boolean load(Object[] props, int offset) {
if (_values.size() > 0) {
int p = _values.size() - 1;
while (_index[p] > _values.get(p).size() - 1 && p > 0) {
_index[p] = 0;
--p;
++_index[p];
}
if (_index[p] > _values.get(p).size() - 1) {
return false;
}
// System.err.print("\t TypedValueGroup.load{");
for (int i = 0; i < _values.size(); ++i) {
props[i + offset] = _values.get(i).get(_index[i]).data;
// System.err.print("[" + _index[i] + "](" + props[i+offset].getClass().getSimpleName() +
// ')');
}
// System.err.println("}");
++_index[_values.size() - 1];
return true;
} else {
return false;
}
}
private QueryResult gatherResultInfoForSelectQuery(
String queryString, int queryNr, boolean sorted, Document doc, String[] rows) {
Element root = doc.getRootElement();
// Get head information
Element child =
root.getChild("head", Namespace.getNamespace("http://www.w3.org/2005/sparql-results#"));
// Get result rows (<head>)
List headChildren =
child.getChildren(
"variable", Namespace.getNamespace("http://www.w3.org/2005/sparql-results#"));
Iterator it = headChildren.iterator();
ArrayList<String> headList = new ArrayList<String>();
while (it.hasNext()) {
headList.add(((Element) it.next()).getAttributeValue("name"));
}
List resultChildren =
root.getChild("results", Namespace.getNamespace("http://www.w3.org/2005/sparql-results#"))
.getChildren(
"result", Namespace.getNamespace("http://www.w3.org/2005/sparql-results#"));
int nrResults = resultChildren.size();
QueryResult queryResult = new QueryResult(queryNr, queryString, nrResults, sorted, headList);
it = resultChildren.iterator();
while (it.hasNext()) {
Element resultElement = (Element) it.next();
String result = "";
// get the row values and paste it together to one String
for (int i = 0; i < rows.length; i++) {
List bindings =
resultElement.getChildren(
"binding", Namespace.getNamespace("http://www.w3.org/2005/sparql-results#"));
String rowName = rows[i];
for (int j = 0; j < bindings.size(); j++) {
Element binding = (Element) bindings.get(j);
if (binding.getAttributeValue("name").equals(rowName))
if (result.equals(""))
result +=
rowName + ": " + ((Element) binding.getChildren().get(0)).getTextNormalize();
else
result +=
"\n"
+ rowName
+ ": "
+ ((Element) binding.getChildren().get(0)).getTextNormalize();
}
}
queryResult.addResult(result);
}
return queryResult;
}
/**
* Returns the content of a JDOM Element detached from it.
*
* @param elt the element to get the content from.
* @return a (possibly empty) list of JDOM nodes, detached from their parent.
*/
private List getDetachedContent(Element elt) {
List content = elt.getContent();
List nodes = new ArrayList(content.size());
while (content.size() != 0) {
Object o = content.remove(0);
nodes.add(o);
}
return (nodes);
}
// returns list of statements
public static void replaceMult(Statement st, NodeFactory f, Map o2n, Collection result)
throws ModelException {
List l1 = new ArrayList();
replaceMultSPO(st, f, o2n, l1, st.subject(), 0);
List l2 = new ArrayList();
for (int i = 0; i < l1.size(); i++)
replaceMultSPO((Statement) l1.get(i), f, o2n, l2, st.predicate(), 1);
for (int i = 0; i < l2.size(); i++)
replaceMultSPO((Statement) l2.get(i), f, o2n, result, st.object(), 2);
}
@Override
public void processingInstruction(String target, String data) throws SAXException {
_logger.fine("Processing Instruction " + target);
_logger.fine("Processing Instruction data: " + data);
if (target.equals("assemble")) {
if (!_stack.isEmpty()) {
ElementInfo element = _stack.get(_stack.size() - 1);
Matcher matcher = PROCESSING_INSTRUCTION.matcher(data);
while (matcher.find()) {
if (matcher.groupCount() == 2) {
String name = matcher.group(1);
if (name.charAt(0) == '@') {
element.inst.put(name, matcher.group(2));
} else {
element.args.add(guessUntypedValue(name, matcher.group(2)));
}
_logger.fine(
"Processing Instruction for "
+ element.data.getClass()
+ "\n\ttarget = "
+ target
+ "\n\t"
+ name
+ "="
+ matcher.group(2));
}
}
}
}
}
public AlbumBean[] getResult() {
synchronized (libraryList) {
AlbumBean[] res;
res = libraryList.toArray(new AlbumBean[libraryList.size()]);
return res;
}
}
/**
* This will output a list of JDOM nodes as a fragment of an XML document, firing off the SAX
* events that have been registered.
*
* <p><strong>Warning</strong>: This method does not call the {@link
* ContentHandler#setDocumentLocator}, {@link ContentHandler#startDocument} and {@link
* ContentHandler#endDocument} callbacks on the {@link #setContentHandler ContentHandler}. The
* user shall invoke these methods directly prior/after outputting the document fragments.
*
* @param nodes <code>List</code> of JDOM nodes to output.
* @throws JDOMException if any error occurred.
* @see #outputFragment(org.jdom2.Content)
*/
public void outputFragment(List<? extends Content> nodes) throws JDOMException {
if ((nodes == null) || (nodes.size() == 0)) {
return;
}
// Output node list as a document fragment.
elementContent(nodes, new NamespaceStack());
}
// returns list of statements
protected static void replaceMultSPO(
Statement st, NodeFactory f, Map o2n, Collection result, RDFNode toReplace, int position)
throws ModelException {
Collection replacements;
if (toReplace instanceof Statement) {
List l = new ArrayList();
replaceMult((Statement) toReplace, f, o2n, l);
if (l.size() == 1 && toReplace == l.get(0)) {
result.add(st);
return; // keep the same
} else replacements = l;
} else {
Object ro = o2n.get(toReplace);
if (ro instanceof Collection) replacements = (Collection) ro;
else if (ro != null) {
replacements = new ArrayList();
replacements.add(ro);
} else { // no replacement needed
result.add(st); // keep the same statement
return;
}
}
for (Iterator it = replacements.iterator(); it.hasNext(); ) {
Statement rs = null;
Object rr = it.next();
switch (position) {
case 0:
rs = f.createStatement((Resource) rr, st.predicate(), st.object());
break;
case 1:
rs = f.createStatement(st.subject(), (Resource) rr, st.object());
break;
case 2:
rs = f.createStatement(st.subject(), st.predicate(), (RDFNode) rr);
break;
}
result.add(rs);
}
}
/**
* Returns the result of an XSL Transformation as a list of JDOM nodes.
*
* <p>If the result of the transformation is a JDOM document, this method converts it into a list
* of JDOM nodes; any subsequent call to {@link #getDocument} will return <code>null</code>.
*
* @return the transformation result as a (possibly empty) list of JDOM nodes (Elements, Texts,
* Comments, PIs...).
*/
public List getResult() {
List nodes = Collections.EMPTY_LIST;
// Retrieve result from the document builder if not set.
this.retrieveResult();
if (result instanceof List) {
nodes = (List) result;
} else {
if ((result instanceof Document) && (queried == false)) {
List content = ((Document) result).getContent();
nodes = new ArrayList(content.size());
while (content.size() != 0) {
Object o = content.remove(0);
nodes.add(o);
}
result = nodes;
}
}
queried = true;
return (nodes);
}
/**
* This will output a list of JDOM nodes as a document, firing off the SAX events that have been
* registered.
*
* <p><strong>Warning</strong>: This method may output ill-formed XML documents if the list
* contains top-level objects that are not legal at the document level (e.g. Text or CDATA nodes,
* multiple Element nodes, etc.). Thus, it should only be used to output document portions towards
* ContentHandlers capable of accepting such ill-formed documents (such as XSLT processors).
*
* @param nodes <code>List</code> of JDOM nodes to output.
* @throws JDOMException if any error occurred.
* @see #output(org.jdom2.Document)
*/
public void output(List<? extends Content> nodes) throws JDOMException {
if ((nodes == null) || (nodes.size() == 0)) {
return;
}
// contentHandler.setDocumentLocator()
documentLocator(null);
// contentHandler.startDocument()
startDocument();
// Process node list.
elementContent(nodes, new NamespaceStack());
// contentHandler.endDocument()
endDocument();
}
public int getCount() {
synchronized (libraryList) {
return libraryList.size();
}
}
/** Receive notification of the start of an element. */
@Override
public void startElement(String uri, String l, String q, Attributes a) {
/*
* 1. Load a class that matches the element name.
* 2. If no class found, assume the element maps to a String.
