package org.apache.lucene.document; /** * Licensed to the Apache Software Foundation (ASF) under one or more * contributor license agreements. See the NOTICE file distributed with * this work for additional information regarding copyright ownership. * The ASF licenses this file to You under the Apache License, Version 2.0 * (the "License"); you may not use this file except in compliance with * the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ import mx.org.pescador.CustomLoggerJ; import org.apache.lucene.index.IndexReader; import org.apache.lucene.search.Hits; import org.apache.lucene.search.Searcher; import java.util.*; // for javadoc /** Documents are the unit of indexing and search. * * A Document is a set of fields. Each field has a name and a textual value. * A field may be {@link Fieldable#isStored() stored} with the document, in which * case it is returned with search hits on the document. Thus each document * should typically contain one or more stored fields which uniquely identify * it. * *
Note that fields which are not {@link Fieldable#isStored() stored} are * not available in documents retrieved from the index, e.g. with {@link * Hits#doc(int)}, {@link Searcher#doc(int)} or {@link * IndexReader#document(int)}. */ public final class Document implements java.io.Serializable { List fields = new Vector(); private float boost = 1.0f; /** Constructs a new document with no fields. */ public Document() {} /** Sets a boost factor for hits on any field of this document. This value * will be multiplied into the score of all hits on this document. * *
Values are multiplied into the value of {@link Fieldable#getBoost()} of * each field in this document. Thus, this method in effect sets a default * boost for the fields of this document. * * @see Fieldable#setBoost(float) */ public void setBoost(float boost) { this.boost = boost; } /** Returns the boost factor for hits on any field of this document. * *
The default value is 1.0. * *
Note: This value is not stored directly with the document in the index. * Documents returned from {@link IndexReader#document(int)} and * {@link Hits#doc(int)} may thus not have the same value present as when * this document was indexed. * * @see #setBoost(float) */ public float getBoost() { return boost; } /** *
Adds a field to a document. Several fields may be added with * the same name. In this case, if the fields are indexed, their text is * treated as though appended for the purposes of search.
*Note that add like the removeField(s) methods only makes sense * prior to adding a document to an index. These methods cannot * be used to change the content of an existing index! In order to achieve this, * a document has to be deleted from an index and a new changed version of that * document has to be added.
*/ public final void add(Fieldable field) { fields.add(field); } /** *Removes field with the specified name from the document. * If multiple fields exist with this name, this method removes the first field that has been added. * If there is no field with the specified name, the document remains unchanged.
*Note that the removeField(s) methods like the add method only make sense * prior to adding a document to an index. These methods cannot * be used to change the content of an existing index! In order to achieve this, * a document has to be deleted from an index and a new changed version of that * document has to be added.
*/ public final void removeField(String name) { Iterator it = fields.iterator(); while (it.hasNext()) { Fieldable field = (Fieldable)it.next(); if (field.name().equals(name)) { it.remove(); return; } } } /** *Removes all fields with the given name from the document. * If there is no field with the specified name, the document remains unchanged.
*Note that the removeField(s) methods like the add method only make sense * prior to adding a document to an index. These methods cannot * be used to change the content of an existing index! In order to achieve this, * a document has to be deleted from an index and a new changed version of that * document has to be added.
