* expand NaN checks in gridfns to "is finite" checks

  (also catchs +/-infinity)
* move final stage of calculation of LHS function H(h)
  out of Maple-generated code (in src/gr.cg/) into hand-written code
  in src/gr/horizon_function.c (comput_H()) so we can check that HD > 0
  before taking its sqrt.
	[HD <= 0 probably means the inerpolated g_ij
	isn't positive definite]
* also systematize CCTK_VWarn() printing for various error conditions
  in horizon_function.c


git-svn-id: http://svn.einsteintoolkit.org/cactus/EinsteinAnalysis/AHFinderDirect/trunk@829 f88db872-0e4f-0410-b76b-b9085cfa78c5

1 files changed, 11 insertions, 7 deletions

diff --git a/param.ccl b/param.ccl
index 5fce375..850438b 100644
--- a/param.ccl
+++ b/param.ccl
@@ -504,15 +504,19 @@ real geometry__Schwarzschild_EF__Delta_xyz \
 
 ########################################
 
-# these are pretty cheap tests to catch assorted wierdness, so it's
-# probably worth leaving them on unless you're trying to squeeze every
-# last nanosecond...
-boolean check_h_for_NaNs \
-  "should we check the horizon shape function h for NaNs?"
+#
+# These tests control whether we check that various angular gridfns
+# are finite (neither NaN nor infinity) at various points in evaluating
+# the H(h) function.  These are pretty cheap tests, and they're quite
+# useful in catching assorted wierdness, so it's probably worth leaving
+# them enabled unless you're trying to squeeze every last nanosecond...
+#
+boolean check_that_h_is_finite \
+  "should we check that horizon shape function h is finite?"
 {
 } "true"
-boolean check_geometry_for_NaNs \
-  "should we check the interpolated geometry variables for NaNs?"
+boolean check_that_geometry_is_finite \
+  "should we check the interpolated geometry variables are finite?"
 {
 } "true"