-
Notifications
You must be signed in to change notification settings - Fork 2.9k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
DUSE_SOFT_BODY_MULTI_BODY_DYNAMICS_WORLD compiler error #1106
Comments
Don't define DUSE_SOFT_BODY_MULTI_BODY_DYNAMICS_WORLD. Does it work then? For pybullet use pip install pybullet. See also the pybullet quickstart guide at pybullet.org |
#1107 is with DUSE_SOFT_BODY_MULTI_BODY_DYNAMICS_WORLD disabled, but that one fails due to separate issues with the CMake configuration. I use ROS and TensorFlow, which means using both the C & Python APIs, and I was hoping to use a single installation for integrating both. |
I think this is likely solved fairly easily by adding whatever lib has a |
Sure, we should fix the cmake build. Why is pybullet not sufficient? pybullet works great with tensorflow. So you want a ROS - pybullet bridge? I think the c - api is not needed for that. |
Which python version are you using and what flags/options do you enable exactly using cmake? cmake with standard out-of-the-box options should build, since it is in Appveyor and Travis. |
I should be able to stick with python as long as I can manage performance bottlenecks, I'll explain why I asked. I work on a project called costar that uses more traditional approaches to planning and control with a combination of C++ and python code. For that reason, accessing a single scene instance from both python and C++ would be convenient but not a blocking requirement. A motivating example where C/C++ provides a substantial benefit would be if bullet could take in high bandwidth data like either a point cloud from an RGBD sensor or a mesh continuously updated via C/C++. An example of such a mesh would be one created with data from SLAM running outside bullet. I'm planning to modify costar to use a combination of something more traditional like the Tasks robot model based control and constrained optimization library alongside machine learning for decision making. Actually since it looks like some of the IK stuff in bullet may be in progress Tasks may be a useful/interesting/relevant resource. The new pybullet functionality looks like it will be a great resource for the machine learning / reinforcement learning component, so thanks for the great work! My cmake config was the same as #1107 I ran into this error first and it was with the addition of I actually encountered this compiler error both in Linux and OS X. #1107 is OS X only. Here were the Linux flags and a script to reproduce, though the soft body flag is removed in the linked code so I can compile on linux:
|
…r_SharedMemory BulletSoftBody was not linking to App_PhysicsServer_SharedMemory when USE_SOFT_BODY_MULTI_BODY_DYNAMICS_WORLD was enabled.
btsoftbody fix #1106, compiler error in App_PhysicsServer_SharedMemory
I'm getting a compiler error on OS X when I define
DUSE_SOFT_BODY_MULTI_BODY_DYNAMICS_WORLD=ON
in my cmake build when trying to build commit 47eae8d, and the error is probably related to #850.Full homebrew gist of installation process steps & output:
https://gist.github.com/ahundt/df791c901f89da9341f7760455c4bd9d
Relevant compiler error snippet:
The text was updated successfully, but these errors were encountered: