vahancho / erkir Goto Github PK

target_include_directories(erkir PUBLIC include) make it easier to use this library.
It will then be enough to write target_link_libraries(<target> erkir) and the include directory will be accessible automatically.
But as it is now one have to specify erkir include dir manually with e.g include_libraries().

Hi!

Just wanted to give you a heads up that while adding support for the new version of the library in Conan Center Index in conan-io/conan-center-index#23224, we realized that the CMake file lists the wrong version of the library from what was later tagged as 2.1.0 in https://github.com/vahancho/erkir/blob/master/CMakeLists.txt#L9 :)

CMake is currently configured to use CMAKE_SOURCE_DIR
Conflicts may arise when the library is imported (for example, using CPM)
It is necessary to change every instance of CMAKE_SOURCE_DIR to CMAKE_CURRENT_SOURCE_DIR in order to prevent the latter

Good morning!

I am using your library and I think I have encountered a discrepancy regarding the heightof a point when converted into a different coordinate standard (i.e. when chaning its datum). In particular, if I define a point as

erkir::ellipsoidal::Point greenwichWGS84(51.47788, -0.00147, 0.0, erkir::ellipsoidal::Datum::Type::WGS84);

and then I convert it to the OSGB36 standard, I obtain an height of -45.9051.

Is this result the correct height, maybe because we are in a different standard? Or is there a mistake somewhere?

Also, in the wikipedia page the Helmert transformation is defined as:

while in your code it is written slightly differently. Is there a reason for that?

Thank you in advance! 🙏

It got some crazy jumps so I just checked with OpenCPN and it concurs with my findings. Setting OpenCPN to degrees, decimal minutes correspons to Point.latitude().degrees()

The documentation for Point.h is inconsistent. It states that a default constructed point is invalid, however the default constructor sets lat lon to 0 and valid to true.