-
Notifications
You must be signed in to change notification settings - Fork 13.7k
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
catkin build fails -- Fails to find headers from ecl/ #9554
Comments
@dagar I will but I can't promise to have it ready this week. |
Now that we're working toward a package to use for ROS sim/testing, do we want to support this additional build system? The idea is to make this more user friendly in the long run and only require building for dev purposes. |
Any conclusion on this issue, its still failing for me with same error while catkin build : "Cannot find ecl.h, not such file or directory" |
I've been using "v1.7.3" until this is fixed. I'm not sure why this is closed. |
Exactly, i am not at all able to catkin build, px4 not prepared to use catkin environment |
@mickey13 this isn't closed. |
Possible fix in #9679. Could someone please test it to see if your issue is fully resolved? |
How do you downgrade to Firmware v1.7.3? |
In the firmware directory you can use: Hoping the issue is resolved soon as I have other incompatibilities with v1.7.3. |
I ended up building a ROS node in my simulation (Gazebo) package for starting instances of the latest PX4 firmware (so that I could use the latest version in my SITL setup):
Note you'll want to blacklist building this with catkin:
This is just a workaround that I'm using until this issue is resolved. |
What is the motivation for tightly-coupling PX4 within your ROS environment? I don't see an advantage to do this. This build system isn't actively supported for this project. Use PX4 as an independent node, launch with roslaunch, and build with the supported system. Is there a problem with the standard approach? |
My requirement (and I believe it's the same for others) is that I want to be able to run PX4 instances from my ROS workspace for each modeled UAV in my simulation environment. I don't consider that requirement "tightly-coupling PX4 within your ROS environment", as I include other non-PX4 packages via wstool and build them with catkin. Since PX4 cannot be built with catkin any longer, nor is there a ROS package that can be downloaded, it's not obvious how to start PX4 processes in a standard simulation environment that uses ROS (i.e. you can't roscd into px4). What the node I mentioned above does is starts up a PX4 process so you can connect to it via MAVROS for any UAVs in your simulation environment. |
You sure can! You just don't have your environment setup as described by the docs. |
Fixed in #10587. |
Ladies and gentleman, i didn't want to upset you, but issue persist in version 1.8.1 Ros version: Kinetic To generate error: # create tmp workspace
mkdir -p ~/tmp_ws/src
cd ~/tmp_ws/src
# clone px4 source
git clone git clone https://github.com/PX4/Firmware.git
cd ~/tmp_ws/src/Firmware
# check out to version 1.8.1
git checkout v1.8.1
# rest should be familiar
git submodule update --init --recursive
cd ~/tmp_ws/
catkin_make Edit: |
@incebellipipo the issue is fixed in PX4 master (what will be v1.9.0). |
Describe the bug
Compiling Firmware as a ROS package using catkin build fails, due to catkin not finding headers located in the src/lib/ecl/ folder.
Firmware itself builds fine when I use standard: mkdir build && cd build && cmake .. && make. I used this approach when first trying out the PX4/Firmware software, but now I was going to proceed to use ROS, so I wanted to build the PX4/Firmware as a ROS package.
To Reproduce
Steps to reproduce the behavior:
0. Clone Firmware inside your ros_workspace/src folder (master branch)
PS.- FYI, I just tried building (catkin build) the Firmware version v1.7.3 and that works fine (I still have to test some launch files, but at least the ROS package compiling finished successfully). I'll report if this version works fine, since that might mean that the compilation error might have been introduced between then and the current master.
The text was updated successfully, but these errors were encountered: