| Time |
Nick |
Message |
| 00:00 |
|
DI3HARD139 joined #minetest-dev |
| 00:16 |
paramat |
merging games 2062 2068 |
| 00:26 |
paramat |
merged, 6 PRs! |
| 00:27 |
ThomasMonroe |
how goes development paramat? =) |
| 00:28 |
paramat |
hi, erm ok i guess =) |
| 00:29 |
paramat |
slow |
| 00:29 |
ThomasMonroe |
heh same here :) |
| 00:29 |
paramat |
my todo list: remove 2, add 2 =/ |
| 00:29 |
ThomasMonroe |
lol, that's the way it goes I guess XD |
| 00:30 |
ThomasMonroe |
we've got some of the foundation done for development on Qub3d :D |
| 00:31 |
paramat |
good stuff |
| 00:32 |
ThomasMonroe |
yeah, I'm getting mad at MSVC, nothing new there. ;) |
| 00:37 |
rubenwardy |
-> #minetest |
| 02:28 |
|
srifqi joined #minetest-dev |
| 03:56 |
|
AndroBuilder joined #minetest-dev |
| 04:28 |
|
YuGiOhJCJ joined #minetest-dev |
| 04:38 |
|
YuGiOhJCJ joined #minetest-dev |
| 04:43 |
|
YuGiOhJCJ joined #minetest-dev |
| 05:38 |
|
srifqi joined #minetest-dev |
| 05:54 |
|
YuGiOhJCJ joined #minetest-dev |
| 08:07 |
|
YuGiOhJCJ joined #minetest-dev |
| 08:34 |
|
SilverLuke joined #minetest-dev |
| 08:36 |
|
fwhcat joined #minetest-dev |
| 09:24 |
|
Gael-de-Sailly joined #minetest-dev |
| 09:32 |
|
Wuzzy joined #minetest-dev |
| 09:40 |
|
YuGiOhJCJ joined #minetest-dev |
| 10:39 |
|
lisac joined #minetest-dev |
| 10:45 |
|
YuGiOhJCJ joined #minetest-dev |
| 11:14 |
|
YuGiOhJCJ joined #minetest-dev |
| 11:30 |
|
Fixer joined #minetest-dev |
| 11:33 |
|
YuGiOhJCJ joined #minetest-dev |
| 11:38 |
|
fwhcat joined #minetest-dev |
| 12:08 |
|
YuGiOhJCJ joined #minetest-dev |
| 12:22 |
|
Krock joined #minetest-dev |
| 12:30 |
|
Beton joined #minetest-dev |
| 12:46 |
|
YuGiOhJCJ joined #minetest-dev |
| 13:26 |
|
fwhcat_ joined #minetest-dev |
| 14:23 |
|
srifqi joined #minetest-dev |
| 14:43 |
|
SilverLuke joined #minetest-dev |
| 14:56 |
|
srifqi left #minetest-dev |
| 15:15 |
|
srifqi joined #minetest-dev |
| 15:37 |
nerzhul |
Krock: i added the test you wanted on #7121 if it's okay for you, tell me |
| 15:37 |
ShadowBot |
https://github.com/minetest/minetest/issues/7121 -- Add Voxelarea unittests by nerzhul |
| 15:51 |
nerzhul |
i push a first change to repair android build, partially outside of #6796 |
| 15:51 |
ShadowBot |
https://github.com/minetest/minetest/issues/6796 -- Android build fixes for c++11 by stujones11 |
| 15:52 |
nerzhul |
(updating tools to recent versions to have a gradle near standard Android Studio ndk builds) |
| 15:52 |
rubenwardy |
maybe 6796 should be merged too? |
| 15:53 |
nerzhul |
i will test it just after |
| 15:53 |
rubenwardy |
cool |
| 15:54 |
nerzhul |
and after this merge i will think about a real proper cmake build using gradle, if it's possible |
| 15:54 |
nerzhul |
it can permit to simplify our process |
| 15:55 |
rubenwardy |
that would be very nice |
| 15:55 |
rubenwardy |
good to get it working ASAP, so better controls can be worked on |
| 15:55 |
nerzhul |
it's not simple, i think it needs to refactor all our build system in src folder :p |
| 15:56 |
nerzhul |
but at least i can easily use src subfolders as they are just targeting files, not build logic |
| 15:56 |
rubenwardy |
it meaning android builds, not gradle/cmake builds |
| 15:56 |
rubenwardy |
also, I really dislike this practice of having CMakeLists.txt for each dir in src/ |
| 15:57 |
nerzhul |
rubenwardy it's relatively standard to have this model |
| 15:57 |
rubenwardy |
it confusing for CLion |
| 15:57 |
nerzhul |
not mine :p |
| 15:57 |
nerzhul |
