Regular expression - splitting a string

Question

I have a long string that I'm trying to split into a series of substrings. I would like each of the substrings to start with "TTL.. I'm fairly certain that I'm missing something very basic here. I've attached my code which yield NO GROUPS. I didn't see another method for returning the text that the regular expression matched.
String finalLongstring="TTL1,clip1+TTL2+clip3,TTL4,clip4,TTL5,clip5+TTL6+"+
   "clip6+TTL7+clip7,TTL8,clip8,TTL9,clip9,TTL10,clip10,TTL11,clip11,TTL12,clip12,"+
   "TTL13,clip13+TTL14+clip14,TTL15,clip15,TTL16,clip16,TTL17,clip17,"+
   "TTL18,clip18,TTL19,clip19,TTL20,clip20,TTL21,clip21,TTL22,clip22,"+
   "TTL23,clip23,TTL24,clip24,TTL25,clip25,TTL26,clip26,TTL27,clip27,"+
   "TTL28,clip28,TTL29,clip29"
List<String> chapters = new ArrayList<String>();
          chapters.clear();
          Pattern chapter=null;
          chapter=Pattern.compile("(TTL\\d+([+,]|clip\\d+)*)");
          //                      ||    |  |  | |  |    | |
          //                      ||    |  |  | |  |    | Repeat (commas pluses and clips group) 0 or more times
          //                      ||    |  |  | |  |    one or more digits following 'clip'
          //                      ||    |  |  | |  clip
          //                      ||    |  |  | or..
          //                      ||    |  |  plus or comma symbols
          //                      ||    |  group the +, and clip information together
          //                      ||    one or more digits
          //                      |Match clips starting with TTL
          //                      |
          Matcher cp = chapter.matcher(finalLongstring);  //NO MATCHES!!
          String [] temp = chapter.split(finalLongstring);  //temp =EMPTY STRING ARRAY
          do{
               String chapterPlus=cp.group(1);
               if(cp.hitEnd()){break;}
               chapters.add(chapterPlus);
          }while(true);Thanks in advance for the help.
Icesurfer

Answer 1

The main reason your matcher didn't work is because you never told it to do anything. You have to call one of the methods matches(), find() or lookingAt(), and make sure it returns true, before you can use the group() methods. When I did that, your regex worked, but then I modified it to demonstrate a better use of capturing groups, as shown here: import java.util.regex.*;
public class Test
  public static void main(String... args)
    String str="TTL1,clip1+TTL2+clip3,TTL4,clip4,TTL5,clip5+TTL6+clip6+"+
       "TTL7+clip7,TTL8,clip8,TTL9,clip9,TTL10,clip10,TTL11,clip11,TTL12,clip12,"+
       "TTL13,clip13+TTL14+clip14,TTL15,clip15,TTL16,clip16,TTL17,clip17,"+
       "TTL18,clip18,TTL19,clip19,TTL20,clip20,TTL21,clip21,TTL22,clip22,"+
       "TTL23,clip23,TTL24,clip24,TTL25,clip25,TTL26,clip26,TTL27,clip27,"+
       "TTL28,clip28,TTL29,clip29";
    Pattern p = Pattern.compile("(TTL\\d+)[+,](clip\\d+)[+,]");
    Matcher m = p.matcher(str);
    while (m.find())
      System.out.printf("%6s %s%n", m.group(1), m.group(2));
}The reason your split() attempt didn't work is because the regex matched all of the text; the split() regex is supposed to match the parts you don't want. In fact, it did split the text, creating a list of empty strings, but then it threw them all away, because split() discards trailing empty fields by default.
Finally, the hitEnd() method is not appropriate in this context. It and the requireEnd() method were added to support the Scanner class in JDK 1.5. If you want to see how they work, look at the source code for Scanner, but for now, just classify them as an advanced topic. When you're iterating through text with the find() method, you stop when find() returns false, plain and simple.