Fix compilation of nav2_smac_planner on Windows

[traversaro]

Basic Info

Info	Please fill out this column
Ticket(s) this addresses	None
Primary OS tested on	Windows
Robotic platform tested on	None, compilation fix.
Does this PR contain AI generated software?	No.

---

Description of contribution in a few bullet points

navigation2 uses the CMAKE_WINDOWS_EXPORT_ALL_SYMBOLS to build shared libraries on Windows, see #1704 for more details. However, CMAKE_WINDOWS_EXPORT_ALL_SYMBOLS can't automatically export public static members, see https://www.kitware.com/create-dlls-on-windows-without-declspec-using-new-cmake-export-all-feature/ .

In case that public static members are needed, it is necessary to manually annotate them with visibility macros. This PR does that for the static members that are actually used in the code, in node_hybrid and node_lattice files.

Description of documentation updates required from your changes

None.

Description of how this change was tested

I compiled the package on Windows.

Future work that may be required in bullet points

None

For Maintainers:

• Check that any new parameters added are updated in docs.nav2.org
• Check that any significant change is added to the migration guide
• Check that any new features OR changes to existing behaviors are reflected in the tuning guide
• Check that any new functions have Doxygen added
• Check that any new features have test coverage
• Check that any new plugins is added to the plugins page
• If BT Node, Additionally: add to BT's XML index of nodes for groot, BT package's readme table, and BT library lists
• Should this be backported to current distributions? If so, tag with backport-*.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Files with missing lines	Coverage Δ
..._planner/include/nav2_smac_planner/node_hybrid.hpp	95.65% <ø> (ø)
...planner/include/nav2_smac_planner/node_lattice.hpp	100.00% <ø> (ø)

... and 3 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[sebatztian]

There seems to be an issue with the global shared variable export on Linux (with kilted at least)
We have 2 instances of the SMAC Hybrid Plugin with different cost_penalty values.
Due to the EXPORT, some variables are shared between the 2 instances, leading to shared parameters. E.g. within motion_table:

// in node_hybrid.hpp
NAV2_SMAC_PLANNER_COMMON_EXPORT static HybridMotionTable motion_table;

The motion_model is initialized within the first call of createPlan(), and thus overwritten by the first createPlan() call of the second plugin.
The cost_penalty, for example, is set within initMotionModel as a variable of motion_model and therefore used by both planners.
We are unsure if this effect occurs for other variables as well.

// in smac_planner_hybrid.cpp
nav_msgs::msg::Path SmacPlannerHybrid::createPlan(...)
{
 ...
  _a_star->setCollisionChecker(&_collision_checker);
 ...

// in a_star.cpp
template<typename NodeT>
void AStarAlgorithm<NodeT>::setCollisionChecker(GridCollisionChecker * collision_checker)
{
  ...
  unsigned int x_size = _costmap->getSizeInCellsX();
  unsigned int y_size = _costmap->getSizeInCellsY();

  clearGraph();

  if (getSizeX() != x_size || getSizeY() != y_size) {
    _x_size = x_size;
    _y_size = y_size;
    NodeT::initMotionModel(_motion_model, _x_size, _y_size, _dim3_size, _search_info);
  }
  _expander->setCollisionChecker(_collision_checker);
}

I don't know what is specifically fixed by the EXPORT in Windows, but it creates unexpected behavior on Linux.

[traversaro]

Hello @sebatztian, can you open a new issue explaining how to reproduce your problem? On Linux all symbols are exported by default, so I do not think it should make a difference to have or not to have NAV2_SMAC_PLANNER_COMMON_EXPORT, but perhaps I am missing somethings.

[mini-1235]

I think this is related to #5118

@sebatztian if you remove NAV2_SMAC_PLANNER_COMMON_EXPORT, does that work for you? From my understanding, the issue is caused by static variables being shared across all plugin instances

[sebatztian]

@mini-1235 @traversaro thank you for the quick reply. Yes, the issue you linked seems to be the same thing.

Sorry, then it is not related to NAV2_SMAC_PLANNER_COMMON_EXPORT, but the static variable itself, which I thought was instance specific.

I don't have a develop Nav2 Environment on the same machine, where I could recreate the problem, so it's not that easy, but judging from the date of the issue, this has to do with the static variable itself rather than the specific export.