new tester; all tests categorized

tests/float/tfloat1.nim

@@ -0,0 +1,15 @@
+discard """
+  file: "tfloat1.nim"
+  outputsub: "Error: unhandled exception: FPU operation caused an overflow [EFloatOverflow]"
+  exitcode: "1"
+"""
+# Test new floating point exceptions
+
+{.floatChecks: on.}
+
+var x = 0.8
+var y = 0.0
+
+echo x / y #OUT Error: unhandled exception: FPU operation caused an overflow [EFloatOverflow]
+
+

tests/float/tfloat2.nim

@@ -0,0 +1,15 @@
+discard """
+  file: "tfloat2.nim"
+  outputsub: "Error: unhandled exception: FPU operation caused a NaN result [EFloatInvalidOp]"
+  exitcode: "1"
+"""
+# Test new floating point exceptions
+
+{.floatChecks: on.}
+
+var x = 0.0
+var y = 0.0
+
+echo x / y #OUT Error: unhandled exception: FPU operation caused a NaN result [EFloatInvalidOp]
+
+

tests/float/tfloat3.nim

@@ -0,0 +1,24 @@
+discard """
+  file: "tfloat3.nim"
+  output: "Nimrod    3.4368930843, 0.3299290698 C double: 3.4368930843, 0.3299290698"
+"""
+
+import math, strutils
+
+{.emit: """
+void printFloats(void) {
+    double y = 1.234567890123456789;
+    
+    printf("C double: %.10f, %.10f ", exp(y), cos(y));
+}
+""".}
+
+proc c_printf(frmt: CString) {.importc: "printf", header: "<stdio.h>", varargs.}
+proc printFloats {.importc, nodecl.}
+
+var x: float = 1.234567890123456789
+c_printf("Nimrod    %.10f, %.10f ", exp(x), cos(x))
+printFloats()
+
+
+