methylDragon

.. I have no idea how to fix that Windows linking error. All the involved objects are templated, and the other platforms don't complain, so what gives :scream:

I'm gonna change around a few methods for const correctness..

* Linux [](http://ci.ros2.org/job/ci_linux/17639/) * Linux-aarch64 [](http://ci.ros2.org/job/ci_linux-aarch64/12198/) * Windows [](http://ci.ros2.org/job/ci_windows/18214/)

* Linux [](http://ci.ros2.org/job/ci_linux/17736/) * Linux-aarch64 [](http://ci.ros2.org/job/ci_linux-aarch64/12277/) * Windows [](http://ci.ros2.org/job/ci_windows/18379/)

The windows instability seems to be unrelated, but I started a rebuild anyway * Windows [](http://ci.ros2.org/job/ci_windows/18382/)

I'll be iterating on the changes incorporated from https://github.com/methylDragon/rclcpp/pull/1

* Linux [](http://ci.ros2.org/job/ci_linux/17826/) * Linux-aarch64 [](http://ci.ros2.org/job/ci_linux-aarch64/12393/) * Windows [](http://ci.ros2.org/job/ci_windows/18495/)

* Linux [](http://ci.ros2.org/job/ci_linux/17835/) * Linux-aarch64 [](http://ci.ros2.org/job/ci_linux-aarch64/12403/) * Windows [](http://ci.ros2.org/job/ci_windows/18502/)

Re-running windows CI with parantheses around the fold expression (the only issue is cppcheck) * Windows [](http://ci.ros2.org/job/ci_windows/18507/)

I realized me adding the parentheses has a ridiculously minuscule chance of changing the behavior for linux machines, but I'm going to run CI just to be sure. * Linux...