CJKV Information Processing, 2nd Edition

Noncoded Versus Coded Character Sets

|

181

SUBstitution) feature is an example of a well-established glyph substitution implementa-

tion that is used by a wide variety of applications that support vertical writing.

Although the JIS standards themselves do not directly encode vertical variants, some

vendor character sets that are based on them, in particular JIS X 0208:1997, do so.

Vertical forms of characters intended for horizontal writing, when compared to their par-

ent (horizontal) forms, may rotate 90˚ clockwise, rotate 90˚ clockwise then ip or be mir-

rored, change their form completely, or reposition. Parentheses and other bracket-like

characters, including arrows, rotate 90˚ clockwise. e wave dash (31) and Japanese

long vowel mark (3) are examples of characters that rotate then ip or are mirrored.

Periods and commas, along the small kana that are specic to Japanese, reposition to the

upper-right corner.

Interestingly, Unicode directly encodes some, but not all, vertical variants by virtue of the

fact that some national character set standards encode them. is is merely for code-point

compatibility. e corresponding Unicode code points and code point ranges are 1

through 19, 3 through 44, 47, and 48.

Some characters deserve special mention because they are used strictly for vertical writ-

ing, and thus do not have a corresponding ...