better public interface and doxygen

1 files changed, 27 insertions, 0 deletions

diff --git a/tests/common/test_utf8.c b/tests/common/test_utf8.c
new file mode 100644
index 0000000..72f8904
--- /dev/null
+++ b/tests/common/test_utf8.c
@@ -0,0 +1,27 @@
+#include "common/utf8.h"
+#include <assert.h>
+#include <stdlib.h>
+#include <string.h>
+
+int main(void)
+{
+    char latin1[256];
+    for (int i = 1; i <= 255; i++) latin1[i - 1] = (char)i;
+    latin1[255] = '\0';
+    size_t len = 0, back_len = 0;
+    char *utf8 = yasl_utf8_from_latin1(latin1, 255, &len);
+    assert(utf8 && yasl_utf8_validate(utf8, len));
+    char *back = yasl_utf8_to_latin1(utf8, &back_len);
+    assert(back && back_len == 255 && memcmp(back, latin1, 256) == 0);
+    free(back);
+    free(utf8);
+    assert(!yasl_utf8_to_latin1("😀", &len));
+    assert(!yasl_utf8_to_latin1("\xc3", &len));
+    assert(!yasl_utf8_validate("\xc3", 1));
+    assert(!yasl_utf8_validate("\xe2\x82", 2));
+    assert(!yasl_utf8_validate("\xf0\x9f\x98", 3));
+    assert(!yasl_utf8_validate("\xed\xa0\x80", 3)); /* Surrogate. */
+    assert(yasl_utf8_validate("😀", sizeof("😀") - 1));
+    assert(yasl_utf8_validate(NULL, 0));
+    return 0;
+}