keywordfinder.java

WAP ide 代码

    return attribArray.length;
  }

  /**
   * This method takes the raw data from a specially formatted file (wml.txt,
   * si.txt, or sl.txt) and then plucks out the special section of the raw data.
   * @param String data - the raw data from one of the specially formatted files.
   * the tags section of the raw data.
   * @return int containing the number of elements in the array.
   */
  private int generateSpecialArray(String data) {
    int specialSize = 0;

    // look for the start of the attribute section
    int startPos = data.indexOf("&");
    int endPos = data.length();

    // Get the whole section with verbose information
    String tempData = data.substring(startPos, endPos - 3);

    // Count the lines
    StringTokenizer st = new StringTokenizer(tempData, "\n");
    while (st.hasMoreTokens()) {
      specialSize++;
      st.nextToken();
    }

    // Create the string array
    int count = 0;
    specialArray = new String[specialSize];
    st = new StringTokenizer(tempData, "\n");
    while (st.hasMoreTokens()) {
      specialArray[count] = st.nextToken();
      specialArray[count] = specialArray[count].substring(0, specialArray[count].indexOf("="));
      count++;
    }
    return specialArray.length;
  }

  /** Get the required file name based on the document type.
   *  @param int docType - the type of document being created.
   *  @return String - the file name to be used, filename.txt.
   */
  private String getFileName(int docType) {
    switch (docType) {
      case 1:
        return "wml12.txt";
      case 2:
        return "si.txt";
      case 3:
        return "sl.txt";
    }
    return "";  // in case of an error or an unsupported document type, such as text, graphics, etc.
  }


  /**
   *  Compares words with those from the arrays and if it matches returns a special code.
   *  The code is as follows:
   *  0 - word not found.
   *  1 - a tag was found (<wml>.
   *  2 - an attribute was found (type="unknown").
   *  3 - a special tag was found (&123 or &nsbp).
   *  @param String word - the word being typed and looked at.
   *  @param int Type - the type of word searched for, (1) tag, (2) attribute, (3) special.
   *  @return int - one of the numbers listed above.
   */
  public int checkKeyword(String word, int type) {
    if (type == 1) {// look for the word in the tagArray
      for (int i = 0; i < tagCount; i++)
        if (word.equals(tagArray[i]))
          return 1;
    }
    if (type == 2) {// look for the word in the attribArray - found in tags
      for (int i = 0; i < attribCount; i++)
        if (word.equals(attribArray[i]))
          return 2;
    }
    if (type == 3) {// look for the word in the specialArray - found in text
      for (int i = 0; i < specialCount; i++)
        if (word.equals(specialArray[i]))
          return 3;
    }

    // Default return type
    return 0;
  }


  /**
   *  Searches the entire document for keywords, used when updating the document
   *  in part.
   *  @param String doc - the whole document.
   *  @param int offset - the strarting point of the insertion.
   *  @return Vector - A string object, comma seperated, containing the start position,
   *  end position, and the type of coloring required.
   */
  public Vector scanDocument(String doc) {

    Vector v = new Vector();

    v = scanTags(doc, v);
    v = scanSpecial(doc, v);
    v = scanComments(doc, v);
    return v;
  }

  /**
   * Searches for all the tags in the document, if it finds a tag it encodes it
   * accordingly.
   * @param String doc - the whole document string.
   * @param Vector v - the vector with position and color code information.
   * @return Vector - vector containing poition and color codes.
   */
  private Vector scanTags(String doc, Vector v) {

    int pos = -1;

    while (pos < doc.length()) {
      pos = doc.indexOf("<", pos);
      if (pos > -1) {
        int start = pos;
        int end = doc.indexOf("<", pos + 1);
        if (end == -1)
          end = doc.length();
        v = parseTag(start, end, doc, v);
      }
      else
        pos = doc.length();

      pos++;
    }

    return v;
  }

  /**
   * Helper function for the scanTags function.  Parses a tag into its respective colors.
   * @param int start - the start of the tag.
   * @param int end - the start of a new tag.
   * @param String doc - the whole document.
   * @param Vector v - the vector containing the formastting data.
   * @return Vector - the vector containing position and color coding data.
   */
  private Vector parseTag(int start, int end, String doc, Vector v) {

