--- a/tests/test_json.c Sat Jan 25 15:22:01 2025 +0100
+++ b/tests/test_json.c Tue Jan 28 18:31:17 2025 +0100
@@ -120,7 +120,8 @@
CX_TEST(test_json_escaped_strings) {
cxstring text = cx_str(
"{\n"
- "\t\"object\":\"{\\n\\t\\\"object\\\":null\\n}\"}\"\n"
+ "\t\"object\":\"{\\n\\t\\\"object\\\":null\\n}\",\n"
+ "\t\"ctrl-chars\":\"\\\\foo\\r\\nbar\\f*ring\\/ring*\\b\"\n"
"}"
);
@@ -138,6 +139,152 @@
cxJsonAsCxString(object),
CX_STR("{\n\t\"object\":null\n}"))
);
+ CxJsonValue *ctrl = cxJsonObjGet(obj, "ctrl-chars");
+ CX_TEST_ASSERT(cxJsonIsString(ctrl));
+ CX_TEST_ASSERT(0 == cx_strcmp(
+ cxJsonAsCxString(ctrl),
+ CX_STR("\\foo\r\nbar\f*ring/ring*\b"))
+ );
+ cxJsonValueFree(obj);
+ }
+ cxJsonDestroy(&json);
+}
+
+CX_TEST(test_json_escaped_unicode_strings) {
+ cxstring text = cx_str(
+ "{\n"
+ "\"ascii\":\"\\u0041\\u0053\\u0043\\u0049\\u0049\",\n"
+ "\"unicode\":\"\\u00df\\u00DF\",\n"
+ "\"mixed\":\"mixed ä ö \\u00e4 \\u00f6\",\n"
+ "\"wide\":\"\\u03a3\\u29b0\",\n"
+ "\"surrogatepair1\":\"\\ud83e\\udff5\",\n"
+ "\"surrogatepair2\":\"test\\ud83e\\udff1AA\"\n,"
+ "\"mixed2\":\"123\\u03a3\\ud83e\\udfc5\\u00df\""
+ "}"
+ );
+
+ CxJson json;
+ cxJsonInit(&json, NULL);
+ CX_TEST_DO {
+ cxJsonFill(&json, text);
+ CxJsonValue *obj;
+ CxJsonStatus result = cxJsonNext(&json, &obj);
+ CX_TEST_ASSERT(result == CX_JSON_NO_ERROR);
+ CX_TEST_ASSERT(cxJsonIsObject(obj));
+
+ CxJsonValue *ascii = cxJsonObjGet(obj, "ascii");
+ CX_TEST_ASSERT(cxJsonIsString(ascii));
+ CX_TEST_ASSERT(0 == cx_strcmp(
+ cxJsonAsCxString(ascii),
+ CX_STR("ASCII"))
+ );
+
+ CxJsonValue *unicode = cxJsonObjGet(obj, "unicode");
+ CX_TEST_ASSERT(cxJsonIsString(unicode));
+ CX_TEST_ASSERT(0 == cx_strcmp(
+ cxJsonAsCxString(unicode),
+ CX_STR("ßß"))
+ );
+
+ CxJsonValue *mixed = cxJsonObjGet(obj, "mixed");
+ CX_TEST_ASSERT(cxJsonIsString(mixed));
+ CX_TEST_ASSERT(0 == cx_strcmp(
+ cxJsonAsCxString(mixed),
+ CX_STR("mixed ä ö ä ö"))
+ );
+
+ CxJsonValue *wide = cxJsonObjGet(obj, "wide");
+ CX_TEST_ASSERT(cxJsonIsString(wide));
+ CX_TEST_ASSERT(0 == cx_strcmp(
+ cxJsonAsCxString(wide),
+ CX_STR("\u03a3\u29b0"))
+ );
+
+ CxJsonValue *surrogatepair1 = cxJsonObjGet(obj, "surrogatepair1");
+ CX_TEST_ASSERT(cxJsonIsString(surrogatepair1));
+ CX_TEST_ASSERT(0 == cx_strcmp(
+ cxJsonAsCxString(surrogatepair1),
+ CX_STR("\xf0\x9f\xaf\xb5"))
+ );
+
+ CxJsonValue *surrogatepair2 = cxJsonObjGet(obj, "surrogatepair2");
+ CX_TEST_ASSERT(cxJsonIsString(surrogatepair2));
+ CX_TEST_ASSERT(0 == cx_strcmp(
+ cxJsonAsCxString(surrogatepair2),
+ CX_STR("test\xf0\x9f\xaf\xb1" "AA"))
+ );
+
+ CxJsonValue *mixed2 = cxJsonObjGet(obj, "mixed2");
+ CX_TEST_ASSERT(cxJsonIsString(mixed2));
