CJKV Information Processing, 2nd Edition

Unicode Encoding Methods

|

219

UTF-7 encoding exampleTable 4-21.

Encoding form String

Characters



UCS-2 6 3 8 5

UCS-2 bit arrays 1111 11111 111 1111

Six-bit segments 1111 111 111 11 111 1

UTF-7 2 62 4 4 4 77 67 2

UTF-7—visual +bLOMwg-

Because UTF-7 is a seven-bit, byte-based encoding whose code units are seven-bit bytes,

there is no need to explicitly indicate byte order. is means that the BOM can eectively

be removed when converting to UTF-7 encoding. Of course, the BOM should be reintro-

duced when converting to the UCS-2, UCS-4, UTF-16, or UTF-32 encoding forms.

UTF-7 encoding, by denition, interoperates only with UCS-2 and UTF-16 encodings,

and is applied in big-endian byte order. UTF-7 encoding can obviously be converted to

UTF-8, UCS-4, or UTF-32 encoding, given that their encoding spaces are a superset of

that of UTF-7 encoding. is means that if UTF-7–encoded data is encountered, it can be

converted to a currently supported Unicode encoding form. e UTF-16 Surrogates are

treated as though they were UCS-2–encoded, meaning no special treatment.

Also note how this example necessarily spans character boundaries. is is because Base64

transformation spans byte boundaries due to the splitting of multiple bytes into six-bit

segments. UTF-7 clearly makes it challenging to perform many common text-processing

tasks, such as character ...