| Time |
Nick |
Message |
| 02:38 |
|
nephele_ joined #minetest-dev |
| 04:01 |
|
Taoki joined #minetest-dev |
| 05:00 |
|
MTDiscord joined #minetest-dev |
| 07:03 |
nerzhul_ |
maybe paramat is like me, busy :D |
| 07:43 |
MTDiscord |
<appguru> Schrödinger's core dev huh? |
| 08:00 |
|
ShadowNinja joined #minetest-dev |
| 08:06 |
nerzhul_ |
sfan5, Krock i'm not sure, but do we validate that 5.5 will be with our own embedded irrlicht ? |
| 08:06 |
nerzhul_ |
if yes we can make effort on it. i suggest that we start from 1.8.4 & backport fixes from 1.9.0 which are required |
| 10:02 |
|
proller joined #minetest-dev |
| 10:19 |
|
calcul0n__ joined #minetest-dev |
| 10:29 |
|
Fleckenstein joined #minetest-dev |
| 11:32 |
|
Fixer joined #minetest-dev |
| 12:00 |
|
appguru joined #minetest-dev |
| 12:18 |
|
Lunatrius` joined #minetest-dev |
| 12:21 |
|
Kray joined #minetest-dev |
| 13:21 |
|
Lunatrius joined #minetest-dev |
| 13:23 |
|
Lunatrius` joined #minetest-dev |
| 13:45 |
|
T4im joined #minetest-dev |
| 13:59 |
|
Thomas-S_ joined #minetest-dev |
| 14:09 |
|
Wuzzy joined #minetest-dev |
| 15:04 |
|
mizux joined #minetest-dev |
| 15:16 |
|
numzero joined #minetest-dev |
| 15:32 |
|
nephele joined #minetest-dev |
| 16:59 |
Krock |
nerzhul_: there are no certain plans yet. if it's in an usable state we can merge it to master I think |
| 16:59 |
Krock |
also yes, backport from 1.9.0 |
| 17:01 |
sfan5 |
why backport instead of just using 1.9.0's development state? |
| 17:02 |
Krock |
rebase onto 1.9.0 might already now result in merge conflicts, I think? otherwise that's obvously a better approach |
| 17:03 |
sfan5 |
can we revert #9025? most people run an updated 3d_armor now and for the people who don't it's better that it's entirely broken than subtly broken |
| 17:03 |
ShadowBot |
https://github.com/minetest/minetest/issues/9025 -- Revert Client::sendPlayerPos optimization (part of 81c7f0a) by ClobberXD |
| 17:03 |
rubenwardy |
revert revert |
| 17:04 |
Krock |
revert the revert commit? uh okay |
| 17:05 |
rubenwardy |
soon: revert revert revert |
| 17:05 |
Zughy[m]1 |
you have to access the 3rd layer of reversion, and plant it safely |
| 17:05 |
Krock |
the FlagDesc and Settings fixes are at a partial 4th layer of reverting and patching |
| 17:06 |
Krock |
but yes I agree to reverting that. The previous code looks correct |
| 17:06 |
Zughy[m]1 |
Minetest 5.4.0: Christopher Nolan |
| 17:26 |
|
fluxflux joined #minetest-dev |
| 18:48 |
|
homthack joined #minetest-dev |
| 18:50 |
|
fluxflux_ joined #minetest-dev |
| 18:55 |
rubenwardy |
super small quality of life change: #10864 |
| 18:55 |
ShadowBot |
https://github.com/minetest/minetest/issues/10864 -- ContentDB: Order installed content first by rubenwardy |
| 19:20 |
|
proller joined #minetest-dev |
| 19:32 |
sofar |
oh wow someone tried to actually register a server on my mt2fa infrastructure, lol |
| 19:32 |
rubenwardy |
What's the recommended way to make a std::(w)string from a (w)char * which needs to be deleted? |
| 19:33 |
rubenwardy |
Or perhaps I could do char *str = get(); auto str_p = std::make_unique<char*>(str); |
| 19:35 |
Krock |
assign & free ? |
| 19:36 |
numzero |
yes, I think. std::string can’t take storage ownership AFAIK |
| 19:36 |
Krock |
if you had access to the internal c_str() you could attempt to std::swap and delete one |
| 19:51 |
|
Wuzzy joined #minetest-dev |
| 19:56 |
MTDiscord |
<Jordach> sfan5, >using a mod built from 2013 era API in 2021 |
| 19:56 |
sfan5 |
yes that's the exact issue |
| 19:57 |
MTDiscord |
<Jordach> i mean this is what happens when you don't refactor anew every once in a while: see dreambuilder |
| 19:58 |
sfan5 |
rubenwardy: you could store a boolean indicating sort state instead of an unsorted copy, right? |
| 20:18 |
rubenwardy |
not without making a new HTTP requested |
| 20:19 |
rubenwardy |
you could store original order in the list I guess |
| 20:19 |
rubenwardy |
the unsorted copy is to fix the order being lost if you install and uninstall a pckage - that results in the package still being at the top of the list |
| 20:20 |
rubenwardy |
wel, I guess you that's a minor issue - it's only in memory |
| 20:21 |
rubenwardy |
ok, explained clearer: mymod is installed -> open CDB, mymod at the top -> uninstall mymod -> close CDB, reopen -> without `unordered`, mymod is still at top |
| 20:21 |
rubenwardy |
closing and reopening the client refetches the list using HTTP so that restores the order |
| 20:21 |
sfan5 |
I see |
| 20:22 |
rubenwardy |
It's a minor issue for a package to be stuck at the top I guess, given that the order is arbitrary anyway |
| 20:23 |
rubenwardy |
I'm happy to add a comment or just remove it |
| 20:24 |
rubenwardy |
"Unordered stores original order from ContentDB API" errr slight naming issue maybe |
| 20:26 |
rubenwardy |
I'm err-ing towardings removing it as it feels a bit messy |
| 21:07 |
|
proller joined #minetest-dev |
| 21:30 |
|
calcul0n_ joined #minetest-dev |
| 22:07 |
|
absurb joined #minetest-dev |
| 22:11 |
rubenwardy |
As an experiment, I tried using a scroll_container in the ContentDB dialog with screenshots disabled. Turns out that a 1000 packages causes it to be super laggy even with screenshots disabled and labels used instead of text areas. I'm not surprised |
| 22:12 |
rubenwardy |
perhaps table could be modified to be better suited to ContentDB, then you could have view recylcing |
| 22:24 |
|
Taoki joined #minetest-dev |
| 23:18 |
MTDiscord |
<exe_virus> Are all core devs using linux in some way? I want to dev a research project for testing out different octree layouts in 2D with SDL, and wanted to know if I have to support cross platform or not, so when I share it in a PR later for spatial indexes and objects, that devs can easily compile and test for themselves. |
| 23:19 |
MTDiscord |
<exe_virus> *Quadtree but you get the idea |
| 23:19 |
sfan5 |
pretty much all linux yes |
