| Time |
Nick |
Message |
| 00:41 |
|
Cornelia joined #minetest-dev |
| 00:43 |
|
paramat joined #minetest-dev |
| 01:01 |
|
ANAND joined #minetest-dev |
| 01:16 |
|
Cornelia joined #minetest-dev |
| 01:20 |
|
Miner_48er joined #minetest-dev |
| 02:22 |
|
Lone-Star joined #minetest-dev |
| 05:09 |
|
Ruslan1 joined #minetest-dev |
| 05:19 |
|
kaeza joined #minetest-dev |
| 06:42 |
|
fwhcat joined #minetest-dev |
| 06:57 |
|
fwhcat joined #minetest-dev |
| 07:42 |
VanessaE |
request: minimal first-press F5 display that ALL players get to see, without needing debug priv. should show the direction you're facing, your coords, and the node you're pointing at. the existing first-press-of-F5 that shows those things along with drawtime, map seed, RTT, and subsequent displays after that, should continue to need debug priv. |
| 07:45 |
|
proller joined #minetest-dev |
| 07:50 |
VanessaE |
https://github.com/minetest/minetest/issues/8472 |
| 08:26 |
|
Beton joined #minetest-dev |
| 08:41 |
|
YuGiOhJCJ joined #minetest-dev |
| 09:45 |
|
karamel joined #minetest-dev |
| 09:51 |
rubenwardy |
actual debug information shouldn't need any privilege |
| 09:51 |
rubenwardy |
only debug information which could use for cheating, such as the wireframe |
| 09:52 |
VanessaE |
rtight |
| 09:52 |
VanessaE |
right* |
| 10:01 |
|
proller joined #minetest-dev |
| 10:12 |
|
karamel joined #minetest-dev |
| 10:22 |
|
proller joined #minetest-dev |
| 10:29 |
|
Fixer joined #minetest-dev |
| 10:36 |
|
Darcidride joined #minetest-dev |
| 12:27 |
|
kaeza joined #minetest-dev |
| 13:28 |
|
twoelk joined #minetest-dev |
| 13:50 |
|
Zeno` joined #minetest-dev |
| 14:22 |
|
calcul0n_ joined #minetest-dev |
| 14:28 |
|
Ruslan1 joined #minetest-dev |
| 14:53 |
|
kaeza joined #minetest-dev |
| 15:38 |
|
YuGiOhJCJ joined #minetest-dev |
| 16:17 |
|
benrob0329 joined #minetest-dev |
| 16:40 |
|
Cornelia joined #minetest-dev |
| 17:13 |
|
fwhcat joined #minetest-dev |
| 17:14 |
|
proller joined #minetest-dev |
| 17:17 |
|
benrob0329 joined #minetest-dev |
| 17:19 |
|
benrob0329 joined #minetest-dev |
| 17:24 |
|
ensonic joined #minetest-dev |
| 18:30 |
|
GreenDimond joined #minetest-dev |
| 18:32 |
GreenDimond |
So my server logs are filled with "ScriptApiBase::objectrefGetOrCreate(): Pushing ObjectRef to removed/deactivated object" warnings. I have no idea where to start debugging this. |
| 18:32 |
GreenDimond |
I am wondering if this is caused by using set_properties() on a missing/deactivated entity |
| 18:43 |
|
Wuzzy joined #minetest-dev |
| 19:49 |
p_gimeno |
GreenDimond: isolating which mod is causing it would be a start |
| 19:49 |
GreenDimond |
Great |
| 19:50 |
GreenDimond |
I have 100 |
| 19:50 |
GreenDimond |
Isolating the mod was kind of a given |
| 19:50 |
GreenDimond |
The question is how can I do it faster |
| 19:50 |
p_gimeno |
bisection? |
| 19:51 |
GreenDimond |
Without restarting my server an inane amount of times |
| 19:51 |
sfan5 |
print the mod name when each callback runs |
| 19:52 |
sofar |
sounds almost like a leak |
| 19:52 |
p_gimeno |
it takes 7 attempts to bisect 100 mods |
| 19:52 |
sofar |
it takes 10 for 1000 |
| 19:52 |
GreenDimond |
sfan5: By adding a print statement to every single mod callback?..... |
| 19:52 |
sfan5 |
correct |
| 19:52 |
sfan5 |
you can do this programatically |
| 19:53 |
sofar |
in C++ |
| 19:53 |
sfan5 |
I was thinking of lua code that wraps each callback but just doing it in the C++ of callbacks is even simpler |
| 19:56 |
GreenDimond |
So I get to recompile Minetest |
| 19:56 |
GreenDimond |
yay |
| 19:59 |
sofar |
"you can do it" |
| 20:43 |
|
paramat joined #minetest-dev |
| 20:43 |
|
Ruslan1 joined #minetest-dev |
| 20:53 |
|
proller joined #minetest-dev |
| 21:28 |
|
ssieb joined #minetest-dev |
| 21:52 |
|
YuGiOhJCJ joined #minetest-dev |
| 21:54 |
p_gimeno |
so in #8461, the emitter is now a scene node, so for attached particles, why not attach it to the object's scene node? |
| 21:54 |
ShadowBot |
https://github.com/minetest/minetest/issues/8461 -- Reimplement particles using the Irrlicht particle system and improve collision handling by DTA7 |
| 21:55 |
p_gimeno |
also, vel is unused (there's no report from the compiler because of the call to rotateXZBy) |
| 22:04 |
p_gimeno |
and rotating p.pos by yaw makes no sense, even less if the scene node gets attached. At that point p.pos is just a random position within a cube, no point in rotating it. |
| 22:05 |
|
proller joined #minetest-dev |
| 22:09 |
sfan5 |
so basically, the PR still needs some work |
| 22:11 |
p_gimeno |
yes I think so |
| 22:22 |
p_gimeno |
is DTA7 on IRC? |
| 22:25 |
|
Lia joined #minetest-dev |
| 23:51 |
|
ANAND joined #minetest-dev |
| 23:55 |
|
twoelk left #minetest-dev |
