elevation wrong

[csawp]

OSI cannot handle this xodr. The elevation chages in a line.
ElevationChangeIn50MLine.zip

wrong shape

[csawp]

Basically OK, but there are a few small flaws. It's not important
But repeatedly load many memory leaks lane and LaneSection

Or, for example

No release

[csawp]

class lane and class LaneSection

Thanks for fixing the bug, but there are some places that haven't been fixed

[eknabevcc]

Thanks for testing - good finding that additional bug!

Committed fixes included in release v2.26.6.

Regarding memory leakage. I fixed the first one. We'll not have time to do more right now due to other prioritizations. But anyone that finds and can fix glitches are welcome to do a pull request. I like that approach since the cred (commit owner) goes to the finder 👍.

[csawp]

Destructor releases memory completely
Road::~Road()
{
for (size_t i = 0; i < geometry_.size(); i++)
{
delete(geometry_[i]);
}
geometry_.clear();
for (size_t i = 0; i < elevation_profile_.size(); i++)
{
delete(elevation_profile_[i]);
}
elevation_profile_.clear();
for (size_t i = 0; i < super_elevation_profile_.size(); i++)
{
delete(super_elevation_profile_[i]);
}
super_elevation_profile_.clear();
for (size_t i = 0; i < super_elevation_shape_profile_.size(); i++)
{
delete(super_elevation_shape_profile_[i]);
}
super_elevation_shape_profile_.clear();
for (size_t i = 0; i < link_.size(); i++)
{
delete(link_[i]);
}
link_.clear();
for (size_t i = 0; i < lane_section_.size(); i++)
{
delete(lane_section_[i]);
}
lane_section_.clear();
for (size_t i = 0; i < lane_offset_.size(); i++)
{
delete(lane_offset_[i]);
}
lane_offset_.clear();
for (size_t i = 0; i < signal_.size(); i++)
{
delete(signal_[i]);
}
signal_.clear();
for (size_t i = 0; i < object_.size(); i++)
{
delete(object_[i]);
}
object_.clear();
}
~LaneSection()
{
for (size_t i = 0; i < lane_.size(); i++)
{
delete lane_[i];
}
lane_.clear();
}
~Lane()
{
for (size_t i = 0; i < link_.size(); i++)
{
delete link_[i];
}
link_.clear();

		for (size_t i = 0; i < lane_width_.size(); i++)
		{
			delete lane_width_[i];
		}
		lane_width_.clear();


		for (size_t i = 0; i < lane_height_.size(); i++)
		{
			delete lane_height_[i];
		}
		lane_height_.clear();

		for (size_t i = 0; i < lane_roadMark_.size(); i++)
		{
			delete lane_roadMark_[i];
		}
		lane_roadMark_.clear();

		delete lane_boundary_;
		lane_boundary_ = 0;
	}

[eknabevcc]

Excellent, thanks for the code!

But instead of me committing it I really want to encourage you to file a pull request - so that the commit will be associated with your GitHub user.

Here's a guide how to do it:
https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/proposing-changes-to-your-work-with-pull-requests/creating-a-pull-request-from-a-fork

(you need write-access to the repo from which you do the pull request, that's why external esmini users need to do it from a fork that they own)

Of course, if you don't want to or don't have the time: Just let me know and I will copy the code and do the commit.

[eknabevcc]

Thanks for the pull request (#336)!
Closing this issue now since the issue is solved within this issue and some memory leaks solved by your pull request.