* 3. Otherwise, construct a new object of the class with element attributes.
*/
_logger.fine(
S.fine(_logger)
? "Consider element " + l + "\n uri " + uri + "\n q " + q
: null);
ElementInfo info = new ElementInfo();
// Record java packages defined on this element as xmlns
for (int i = 0; i < a.getLength(); ++i) {
_logger.fine(
S.fine(_logger)
? " attr "
+ a.getQName(i)
+ "="
+ a.getValue(i)
+ "\n "
+ a.getQName(i)
+ ":"
+ a.getURI(i)
: null);
if (a.getQName(i).startsWith("xmlns:") && a.getValue(i).startsWith("java://")) {
info.pkgs.put(a.getQName(i).substring(6), a.getValue(i).substring(7));
}
}
// Resolve the package name of this element, which could be empty (default package)
int colon = q.indexOf(':');
if (colon > 0) {
String xmlns = q.substring(0, colon);
// is it defined right here?
info.jpkg = info.pkgs.get(xmlns);
// find a matching namespace from ancesters
if (info.jpkg == null && !_stack.isEmpty()) {
for (int i = _stack.size() - 1; i >= 0; --i) {
info.jpkg = _stack.get(i).pkgs.get(xmlns);
if (info.jpkg != null) {
break;
}
}
}
} else if (isPrimitiveType(q)) {
info.jpkg = "java.lang";
} else if (!_stack.isEmpty()) {
info.jpkg = _stack.get(_stack.size() - 1).jpkg;
} else {
info.jpkg = _jpkg;
}
_logger.fine("to create element with package = " + info.jpkg);
try {
info.name =
(info.jpkg != null) ? info.jpkg + '.' + Strings.toCamelCase(l) : Strings.toCamelCase(l);
try {
if (info.name.endsWith("...")) {
// Array construction
info.type = Class.forName(info.name.substring(0, info.name.length() - 3));
info.data = new ArrayList<Object>();
} else {
// Non-array construction
int size = a.getLength();
TypedValueGroup arguments = new TypedValueGroup();
for (int i = 0; i < size; ++i) {
if (!a.getQName(i).startsWith("xmlns:") && !a.getQName(i).equals("xmlns")) {
arguments.add(guessUntypedValue(a.getQName(i), a.getValue(i)));
}
}
arguments.complete();
_logger.fine(S.fine(_logger) ? "arguments=" + arguments : null);
if (arguments.size() > 0) {
if (arguments.size() == 1 && "java.lang".equals(info.jpkg)) {
info.inst.put(
"@as",
Strings.toCamelCase(
arguments.get(0).name, '-', false)); // respect original spelling
info.data = arguments.get(0).get(0).data;
info.type = arguments.get(0).get(0).type;
} else {
Exception last = null;
Object[] args = new Object[arguments.size()];
while (arguments.load(args, 0)) {
try {
_logger.fine(
S.fine(_logger)
? "to create " + info.name + " with args: " + args.length + args(args)
: null);
info.data = _factory.create(info.name, args);
info.type = info.data.getClass();
break;
} catch (InvocationTargetException x) {
throw x;
} catch (Exception x) {
last = x;
_logger.fine(
"failure in creating " + info.name + ": probing for other constructors");
}
}
if (info.data == null) {
throw last;
}
}
} else {
_logger.fine("Create " + info.name + " with the default constructor");
info.data = _factory.create(info.name);
info.type = info.data.getClass();
}
}
} catch (ClassNotFoundException x) {
// no class by the element name is found, assumed String
if (!_lenient) {
throw new BeanAssemblyException("No class associated with element " + q);
} else {
_logger.log(Level.WARNING, "can't find class " + info.name, x);
}
}
_stack.add(info);
// _logger.fine(">>ElementInfo: " + info.type.getName() + " in " + info);
// all other exceptions indicate mismatches between the beans and the XML schema
} catch (Exception x) {
if (!_lenient) {
throw new BeanAssemblyException("Failed to assemble bean from element " + q, x);
} else {
_logger.log(Level.SEVERE, "can't create object for this element", x);
}
}
}
protected void init(Element root) {
NodeList children = root.getChildNodes();
for (int i = 0; i < children.getLength(); i++) {
Node child = children.item(i);
if (child.getNodeName().equals("meta")) {
Element meta = (Element) child;
String value = XMLUtil.getText(meta);
this.metaAttributes.put(meta.getAttribute("name"), value);
}
}
// handle registers - OPTIONAL
