What makes the code significant is that it's the first time that results from equations of gravity (using values for G, M, and r) match results from special relativity (SR) for any inputs for such test, insofar as finite precision allows. This makes the EP fully bidirectional. Click the Run button to see matching results.
Most tests of general relativity (GR) have been tests of the Schwarzschild metric, which limits the height of an approximately uniform gravitational field. It predicts that they can be only so tall in nature. For instance, you might be able to create a 1.0 g field with a height of 1 AU, but not 1 light year. This height is unlimited in SR, which handles uniform gravity tests. Thus the metric can't do a comparison test to SR using the default inputs in the code.
To fix this issue a logical error in the derivation of the metric was corrected, to derive a new metric. Three other issues had to be addressed, so it was like a bug with four causes. The new metric is fully experimentally confirmed (e.g. it predicts 42.98 arcseconds per century for Mercury's Schwarzschild precession) and solves several major problems in physics, like the black hole information paradox.
I have a write-up that explains everything and with more code; here it's just one unit test that other techies might find interesting. Though my findings are considered crackpot theory and so are throwaway, I'm happy with this code in particular. To me the proof is always in the pudding, with experimental confirmation, logical consistency, and passing of unit tests. I learned that the universe is more interesting than we think, with fewer assumptions needed to explain observations.