databaseloader.java

MacroWeka扩展了著名数据挖掘工具weka

                    rs1.close();
                    break;
                case BOOL:
                    //System.err.println("boolean --> nominal");
                    attributeTypes[i - 1] = Attribute.NOMINAL;
                    m_nominalIndexes[i - 1] = new Hashtable();
                    m_nominalIndexes[i - 1].put("false", new Double(0));
                    m_nominalIndexes[i - 1].put("true", new Double(1));
                    m_nominalStrings[i - 1] = new FastVector();
                    m_nominalStrings[i - 1].addElement("false");
                    m_nominalStrings[i - 1].addElement("true");
                    break;
                case DOUBLE:
                    //System.err.println("BigDecimal --> numeric");
                    attributeTypes[i - 1] = Attribute.NUMERIC;
                    break;
                case BYTE:
                    //System.err.println("byte --> numeric");
                    attributeTypes[i - 1] = Attribute.NUMERIC;
                    break;
                case SHORT:
                    //System.err.println("short --> numeric");
                    attributeTypes[i - 1] = Attribute.NUMERIC;
                    break;
                case INTEGER:
                    //System.err.println("int --> numeric");
                    attributeTypes[i - 1] = Attribute.NUMERIC;
                    break;
                case LONG:
                    //System.err.println("long --> numeric");
                    attributeTypes[i - 1] = Attribute.NUMERIC;
                    break;
                case FLOAT:
                    //System.err.println("float --> numeric");
                    attributeTypes[i - 1] = Attribute.NUMERIC;
                    break;
                case DATE:
                    attributeTypes[i - 1] = Attribute.DATE;
                    break;
                default:
                    //System.err.println("Unknown column type");
                    attributeTypes[i - 1] = Attribute.STRING;
            }
        }
        FastVector attribInfo = new FastVector();
        for (int i = 0; i < numAttributes; i++) {
            /* Fix for databases that uppercase column names */
            //String attribName = attributeCaseFix(md.getColumnName(i + 1));
            String attribName = md.getColumnName(i + 1);
            switch (attributeTypes[i]) {
                case Attribute.NOMINAL:
                    attribInfo.addElement(new Attribute(attribName, m_nominalStrings[i]));
                    break;
                case Attribute.NUMERIC:
                    attribInfo.addElement(new Attribute(attribName));
                    break;
                case Attribute.STRING:
                    attribInfo.addElement(new Attribute(attribName, (FastVector)null));
                    break;
                case Attribute.DATE:
                    attribInfo.addElement(new Attribute(attribName, (String)null));
                    break;
                default:
                    throw new IOException("Unknown attribute type");
            }
        }
        m_structure = new Instances(endOfQuery(true), attribInfo,0);
        //get rid of m_idColumn
        if(m_DataBaseConnection.getUpperCase())
              m_idColumn = m_idColumn.toUpperCase();
        //System.out.println(m_structure.attribute(0).name().equals(idColumn));
        if(m_structure.attribute(0).name().equals(m_idColumn)){
            m_oldStructure = new Instances(m_structure,0);
            m_oldStructure.deleteAttributeAt(0);
            //System.out.println(m_structure);
        }
        else
            m_oldStructure = new Instances(m_structure,0);
    }
    else{
        if(m_oldStructure == null)
            m_oldStructure = new Instances(m_structure,0);
    }
    m_DataBaseConnection.disconnectFromDatabase();
    }
    catch(Exception ex) {
        ex.printStackTrace();
	printException(ex);
    } 
    return m_oldStructure;
    
  }
  
  

  /**
   * Return the full data set in batch mode (header and all intances at once).
   *
   * @return the structure of the data set as an empty set of Instances
   * @exception IOException if there is no source or parsing fails
   */
  public Instances getDataSet() throws IOException {

    if (m_DataBaseConnection == null) {
      throw new IOException("No source database has been specified");
    }
    if (getRetrieval() == INCREMENTAL) {
      throw new IOException("Cannot mix getting Instances in both incremental and batch modes");
    }
    setRetrieval(BATCH);
    connectToDatabase();
    
