Implement PR feedback

Author: AngusGMorrison

crates/lox_core/src/bodies/fundamental.rs

@@ -172,7 +172,8 @@ mod tests {
 
     // Relative error tolerance for float_eq assertions.
     // This is somewhat loose, being based on observations of how closely our implementations
-    // match ERFA outputs rather than any target TOLERANCE.
+    // match ERFA outputs rather than any target tolerance.
+    // See https://github.com/lox-space/lox/pull/23#discussion_r1398485509
     const TOLERANCE: f64 = 1e-11;
 
     // Test cases for t.
@@ -185,7 +186,7 @@ mod tests {
         assert_float_eq!(
             general_accum_precession_in_longitude(T_ZERO),
             0.0,
-            rel <= TOLERANCE
+            abs <= TOLERANCE
         );
         assert_float_eq!(
             general_accum_precession_in_longitude(T_POSITIVE),

crates/lox_core/src/bodies/nutation.rs

@@ -35,11 +35,11 @@ struct Coefficients {
     /// `Ω`.
     om: f64,
     /// Longitude sine.
-    long_sin_1: f64,
-    long_sin_t: f64,
+    sin_psi_1: f64,
+    sin_psi_t: f64,
     /// Obliquity cosine.
-    obl_cos_1: f64,
-    obl_cos_t: f64,
+    cos_eps_1: f64,
+    cos_eps_t: f64,
 }
 
 /// Calculate nutation coefficients at `epoch` using the given [Model].