Name: gm110360 Date: 02/11/2004


A DESCRIPTION OF THE REQUEST :
In the Unicode 4.0 support introduced as part of JSR204, the Character.isHighSurrogate and isLowSurrogate methods were introduced. But they accept char parameters. They should accept int parameters because a code point can refer to a surrogate.

JUSTIFICATION :
The justification for this is that codePointAt, codePointBefore, and codePointAfter all return int values. If there is a malformed surrogate pair there consisting of only half of a surrogate pair then these methods will return the actual surrogate value. The question then becomes how does one then detect this without hardcoding the surrogate values into our code. You would like to call isHighSurrogate or isLowSurrogate (shouldn't there also be a combined isSurrogate that returns true if it is a high or low surrogate) with the value you got from the codePointXxxx method. But you can't because you have an int value and they accept char.
(Incident Review ID: 238390)
======================================================================