    
    Instances result = null;
    try{
    if (m_DataBaseConnection.execute(m_query) == false) 
      throw new Exception("Query didn't produce results");
    ResultSet rs = m_DataBaseConnection.getResultSet();
    ResultSetMetaData md = rs.getMetaData();
    // Determine structure of the instances
    int numAttributes = md.getColumnCount();
    int [] attributeTypes = new int [numAttributes];
    m_nominalIndexes = new Hashtable [numAttributes];
    m_nominalStrings = new FastVector [numAttributes];
    for (int i = 1; i <= numAttributes; i++) {
      switch (m_DataBaseConnection.translateDBColumnType(md.getColumnTypeName(i))) {
	
      case STRING :
        ResultSet rs1;
        String columnName = md.getColumnName(i);
        if(m_DataBaseConnection.getUpperCase())
            columnName = columnName.toUpperCase();
        String end = endOfQuery(false);
        m_nominalIndexes[i - 1] = new Hashtable();
        m_nominalStrings[i - 1] = new FastVector();
        if(m_DataBaseConnection.execute("SELECT DISTINCT ( "+columnName+" ) FROM "+ end) == false){
            throw new Exception("Nominal values cannot be retrieved");
        }
        rs1 = m_DataBaseConnection.getResultSet();
        attributeTypes[i - 1] = Attribute.NOMINAL;
        stringToNominal(rs1,i);
        rs1.close();  
	break;
      case BOOL:
	//System.err.println("boolean --> nominal");
	attributeTypes[i - 1] = Attribute.NOMINAL;
	m_nominalIndexes[i - 1] = new Hashtable();
	m_nominalIndexes[i - 1].put("false", new Double(0));
	m_nominalIndexes[i - 1].put("true", new Double(1));
	m_nominalStrings[i - 1] = new FastVector();
	m_nominalStrings[i - 1].addElement("false");
	m_nominalStrings[i - 1].addElement("true");
	break;
      case DOUBLE:
	//System.err.println("BigDecimal --> numeric");
	attributeTypes[i - 1] = Attribute.NUMERIC;
	break;
      case BYTE:
	//System.err.println("byte --> numeric");
	attributeTypes[i - 1] = Attribute.NUMERIC;
	break;
      case SHORT:
	//System.err.println("short --> numeric");
	attributeTypes[i - 1] = Attribute.NUMERIC;
	break;
      case INTEGER:
	//System.err.println("int --> numeric");
	attributeTypes[i - 1] = Attribute.NUMERIC;
	break;
      case LONG:
	//System.err.println("long --> numeric");
	attributeTypes[i - 1] = Attribute.NUMERIC;
	break;
      case FLOAT:
	//System.err.println("float --> numeric");
	attributeTypes[i - 1] = Attribute.NUMERIC;
	break;
      case DATE:
	attributeTypes[i - 1] = Attribute.DATE;
	break;
      default:
	//System.err.println("Unknown column type");
	attributeTypes[i - 1] = Attribute.STRING;
      }
    }

    // Step through the tuples
    //System.err.println("Creating instances...");
    FastVector instances = new FastVector();
    while(rs.next()) {
      double[] vals = new double[numAttributes];
      for(int i = 1; i <= numAttributes; i++) {
	switch (m_DataBaseConnection.translateDBColumnType(md.getColumnTypeName(i))) {
	case STRING :
	  String str = rs.getString(i);
	  