but old versions yes |
| 15:57 |
rubenwardy |
oh lol |
| 15:57 |
rubenwardy |
I mean, operator overloading with templates just completely breaks CLion, so it's not exactly solid |
| 15:58 |
nerzhul |
okay it's nice to see the patch properly build initialization which was totally broken :p |
| 15:58 |
nerzhul |
properly fix* |
| 15:58 |
nerzhul |
currently compiling |
| 15:58 |
nerzhul |
operator overloading is not well supported in clion, it's better in 2017.3 but it's not perfect |
| 16:00 |
nerzhul |
okay we have new build problems due to recents commits on the guiConfirmRegistration |
| 16:00 |
nerzhul |
i will try to compile all files except this at least |
| 16:02 |
nerzhul |
okay, i push #6796, it fixes all problems except recent changes on registration which are more recent than this PR |
| 16:02 |
ShadowBot |
https://github.com/minetest/minetest/issues/6796 -- Android build fixes for c++11 by stujones11 |
| 16:02 |
rubenwardy |
nice |
| 16:03 |
rubenwardy |
+1 for me |
| 16:03 |
rubenwardy |
*from |
| 16:03 |
nerzhul |
now i need to find how to solve the build issue heh |
| 16:03 |
nerzhul |
it's a virtual problem on the registration dialog |
| 16:04 |
sfan5 |
well this was pretty successful |
| 16:04 |
sfan5 |
I said I would test the android build fixes later today |
| 16:04 |
nerzhul |
it's better than before heh |
| 16:04 |
sfan5 |
but it looks like that's not longer needed |
| 16:05 |
nerzhul |
sfan5 i need to fix the last point but yes :) |
| 16:06 |
nerzhul |
sfan5, rubenwardy can you check #7121 please ? i fixed the Krock point :) (i will push the lint fix too) |
| 16:06 |
ShadowBot |
https://github.com/minetest/minetest/issues/7121 -- Add Voxelarea unittests by nerzhul |
| 16:08 |
rubenwardy |
wow, I must be tired |
| 16:08 |
rubenwardy |
thought adareas said address |
| 16:10 |
srifqi |
nerzhul: https://github.com/minetest/minetest/commit/57418111a8352f2d6dabf1d3e43c1719f4b118c0#commitcomment-28035214 |
| 16:10 |
srifqi |
Is it like that? |
| 16:12 |
nerzhul |
yes it's that problem |
| 16:13 |
nerzhul |
it's due to the fact we passed to another model of define, which is not compatible with android it seems |
| 16:22 |
nerzhul |
okay i got the fix, i ported one similar fix from the stujones11 PR to intleditbox that should do the trick |
| 16:25 |
Krock |
nerzhul, interesting choice of indents you've got there |
| 16:26 |
nerzhul |
not my choice xD |
| 16:30 |
nerzhul |
build is nearly fixed, i just need to ensure we build intlGUIEditBox.cpp it seems ndk doesn't want... |
| 16:33 |
|
Gael-de-Sailly joined #minetest-dev |
| 16:33 |
|
Gael-de-Sailly joined #minetest-dev |
| 16:33 |
Megaf |
updated |
| 16:33 |
Megaf |
web#121 |
| 16:33 |
ShadowBot |
https://github.com/minetest/minetest.github.io/issues/121 -- Channel description improvements 2. by Megaf |
| 16:34 |
|
srifqi joined #minetest-dev |
| 16:34 |
nerzhul |
wtf why this doesn't pass |
| 16:35 |
srifqi |
Any idea on how you did it, nerzhul? |
| 16:36 |
Megaf |
nerzhul: define this |
| 16:36 |
nerzhul |
the build fix ? yeah i added the missing functions with the proper outputs, but i need to fix the linking issue |
| 16:36 |
srifqi |
Okay. |
| 16:44 |
|
Gael-de-Sailly joined #minetest-dev |
| 16:45 |
nerzhul |
I push the build fix, but not the linker fix, i didn't find it atm, but at least we go ahead :) |
| 16:46 |
nerzhul |
See https://github.com/minetest/minetest/commit/a1c5a011421840f4b2f9d8b6cc9b9724cf333168 i also fix a variable name overloading issue |
| 16:47 |
nerzhul |
