Move examples into subfolders, add READMEs

1 files changed, 56 insertions, 0 deletions

diff --git a/examples/using_double_on_avr/test_conversions.c b/examples/using_double_on_avr/test_conversions.c
new file mode 100644
index 00000000..22620a6a
--- /dev/null
+++ b/examples/using_double_on_avr/test_conversions.c
@@ -0,0 +1,56 @@
+#include "double_conversion.h"
+#include <math.h>
+#include <stdio.h>
+
+static const double testvalues[] = {
+           0.0,        -0.0,         0.1,         -0.1,
+          M_PI,       -M_PI,  123456.789,  -123456.789,
+      INFINITY,   -INFINITY,         NAN, INFINITY - INFINITY,
+          1e38,       -1e38,        1e39,        -1e39,
+         1e-38,      -1e-38,       1e-39,       -1e-39,
+   3.14159e-37,-3.14159e-37, 3.14159e-43, -3.14159e-43,
+         1e-60,      -1e-60,       1e-45,       -1e-45,
+    0.99999999999999, -0.99999999999999, 127.999999999999, -127.999999999999
+};
+
+#define TESTVALUES_COUNT (sizeof(testvalues)/sizeof(testvalues[0]))
+
+int main()
+{
+    int status = 0;
+    int i;
+    for (i = 0; i < TESTVALUES_COUNT; i++)
+    {
+        double orig = testvalues[i];
+        float expected_float = (float)orig;
+        double expected_double = (double)expected_float;
+        
+        float got_float = double_to_float(*(uint64_t*)&orig);
+        uint64_t got_double = float_to_double(got_float);
+        
+        uint32_t e1 = *(uint32_t*)&expected_float;
+        uint32_t g1 = *(uint32_t*)&got_float;
+        uint64_t e2 = *(uint64_t*)&expected_double;
+        uint64_t g2 = got_double;
+        
+        if (g1 != e1)
+        {
+            printf("%3d double_to_float fail: %08x != %08x\n", i, g1, e1);
+            status = 1;
+        }
+        
+        if (g2 != e2)
+        {
+            printf("%3d float_to_double fail: %016llx != %016llx\n", i,
+                (unsigned long long)g2,
+                (unsigned long long)e2);
+            status = 1;
+        }
+    }
+
+    return status;
+}
+
+    
+    
+