	  if (rs.wasNull()) {
	    vals[i - 1] = Instance.missingValue();
            } else {
                Double index = (Double)m_nominalIndexes[i - 1].get(str);
                if (index == null) {
                    index = new Double(m_structure.attribute(i-1).addStringValue(str));
                }
                vals[i - 1] = index.doubleValue();
            }
	  break;
	case BOOL:
	  boolean boo = rs.getBoolean(i);
	  if (rs.wasNull()) {
	    vals[i - 1] = Instance.missingValue();
	  } else {
	    vals[i - 1] = (boo ? 1.0 : 0.0);
	  }
	  break;
	case DOUBLE:
	  double dd = rs.getDouble(i);
	  if (rs.wasNull()) {
	    vals[i - 1] = Instance.missingValue();
	  } else {
	    vals[i - 1] =  dd;
	  }
	  break;
	case BYTE:
	  byte by = rs.getByte(i);
	  if (rs.wasNull()) {
	    vals[i - 1] = Instance.missingValue();
	  } else {
	    vals[i - 1] = (double)by;
	  }
	  break;
	case SHORT:
	  short sh = rs.getByte(i);
	  if (rs.wasNull()) {
	    vals[i - 1] = Instance.missingValue();
	  } else {
	    vals[i - 1] = (double)sh;
	  }
	  break;
	case INTEGER:
	  int in = rs.getInt(i);
	  if (rs.wasNull()) {
	    vals[i - 1] = Instance.missingValue();
	  } else {
	    vals[i - 1] = (double)in;
	  }
	  break;
	case LONG:
	  long lo = rs.getLong(i);
	  if (rs.wasNull()) {
	    vals[i - 1] = Instance.missingValue();
	  } else {
	    vals[i - 1] = (double)lo;
	  }
	  break;
	case FLOAT:
	  float fl = rs.getFloat(i);
	  if (rs.wasNull()) {
	    vals[i - 1] = Instance.missingValue();
	  } else {
	    vals[i - 1] = (double)fl;
	  }
	  break;
	case DATE:
          Date date = rs.getDate(i);
          if (rs.wasNull()) {
	    vals[i - 1] = Instance.missingValue();
	  } else {
            // TODO: Do a value check here.
            vals[i - 1] = (double)date.getTime();
          }
          break;
	default:
	  vals[i - 1] = Instance.missingValue();
	}
      }
      Instance newInst;
      newInst = new Instance(1.0, vals);
      instances.addElement(newInst);
    }   
    
    // Create the header and add the instances to the dataset
    //System.err.println("Creating header...");
    FastVector attribInfo = new FastVector();
    for (int i = 0; i < numAttributes; i++) {
      /* Fix for databases that uppercase column names */
      //String attribName = attributeCaseFix(md.getColumnName(i + 1));
      String attribName = md.getColumnName(i + 1);
      switch (attributeTypes[i]) {
      case Attribute.NOMINAL:
	attribInfo.addElement(new Attribute(attribName, m_nominalStrings[i]));
	break;
      case Attribute.NUMERIC:
	attribInfo.addElement(new Attribute(attribName));
	break;
      case Attribute.STRING:
	attribInfo.addElement(new Attribute(attribName, (FastVector)null));
	break;
      case Attribute.DATE:
	attribInfo.addElement(new Attribute(attribName, (String)null));
	break;
      default:
	throw new IOException("Unknown attribute type");
      }
    }
    result = new Instances(endOfQuery(true), attribInfo, 
				     instances.size());
    for (int i = 0; i < instances.size(); i++) {
      result.add((Instance)instances.elementAt(i));
    }
    rs.close();
    m_DataBaseConnection.disconnectFromDatabase();
    //get rid of m_idColumn
    if(m_DataBaseConnection.getUpperCase())
        m_idColumn = m_idColumn.toUpperCase();
    if(result.attribute(0).name().equals(m_idColumn)){
        result.deleteAttributeAt(0);
    }
    m_structure = new Instances(result,0);
    }
    catch(Exception ex) {
	printException(ex);
        StringBuffer text = new StringBuffer();
        if(m_query.equals("Select * from Results0")){
            text.append("\n\nDatabaseLoader options:\n");
            Enumeration enumi = listOptions();
            while (enumi.hasMoreElements()) {
                Option option = (Option)enumi.nextElement();
                text.append(option.synopsis()+'\n');
                text.append(option.description()+'\n');
            }
            System.out.println(text);
        }
    }
    //System.out.println(result);
    return result;