package LBJ2.nlp;

import LBJ2.parse.LinkedVector;


/**
  * Use this class to represent a natural language document.
  * <code>SentenceSplitter</code> and <code>Sentence.wordSplit()</code> are
  * used to represent the text of the document internally as a collection of
  * vectors of words.  As such, the text of the document is assumed plain,
  * i.e. there should not be any mark-up.
  *
  * @author Nick Rizzolo
 **/
public class NLDocument extends LinkedVector
{
  /** The name of the file this document came from. */
  private String fileName;


  /**
    * This constructor takes the entire text of the document in a String array
    * as input and initializes the representation.
    *
    * @param text The entire text of the document.  Each element of this array
    *             should represent a line of input without any line
    *             termination characters.
   **/
  public NLDocument(String[] text) { this(null, text); }


  /**
    * This constructor takes the entire text of the document in a String array
    * as input and initializes the representation.
    *
    * @param p    The previous child in the parent vector.
    * @param text The entire text of the document.  Each element of this array
    *             should represent a line of input without any line
    *             termination characters.
   **/
  public NLDocument(NLDocument p, String[] text) {
    super(p);
    addAll(new SentenceSplitter(text));
  }


  /**
    * Creates a document from the contents of the named file.
    *
    * @param file The name of the file containing a natural language, plain
    *             text document.
   **/
  public NLDocument(String file) { this(null, file); }


  /**
    * Creates a document from the contents of the named file.
    *
    * @param p    The previous child in the parent vector.
    * @param file The name of the file containing a natural language, plain
    *        text document.
   **/
  public NLDocument(NLDocument p, String file) {
    super(p);
    fileName = file;
    addAll(new SentenceSplitter(file));
  }


  /**
    * Returns the name of the file this document came from, or
    * <code>null</code> if one was not specified.
   **/
  public String getFileName() { return fileName; }


  /**
    * Adds all the sentences that come from the argument sentence splitter to
    * this document after using a word splitter to chop them up.
    *
    * @param splitter A sentence splitter.
   **/
  public void addAll(SentenceSplitter splitter) {
    Sentence[] rawSentences = splitter.splitAll();
    for (int i = 0; i < rawSentences.length; ++i)
      add(rawSentences[i].wordSplit());
  }
}