我们在创建lucene索引的时候经常是从数据库取出相关的记录封装成一个JavaBean,然后将JavaBean的相关字段分别再创建索引.如果JavaBean增加和删除一个字段的话,我们必须修改我们创建索引的程序对应的增加和删除索引字段.如果我们索引创建分布不同程序中,这样修改就比较麻烦,下面是我运用Annotation写的一个lucene 创建索引工具类.希望对学习Annotation和创建索引有用.
索引字段声明注解类,可以扩展此类来增加自己想要的相关属性
//工具类--解析注解。生成索引字段
索引对应javabean定义
创建索引
索引字段声明注解类,可以扩展此类来增加自己想要的相关属性
@Retention(RetentionPolicy.RUNTIME)
@Target( { ElementType.FIELD })
public @interface IndexAnnotation {
//存储
public boolean store() default false;
//分词
public boolean analyse() default false;
//删除HTML代码
public boolean parseHtml() default true;
//权重分
public float boost() default 10;
}
//工具类--解析注解。生成索引字段
@SuppressWarnings("unchecked")
public final class IndexDocumentUtils {
private final static Logger log = LoggerFactory
.getLogger(IndexDocumentUtils.class);
/**
* 创建索引
*
* @param idataIndex
* @return
*/
public static Document createDocument(IdataIndex idataIndex) {
Class clzss = idataIndex.getClass();
Document doc = new Document();
Field[] fields = clzss.getDeclaredFields();
for (Field field : fields) {
if (field.getName().equals("serialVersionUID"))
continue;
String value = getFieldValue(idataIndex, field.getName());
org.apache.lucene.document.Field indexField = new org.apache.lucene.document.Field(
field.getName(), value, getStore(idataIndex, field
.getName()), getIndex(idataIndex, field.getName()));
//设置权重值
indexField.setBoost(getBoost(idataIndex, field.getName()));
doc.add(indexField);
}
return doc;
}
/**
* 通过反射获取字段值
*
* @param idataIndex
* @param fieldName
* @return
*/
private static Pattern tagPattern = Pattern
.compile("<.*?>", Pattern.DOTALL);
private static String getFieldValue(IdataIndex idataIndex, String fieldName) {
try {
boolean isMatcher = false;
String value = StringUtil.defaultIfEmpty(BeanUtils.getProperty(
idataIndex, fieldName));
StringBuffer sb = new StringBuffer();
if (isParseHtml(idataIndex, fieldName)) {// 是否解析html内容
if (StringUtils.isNotEmpty(value)) {
Matcher matcher = tagPattern.matcher(value);
while (matcher.find()) {
isMatcher = true;
matcher.appendReplacement(sb, "");
}
matcher.appendTail(sb);
} else {
return "";
}
}
return isMatcher ? sb.toString() : value;
} catch (Exception e) {
log.error(e);
return "";
}
}
/**
* 返回索引字段是否存储
*
* @param idataIndex
* @param fieldName
* @return
*/
private static Store getStore(IdataIndex idataIndex, String fieldName) {
Class clzss = idataIndex.getClass();
try {
Field field = clzss.getDeclaredField(fieldName);
IndexAnnotation ia = field.getAnnotation(IndexAnnotation.class);
if (ia != null) {// 检查注解的值
if (ia.store()) {
return org.apache.lucene.document.Field.Store.YES;
}
}
} catch (Exception e) {
log.error(e);
}
return org.apache.lucene.document.Field.Store.NO;
}
/**
* 返回索引字段是否索引
*
* @param idataIndex
* @param fieldName
* @return
*/
private static Index getIndex(IdataIndex idataIndex, String fieldName) {
Class clzss = idataIndex.getClass();
try {
Field field = clzss.getDeclaredField(fieldName);
IndexAnnotation ia = field.getAnnotation(IndexAnnotation.class);
if (ia != null) {// 检查注解的值
if (ia.analyse()) {
return org.apache.lucene.document.Field.Index.ANALYZED;
}
}
} catch (Exception e) {
log.error(e);
}
return org.apache.lucene.document.Field.Index.ANALYZED;
}
/**
* 返回索引字段是否解析HTML
*
* @param idataIndex
* @param fieldName
* @return
*/
private static boolean isParseHtml(IdataIndex idataIndex, String fieldName) {
Class clzss = idataIndex.getClass();
try {
Field field = clzss.getDeclaredField(fieldName);
IndexAnnotation ia = field.getAnnotation(IndexAnnotation.class);
if (ia != null) {// 检查注解的值
return ia.parseHtml();
}
} catch (Exception e) {
log.error(e);
}
return true;
}
/**
* 返回权重值
* @param idataIndex
* @param fieldName
* @return
*/
private static float getBoost(IdataIndex idataIndex, String fieldName) {
Class clzss = idataIndex.getClass();
try {
Field field = clzss.getDeclaredField(fieldName);
IndexAnnotation ia = field.getAnnotation(IndexAnnotation.class);
if (ia != null) {// 检查注解的值
return ia.boost();
}
} catch (Exception e) {
log.error(e);
}
return 10;
}
}
索引对应javabean定义
public class SearchIndex implements Serializable {
/**
*
*/
private static final long serialVersionUID = 153648837940506749L;
//索引编号
@IndexAnnotation(store = true)
private String id;
//资源ID
@IndexAnnotation(store = true)
private String resourceId;
//标题
@IndexAnnotation(store = true, analyse = true,boost=100)
private String title;
//索引内容说明
@IndexAnnotation(store = true, analyse = true,boost=50)
private String content;
public String getId() {
return id;
}
public void setId(String id) {
this.id = id;
}
public String getResourceId() {
return resourceId;
}
public void setResourceId(String resourceId) {
this.resourceId = resourceId;
}
public String getTitle() {
return title;
}
public void setTitle(String title) {
this.title = title;
}
public String getContent() {
return content;
}
public void setContent(String content) {
this.content = content;
}
public String toString() {
return ToStringBuilder.reflectionToString(this, ToStringStyle.MULTI_LINE_STYLE);
}
}
创建索引
fsWriter.addDocument(IndexDocumentUtils.createDocument(searchIndex));