@Minlen(1) on StringTokenizer.nextToken

[panacekcz]

StringTokenizer.nextToken always returns non-empty strings, so its return type could be @MinLen(1).

[panacekcz]

Currently, the following code causes a upper bound checker warning, although it is safe:

import java.util.StringTokenizer;
class StringTokenizerMinLen {
 void test(String str, String delim, boolean returnDelims) {
  StringTokenizer st = new StringTokenizer(str, delim, returnDelims);
  while (st.hasMoreTokens()) {
   char c = st.nextToken().charAt(0);
  }
 }
}

[panacekcz]

Because of kelloggm#165, char c = st.nextToken().charAt(0); will not work even after this annotation is added. However, String token = st.nextToken(); char c = token.charAt(0); will work.