Element r = XMLUtil.getChildElement(root, "registers");
if (r != null) {
List registers = XMLUtil.getChildElements(r, "register");
for (int i = 0; i < registers.size(); i++) {
Element register = (Element) registers.get(i);
RegisterDescriptor registerDescriptor =
DescriptorFactory.getFactory().createRegisterDescriptor(register);
registerDescriptor.setParent(this);
this.registers.add(registerDescriptor);
}
}
// handle global-conditions - OPTIONAL
Element globalConditionsElement = XMLUtil.getChildElement(root, "global-conditions");
if (globalConditionsElement != null) {
Element globalConditions = XMLUtil.getChildElement(globalConditionsElement, "conditions");
ConditionsDescriptor conditionsDescriptor =
DescriptorFactory.getFactory().createConditionsDescriptor(globalConditions);
conditionsDescriptor.setParent(this);
this.globalConditions = conditionsDescriptor;
}
// handle initial-steps - REQUIRED
Element intialActionsElement = XMLUtil.getChildElement(root, "initial-actions");
List initialActions = XMLUtil.getChildElements(intialActionsElement, "action");
for (int i = 0; i < initialActions.size(); i++) {
Element initialAction = (Element) initialActions.get(i);
ActionDescriptor actionDescriptor =
DescriptorFactory.getFactory().createActionDescriptor(initialAction);
actionDescriptor.setParent(this);
this.initialActions.add(actionDescriptor);
}
// handle global-actions - OPTIONAL
Element globalActionsElement = XMLUtil.getChildElement(root, "global-actions");
if (globalActionsElement != null) {
List globalActions = XMLUtil.getChildElements(globalActionsElement, "action");
for (int i = 0; i < globalActions.size(); i++) {
Element globalAction = (Element) globalActions.get(i);
ActionDescriptor actionDescriptor =
DescriptorFactory.getFactory().createActionDescriptor(globalAction);
actionDescriptor.setParent(this);
this.globalActions.add(actionDescriptor);
}
}
// handle common-actions - OPTIONAL
// - Store actions in HashMap for now. When parsing Steps, we'll resolve
// any common actions into local references.
Element commonActionsElement = XMLUtil.getChildElement(root, "common-actions");
if (commonActionsElement != null) {
List commonActions = XMLUtil.getChildElements(commonActionsElement, "action");
for (int i = 0; i < commonActions.size(); i++) {
Element commonAction = (Element) commonActions.get(i);
ActionDescriptor actionDescriptor =
DescriptorFactory.getFactory().createActionDescriptor(commonAction);
actionDescriptor.setParent(this);
addCommonAction(actionDescriptor);
}
}
// handle timer-functions - OPTIONAL
Element timerFunctionsElement = XMLUtil.getChildElement(root, "trigger-functions");
if (timerFunctionsElement != null) {
List timerFunctions = XMLUtil.getChildElements(timerFunctionsElement, "trigger-function");
for (int i = 0; i < timerFunctions.size(); i++) {
Element timerFunction = (Element) timerFunctions.get(i);
Integer id = new Integer(timerFunction.getAttribute("id"));
FunctionDescriptor function =
DescriptorFactory.getFactory()
.createFunctionDescriptor(XMLUtil.getChildElement(timerFunction, "function"));
function.setParent(this);
this.timerFunctions.put(id, function);
}
}
// handle steps - REQUIRED
Element stepsElement = XMLUtil.getChildElement(root, "steps");
List steps = XMLUtil.getChildElements(stepsElement, "step");
for (int i = 0; i < steps.size(); i++) {
Element step = (Element) steps.get(i);
StepDescriptor stepDescriptor =
DescriptorFactory.getFactory().createStepDescriptor(step, this);
this.steps.add(stepDescriptor);
}
// handle splits - OPTIONAL
Element splitsElement = XMLUtil.getChildElement(root, "splits");
if (splitsElement != null) {
List split = XMLUtil.getChildElements(splitsElement, "split");
for (int i = 0; i < split.size(); i++) {
Element s = (Element) split.get(i);
SplitDescriptor splitDescriptor = DescriptorFactory.getFactory().createSplitDescriptor(s);
splitDescriptor.setParent(this);
this.splits.add(splitDescriptor);
}
}
// handle joins - OPTIONAL:
Element joinsElement = XMLUtil.getChildElement(root, "joins");
if (joinsElement != null) {
List join = XMLUtil.getChildElements(joinsElement, "join");
for (int i = 0; i < join.size(); i++) {
Element s = (Element) join.get(i);
JoinDescriptor joinDescriptor = DescriptorFactory.getFactory().createJoinDescriptor(s);
joinDescriptor.setParent(this);
this.joins.add(joinDescriptor);
}
}
}
public void writeXML(PrintWriter out, int indent) {
XMLUtil.printIndent(out, indent++);
out.println("<workflow>");
Iterator iter = metaAttributes.entrySet().iterator();
while (iter.hasNext()) {
Map.Entry entry = (Map.Entry) iter.next();
XMLUtil.printIndent(out, indent);
out.print("<meta name=\"");
out.print(XMLUtil.encode(entry.getKey()));
out.print("\">");
out.print(XMLUtil.encode(entry.getValue()));
out.println("</meta>");
}
if (registers.size() > 0) {
XMLUtil.printIndent(out, indent++);
out.println("<registers>");
for (int i = 0; i < registers.size(); i++) {
RegisterDescriptor register = (RegisterDescriptor) registers.get(i);
register.writeXML(out, indent);
}
XMLUtil.printIndent(out, --indent);
out.println("</registers>");
}
if (timerFunctions.size() > 0) {
XMLUtil.printIndent(out, indent++);
out.println("<trigger-functions>");
Iterator iterator = timerFunctions.entrySet().iterator();
while (iterator.hasNext()) {
Map.Entry entry = (Map.Entry) iterator.next();
XMLUtil.printIndent(out, indent++);
out.println("<trigger-function id=\"" + entry.getKey() + "\">");
FunctionDescriptor trigger = (FunctionDescriptor) entry.getValue();
trigger.writeXML(out, indent);
XMLUtil.printIndent(out, --indent);
out.println("</trigger-function>");
}
while (iterator.hasNext()) {}
XMLUtil.printIndent(out, --indent);
out.println("</trigger-functions>");
}
if (getGlobalConditions() != null) {
XMLUtil.printIndent(out, indent++);
out.println("<global-conditions>");
getGlobalConditions().writeXML(out, indent);
out.println("</global-conditions>");
}
XMLUtil.printIndent(out, indent++);
out.println("<initial-actions>");
for (int i = 0; i < initialActions.size(); i++) {
ActionDescriptor action = (ActionDescriptor) initialActions.get(i);
action.writeXML(out, indent);
}
XMLUtil.printIndent(out, --indent);
out.println("</initial-actions>");
if (globalActions.size() > 0) {
XMLUtil.printIndent(out, indent++);
out.println("<global-actions>");
for (int i = 0; i < globalActions.size(); i++) {
ActionDescriptor action = (ActionDescriptor) globalActions.get(i);
action.writeXML(out, indent);
}
XMLUtil.printIndent(out, --indent);
out.println("</global-actions>");
}
if (commonActions.size() > 0) {
XMLUtil.printIndent(out, indent++);
out.println("<common-actions>");
Iterator iterator = getCommonActions().values().iterator();
while (iterator.hasNext()) {
ActionDescriptor action = (ActionDescriptor) iterator.next();
action.writeXML(out, indent);
}
XMLUtil.printIndent(out, --indent);
out.println("</common-actions>");
}
XMLUtil.printIndent(out, indent++);
out.println("<steps>");
for (int i = 0; i < steps.size(); i++) {
StepDescriptor step = (StepDescriptor) steps.get(i);
step.writeXML(out, indent);
}
XMLUtil.printIndent(out, --indent);
out.println("</steps>");
if (splits.size() > 0) {
XMLUtil.printIndent(out, indent++);
out.println("<splits>");
for (int i = 0; i < splits.size(); i++) {
SplitDescriptor split = (SplitDescriptor) splits.get(i);
split.writeXML(out, indent);
}
XMLUtil.printIndent(out, --indent);
out.println("</splits>");
}
if (joins.size() > 0) {
XMLUtil.printIndent(out, indent++);
out.println("<joins>");
for (int i = 0; i < joins.size(); i++) {
JoinDescriptor join = (JoinDescriptor) joins.get(i);
join.writeXML(out, indent);
}
XMLUtil.printIndent(out, --indent);
out.println("</joins>");
}
XMLUtil.printIndent(out, --indent);
out.println("</workflow>");
}
/** Receive notification of the end of an element. */
@Override
public void endElement(String uri, String l, String q) {
/*
* 1. If current element is a String, update its value from the string buffer.
