
com.jaeksoft.searchlib.parser.ParserFieldTarget Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of opensearchserver Show documentation
Show all versions of opensearchserver Show documentation
OpenSearchServer is a powerful, enterprise-class, search engine program. Using the web user interface,
the crawlers (web, file, database, ...) and the REST/RESTFul API you will be able to integrate quickly and
easily advanced full-text search capabilities in your application. OpenSearchServer runs on Windows and
Linux/Unix/BSD.
/**
* License Agreement for OpenSearchServer
*
* Copyright (C) 2012-2014 Emmanuel Keller / Jaeksoft
*
* http://www.open-search-server.com
*
* This file is part of OpenSearchServer.
*
* OpenSearchServer is free software: you can redistribute it and/or
* modify it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* OpenSearchServer is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with OpenSearchServer.
* If not, see .
**/
package com.jaeksoft.searchlib.parser;
import java.io.IOException;
import java.util.ArrayList;
import java.util.List;
import java.util.regex.Pattern;
import org.apache.commons.lang.StringEscapeUtils;
import org.w3c.dom.Node;
import org.xml.sax.SAXException;
import com.jaeksoft.searchlib.index.IndexDocument;
import com.jaeksoft.searchlib.schema.FieldValueItem;
import com.jaeksoft.searchlib.util.DomUtils;
import com.jaeksoft.searchlib.util.RegExpUtils;
import com.jaeksoft.searchlib.util.StringUtils;
import com.jaeksoft.searchlib.util.XmlWriter;
import com.jaeksoft.searchlib.util.map.TargetField;
public class ParserFieldTarget extends TargetField {
private String captureRegexp;
private boolean removeTag;
private boolean convertHtmlEntities;
private Pattern captureRegexpPattern;
public ParserFieldTarget(String name, String captureRegexp,
String analyzer, boolean removeTag) {
super(name, analyzer, null, null);
this.captureRegexp = captureRegexp;
this.removeTag = removeTag;
this.checkRegexpPattern();
}
public ParserFieldTarget(String name, Node node) {
super(name, node);
List nl = DomUtils.getNodes(node, "captureRegexp");
if (nl.size() > 0)
captureRegexp = StringEscapeUtils.unescapeXml(nl.get(0)
.getTextContent());
nl = DomUtils.getNodes(node, "removeTag");
removeTag = nl != null && nl.size() > 0;
nl = DomUtils.getNodes(node, "convertHtmlEntities");
convertHtmlEntities = nl != null && nl.size() > 0;
checkRegexpPattern();
}
public void writeXml(XmlWriter xmlWriter) throws SAXException {
if (captureRegexp != null) {
xmlWriter.startElement("captureRegexp");
xmlWriter.textNode(StringEscapeUtils.escapeXml(captureRegexp));
xmlWriter.endElement();
}
if (removeTag) {
xmlWriter.startElement("removeTag");
xmlWriter.endElement();
}
if (convertHtmlEntities) {
xmlWriter.startElement("convertHtmlEntities");
xmlWriter.endElement();
}
}
public final boolean hasRegexpPattern() {
return (captureRegexp != null);
}
private void checkRegexpPattern() {
if (captureRegexp != null)
if (captureRegexp.trim().length() == 0)
captureRegexp = null;
captureRegexpPattern = captureRegexp == null ? null : Pattern
.compile(captureRegexp);
}
/**
* @return the captureRegexp
*/
public String getCaptureRegexp() {
return captureRegexp;
}
/**
* @param replaceRegexpTag
* the replaceRegexpTag to set
*/
public void setCaptureRegexp(String captureRegexp) {
this.captureRegexp = captureRegexp;
}
@Override
final public void addFieldValueItems(List fieldValueItems,
IndexDocument targetDocument) throws IOException {
if (fieldValueItems == null)
return;
List values = new ArrayList(0);
if (captureRegexpPattern == null) {
for (FieldValueItem fieldValueItem : fieldValueItems)
values.add(fieldValueItem.getValue());
} else {
synchronized (captureRegexpPattern) {
for (FieldValueItem fieldValueItem : fieldValueItems)
for (String value : RegExpUtils.getGroups(
captureRegexpPattern, fieldValueItem.getValue()))
values.add(value);
}
}
if (removeTag || convertHtmlEntities) {
int pos = 0;
for (String value : values) {
if (removeTag)
value = StringUtils.removeTag(value);
if (convertHtmlEntities)
value = StringEscapeUtils.unescapeHtml(value);
values.set(pos++, value);
}
}
addValues(values, targetDocument);
}
/**
* @return the removeTag
*/
public boolean isRemoveTag() {
return removeTag;
}
/**
* @param removeTag
* the removeTag to set
*/
public void setRemoveTag(boolean removeTag) {
this.removeTag = removeTag;
}
/**
* @param convertHtmlEntities
* the convertHtmlEntities to set
*/
public void setConvertHtmlEntities(boolean convertHtmlEntities) {
this.convertHtmlEntities = convertHtmlEntities;
}
/**
* @return the convertHtmlEntities
*/
public boolean isConvertHtmlEntities() {
return convertHtmlEntities;
}
}
© 2015 - 2025 Weber Informatics LLC | Privacy Policy