-
When using characters that aren't basic ASCII characters, you must save the source code file as UTF-8 or UTF-16 to prevent character corruption. The endianness of a UTF-16 source code file is irrelevant.
-
When in a source code file saved as UTF-8, use the narrow string constant " " when dealing with string or character literals. That ensures that no code units get chopped off or misinterpreted as UTF-16.
-
When in a source code file saved as UTF-8, the only time you should use a wide string constant L" " is when specifically creating a UTF-16 string literal using only 16-bit escape codes (\xhhhh).
-
When in a source code file saved as UTF-8, it is easier to simply assign a UTF-8 string constant to a
Utf16Stringobject than to create a UTF-16 string literal. The conversion to UTF-16 is done automatically. -
When in a source code file saved as UTF-16, use the wide string constant L" " when dealing with string or character literals. That ensures that no code units are cut in half and interpreted as UTF-8.
-
When in a source code file is saved as UTF-16, the only time you should use a narrow string constant " " is when specifically creating a UTF-8 string literal using only 8-bit escape codes (\xhh).
-
When a source code file is saved as UTF-16, it is easier to simply assign a UTF-16 string constant to a
Utf8Stringobject rather than create a UTF-8 string literal. The conversion to UTF-8 is done automatically.