+ CX_TEST_ASSERT(0 == cx_strcmp(
+ cxJsonAsCxString(mixed2),
+ CX_STR("123\u03a3\xf0\x9f\xaf\x85ß"))
+ );
+
+ cxJsonValueFree(obj);
+ }
+ cxJsonDestroy(&json);
+}
+
+CX_TEST(test_json_escaped_unicode_malformed) {
+ CxJson json;
+ cxJsonInit(&json, NULL);
+ CxJsonValue *obj;
+ CxJsonStatus result;
+ CX_TEST_DO {
+ cxJsonFill(&json, "\"too few \\u123 digits\"");
+ result = cxJsonNext(&json, &obj);
+ CX_TEST_ASSERT(result == CX_JSON_NO_ERROR);
+ CX_TEST_ASSERT(cxJsonIsString(obj));
+ CX_TEST_ASSERT(0 == cx_strcmp(
+ cxJsonAsCxString(obj),
+ CX_STR("too few \\u123 digits")
+ ));
+ cxJsonValueFree(obj);
+ cxJsonFill(&json, "\"too many \\u00E456 digits\"");
+ result = cxJsonNext(&json, &obj);
+ CX_TEST_ASSERT(result == CX_JSON_NO_ERROR);
+ CX_TEST_ASSERT(cxJsonIsString(obj));
+ CX_TEST_ASSERT(0 == cx_strcmp(
+ cxJsonAsCxString(obj),
+ CX_STR("too many ä56 digits")
+ ));
+ cxJsonValueFree(obj);
+ cxJsonFill(&json, "\"only high \\uD800 surrogate\"");
+ result = cxJsonNext(&json, &obj);
+ CX_TEST_ASSERT(result == CX_JSON_NO_ERROR);
+ CX_TEST_ASSERT(cxJsonIsString(obj));
+ CX_TEST_ASSERT(0 == cx_strcmp(
+ cxJsonAsCxString(obj),
+ CX_STR("only high \\uD800 surrogate")
+ ));
+ cxJsonValueFree(obj);
+ cxJsonFill(&json, "\"only low \\uDC00 surrogate\"");
+ result = cxJsonNext(&json, &obj);
+ CX_TEST_ASSERT(result == CX_JSON_NO_ERROR);
+ CX_TEST_ASSERT(cxJsonIsString(obj));
+ CX_TEST_ASSERT(0 == cx_strcmp(
+ cxJsonAsCxString(obj),
+ CX_STR("only low \\uDC00 surrogate")
+ ));
+ cxJsonValueFree(obj);
+ cxJsonFill(&json, "\"two high \\uD800\\uD800 surrogates\"");
+ result = cxJsonNext(&json, &obj);
+ CX_TEST_ASSERT(result == CX_JSON_NO_ERROR);
+ CX_TEST_ASSERT(cxJsonIsString(obj));
+ CX_TEST_ASSERT(0 == cx_strcmp(
+ cxJsonAsCxString(obj),
+ CX_STR("two high \\uD800\\uD800 surrogates")
+ ));
+ cxJsonValueFree(obj);
+ cxJsonFill(&json, "\"high plus bullshit \\uD800\\u567 foo\"");
+ result = cxJsonNext(&json, &obj);
+ CX_TEST_ASSERT(result == CX_JSON_NO_ERROR);
+ CX_TEST_ASSERT(cxJsonIsString(obj));
+ CX_TEST_ASSERT(0 == cx_strcmp(
+ cxJsonAsCxString(obj),
+ CX_STR("high plus bullshit \\uD800\\u567 foo")
+ ));
cxJsonValueFree(obj);
}
cxJsonDestroy(&json);
@@ -1042,6 +1189,8 @@
cx_test_register(suite, test_json_init_default);
cx_test_register(suite, test_json_simple_object);
cx_test_register(suite, test_json_escaped_strings);
+ cx_test_register(suite, test_json_escaped_unicode_strings);
+ cx_test_register(suite, test_json_escaped_unicode_malformed);
cx_test_register(suite, test_json_escaped_end_of_string);
cx_test_register(suite, test_json_object_incomplete_token);
cx_test_register(suite, test_json_token_wrongly_completed);