* 2. Add the element to parent.
*/
ElementInfo element = _stack.remove(_stack.size() - 1);
_logger.fine("endElement " + element);
if (element.type == null) {
_logger.warning("Element " + element.name + " not created ");
return;
} else if (_chars.length() > 0) {
try {
injectProperty(element.data, String.class, _chars.toString(), null, null);
} catch (Exception x) {
if (!_lenient) {
throw new BeanAssemblyException(
"Failed to set characters to object " + element.type.getName(), x);
} else {
_logger.warning("Failed to set characters to parent " + element.data);
}
}
}
_chars.setLength(0);
_logger.fine(
"<<ElementInfo: "
+ element.type.getName()
+ " in "
+ element
+ "\n @as is "
+ element.inst.get("@as")
+ "\n @id is "
+ element.inst.get("@id"));
if (List.class.isAssignableFrom(element.data.getClass()) && element.name.endsWith("...")) {
List<?> list = (List<?>) element.data;
Object array = Array.newInstance(element.type, list.size());
for (int i = 0; i < list.size(); ++i) {
Array.set(array, i, list.get(i));
}
element.data = array;
}
String id = element.inst.get("@id");
if (id != null) {
// locally stored object - not added to the parent
_local.put(id, element);
} else if (!_stack.isEmpty()) {
// inject into the parent as a property
ElementInfo parent = _stack.get(_stack.size() - 1);
_logger.fine("Parent is " + parent.data.getClass().getName());
try {
String as = element.inst.get("@as");
if (as != null) {
injectProperty(
parent.data,
element.type,
element.data,
Strings.toCamelCase(as, '-', false),
element.args.complete());
} else {
injectProperty(parent.data, element.type, element.data, null, element.args.complete());
}
} catch (Exception x) {
if (!_lenient) {
throw new BeanAssemblyException(
"Failed to set value " + element.data + " to parent " + parent.data, x);
} else {
_logger.log(
Level.WARNING,
"Failed to set value " + element.data + " to parent " + parent.data,
x);
}
}
}
_top = element.data;
}
public void startElement(String namespaceURI, String localName, String qName, Attributes atts)
throws SAXException {
if (qName.equals("patent-document") || qName.equals("fulltext-document")) {
nbNPLRef = 0;
nbPatentRef = 0;
nbAllRef = 0;
int length = atts.getLength();
// Process each attribute
for (int i = 0; i < length; i++) {
// Get names and values for each attribute
String name = atts.getQName(i);
String value = atts.getValue(i);
if (name != null) {
if (name.equals("lang")) {
// Global_Language_Code = value.toLowerCase();
}
if (name.equals("doc-number")) {
PatentNumber = "EP" + value;
}
if (name.equals("kind")) {
CodeType = value;
}
if (name.equals("date")) {
PublicDate = value;
}
}
}
CitedPatentNumber = new ArrayList<String>();
accumulatedText = new StringBuffer();
allContent = new StringBuffer();
accumulator.setLength(0);
} else if (qName.equals("description")) {
accumulator.setLength(0);
} else if (qName.equals("ref") || qName.equals("bibl")) {
int length = atts.getLength();
nbAllRef++;
// Process each attribute
for (int i = 0; i < length; i++) {
// Get names and values for each attribute
String name = atts.getQName(i);
String value = atts.getValue(i);
if (name != null) {
if (name.equals("type") || name.equals("typ")) {
if (value.equals("npl") || value.equals("book") || value.equals("journal")) {
String content = getText();
// we output what has been read so far in the description
// we tokenize the text
// ArrayList<String> tokens =
// StringTokenizer st = new StringTokenizer(content, delimiters, true);
List<String> tokenizations = new ArrayList<String>();
try {
// TBD: pass a language object to the tokenize method call
tokenizations = analyzer.tokenize(content);
} catch (Exception e) {
LOGGER.debug("Tokenization for XML patent document has failed.");
}
// int nbTokens = st.countTokens();
int nbTokens = tokenizations.size();
int j = 0;
// while (st.hasMoreTokens()) {
for (String token : tokenizations) {
// String token = st.nextToken().trim();
if ((token.trim().length() == 0)
|| (token.equals(" "))
|| (token.equals("\t"))
|| (token.equals("\n"))
|| (token.equals("\r"))) {
continue;
