This is an amendment to 4bfc2c0311.
Add an autolimits compiler option. When true, *limited will be inferred from the presence of the *range attributes.
When false, models where limited is not specified, and range is will fail to compile.
PiperOrigin-RevId: 472438650
Change-Id: I489098e879ee695030f44e0d8cd969787e75fb90
Commit e3a82247c2 removed MJAPI in mujoco.h but not in engine_derivative.h, causing a linking error when building on Windows with MSVC.
PiperOrigin-RevId: 471639915
Change-Id: I6b965e23a19faee6f9f8f9c0168dc2774b19f310
- Added `refsite` attribute to actuators with `site` transmission.
- Such actuators now have a well defined length and can be used for e.g., Cartesian end-effector control.
- Added example models and documentation describing the new feature.
PiperOrigin-RevId: 471561218
Change-Id: I538f09af9600d5c53992c4a9fed325648aab4fa3
- Improved time-sync logic readability in `PhysicsLoop`.
- Fixed unattainable condition for breaking from stepping loop if data was reset.
- Added measurement of actual real-time tracking.
- Moved slowdown overlay to top left.
- Report slowdown as % of real-time, rather than fraction.
- Report actual slowdown if different than requested (very small timestep, PhysicsLoop cannot keep up) by more than 10%.
- Replaced slowdown increments of 2 with decimal-rounded increments of 10^(1/10). This is a finer-grained scale (approximately 3x as fine) and well suited for decimal percentage representation.
- Maximum slowdown increased to 1000.
- Added `Simulate.refreshrate` with a default of 60Hz, in case videomode refreshrate is 0 (can occur e.g. when forwarding over X11).
- Remove realtime reporting from regular info overlay.
- Various cleanups.
PiperOrigin-RevId: 471547481
Change-Id: Idd2514a4defb1dc431617f3a515cc747c4fd8971
This allows us to use GLFW without needing to introduce a link-time dependency on the library, which in turn enables us to get GLFW through Python.
PiperOrigin-RevId: 470100999
Change-Id: I27c84ad3310afdbebd582ab0033b6a963ba6307e
- `anchor` determines the point of wrench application, in the frame of body2.
- `tfratio` scales applied torques relative to applied forces.
- Add visualisation of both anchor points to both weld and connect constraints.
- Add a test model showing how the new weld parameters behave.
PiperOrigin-RevId: 470021901
Change-Id: I3faba0e5dcab2e29ee81096f5ff71ff66d99758c
- `anchor` determines the point of wrench application, in the frame of body2.
- `tfratio` scales applied torques relative to applied forces.
- Add visualisation of both anchor points to both weld and connect constraints.
- Add a test model showing how the new weld parameters behave.
PiperOrigin-RevId: 469228149
Change-Id: I836b0791f10fb624607a12ef3c687da991c21789
- `anchor` determines the point of wrench application, in the frame of body2.
- `tfratio` scales applied torques relative to applied forces.
- Add visualisation of both anchor points to both weld and connect constraints.
- Add a test model showing how the new weld parameters behave.
PiperOrigin-RevId: 469190483
Change-Id: I20f6da85b09cd2c4c0b5d29eb8003eabb524e55e
* If output file exists, ask user whether to overwrite
* Fix erroneous check for whether XML file was successfully saved
PiperOrigin-RevId: 468743682
Change-Id: I06058c498328d333792e4b40d65c3a99ca1842e4
- `distance` constraints were not very useful and increased the contact buffer within the constraint solver, violating a core design principle. Therefore they have been removed.
- The enum value `mjEQ_DISTANCE` is retained, but errors will be thrown in the parser and engine if these constraints are selected.
PiperOrigin-RevId: 467050101
Change-Id: Ie72a07e4fe338eea5107004a9fc0f5e47eb0c3b7
This is a very subtle leak caused interaction between longjmp and compiler optimisation. Specifically, at the point where the setjmp takes places, these pointers have never been reassigned from its nullptr initialization. Without the volatile keyword, the compiler is free to assume that these pointers remain nullptr when the setjmp returns, and therefore to pass nullptr directly to the mj_deleteModel and mj_deleteData calls in the subsequent catch block, without ever reading the actual pointer values.
PiperOrigin-RevId: 466649447
Change-Id: I51e82ec8b566deb144221e11a99d5124a8267de4
The memory handling in the asset loading logic in user_mesh.cc is currently flipped: it frees the asset content buffer when reading from VFS but not from the filesystem. This causes the VFS object to leave these functions in an invalid state, while leaking locally allocated buffers that hold contents read from the filesystem.
PiperOrigin-RevId: 466643390
Change-Id: I162221e019861f23fb61308facd63dbb0422e113
Saran Tunyasuvunakool