This is the remaining issue |
| 16:47 |
nerzhul |
jni/../jni/src/gui/guiConfirmRegistration.cpp:102: error: undefined reference to 'irr::gui::intlGUIEditBox::intlGUIEditBox(wchar_t const*, bool, irr::gui::IGUIEnvironment*, irr::gui::IGUIElement*, int, irr::core::rect<int> const&, bool, bool)' |
| 16:47 |
nerzhul |
(it's the correct constructor...) |
| 16:51 |
nerzhul |
oh... i think i found it |
| 16:52 |
nerzhul |
i missed a ifdef :p |
| 16:52 |
nerzhul |
yes it's fixed nice |
| 16:52 |
nerzhul |
i push the last fix |
| 16:52 |
nerzhul |
which is due to a ifdef in the header |
| 16:53 |
nerzhul |
in the cpp file * |
| 16:53 |
nerzhul |
(intlGUIEditBox.cpp) |
| 16:53 |
nerzhul |
https://github.com/minetest/minetest/commit/f7938fab9fa675c60bfa249f01b2b608196cdbc0 |
| 16:57 |
nerzhul |
the compilation is okay, but the android linking is not good yet |
| 17:11 |
|
paramat joined #minetest-dev |
| 17:21 |
nerzhul |
everything is now okay :) |
| 17:21 |
nerzhul |
i push a typo in the makefile due to too update |
| 17:23 |
nerzhul |
the confirmation dialog is just wow :p |
| 17:23 |
nerzhul |
unreadable and crash when clicking on it |
| 17:27 |
nerzhul |
i reported #7122 |
| 17:27 |
ShadowBot |
https://github.com/minetest/minetest/issues/7122 -- Android: confirmation dialog is... unreable and crashes |
| 17:27 |
nerzhul |
but at least is compiles and launches :p |
| 17:27 |
nerzhul |
the serverlist is also unreable |
| 17:28 |
nerzhul |
UI is too little, and not properly sized to client UI size |
| 17:31 |
|
stujones11 joined #minetest-dev |
| 17:33 |
stujones11 |
nerzhul: I am looking into the remaining linking issues but I noticed the line endings are broken again in the irrlicht patch #6750 |
| 17:33 |
ShadowBot |
https://github.com/minetest/minetest/issues/6750 -- Android: Irrlicht patch has wrong line endings |
| 17:33 |
stujones11 |
Could this have something to do with .gitignore? maybe it requires -f |
| 17:47 |
stujones11 |
nerzhul: I agree, there is still much to be done but I am happy to spend some time on this once we have a working build system |
| 17:48 |
stujones11 |
Also, +1 for moving to cmake + gradle if that is at all possible :) |
| 17:59 |
nerzhul |
stujones11, linking is fixed, for the irrlicht line endine we can fix it |
| 17:59 |
nerzhul |
i fixed the code manually |
| 17:59 |
nerzhul |
stujones11, i think it's possible, but not easy at it needs to rewrite the whole system :p |
| 18:00 |
nerzhul |
i will try to compile minetest lib at least with gradle |
| 18:01 |
stujones11 |
nerzhul: thanks, iirc the latter does work if you use the right version ;) |
| 18:15 |
|
Lone-Star joined #minetest-dev |
| 18:27 |
|
fwhcat__ joined #minetest-dev |
| 18:27 |
nerzhul |
the begin is promising but there are many works to do as i need to re-do all from scratch |
| 18:33 |
stujones11 |
Yeah, a lot of work I know so maybe not a high priority atm, definitely worth keeping in mind though |
| 18:38 |
nerzhul |
yeah :p |
| 18:38 |
nerzhul |
but at least i can share some cmakelists parts heh |
| 18:39 |
nerzhul |
i don't understand where does android_native_app_glue.h comes from |
| 18:39 |
nerzhul |
oh it's in the ndk |
| 18:43 |
nerzhul |
oh nice some parts are correctly compiling heh, maybe i can succeed in MT build without buildings deps at least |
| 18:46 |
|
proller joined #minetest-dev |
| 18:49 |
nerzhul |
wow nice, i have all MT src/ subfolders built heh |
| 18:49 |
nerzhul |
i need the root folder now |
| 18:51 |
Krock |
