Question

I have `a string example = “this site holds all the examples from The Java Developers Almanac and more. Copy and paste these examples directly into your applications”

after token and do some i want on string example i have arraylist like :

ArrayList <token > arl = " "this site holds ", "holds all the examples ", "the examples from The Java Developers", " Copy and paste " )

“this site holds “, i know position start and end in string test : star = 1 end = 3
” holds all the examples “, i know position stat = 3 end = 6,
“the examples from The Java Developers”, i know position stat = 5 end =10,
“Copy and paste” i know position stat = 14 end = 17,

we can see,some element in arl overlaping :”this site holds “,”holds all the examples “,”the examples from The Java Developers”.

The problem here is how can i merge overlaping element to recived arraylist like

ArrayList result =”” this site holds all the examples from The Java Developers”,”” Copy and paste””;

Here my code : but it only merge fist elecment if check is element overloaping

public ArrayList<TextChunks> finalTextChunks(ArrayList<TextChunks> textchunkswithkeyword) {
        ArrayList<TextChunks > result = (ArrayList<TextChunks>) textchunkswithkeyword.clone();
            //System.out.print(result.size());
            int j;
            for(int i=0;i< result.size() ;i++) {
                int index = i;
                if(i+1>=result.size()){
                    break;
                }
                j=i+1;
                if(result.get(i).checkOverlapingTwoTextchunks(result.get(j))== true) {
                    TextChunks temp = new TextChunks();
                    temp = handleOverlaping(textchunkswithkeyword.get(i),textchunkswithkeyword.get(j),resultSearchEngine);
                    result.set(i, temp);
                    result.remove(j);
                    i = index;
                    continue;
            }
        }
        return result;  
    }
}

Thanks in avadce

Answer 1

The following should do it or at least illustrates an idea for merging the chunks. Basically I’m destroying the existing chunks and recreate new ones. Sounds horrible but simplifies a lot. I just store the words in a List and iterate over that word list to build new (merged!) chunks.

private List<TextChunks> finalTextChunks(List<TextChunks> textchunkswithkeyword) {

   private List<TextChunks> result = new ArrayList<TextChunk>();
   private List<String> wordList = new ArrayList<String>();

   // store all words in an arraylist, words are stored at their correct positions,
   // ignored words from the original text are represented by null entries
   for (TextChunks chunk : textchunkswithkeyword) {
     int start = chunk.getStartTextchunks();
     List<Token> tokens = chunk.getTokens(); // TODO - implement getTokens() in TextChunks class
     for (int i = 0; i < tokens.length; i++) {
        wordList.set(start+i, tokens.get(i).toString()); // TODO - overwrite toString() in Token class
     }
   }

   // recreate the chunks
   int start = 0;
   boolean isChunk = false;
   StringBuilder chunkBuilder;

   for (int i = 0; i < wordList.size(); i++) {
     String word = wordList.get(i);
     if (word == null) {
       if (isChunk) {
         // end of chunk detected
         TextChunk chunk = new TextChunk(chunkBuilder.toString().split(" "), start, i);
         result.add(chunk);
         isChunk = false;
       } else {
         // do nothing
       }
     } else {
       if (isChunk) {
         // chunk gets longer by one word
         chunkBuilder.append(" ").append(word);
       } else {
         // new chunk starts here
         chunkBuilder = new StringBuilder(word);
         start = i;
         isChunk = true;
       }
   }
   if (isChunk) {
     // create and add the last chunk
     TextChunks chunk = new TextChunk(chunkBuilder.toString(), start, wordList.size()-1);
     result.add(chunk);
   }
   return result;
}

(Warning – absolutely not tested, I have neither an IDE nor a compiler at hand)

EDIT

changed the code – you said, that the TextChunk class holds a token (words?) array. It was just three simple modifications.

EDIT 2

Final edit – I partially adapted my code to your classes. What you need to do:

1. implement a getTokens() method in TextChunks that simply returns the arrt field
2. implement a TextChunks constructor that takes a String (with space-separated words), the start and the end. Your Token class already provides a static method to convert the String in an arraylist of tokens
3. overwrite toString() method in Token class so that simply returns the token String.