com.jaeksoft.searchlib.parser.ExternalParser 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) 2013-2014 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 results;
public final List links;
public final String error;
public Results() {
results = null;
links = null;
error = null;
}
public Results(Throwable t) {
this.error = t.getMessage();
results = null;
links = null;
}
public Results(List resultItems,
Set detectedLinks, Throwable error) {
this.results = resultItems != null ? new ArrayList(
resultItems.size()) : null;
if (resultItems != null)
for (ParserResultItem resultItem : resultItems)
results.add(resultItem.getNewExternalResult());
this.links = detectedLinks != null ? new ArrayList(
detectedLinks.size()) : null;
if (detectedLinks != null)
this.links.addAll(detectedLinks);
this.error = error == null ? null : error.getMessage();
}
}
public final static void doParserContent(Parser parser, File tempDir,
final IndexDocument sourceDocument,
final StreamLimiter streamLimiter, final LanguageEnum lang)
throws IOException, SearchLibException,
TransformerConfigurationException, SAXException {
PrintWriter configWriter = null;
ByteArrayOutputStream err = null;
ByteArrayOutputStream out = null;
try {
// Prepare the files JSON and XML
Command command = new Command(sourceDocument, streamLimiter, lang);
File commandFile = new File(tempDir, FILE_PARSER_COMMAND);
JsonUtils.jsonToFile(command, commandFile);
File configFile = new File(tempDir, FILE_PARSER_CONFIG);
configWriter = new PrintWriter(configFile, FILE_PARSER_ENCODING);
XmlWriter xmlWriter = new XmlWriter(configWriter,
FILE_PARSER_ENCODING);
parser.writeXmlConfig(xmlWriter);
xmlWriter.endDocument();
configWriter.close();
// Execute
out = new ByteArrayOutputStream();
err = new ByteArrayOutputStream();
String classPath = ExecuteUtils.getClassPath();
if (!StringUtils.isEmpty(ADDITIONAL_CLASSPATH))
classPath = StringUtils.fastConcat(classPath,
SystemUtils.PATH_SEPARATOR, ADDITIONAL_CLASSPATH);
int statusCode = ExecuteUtils.command(tempDir, "java", classPath,
true, out, err, 3600000L, ExternalParser.class.getName());
if (statusCode != 0)
throw new SearchLibException(err.toString("UTF-8"));
File fileParserResults = new File(tempDir, FILE_PARSER_RESULTS);
if (!fileParserResults.exists())
return;
Results results = JsonUtils.getObject(fileParserResults,
Results.class);
if (!StringUtils.isEmpty(results.error)) {
Logging.warn("External parser error: " + err.toString());
throw new SearchLibException.ExternalParserException(
results.error);
}
parser.setExternalResults(results);
} catch (ExecuteException e) {
if (err != null)
Logging.error(err.toString("UTF-8"));
throw e;
} finally {
IOUtils.close(configWriter, err, out);
}
}
public final static void main(String[] args) {
Results results = null;
try {
File fileParserConfig = new File(FILE_PARSER_CONFIG);
if (!fileParserConfig.exists())
throw new FileNotFoundException(
fileParserConfig.getAbsolutePath());
XPathParser xpp = new XPathParser(fileParserConfig);
Parser parser = (Parser) ParserFactory.create(null, xpp,
xpp.getNode("/parser"));
if (parser == null)
throw new SearchLibException("Parser not found");
File fileParserCommand = new File(FILE_PARSER_COMMAND);
if (!fileParserCommand.exists())
throw new FileNotFoundException(
fileParserCommand.getAbsolutePath());
Command command = JsonUtils.getObject(fileParserCommand,
Command.class);
if (command == null)
throw new SearchLibException("Not parsing command found");
StartStopListener.OPENSEARCHSERVER_DATA_FILE = new File(
command.opensearchserver_data_path);
IndexDocument indexSourceDocument = command.sourceDocument == null ? null
: DocumentUpdate.getIndexDocument(command.sourceDocument);
StreamLimiterFile streamLimiterFile = new StreamLimiterFile(
parser.getSizeLimit(), new File(command.filePath),
command.originalURL);
parser.doParserContent(null, indexSourceDocument,
streamLimiterFile, command.lang);
results = parser.getExternalResults();
} catch (Exception e) {
results = new Results(e);
} finally {
File fileParserResults = new File(FILE_PARSER_RESULTS);
if (fileParserResults.exists())
fileParserResults.delete();
try {
JsonUtils.jsonToFile(results, fileParserResults);
} catch (Exception e) {
e.printStackTrace();
System.exit(1);
}
}
System.exit(0);
}
}