nerzhul, running the unittests locally now. will approve on success |
| 18:51 |
stujones11 |
It would also be nice if someone could update the dev-wiki http://dev.minetest.net/Android |
| 18:51 |
stujones11 |
the minimum supported ndk should now be ndk-r16b, though later ndk-r15x versions may also work |
| 18:52 |
stujones11 |
It may also help and encourage others to contribute |
| 18:53 |
nerzhul |
currently i just use the last NDK, previous NDK has no real sense |
| 18:55 |
stujones11 |
Me too, but dev-wiki is suggesting to use ndk-r11, which will definitely not work :p |
| 18:55 |
sfan5 |
the only supported NDK should be the latewst one |
| 18:55 |
sfan5 |
also I think it'd be more encouraging for devs if our Android build system wasn't such a dumpster fire |
| 18:55 |
stujones11 |
^ |
| 18:56 |
nerzhul |
sfan5 it's why i'm working on making him being standard :p |
| 18:56 |
sfan5 |
I'm not only talking about the 1000 line Makefile, using Android's ndk-build stuff is stupid too |
| 18:56 |
nerzhul |
i'm not using that :p |
| 18:56 |
nerzhul |
i'm using google standard for NDK builds using gradle & cmake |
| 18:57 |
sfan5 |
a clean solution would be to have cmake generate a shared library, which the Android makefile just copies |
| 18:57 |
nerzhul |
sfan5 i think i'm on that way |
| 18:57 |
sfan5 |
yeah, or the native cmake support (though i've never used that) |
| 18:57 |
nerzhul |
https://github.com/minetest/minetest/pull/7123 |
| 18:57 |
nerzhul |
#7123 |
| 18:57 |
ShadowBot |
https://github.com/minetest/minetest/issues/7123 -- Android: switch to cmake based standard builds by nerzhul |
| 18:57 |
nerzhul |
currently only src subfolders are compiling, i'm looking for other sources currently |
| 18:58 |
nerzhul |
having proper subfolders in src will help greatly to include those properly in android cmake build heh |
| 19:00 |
Wayward_One |
the wiki has been updated |
| 19:08 |
nerzhul |
nice , only 10 files are not built with this method and needs fixes :) |
| 19:48 |
nerzhul |
yes, everything is built heh |
| 19:59 |
stujones11 |
nerzhul: good work, could this make it possible for travis-ci check the android build status? |
| 19:59 |
stujones11 |
possibly using pre-built dependencies? A little off-topic, I know. |
| 20:02 |
|
Darcidride_ joined #minetest-dev |
| 20:15 |
nerzhul |
merging #7121 ty Krock |
| 20:15 |
ShadowBot |
https://github.com/minetest/minetest/issues/7121 -- Add Voxelarea unittests by nerzhul |
| 20:15 |
nerzhul |
stujones11: build is not complete yet |
| 20:16 |
nerzhul |
and we can do it but NDK is not easy to embed in CI, and build is very very very long too on a single core |
| 20:16 |
nerzhul |
but after refactoring the build maybe we can do it |
| 20:16 |
stujones11 |
Thanks, I realize why it not practical atm =) |
| 20:27 |
|
Fritigern joined #minetest-dev |
| 20:35 |
stujones11 |
Interestingly, for me the android version seems to compile slightly faster than the regular build, even using a single core |
| 20:35 |
|
Gael-de-Sailly joined #minetest-dev |
| 20:35 |
stujones11 |
excluding the dependencies, of course. |
| 20:44 |
paramat |
thanks for working on Android |
| 20:44 |
paramat |
game#2080 |
| 20:44 |
ShadowBot |
https://github.com/minetest/minetest_game/issues/2080 -- New pine needles texture by Splizard by paramat |
| 20:48 |
stujones11 |
paramat: yw, it is something I care about and almost feel partly responsible for its current state ;) |
| 21:02 |
|
Cornelia joined #minetest-dev |
| 21:13 |
paramat |
