| Commit message (Collapse) | Author | Age |
|---|
| |
|
|
|
|
|
|
|
|
|
|
| |
Previously, the biomemap was 'BIOME_NONE' for a mapchunk column if a
stone surface was not found in it, causing water surface decorations
and water surface dust to fail.
Store the biome ID of the biome calculated at a water surface and add
it to the biomemap if the biomemap is 'BIOME_NONE' for the mapchunk
column. The biome calculated at a stone surface still has priority
for the biomemap entry, as it should.
Edit an incorrect comment.
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
| |
Revert commit 99143f494711034068685b6ee845ce19fa09d7d9 and commit
f4ca830abe1aa22875c99b31bf2ee56e26f83f05.
These commits caused biome dust to be applied even when there was no core
mapgen terrain in a mapchunk column. So the dust, which overgenerates,
then appeared on structures added by mods in 'on_generated', such as
floatlands, asteroids or above-surface realms.
|
| | |
|
| | |
|
| |
|
|
|
|
| |
Adding support for _NET_WM_PID as defined in Extended Window Manager Hints
Move verbose messaging to setupXorgTopLevelWindow method as Xorg messages should only occur when running in Xorg env.
Irrlicht returns the XDisplay as a void* and XWindow as an unsigned long so reinterpret those as the appropriate type. Also fixed a spaces for tab formating issue
|
| |
|
|
| |
* Version scheme change: 0.5.0 -> 5.0.0
|
| | |
|
| |
|
| |
client_lua_api.md -> client_lua_api.txt
|
| |
|
|
|
|
|
|
|
| |
* Minimal: Explicit log levels
* Minimal: Formspec test now disabled by default
This is done because the formspec test was disruptive.
This test can now be toggled with server command test_formspec first.
|
| |
|
|
| |
* Server: move shutdown parts to a specific shutdown state object
|
| | |
|
| | |
|
| |
|
|
|
| |
Require 'basic_privs' priv
Only the online players are listed.
|
| |
|
|
| |
Also update wrapper
|
| |
|
| |
Resync with stable-0.4
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
MT doesn't launch without that
|
| |
|
|
| |
It remains one issue with MT itself and rtti
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
| |
Previously the only way to disable biomes was to 'clear' the registered
biomes in a mod, but this method causes large amounts of unnecessary
processing:
1. Calculation of 4 2D noises.
2. Looping through all nodes of a mapchunk replacing nodes with identical
nodes.
The new flag disables those operations.
|
| | |
|
| | |
|
| |
|
|
|
| |
* CSM/SSM: Add on_mods_loaded callback
|
| |
|
|
| |
* Fix builtin lua function os.tempfolder
|
| | |
|
| |
|
|
|
|
|
| |
* Modernize lua read (part 1): C++ templating assurance
Implement the float reader
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Node definition:
Clearly document custom selection box/collision box (collision box
documentation was missing).
Remove incorrect light attenuation statement and duplicated light
source statement.
Nodes:
Document which drawtypes require 'paramtype = "light"' to avoid
appearing black.
Remove incorrect light attenuation statement.
HUD: Remove 'HUD API is experimental' text.
Noise params: Spread of every octave must exceed 1.
Particles and spawners: Document glow values.
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
| |
'generateBiomes()' constructs the biomemap as it generates biomes.
The biome calculated at first stone surface encountered is added to
the biomemap.
Previously, if no stone surface was encountered in a mapchunk column
the biomemap was left empty for that (x, z) position, causing biome
dust and water surface decoration placement to fail.
If at the base of a mapchunk column the biomemap is empty, add the
currently active biome to the biomemap, or if biome is NULL calculate
it for this position and add it to the biomemap.
|
| |
|
|
| |
* Fix isNan on setYaw Lua call
|
| |
|
|
|
| |
Only allow river generation to replace c_stone.
This also acts as an optimisation by being placed before canyon shape
calculation.
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
| |
Move code to pkgmgr
|
| |
|
|
|
|
| |
a78659ed0529226ad937f5e241ad72ba93702959
Added more strong unittests on that part to prevent future regression
|
| |
|
|
|
| |
In some configurations, such as when using the runit supervisor and its
tool chpst, the HOME variable might be unset. This resulted in an
unclear error message that was hard to pin down.
|
| |
|
|
|
|
|
|
|
|
| |
1 master
Fix 3 warnings reported by GCC 8.1 of the following type
```src/client/gameui.cpp:191:43: warning: « void* memset(void*, int, size_t) » effacement d'un objet du type non trivial « struct GameUI::Flags »; use assignment or value-initialization instead [-Wclass-memaccess]
memset(&m_flags, 0, sizeof(GameUI::Flags));
```
|
| |
|
|
|
|
|
| |
```
src/translation.cpp:43:16: warning: interception du type polymorphique « class std::out_of_range » par valeur [-Wcatch-value=]
} catch (std::out_of_range) {
```·
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
| |
Because vein ore uses 3D noise (all the other ores use 2D noise) the
perlinmap Y size can be different in different mapchunks when close
to the ore Y limits.
Previously this caused bugs in the vein structure because changes in
perlinmap Y size did not recreate the noise objects.
Delete and recreate the noise objects with the new Y size if Y size
has changed.
|