*/ public final void removeFields(String name) { Iterator it = fields.iterator(); while (it.hasNext()) { Fieldable field = (Fieldable)it.next(); if (field.name().equals(name)) { it.remove(); } } } /** Returns a field with the given name if any exist in this document, or * null. If multiple fields exists with this name, this method returns the * first value added. * Do not use this method with lazy loaded fields. */ public final Field getField(String name) { for (int i = 0; i < fields.size(); i++) { Field field = (Field)fields.get(i); if (field.name().equals(name)) return field; } return null; } /** Returns a field with the given name if any exist in this document, or * null. If multiple fields exists with this name, this method returns the * first value added. */ public Fieldable getFieldable(String name) { for (int i = 0; i < fields.size(); i++) { Fieldable field = (Fieldable)fields.get(i); if (field.name().equals(name)) return field; } return null; } /** Returns the string value of the field with the given name if any exist in * this document, or null. If multiple fields exist with this name, this * method returns the first value added. If only binary fields with this name * exist, returns null. */ public final String get(String name) { for (int i = 0; i < fields.size(); i++) { Fieldable field = (Fieldable)fields.get(i); if (field.name().equals(name) && (!field.isBinary())) return field.stringValue(); } return null; } /** Returns an Enumeration of all the fields in a document. * @deprecated use {@link #getFields()} instead */ public final Enumeration fields() { return ((Vector)fields).elements(); } /** Returns a List of all the fields in a document. *Note that fields which are not {@link Fieldable#isStored() stored} are
* not available in documents retrieved from the index, e.g. with {@link
* Hits#doc(int)}, {@link Searcher#doc(int)} or {@link IndexReader#document(int)}.
*/
public final List getFields() {
return fields;
}
/**
* Returns an array of {@link Field}s with the given name.
* This method can return null.
* Do not use with lazy loaded fields.
*
* @param name the name of the field
* @return a Field[] array
*/
public final Field[] getFields(String name) {
List result = new ArrayList();
for (int i = 0; i < fields.size(); i++) {
Field field = (Field)fields.get(i);
if (field.name().equals(name)) {
result.add(field);
}
}
if (result.size() == 0)
return null;
return (Field[])result.toArray(new Field[result.size()]);
}
/**
* Returns an array of {@link Fieldable}s with the given name.
* This method can return null.
*
* @param name the name of the field
* @return a Fieldable[] array or null
*/
public Fieldable[] getFieldables(String name) {
List result = new ArrayList();
for (int i = 0; i < fields.size(); i++) {
Fieldable field = (Fieldable)fields.get(i);
if (field.name().equals(name)) {
result.add(field);
}
}
if (result.size() == 0)
return null;
return (Fieldable[])result.toArray(new Fieldable[result.size()]);
}
/**
* Returns an array of values of the field specified as the method parameter.
* This method can return null.
*
* @param name the name of the field
* @return a String[] of field values or null
*/
public final String[] getValues(String name) {
List result = new ArrayList();
for (int i = 0; i < fields.size(); i++) {
Fieldable field = (Fieldable)fields.get(i);
if (field.name().equals(name) && (!field.isBinary()))
result.add(field.stringValue());
}
if (result.size() == 0)
return null;
return (String[])result.toArray(new String[result.size()]);
}
/**
* Returns an array of byte arrays for of the fields that have the name specified
* as the method parameter. This method will return null if no
* binary fields with the specified name are available.
*
* @param name the name of the field
* @return a byte[][] of binary field values or null
*/
public final byte[][] getBinaryValues(String name) {
List result = new ArrayList();
for (int i = 0; i < fields.size(); i++) {
Fieldable field = (Fieldable)fields.get(i);
if (field.name().equals(name) && (field.isBinary()))
result.add(field.binaryValue());
}
if (result.size() == 0)
return null;
return (byte[][])result.toArray(new byte[result.size()][]);
}
/**
* Returns an array of bytes for the first (or only) field that has the name
* specified as the method parameter. This method will return null
* if no binary fields with the specified name are available.
* There may be non-binary fields with the same name.
*
* @param name the name of the field.
* @return a byte[] containing the binary field value or null
*/
public final byte[] getBinaryValue(String name) {
for (int i=0; i < fields.size(); i++) {
Fieldable field = (Fieldable)fields.get(i);
if (field.name().equals(name) && (field.isBinary()))
return field.binaryValue();
}
return null;
}
/** Prints the fields of a document for human consumption. */
public final String toString() {
StringBuffer buffer = new StringBuffer();
buffer.append("Document<");
for (int i = 0; i < fields.size(); i++) {
Fieldable field = (Fieldable)fields.get(i);
buffer.append(field.toString());
if (i != fields.size()-1)
buffer.append(" ");
}
buffer.append(">");
return buffer.toString();
}
}