please can anyone review #6803 ? significant optimisation that is also holding up another PR |
| 21:14 |
ShadowBot |
https://github.com/minetest/minetest/issues/6803 -- Optimize wieldmesh by numberZero |
| 21:16 |
nerzhul |
stujones11, maybe the build is not optimized |
| 21:22 |
stujones11 |
nerzhul: yeah, I am comparing a debug build here, so that is probably why :/ |
| 21:24 |
nerzhul |
well, i got all mt repository lib using cmake in my build, now i need to find how to integrate external non cmake libs |
| 21:25 |
nerzhul |
gradle build has 6 targets, then the current makefile could not apply properl |
| 21:25 |
nerzhul |
properly* |
| 23:05 |
paramat |
merging #7117 |
| 23:05 |
ShadowBot |
https://github.com/minetest/minetest/issues/7117 -- Biome API: Add 'get_biome_name(biome_id)' API by paramat |
| 23:10 |
paramat |
merged. 93PRs, good to not be stuck at 100 |
| 23:11 |
paramat |
might do some more long line splitting in lua_api tonight |
| 23:17 |
|
stujones11 left #minetest-dev |
| 23:31 |
Gael-de-Sailly |
paramat: so we can't allow mapgen parameters to be chosen at world creation? |
| 23:36 |
paramat |
hi |
| 23:36 |
paramat |
btw everyone i'm now editing lua_api.txt so please no conflicting merges thanks :] |
| 23:37 |
paramat |
Gael some are ok, the clobal ones |
| 23:37 |
paramat |
*global |
| 23:40 |
paramat |
by that i mean formspec type options |
| 23:40 |
paramat |
in advanced settings you can set any mapgen parameters before world creation |
| 23:41 |
Gael-de-Sailly |
yes but in practice players don't do that. |
| 23:41 |
Gael-de-Sailly |
very few people are able to understand the meaning of these parameters |
| 23:42 |
Gael-de-Sailly |
even myself I sometimes don't remember what I made |
| 23:43 |
Gael-de-Sailly |
so to avoid this confusion we could find 5-10 good presets and the player choses within them |
| 23:44 |
paramat |
possibly |
| 23:44 |
paramat |
#7124 will merge in a moment |
| 23:44 |
ShadowBot |
https://github.com/minetest/minetest/issues/7124 -- Lua_api.txt: Split long lines part 3 by paramat |
| 23:44 |
celeron55 |
not necessarily a bad idea if it's actually possible to make good presets |
| 23:45 |
Gael-de-Sailly |
we have already some ones |
| 23:45 |
paramat |
i think hmmmm intended this |
| 23:45 |
paramat |
but never coded it |
| 23:45 |
celeron55 |
what i remember from tweaking mapgens is that it's difficult enough to have one set of parameters that make any sense 8) |
| 23:46 |
paramat |
i can help of course but am not interested in doing the coding work |
| 23:48 |
Gael-de-Sailly |
voxelproof is playing a lot with mapgen parameters, and found some interesting combinations |
| 23:48 |
Gael-de-Sailly |
some could be added I think |
| 23:48 |
Gael-de-Sailly |
his work on mapgen v5 is interesting |
| 23:50 |
Gael-de-Sailly |
I've also worked today on valleysmg, to scale rivers and mountains up |
| 23:54 |
paramat |
merged |
| 23:54 |
paramat |
good to hear |
| 23:55 |
paramat |
might be best to first get the sets of parameters for each mapgen before coding it, as finding good params will be the longer and more difficult job |
| 23:56 |
celeron55 |
any and all mapgen work is always interesting |
| 23:56 |
celeron55 |
to follow |
| 23:56 |
paramat |
i'm so productive recently, i'm amazing =] |
| 23:58 |
paramat |
celeron55 maybe try this branch sometime https://github.com/paramat/minetest/tree/mgwatershed |
| 23:58 |
Gael-de-Sailly |
for the coding part, since it's 100% lua, I can try |
| 23:59 |
|
timcu joined #minetest-dev |