}
if ((j > (nbTokens - N) && (N != -1)) || (refFound && (j < N) && (N != -1))) {
try {
accumulatedText.append(token + "\t" + "<other>\n");
allContent.append(token + " ");
} catch (Exception e) {
// e.printStackTrace();
throw new GrobidException("An exception occured while running Grobid.", e);
}
} else {
try {
accumulatedText.append(token + "\t" + "<ignore>\n");
allContent.append(token + " ");
} catch (Exception e) {
// e.printStackTrace();
throw new GrobidException("An exception occured while running Grobid.", e);
}
}
j++;
}
accumulator.setLength(0);
npl = true;
ref = true;
} else if (value.equals("patent") || value.equals("pl")) {
String content = getText();
// we output what has been read so far in the description
// we tokenize the text
// ArrayList<String> tokens =
// TextUtilities.segment(content,"[("+TextUtilities.punctuations);
// StringTokenizer st = new StringTokenizer(content, delimiters, true);
List<String> tokenizations = new ArrayList<String>();
try {
// TBD: pass a language object to the tokenize method call
tokenizations = analyzer.tokenize(content);
} catch (Exception e) {
LOGGER.debug("Tokenization for XML patent document has failed.");
}
// int nbTokens = st.countTokens();
int nbTokens = tokenizations.size();
int j = 0;
for (String token : tokenizations) {
// while (st.hasMoreTokens()) {
// String token = st.nextToken().trim();
if ((token.trim().length() == 0)
|| (token.equals(" "))
|| (token.equals("\t"))
|| (token.equals("\n"))
|| (token.equals("\r"))) {
continue;
}
if ((j > (nbTokens - N)) | (refFound & (j < N))) {
try {
accumulatedText.append(token + "\t" + "<other>\n");
allContent.append(token + " ");
} catch (Exception e) {
// e.printStackTrace();
throw new GrobidException("An exception occured while running Grobid.", e);
}
} else {
try {
accumulatedText.append(token + "\t" + "<ignore>\n");
allContent.append(token + " ");
} catch (Exception e) {
// e.printStackTrace();
throw new GrobidException("An exception occured while running Grobid.", e);
}
}
j++;
}
accumulator.setLength(0);
npl = false;
ref = true;
} else {
System.out.println("Warning: unknown attribute value for ref or bibl: " + value);
ref = false;
npl = false;
}
}
}
}
accumulatorRef.setLength(0);
} else if (qName.equals("claim")) {
accumulator.setLength(0);
} else if (qName.equals("invention-title")) {
accumulator.setLength(0);
} else if (qName.equals("patcit")) {
int length = atts.getLength();
// Process each attribute
for (int i = 0; i < length; i++) {
// Get names and values for each attribute
String name = atts.getQName(i);
String value = atts.getValue(i);
if (name != null) {
if (name.equals("ucid")) {
cited_number = value;
// we normally need to normalize a little bit this patent nummer
}
}
}
}
}
int size() {
return _values.size();
}
public String getText() {
String text = accumulator.toString();
if (text.trim().length() == 0) {
return "";
}
/*text = text.replace("\n", " ");
text = text.replace(" ", " ");*/
if (counting) {
/*
StringTokenizer st = new StringTokenizer(text, delimiters, true);
int count = 0;
while(st.hasMoreTokens()) {
String token = st.nextToken().trim();
if (token.length() == 0) {
continue;
}
count++;
}
*/
int i = currentPatentIndex;
int count = text.length();
while (i < patents.size()) {
PatentItem currentPatent = patents.get(i);
if (currentPatent != null) {
int startOffset = currentPatent.getOffsetBegin();
int endOffset = currentPatent.getOffsetEnd();
if ((startOffset >= offset) && (endOffset <= offset + count)) {
String context = currentPatent.getContext();
/*System.out.println("OFFSET: " + offset);
System.out.println("count: " + count);
System.out.println("startOffset: " + startOffset);
System.out.println("endOffset: " + endOffset);
System.out.println("context: " + context);
System.out.println("text: " + text);*/
String target = "";
if (context.charAt(0) == ' ') {
target = " <ref type=\"patent\">" + context.substring(1, context.length()) + "</ref>";
} else {
target = "<ref type=\"patent\">" + context + "</ref>";
}
text = text.replace(context, target);
currentPatentIndex = i;
