/** Convert an XML string to <code>org.simpleframework.xml</code>-enabled object. */
public static <T> T fromString(Class<T> clazz, String xml) throws IOException {
try {
final StringReader r = new StringReader(xml);
return new Persister().read(clazz, r);
} catch (Exception e) {
throw ExceptionUtils.wrapAs(IOException.class, e);
}
}
public void process() throws ProcessingException {
try {
final SearchEngineResponse response = fetchSearchResponse();
documents = response.results;
resultsTotal = response.getResultsTotal();
} catch (Exception e) {
throw ExceptionUtils.wrapAs(ProcessingException.class, e);
}
}
/** Creates an {@link XMLMemento} from Simple XML-annotated bean. */
static IMemento toMemento(Object benchmarkSettings) throws IOException {
try {
final StringWriter w = new StringWriter();
new Persister().write(benchmarkSettings, w);
XMLMemento memento = XMLMemento.createReadRoot(new StringReader(w.toString()));
return memento;
} catch (Exception e) {
throw ExceptionUtils.wrapAs(IOException.class, e);
}
}
/** Convert any <code>org.simpleframework.xml</code>-enabled object to an XML string. */
public static String toString(Object object) throws IOException {
checkObject(object);
try {
final StringWriter w = new StringWriter();
new Persister().write(object, w);
return w.toString();
} catch (Exception e) {
throw ExceptionUtils.wrapAs(IOException.class, e);
}
}
/**
* Reads an object from a {@link IMemento}. The memento's type (root) must equal the bean's {@link
* Root} annotation name attribute.
*/
static <T> T fromMemento(Class<T> clazz, IMemento memento) throws IOException {
try {
final StringWriter sw = new StringWriter();
final XMLMemento m = XMLMemento.createWriteRoot(memento.getType());
m.putMemento(memento);
m.save(sw);
return new Persister().read(clazz, new StringReader(sw.toString()));
} catch (Exception e) {
throw ExceptionUtils.wrapAs(IOException.class, e);
}
}
/** Open or retrieve an open handle to an {@link IndexSearcher}. */
private IndexSearcher indexOpen(Directory directory) throws ProcessingException {
synchronized (context) {
IndexSearcher searcher = openIndexes.get(directory);
if (searcher == null) {
try {
searcher = new IndexSearcher(DirectoryReader.open(directory));
openIndexes.put(directory, searcher);
} catch (IOException e) {
throw ExceptionUtils.wrapAs(ProcessingException.class, e);
}
}
return searcher;
}
}
/** Performs tokenization and saves the results to the <code>context</code>. */
public void tokenize(PreprocessingContext context) {
// Documents to tokenize
final List<Document> documents = context.documents;
// Fields to tokenize
final String[] fieldNames = documentFields.toArray(new String[documentFields.size()]);
if (fieldNames.length > 8) {
throw new ProcessingException("Maximum number of tokenized fields is 8.");
}
// Prepare arrays
images = Lists.newArrayList();
tokenTypes = new ShortArrayList();
documentIndices = new IntArrayList();
fieldIndices = new ByteArrayList();
final Iterator<Document> docIterator = documents.iterator();
int documentIndex = 0;
final ITokenizer ts = context.language.getTokenizer();
final MutableCharArray wrapper = new MutableCharArray(CharArrayUtils.EMPTY_ARRAY);
while (docIterator.hasNext()) {
final Document doc = docIterator.next();
boolean hadTokens = false;
for (int i = 0; i < fieldNames.length; i++) {
final byte fieldIndex = (byte) i;
final String fieldName = fieldNames[i];
final String fieldValue = doc.getField(fieldName);
if (!StringUtils.isEmpty(fieldValue)) {
try {
short tokenType;
ts.reset(new StringReader(fieldValue));
if ((tokenType = ts.nextToken()) != ITokenizer.TT_EOF) {
if (hadTokens) addFieldSeparator(documentIndex);
do {
ts.setTermBuffer(wrapper);
add(documentIndex, fieldIndex, context.intern(wrapper), tokenType);
} while ((tokenType = ts.nextToken()) != ITokenizer.TT_EOF);
hadTokens = true;
}
} catch (IOException e) {
// Not possible (StringReader above)?
throw ExceptionUtils.wrapAsRuntimeException(e);
}
}
}
if (docIterator.hasNext()) {
addDocumentSeparator();
}
documentIndex++;
}
addTerminator();
// Save results in the PreprocessingContext
context.allTokens.documentIndex = documentIndices.toArray();
context.allTokens.fieldIndex = fieldIndices.toArray();
context.allTokens.image = images.toArray(new char[images.size()][]);
context.allTokens.type = tokenTypes.toArray();
context.allFields.name = fieldNames;
// Clean up
images = null;
fieldIndices = null;
tokenTypes = null;
documentIndices = null;
}
