test: add testcases for low-level utf8 routines.

test/CMakeLists.txt

@@ -7,6 +7,7 @@ set(SOURCES
     red_black_tree.c
     list.c
     structures.c
+    utf8.c
     )
 
 

test/main.c

@@ -6,6 +6,7 @@ void cry_test_arrays();
 void cry_test_red_black_trees();
 void cry_test_maps();
 void cry_test_lists();
+void cry_test_utf8();
 
 int 
 main(int argc, char** argv)
@@ -17,6 +18,7 @@ main(int argc, char** argv)
 	cry_test_red_black_trees();
         cry_test_lists();
 	cry_test_maps();
+	cry_test_utf8();
 
 	cry_unittest_finalize();
 

test/utf8.c

+#include "unittest.h"
+#include "utf8.h"
+#include <assert.h>
+#include <stdio.h>
+
+// simple ascii string (length = 10)
+static const byte* ascii = "teststring";
+
+// simple latin1 string in german (length = 16)
+static const byte* latin1 = "Ger\xC3\xA4t\xC3\xBC""berhitzung";
+
+// alpha beta gamma pi (length = 4)
+static const byte* greek = "\xCE\xB1\xCE\xB2\xCE\xB3\xCE\x80";
+
+// hiragana: ze yo vu ba (length = 4)
+static const byte* hiragana = "\xE3\x81\x9C\xE3\x82\x88\xE3\x82\x94\xE3\x81\xB0";
+static const unicode uc_hiragana[] = { 0x305C, 0x3088, 0x3094, 0x3070 };
+
+// mixed with codepoints of different sizes (length = 9)
+static const byte* mixed = "tz\xC3\xA4 \xE4\xBB\x90 \xF0\x90\xB9\xA0 \xEF\xBB\x85";
+static const unicode uc_mixed[] = { 't', 'z', 0x00E4, ' ', 0x4ED0, ' ', 0x10E60, ' ', 0xFEC5};
+
+
+static void test_utf8_validate(const_pointer data)
+{
+	assert(cry_utf8_validate(ascii) == CRY_OKAY);
+	assert(cry_utf8_validate(latin1) == CRY_OKAY);
+	assert(cry_utf8_validate(greek) == CRY_OKAY);
+	assert(cry_utf8_validate(hiragana) == CRY_OKAY);
+        assert(cry_utf8_validate(mixed) == CRY_OKAY);
+
+        assert(cry_utf8_validate("Non Valid \xE3 utf8string") == CRY_FAIL);
+}
+
+
+static void test_utf8_length(const_pointer data)
+{
+	test_utf8_validate(data);
+
+	assert(cry_utf8_strlen(ascii) == 10);
+	assert(cry_utf8_strlen(latin1) == 16);
+	assert(cry_utf8_strlen(greek) == 4);
+	assert(cry_utf8_strlen(hiragana) == 4);
+	assert(cry_utf8_strlen(mixed) == 9);
+
+	assert(cry_utf8_strsize(ascii) == 10);
+	assert(cry_utf8_strsize(latin1) == 18);
+	assert(cry_utf8_strsize(greek) == 8);
+	assert(cry_utf8_strsize(hiragana) == 12);
+	assert(cry_utf8_strsize(mixed) == 17);
+}
+
+
+static void test_utf8_codepoints(const_pointer data)
+{
+	test_utf8_validate(data);
+
+	const byte* str; 
+	size_t index;
+
+	str = ascii;
+	index = 0;
+	while(*str != 0 && index < 10) {
+		assert(cry_utf8_get(str) == *str);
+                assert(cry_utf8_codepoints(str) == 1);
+		str = cry_utf8_next(str);
+		++index;
+	}
+	assert(index == 10);
+
+	str = hiragana;
+	index = 0;
+	while(*str != 0 && index < 4) {
+		assert(cry_utf8_get(str) == uc_hiragana[index++]);
+                assert(cry_utf8_codepoints(str) == 3);
+		str = cry_utf8_next(str);
+	}
+	assert(index == 4);
+
+	str = mixed;
+	index = 0;
+	while(*str != 0 && index < 9) {
+		assert(cry_utf8_get(str) == uc_mixed[index++]);
+		str = cry_utf8_next(str);
+	}
+	assert(index == 9);
+}
+
+
+static void test_utf8_search(const_pointer data)
+{
+	test_utf8_validate(data);
+
+	// forward search
+	assert(cry_utf8_strchr(ascii, 's') == ascii + 2);
+	assert(cry_utf8_strchr(ascii, 'g') == ascii + 9);
+	assert(cry_utf8_strchr(mixed, 'z') == mixed + 1);
+	assert(cry_utf8_strchr(mixed, 0x00E4) == mixed + 2);
+	assert(cry_utf8_strchr(mixed, 0x4ED0) == mixed + 5);
+	assert(cry_utf8_strchr(mixed, 0xFEC5) == mixed + 14);
+        assert(cry_utf8_strchr(ascii, 'P') == 0);
+
+	// reverse search
+	assert(cry_utf8_strrchr(ascii, 's') == ascii + 4);
+	assert(cry_utf8_strrchr(latin1, 'G') == latin1);
+	assert(cry_utf8_strrchr(hiragana, 0x305C) == hiragana);
+	assert(cry_utf8_strrchr(mixed, 0x00E4) == mixed + 2);
+        assert(cry_utf8_strrchr(ascii, 'P') == 0);
+}
+
+
+void
+cry_test_utf8(void)
+{
+	cry_unittest_run("utf8.length", test_utf8_length, 0, 10);
+	cry_unittest_run("utf8.codepoints", test_utf8_codepoints, 0, 10);
+	cry_unittest_run("utf8.search", test_utf8_search, 0, 10);
+}