    // Clean up the tag and see where it ends.
    String theTag = doc.substring(start, end);
    int checkClose = theTag.indexOf(">");
    if (checkClose > -1)
      theTag = theTag.substring(0, checkClose + 1);
    if (theTag.indexOf("/>") > -1)
      theTag = theTag.substring(0, theTag.length() - 2);
    if (theTag.indexOf("?>") > -1)
      theTag = theTag.substring(0, theTag.length() - 2);
    if (theTag.indexOf(">") > -1)
      theTag = theTag.substring(0, theTag.length() - 1);

    // Now that it is cleaned up, identify the tag being referred too.
    StringTokenizer st = new StringTokenizer(theTag, " ");
    String tag = st.nextToken();
    if (tag.indexOf("<?") == 0)  //just a safety precaution
      tag = tag.substring(2);
    if (tag.indexOf("<!") == 0)  //just a safety precaution
      tag = tag.substring(2);
    if (tag.indexOf("</") == 0)  //just a safety precaution
      tag = tag.substring(2);
    if (tag.indexOf("<") == 0)  //just a safety precaution
      tag = tag.substring(1);

    // Color the tag accordingly, if it turns out to be a tag.
    int keyword = checkKeyword(tag, 1);
    if (keyword == 1) {
      v.add(new Integer(theTag.indexOf(tag) + start));
      v.add(new Integer(tag.length()));
      v.add(new Integer(1));
    }

    StringTokenizer attribString = new StringTokenizer(" ", " ");

    // Now check the attributes (use style 2(attrib) and 5(string))
    int pos = tag.length() + 1;
    while (st.hasMoreTokens()) {
      attribString = new StringTokenizer(st.nextToken() , "=");
      while (attribString.hasMoreTokens()) {
        String attrib = attribString.nextToken();
        if (attrib.indexOf("\"") == -1) { //check it, its an attribute
          keyword = checkKeyword(attrib, 2);
          if (keyword == 2) {
            v.add(new Integer(theTag.indexOf(attrib, pos) + start));
            v.add(new Integer(attrib.length()));
            v.add(new Integer(2));
          }
        }
        else { // its a comment code accordingly
            v.add(new Integer(theTag.indexOf(attrib, pos) + start));
            v.add(new Integer(attrib.length()));
            v.add(new Integer(5));
        }
        pos = theTag.indexOf(attrib, pos);
      }
    }

    return v;
  }

  /**
   * Searches for all the  special (&) tags in the document, if it finds a tag it encodes it
   * accordingly.
   * @param String doc - the whole document string.
   * @param Vector v - the vector with position and color code information.
   * @return Vector - vector containing poition and color codes.
   */
  private Vector scanSpecial(String doc, Vector v) {

    int pos = -1;

    while (pos < doc.length()) {
      pos = doc.indexOf("&", pos);
      if (pos > -1) {
        int start = pos;
        int end = doc.indexOf(";", pos + 1);
        if (end == -1)
          end = doc.length();
        else
          end++;
        String special = doc.substring(start, end);
        int keyword = checkKeyword(special, 3);
        if (keyword == 3) {
          v.add(new Integer(start));
          v.add(new Integer(special.length()));
          v.add(new Integer(3));
        }
      }
      else
        pos = doc.length();

      pos++;
    }

    return v;

  }


  /**
   * Searches for all the comment tags in the document, if it finds a tag it encodes it
   * accordingly.
   * @param String doc - the whole document string.
   * @param Vector v - the vector with position and color code information.
   * @return Vector - vector containing poition and color codes.
   */
  private Vector scanComments(String doc, Vector v) {

      int pos = -1;

    while (pos < doc.length()) {
      pos = doc.indexOf("<!--", pos);
      if (pos > -1) {
        int start = pos;
        int end = doc.indexOf("-->", pos + 1);
        if (end == -1)
          end = doc.length();
        else
          end = end + 3;
        String comment = doc.substring(start, end);
        v.add(new Integer(start));
        v.add(new Integer(comment.length()));
        v.add(new Integer(4));
      }
      else
        pos = doc.length();

      pos++;
    }

    return v;
  }

}