com.jaeksoft.searchlib.crawler.FieldMapGeneric 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) 2010-2013 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
extends GenericMap {
private File mapFile;
protected FieldMapGeneric() {
mapFile = null;
}
protected FieldMapGeneric(Node parentNode) throws XPathExpressionException {
mapFile = null;
load(parentNode);
}
protected FieldMapGeneric(File mapFile, String rootXPath)
throws ParserConfigurationException, SAXException, IOException,
XPathExpressionException {
this.mapFile = mapFile;
if (!mapFile.exists())
return;
XPathParser xpp = new XPathParser(mapFile);
load(xpp.getNode(rootXPath));
}
protected abstract T loadTarget(String targetName, Node node);
protected abstract S loadSource(String source);
public void load(Node parentNode) throws XPathExpressionException {
synchronized (this) {
if (parentNode == null)
return;
List nodeList = DomUtils.getNodes(parentNode, "link");
for (Node node : nodeList) {
String sourceName = StringEscapeUtils.unescapeXml(DomUtils
.getAttributeText(node, "source"));
S source = loadSource(sourceName);
if (source == null)
continue;
String targetName = DomUtils.getAttributeText(node, "target");
T target = loadTarget(targetName, node);
if (target == null)
continue;
add(source, target);
}
}
}
protected abstract void writeTarget(XmlWriter xmlWriter, T target)
throws SAXException;
public void store(XmlWriter xmlWriter) throws SAXException {
for (GenericLink link : getList()) {
T target = link.getTarget();
xmlWriter.startElement(
"link",
"source",
link.getSource().toXmlAttribute(),
"target",
target.toXmlAttribute(),
"analyzer",
target.getAnalyzer(),
"boost",
target.getBoost() == null ? null : Float.toString(target
.getBoost()));
writeTarget(xmlWriter, link.getTarget());
xmlWriter.endElement();
}
}
public void store() throws TransformerConfigurationException, SAXException,
IOException {
synchronized (this) {
if (!mapFile.exists())
mapFile.createNewFile();
PrintWriter pw = new PrintWriter(mapFile);
try {
XmlWriter xmlWriter = new XmlWriter(pw, "UTF-8");
xmlWriter.startElement("map");
store(xmlWriter);
xmlWriter.endElement();
xmlWriter.endDocument();
} finally {
pw.close();
}
}
}
final protected void mapFieldTarget(WebCrawlMaster webCrawlMaster,
ParserSelector parserSelector, LanguageEnum lang,
CommonFieldTarget dfTarget, String content, IndexDocument target)
throws SearchLibException, IOException, ParseException,
SyntaxError, URISyntaxException, ClassNotFoundException,
InterruptedException, InstantiationException,
IllegalAccessException {
if (dfTarget == null)
return;
if (StringUtils.isEmpty(content))
return;
if (dfTarget.isFilePath()) {
File file = new File(dfTarget.getFilePath(content));
if (file.exists()) {
Parser parser = parserSelector.parseFile(null, file.getName(),
null, null, file, lang);
if (parser != null)
parser.popupateResult(0, target);
} else {
Logging.error("File don't exist:" + file.getAbsolutePath());
}
}
if (dfTarget.isCrawlUrl()) {
WebCrawlThread crawlThread = webCrawlMaster.manualCrawl(
LinkUtils.newEncodedURL(content), ListType.DBCRAWL);
crawlThread.waitForStart(60);
crawlThread.waitForEnd(60);
Crawl crawl = crawlThread.getCurrentCrawl();
if (crawl != null) {
IndexDocument targetIndexDocument = crawl
.getTargetIndexDocument(0);
if (targetIndexDocument != null)
target.add(targetIndexDocument);
}
}
if (dfTarget.isConvertHtmlEntities())
content = StringEscapeUtils.unescapeHtml(content);
if (dfTarget.isRemoveTag())
content = StringUtils.removeTag(content);
if (dfTarget.hasRegexpPattern())
content = dfTarget.applyRegexPattern(content);
target.add(dfTarget.getName(), new FieldValueItem(
FieldValueOriginEnum.EXTERNAL, content));
}
}
© 2015 - 2025 Weber Informatics LLC | Privacy Policy