| Time |
Nick |
Message |
| 00:13 |
sfan5 |
okay https://github.com/minetest/irrlicht now runs on Linux, Android, macOS and Windows |
| 00:13 |
|
MTDiscord joined #minetest-dev |
| 00:16 |
sfan5 |
in other words there is nothing preventing us from moving over |
| 00:17 |
MTDiscord |
<Jordach> macOS (when built on Big Sur) no longer fail |
| 00:20 |
|
Taoki joined #minetest-dev |
| 00:27 |
|
rom1504 joined #minetest-dev |
| 05:00 |
|
MTDiscord joined #minetest-dev |
| 06:54 |
|
olliy joined #minetest-dev |
| 08:00 |
|
ShadowNinja joined #minetest-dev |
| 10:20 |
|
proller joined #minetest-dev |
| 10:21 |
sfan5 |
/home/runner/work/minetest/minetest/build/android/native/jni/../../../../src/collision.cpp:36:2: warning: "-ffast-math is known to cause bugs in collision code, do not use!" [-W#warnings] |
| 10:21 |
sfan5 |
#warning "-ffast-math is known to cause bugs in collision code, do not use!" |
| 10:21 |
sfan5 |
lol @ our android build flags |
| 10:23 |
Krock |
haha |
| 10:26 |
|
calcul0n joined #minetest-dev |
| 11:02 |
|
tech_exorcist joined #minetest-dev |
| 11:09 |
|
tech_exorcist joined #minetest-dev |
| 11:17 |
|
tech_exorcist joined #minetest-dev |
| 11:28 |
|
Fixer joined #minetest-dev |
| 11:31 |
|
Fixer_ joined #minetest-dev |
| 12:13 |
sfan5 |
suggestions? https://0x0.st/-Pgs.png (irrlicht fork readme) |
| 12:20 |
MTDiscord |
<appguru> someone on Discord is suggesting you add Irrlicht as a git-submodule? |
| 12:20 |
sfan5 |
rejected |
| 12:20 |
MTDiscord |
<appguru> alright |
| 12:22 |
|
troller joined #minetest-dev |
| 12:39 |
|
numzero joined #minetest-dev |
| 12:39 |
|
troller joined #minetest-dev |
| 12:51 |
nerzhul |
rejected too |
| 12:51 |
nerzhul |
git submodule are horrible to manage |
| 12:51 |
nerzhul |
D |
| 12:51 |
nerzhul |
nice README, sfan5, don't see built it as a non shared lib by default & drop the sharedlib model ? |
| 12:52 |
sfan5 |
by default it only builds a .so |
| 12:52 |
nerzhul |
i suggest that we just use static library compilation, it prevent packagers problems with 2 irrlicht on system i think |
| 12:52 |
sfan5 |
that's up to distributions maintainers |
| 12:52 |
nerzhul |
it build a .a by default on 1.8, they changed that ? |
| 12:52 |
sfan5 |
our CI will use static |
| 12:52 |
nerzhul |
for me as it will be painful, just do static compilation |
| 12:52 |
sfan5 |
dunno, I rewrote the build system in cmake anyway |
| 12:53 |
nerzhul |
it will also permit compiler to optimize more the binary |
| 12:53 |
nerzhul |
perfect |
| 12:54 |
nerzhul |
as we may break some irrlicht ABI we should just optimize it for our own usage |
| 12:54 |
nerzhul |
MT is working properly with the forked version ? is master using it in our build & tests ? |
| 12:55 |
sfan5 |
https://github.com/sfan5/minetest/tree/irr19 |
| 12:55 |
sfan5 |
not in master yet, i'll put it in the same PR as the other fork changes |
| 12:55 |
nerzhul |
okay |
| 12:55 |
nerzhul |
ping me when you need revieww for the master integration |
| 12:56 |
sfan5 |
sure |
| 12:56 |
nerzhul |
after the merge i will push some of our irrlicht code changes inside the lib, from the base repo |
| 13:00 |
Zughy[m] |
"[full | quick]=" if these are options for the /clearobjects command, why are they exactly in the translate template? |
| 13:00 |
Zughy[m] |
https://github.com/minetest/minetest/pull/11032/files#diff-31a82a4cf460e5ea9cdea738ab9666cdcc2e9959c1c2fbcd66ed3ac72e73fd6cR155 |
| 13:00 |
Zughy[m] |
in fact not even Wuzzy has translated them |
| 13:01 |
|
calcul0n_ joined #minetest-dev |
| 13:17 |
|
z812 joined #minetest-dev |
| 13:21 |
sfan5 |
merging #11030, #11027 in 5m |
| 13:21 |
ShadowBot |
https://github.com/minetest/minetest/issues/11030 -- Predict failing placement of ignore nodes by Wuzzy2 |
| 13:21 |
ShadowBot |
https://github.com/minetest/minetest/issues/11027 -- Set ENABLE_SYSTEM_JSONCPP to TRUE by default by sfan5 |
| 13:29 |
sfan5 |
Zughy[m]: no idea, maybe ask him what he thinks about removing them from the translations |
| 14:03 |
sfan5 |
nerzhul, rubenwardy, Krock: https://github.com/minetest/minetest/pull/10970 ready for review |
