UnicodeString is an encoding-neutral class intended to be used when the
encoding of a string will be only be known at runtime. Typical situations are developing an encoding-neutral API or
reading from a text file that may be formatted as either UTF-8 or UTF-16. In such situations, UnicodeString is
the class to use.
UnicodeString, however, is not intended as a general-use string class to be
used all throughout your application. UnicodeString is really a wrapper class
that contains an instance of Utf8String orUtf16String, depending on the conditions at
runtime. As such, it is not as efficient as Utf8String or Utf16String,
which contain more efficient encoding-specific logic. UnicodeString also
doesn't offer quite as much functionality as Utf8String or
Utf16String. Code units, for example cannot cannot be directly retrieved from
a UnicodeString and UnicodeString functions do not accept string constants.
UnicodeString does has string encoding-neutral manipulation functions that
can be used (appending, removing, substrings, etc.), but its primary purpose is to enable easy
conversion to an encoding-specific format. As such, it can be easily cast to a UtfString
encoding-specific class.
//This will give us the same string in UTF-8 and UTF-16 encodings
UnicodeString unicodeString = Utf8String("I'm a string!");
Utf8String utf8String = (Utf8String)unicodeString;
Utf16String utf16String = (Utf16String)unicodeString;UnicodeString was designed to work with encoding-specific UtfString classes
such as Utf8String and Utf16String,
and will not work with string constants or classes like std::string or std::wstring.
//Will compile
UnicodeString unicodeString;
unicodeString = Utf8String("I'm a string");
unicodeString = Utf16String(L"I'm a wide string");
//Won't compile
unicodeString = "I'm a string";
unicodeString = L"I'm a wide string";
Due to UnicodeString being less efficient and less flexible than the
encoding-specific classes, it is best to use encoding-specific classes whenever practical. It would be wise to
pick one particular encoding to be used internally in the application or library you are developing, and assign
the UnicodeString instance to a Utf8String or Utf16String instance as soon as possible. When in doubt, go for UTF-8. It has the advantage of being very
memory-efficient with standard ASCII characters and is backward-compatible with code written for ASCII characters.
Most software uses UTF-8 as the internal encoding.