, it's pretty much fixed |
18:54 |
Desour |
the '/' bind is a shorthand for t, then /. people produce / in weird ways may not get the thing back, but idk if such ppl exist |
18:54 |
Krock |
heh sfan5 was faster to submit the answer :3 |
19:03 |
sfan5 |
so does that mean feature freeze or not? |
19:08 |
Krock |
Yes. Would you please be so nice to update /topic ? |
19:09 |
Krock |
I don't think I have the permissions to do that |
19:09 |
Krock |
> -ChanServ- You are not authorized to perform this operation. |
19:10 |
sfan5 |
-ChanServ- +t - Enables use of the topic and topicappend commands. |
19:10 |
sfan5 |
i guess you can't give yourself op however |
19:11 |
Krock |
makes sense |
19:11 |
|
Topic for #luanti-dev is now FEATURE FREEZE IN EFFECT. Luanti (fka Minetest) core development and maintenance. Chit chat goes to #luanti. https://dev.luanti.org/ https://irc.luanti.org/ https://github.com/luanti-org |
19:15 |
Krock |
Okay. Are there any other topics to discuss? |
19:22 |
Krock |
Thank you for participating. |
20:45 |
|
MTDiscord joined #luanti-dev |
21:16 |
|
YuGiOhJCJ joined #luanti-dev |
22:33 |
|
panwolfram joined #luanti-dev |
23:05 |
|
Eragon joined #luanti-dev |
23:09 |
|
Eragon_ joined #luanti-dev |