Java StringTokenizer奇怪的行为

Question

I'm trying to extract only lowercase alphanumerical characters from a document with this:

String delim = "abcdefghijklmnopqrstuvwxyz0123456789";

StringTokenizer strtok = new StringTokenizer(str, delim, true);

String newstr = "";

while (strtok.hasMoreTokens()) {
    newstr = newstr + strtok.nextToken();
}

return newstr;

Note that the document is already lowercase only. But for some reason all of the punctuation characters are still being returned along with parethesis and /'s, etc.

I thought using the true boolean in the creation of the tokenizer would count delimiters as tokens?

Answer 1

The delim argument is a delimiter. You're basically asking for each token to be "whatever is between lower case letters". Then the 'true' argument says "give me those letters on the edges too". Were you looking for replaceAll("[^abcdefghijklmnopqrstuvwxyz0123456789]","") ?