| 14:06 |
|
Fleckenstein joined #minetest-dev |
| 15:01 |
nerzhul |
sfan5: approved, go ahead on this project |
| 15:01 |
sfan5 |
thanks, I'll wait for some more feedback since this is a major change |
| 15:09 |
MTDiscord |
<Jordach> sfan5, did some QA testing on aarch64 macOS: open user dir and open website link works fine |
| 15:46 |
|
Wuzzy joined #minetest-dev |
| 15:54 |
MTDiscord |
<Jordach> there's a small revision required for the UNIX readme for Irrlicht |
| 15:54 |
MTDiscord |
<Jordach> the problem with that is that UNIX systems (including macOS) don't have X11 but do have valid OpenGL bindings |
| 15:55 |
sfan5 |
just pretend that macOS isn't unix |
| 15:55 |
sfan5 |
(cmake does it correctly anyway, this is just a textual mistake) |
| 15:56 |
MTDiscord |
<Jordach> lmao i did write some build notes for people who want to wrangle with macOS |
| 16:03 |
MTDiscord |
<Jordach> i did detail specifics like binary signing with the adhoc cert |
| 16:03 |
MTDiscord |
<Jordach> a bug in how APFS handles file copies and inodes causes xcode to fail to sign the binary |
| 16:03 |
MTDiscord |
<Jordach> (it's more xcode not being updated for the newer filesystem) |
| 16:06 |
MTDiscord |
<Jordach> sfan5, guide here https://gist.github.com/Jordach/4f3e933f9be6b09283e30a29ddbcf833 |
| 16:06 |
MTDiscord |
<Jordach> it's got macOS specific steps at the bottom |
| 16:08 |
Krock |
will rebase/merge #11035 in 10 minutes |
| 16:08 |
ShadowBot |
https://github.com/minetest/minetest/issues/11035 -- [NO SQUASH] Additional inventory access protections by SmallJoker |
| 16:17 |
Krock |
merging |
| 16:30 |
sfan5 |
even though my ryzen 2600 builds MT in under two minutes waiting for it often feels like a bother... |
| 16:36 |
sfan5 |
@Jordach found the fix for missing stars btw |
| 16:38 |
MTDiscord |
<Jordach> mind pushing to sfan5/irr19 |
| 16:41 |
MTDiscord |
<Jordach> which branch, irrfork or irr19 |
| 16:43 |
sfan5 |
pick any |
| 16:44 |
MTDiscord |
<Jordach> rephrased: which has the stars fix |
| 16:45 |
sfan5 |
both |
| 16:46 |
MTDiscord |
<Jordach> thank |
| 16:47 |
MTDiscord |
<Jordach> okay, how does MT detect the forked irrlicht |
| 16:47 |
MTDiscord |
<Jordach> (well cmake) |
| 16:48 |
sfan5 |
looks for a macro defined in a header file |
| 16:49 |
sfan5 |
wasn't there yesterday so you'll have to pull irrlicht again if you're hitting that |
| 16:49 |
MTDiscord |
<Jordach> spose so |
| 16:49 |
MTDiscord |
<Jordach> rebuilding |
| 16:50 |
MTDiscord |
<Jordach> fwiw, my 3600 compiles MT under debian in about 55 secs |
| 16:51 |
MTDiscord |
<Jordach> there, building |
| 16:55 |
|
proller joined #minetest-dev |
| 17:17 |
|
sofar_ joined #minetest-dev |
| 17:18 |
|
rubenwardy joined #minetest-dev |
| 17:19 |
|
JTE joined #minetest-dev |
| 17:19 |
|
Kray joined #minetest-dev |
| 17:19 |
|
T4im joined #minetest-dev |
| 17:32 |
|
freshreplicant[m joined #minetest-dev |
| 17:48 |
|
numzero joined #minetest-dev |
| 18:01 |
|
giov4[m] joined #minetest-dev |
| 18:05 |
|
Newbyte joined #minetest-dev |
| 18:11 |
MTDiscord |
<Jordach> right some minor issues |
| 18:11 |
MTDiscord |
<Jordach> mouse inputs seem to get stuck under macOS |
| 18:12 |
MTDiscord |
<Jordach> i've tested with my desktop's mouse plugged in and it still occurs |
| 18:28 |
|
Zughy[m] joined #minetest-dev |
| 18:30 |
MTDiscord |
<Jordach> ok no, it's just the terrible mouse driver macOS ships |
| 18:33 |
MTDiscord |
<Jordach> just tested a non-free driver, that fixes it |
| 18:49 |
|
Fixer joined #minetest-dev |
| 18:51 |
|
kb1000 joined #minetest-dev |
| 19:16 |
|
hlqkj joined #minetest-dev |
| 19:48 |
|
z812 joined #minetest-dev |
| 20:11 |
|
numzero joined #minetest-dev |
| 21:35 |
|
proller joined #minetest-dev |
| 21:39 |
MTDiscord |
<appguru> the changelog mentions both me (Lars Müller, spelt Mueller once) and Lars Hofhansl (simply called "Lars"). This is quite confusing. https://dev.minetest.net/Changelog#5.3.0_.E2.86.92_5.4.0 |
| 21:41 |
|
appguru joined #minetest-dev |
| 21:54 |
|
DS-minetest joined #minetest-dev |
