String类里的compareToIngoreCase是这样的：

    public static final Comparator<String> CASE_INSENSITIVE_ORDER = new CaseInsensitiveComparator();

public int compareToIgnoreCase(String str)
{
        return CASE_INSENSITIVE_ORDER.compare(this, str);
}

private static class CaseInsensitiveComparator
                         implements Comparator<String>, java.io.Serializable {
// use serialVersionUID from JDK 1.2.2 for interoperability
private static final long serialVersionUID = 8575799808933029326L;

        public int compare(String s1, String s2) {
            int n1=s1.length(), n2=s2.length();
            for (int i1=0, i2=0; i1<n1 && i2<n2; i1++, i2++) {
                char c1 = s1.charAt(i1);
                char c2 = s2.charAt(i2);
                if (c1 != c2) {
                    c1 = Character.toUpperCase(c1);
                    c2 = Character.toUpperCase(c2);
                    if (c1 != c2) {
                        c1 = Character.toLowerCase(c1);
                        c2 = Character.toLowerCase(c2);
                        if (c1 != c2) {
                            return c1 - c2;
                        }
                    }
                }
            }
            return n1 - n2;
        }
    }

这是JDK源文件里的

比较的方法就是把两个String截成一个一个char的比较，每个char比较另一个String对应位置里的char，如果两个char都换成大写和都换成小写是一样的话，比较下一个，直到找到两个不一样的char，都转换成小写，然后return相差（ASCII里的值）。如果其中一个或者两个String都比较完了还没有同样的char的话，那就return两个String的长度差距。