Time  Nick        Message
00:05 sfan5       you've probably read the function comment, nobody knows and I don't know of any
00:05 DS-minetest xD sounds very... mysterious
15:02 Wuzzy       #9588
15:02 ShadowBot   https://github.com/minetest/minetest/issues/9588 -- Add relative numbers for commands by prepending ~ by Wuzzy2
17:26 sfan5       still hoping for more reviews to ##10970
17:26 ShadowBot   https://github.com/minetest/minetest/issues/10970 -- [DONT SQUASH] Switch Irrlicht dependency to our own fork by sfan5
17:51 Krock_      sfan5: is libbz2-dev no longer required by Irrlicht 1.9-dev?
17:51 sfan5       yes
17:54 Krock_      I cannot do more than testing it on my machine. The code looks good and the build passes (except Android due to outdated libs)
17:55 Krock_      > sfan5 added 3 commits on 17 May 2020
17:55 Krock_      github seems to be drunk somehow
17:56 sfan5       the first commit is technically from May
17:56 sfan5       but some time ago github switched to showing the commit times, not the author times (don't know the correct terminology)
18:01 Krock       the Irrlicht paths are still only visible in the Advanced cmake options. It's not so important but marking those as regular fields would help
18:02 sfan5       good idea
18:03 Krock       looks like the "CACHE" keyword exposes that
18:04 sfan5       I thought the function for that was called mark_as_advanced
18:05 sfan5       no idea if we use it
18:05 Krock       also the descriptions could be improved. "IRRLICHT_INCLUDE_DIR" currently states "Path to a file."
18:05 Krock       hmm
18:05 Krock       oh right. that's also used within FindIrrlicht.cmake
18:10 Krock       and "IRRLICHT_SOURCE_DIR" needs documentation in "Library specific options"
18:10 Krock       (readme.md)
18:11 sfan5       I removed that one
18:11 Krock       oh. In this case it should also be removed from the Windows section
18:12 Krock       * Build section
18:12 Krock       never mind me
18:12 Krock       I'm just stupid
18:14 Krock       hmm weird. Even after specifying the paths (existing cache file), it still seems to check for system only. https://i.postimg.cc/q7mXZP41/grafik.png
18:15 sfan5       you might have to reset the HAS_FORKED_IRRLICHT value
18:15 sfan5       cmake remembers it even if the prerequisites changed
18:17 Krock       HAS_FORKED_IRRLICHT:INTERNAL=1  <-- manual change. this does the trick.
18:17 Krock       strange. this value shouldn't be cached
18:18 sfan5       it's possible to prevent it from being cached but then the test would be performed each time
18:18 Krock       only every time when the CMakeLists change
18:21 sfan5       hm right
18:22 Krock       noted all suggestions -> comment.
18:30 MTDiscord   <Jordach> I’m happy to test irrlicht CMake changes shortly
18:31 MTDiscord   <Jordach> Under macOS to ensure that building still works as expected
18:36 sfan5       Krock: changed stuff as suggested
20:35 sfan5       merging #10970 in 15m
20:35 ShadowBot   https://github.com/minetest/minetest/issues/10970 -- [DONT SQUASH] Switch Irrlicht dependency to our own fork by sfan5
23:25 kilbith     compiled irrlicht fork w/ -DBUILD_SHARED_LIBS=ON and minetest w/ -DIRRLICHT_LIBRARY=/usr/lib/x86_64-linux-gnu/libIrrlicht.so and still getting cmake error
23:27 kilbith     ideally I shouldn't have to specify -DIRRLICHT_LIBRARY
23:35 sfan5       did you install it system-wide?
23:36 kilbith     yes
23:37 kilbith     if-DBUILD_SHARED_LIB is on, I guess it's system-wide?
23:38 sfan5       if you ran sudo make install, sure
23:38 sfan5       but if you already have irrlicht 1.8.4 installed who knows which it'll pick
23:39 kilbith     ah yes, that makes it
23:39 kilbith     documentation missing