}
}
i++;
}
// i = currentArticleIndex;
i = 0;
while (i < articles.size()) {
BibDataSet currentArticle = articles.get(i);
if (currentArticle != null) {
List<Integer> offsets = currentArticle.getOffsets();
int startOffset = -1;
int endOffset = -1;
String context = currentArticle.getRawBib().trim();
if (offsets.size() > 0) {
if (offsets.get(0) != null) {
startOffset = offsets.get(0).intValue();
/*StringTokenizer stt = new StringTokenizer(context, delimiters, true);
int count2 = 0;
while(stt.hasMoreTokens()) {
String token2 = stt.nextToken().trim();
if (token2.length() == 0) {
continue;
}
count2++;
}*/
// endOffset = offsets.get(1).intValue();
endOffset = startOffset + context.length();
}
}
// if ( (startOffset >= offset) && (endOffset <= offset+count) ) {
if ((startOffset >= offset)) {
/*System.out.println("OFFSET: " + offset);
System.out.println("count: " + count);
System.out.println("startOffset: " + startOffset);
System.out.println("endOffset: " + endOffset);
System.out.println("context: " + context);
System.out.println("text: " + text);*/
String target = " <ref type=\"npl\">" + context + "</ref> ";
text = text.replace(context, target);
currentArticleIndex = i;
}
}
i++;
}
offset += count;
}
return text;
}
/**
* This reports the occurrence of an actual element. It will include the element's attributes,
* with the exception of XML vocabulary specific attributes, such as <code>
* xmlns:[namespace prefix]</code> and <code>xsi:schemaLocation</code>.
*
* @param namespaceURI <code>String</code> namespace URI this element is associated with, or an
* empty <code>String</code>
* @param localName <code>String</code> name of element (with no namespace prefix, if one is
* present)
* @param qName <code>String</code> XML 1.0 version of element name: [namespace
* prefix]:[localName]
* @param atts <code>Attributes</code> list for this element
* @throws SAXException when things go wrong
*/
public void startElement(String namespaceURI, String localName, String qName, Attributes atts)
throws SAXException {
if (suppress) return;
Element element = null;
if ((namespaceURI != null) && (!namespaceURI.equals(""))) {
String prefix = "";
// Determine any prefix on the Element
if (!qName.equals(localName)) {
int split = qName.indexOf(":");
prefix = qName.substring(0, split);
}
Namespace elementNamespace = Namespace.getNamespace(prefix, namespaceURI);
element = factory.element(localName, elementNamespace);
} else {
element = factory.element(localName);
}
// Take leftover declared namespaces and add them to this element's
// map of namespaces
if (declaredNamespaces.size() > 0) {
transferNamespaces(element);
}
// Handle attributes
for (int i = 0, len = atts.getLength(); i < len; i++) {
Attribute attribute = null;
String attLocalName = atts.getLocalName(i);
String attQName = atts.getQName(i);
int attType = getAttributeType(atts.getType(i));
// Bypass any xmlns attributes which might appear, as we got
// them already in startPrefixMapping().
// This is sometimes necessary when SAXHandler is used with
// another source than SAXBuilder, as with JDOMResult.
if (attQName.startsWith("xmlns:") || attQName.equals("xmlns")) {
continue;
}
// First clause per http://markmail.org/message/2p245ggcjst27xe6
// patch from Mattias Jiderhamn
if ("".equals(attLocalName) && attQName.indexOf(":") == -1) {
attribute = factory.attribute(attQName, atts.getValue(i), attType);
} else if (!attQName.equals(attLocalName)) {
String attPrefix = attQName.substring(0, attQName.indexOf(":"));
Namespace attNs = Namespace.getNamespace(attPrefix, atts.getURI(i));
attribute = factory.attribute(attLocalName, atts.getValue(i), attType, attNs);
} else {
attribute = factory.attribute(attLocalName, atts.getValue(i), attType);
}
factory.setAttribute(element, attribute);
}
flushCharacters();
if (atRoot) {
document.setRootElement(element); // XXX should we use a factory call?
atRoot = false;
} else {
factory.addContent(getCurrentElement(), element);
}
currentElement = element;
}
TypedValueGroup complete() {
_index = new int[_values.size()];
return this;
}