Java Native Interfaces use modified UTF-8 instead than standard UTF-8. The difference can usually be ignored, except if the string contains supplementary Unicode characters. It could be that some components created by following GIGS test (Geospatial Integrity of Geoscience Software) have such supplementary characters:
GIGS series 2004, "Douala" test (EPSG:6192)
Currently, attempts to run this test produce the following error message:
AssertionError: CharSequence: High surrogate shall be followed by low surrogate.
Commit aac33e8344509fe894c1dcad4ce6c696a38b2bac has shown that this was a bug in our handling of C++ std::string construct in some places rather than a Unicode encoding issue.
Java Native Interfaces use modified UTF-8 instead than standard UTF-8. The difference can usually be ignored, except if the string contains supplementary Unicode characters. It could be that some components created by following GIGS test (Geospatial Integrity of Geoscience Software) have such supplementary characters:
Currently, attempts to run this test produce the following error message: