com.jaeksoft.searchlib.crawler.FieldMap 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.
The newest version!
/**
* License Agreement for OpenSearchServer
*
* Copyright (C) 2008-2014 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 {
final private char concatSeparator;
public FieldMap() {
concatSeparator = '|';
}
public FieldMap(String multilineText, char fieldSeparator, char concatSeparator) throws IOException {
StringReader sr = null;
BufferedReader br = null;
this.concatSeparator = concatSeparator;
try {
sr = new StringReader(multilineText);
br = new BufferedReader(sr);
String line;
while ((line = br.readLine()) != null) {
String[] cols = StringUtils.split(line, fieldSeparator);
if (cols == null || cols.length < 2)
continue;
String source = cols[0];
String target = cols[1];
String analyzer = cols.length > 2 ? cols[2] : null;
Float boost = cols.length > 3 ? Float.parseFloat(cols[3]) : null;
add(new SourceField(source, concatSeparator), new TargetField(target, analyzer, boost, null));
}
} finally {
IOUtils.close(br, sr);
}
}
public FieldMap(File file)
throws XPathExpressionException, ParserConfigurationException, SAXException, IOException {
super(file, "/map");
concatSeparator = '|';
}
public FieldMap(Node node) throws XPathExpressionException {
super(node);
concatSeparator = '|';
}
@Override
protected TargetField loadTarget(String targetName, Node node) {
return new TargetField(targetName, node);
}
@Override
protected SourceField loadSource(String sourceName) {
return new SourceField(sourceName, concatSeparator);
}
@Override
protected void writeTarget(XmlWriter xmlWriter, TargetField target) throws SAXException {
}
private void addFieldContent(FieldContent fc, TargetField targetField, IndexDocument target) throws IOException {
if (fc == null)
return;
targetField.addFieldValueItems(fc.getValues(), target);
}
public void mapIndexDocument(IndexDocument source, IndexDocument target) throws IOException {
for (GenericLink link : getList()) {
SourceField sourceField = link.getSource();
if (sourceField.isUnique()) {
FieldContent fc = sourceField.getUniqueString(source);
if (fc == null)
fc = sourceField.getUniqueString(target);
addFieldContent(fc, link.getTarget(), target);
} else {
String value = sourceField.getConcatString(source, target);
link.getTarget().addValue(value, target);
}
}
}
public void mapIndexDocument(ResultDocument source, IndexDocument target) throws IOException {
for (GenericLink link : getList()) {
SourceField sourceField = link.getSource();
if (sourceField.isUnique()) {
List fvi = sourceField.getUniqueString(source);
link.getTarget().addFieldValueItems(fvi, target);
} else {
String value = sourceField.getConcatString(source, target);
link.getTarget().addValue(value, target);
}
}
}
public void mapIndexDocumentJson(String target, ResultDocument sourceDocument, IndexDocument targetDocument)
throws JsonProcessingException {
Map> map = new TreeMap>();
for (GenericLink link : getList()) {
TargetField targetField = link.getTarget();
if (!target.equals(targetField.getName()))
continue;
SourceField sourceField = link.getSource();
String source = sourceField.getUniqueName();
FieldValue fieldValue = sourceDocument.getReturnFields().get(source);
if (fieldValue == null)
continue;
if (fieldValue.getValuesCount() == 0)
continue;
List list = map.get(source);
if (list == null) {
list = new ArrayList(fieldValue.getValuesCount());
map.put(source, list);
}
fieldValue.populate(list);
}
String json = JsonUtils.toJsonString(map);
targetDocument.setString(target, json);
}
public void cacheAnalyzers(AnalyzerList analyzerList, LanguageEnum lang) throws SearchLibException {
for (GenericLink link : getList()) {
TargetField target = link.getTarget();
if (target == null)
throw new SearchLibException("No target field for " + link.getSource());
target.setCachedAnalyzer(analyzerList, lang);
}
}
public boolean contains(String source, String target) {
if (source == null || target == null)
return false;
for (GenericLink link : getList())
if (target.equals(link.getTarget().getName()))
if (link.getSource().contains(source))
return true;
return false;
}
/**
* Return a map with all different boost for one target name
*
* @param target
* the target field
* @param boostMap
* a boost map
*/
public void populateBoosts(String target, Map boostMap) {
if (target == null || boostMap == null)
return;
for (GenericLink link : getList()) {
TargetField targetField = link.getTarget();
if (target.equals(targetField.getName())) {
Float boost = targetField.getBoost();
if (boost != null && boost != 1.0F)
boostMap.put(boost, targetField);
}
}
}
public boolean isMapped(String target) {
if (target == null)
return false;
for (GenericLink link : getList())
if (target.equals(link.getTarget().getName()))
return true;
return false;
}
public boolean isMapped(String source, String target) {
if (source == null || target == null)
return false;
for (GenericLink link : getList())
if (target.equals(link.getTarget().getName()))
if (link.getSource().contains(source))
return true;
return false;
}
}