private String getConfigValue(String setting, String name, Document doc)
throws XPathExpressionException {
XPathFactory xPathfactory = XPathFactory.newInstance();
XPath xpath = xPathfactory.newXPath();
XPathExpression userExpr =
xpath.compile(
"wikipediaminer/setting[@name=\""
+ setting
+ "\"]/param[@name=\""
+ name
+ "\"]/@value");
return userExpr.evaluate(doc);
}
private NodeList findNodes(Document doc, Node operation) {
List<Node> xpaths = getChildNodes(operation, "xpath");
if (xpaths.isEmpty()) {
return null;
}
String xpathExpression = xpaths.get(0).getTextContent();
if (xpathExpression == null) {
return null;
}
XPathFactory xPathfactory = XPathFactory.newInstance();
XPath xpath = xPathfactory.newXPath();
NodeList nl = null;
try {
XPathExpression expr = xpath.compile(xpathExpression);
nl = (NodeList) expr.evaluate(doc, XPathConstants.NODESET);
} catch (XPathExpressionException ex) {
Utils.onError(new Error.WrongXpathExpression(xpathExpression));
}
return nl;
}
public static void main(String[] args) {
try {
DocumentBuilderFactory db = DocumentBuilderFactory.newInstance();
DocumentBuilder builder = db.newDocumentBuilder();
Document dom = builder.parse("data/geographic-area-data.html");
XPathFactory xpath = XPathFactory.newInstance();
XPath path = xpath.newXPath();
XPathExpression table =
path.compile("//div[@id='mw-content-text']/table[contains(@class,'wikitable')]/tr");
NodeList wikiData = (NodeList) table.evaluate(dom, XPathConstants.NODESET);
NodeList children;
String currentData, cleanData;
/* Open output stream */
FileWriter fstream = new FileWriter("data/parsed.yaml");
BufferedWriter out = new BufferedWriter(fstream);
for (int i = 0; i < wikiData.getLength(); i++) {
if (i == 0) {
continue;
}
out.write(new Integer(i).toString() + ":\n");
children = wikiData.item(i).getChildNodes();
for (int j = 0; j < children.getLength(); j++) {
currentData = (String) children.item(j).getTextContent();
switch (j) {
case 0:
/* Current Data is empty */
break;
case 1:
cleanData = decompose(currentData).trim().replaceAll("[^a-zA-Z\\s]+", "");
out.write("\t\"Geographic entity\": \"" + cleanData + "\"\n");
break;
case 2:
/* Current Data is empty */
break;
case 3:
cleanData = decompose(currentData).trim().replaceAll(",", "");
out.write("\t\"Area\": \"" + cleanData + "\"\n");
break;
case 4:
/* Current Data is empty */
break;
case 5:
cleanData = decompose(currentData).trim();
out.write("\t\"Notes\": \"" + cleanData + "\"\n");
break;
case 6:
/* Current Data is empty */
break;
default:
/* System.out.println("[" + j + "] Hit default case statement. Current Data is: " + currentData); */
break;
}
}
}
/* Close output stream */
out.close();
} catch (Exception e) {
System.out.println(e);
}
}
protected Document parse(InputStream inputStream) throws LibraryException {
Map<Integer, SongBean> songMap = new HashMap<Integer, SongBean>();
Map<Integer, AlbumBean> albumMap = new HashMap<Integer, AlbumBean>();
Map<Integer, AlbumBean> secondaryAlbumMap = new HashMap<Integer, AlbumBean>();
warningList.clear();
try {
// Create a builder factory
DocumentBuilderFactory factory = DocumentBuilderFactory.newInstance();
factory.setValidating(false);
factory.setNamespaceAware(true);
factory.setIgnoringElementContentWhitespace(true);
factory.setIgnoringComments(true);
// Create the builder and parse the file
DocumentBuilder builder = factory.newDocumentBuilder();
// Set an error listener and parse the document
builder.setErrorHandler(new iTradeTunesLibraryErrorHandler());
builder.setEntityResolver(new iTradeTunesLibraryResolver());
Document document = builder.parse(inputStream);
synchronized (libraryList) {
// Query the library document and build the library list
XPath xPath = XPathFactory.newInstance().newXPath();
XPathExpression xPathExpression = xPath.compile(XPATH_LIBRARY_LIST);
NodeList nodelist = (NodeList) xPathExpression.evaluate(document, XPathConstants.NODESET);
// Process the elements in the nodelist
SongBean song = null;
for (int i = 0; i < nodelist.getLength(); i++) {
boolean isTrackID = false;
// Get element and child nodes
Element elem = (Element) nodelist.item(i);
NodeList list = elem.getChildNodes();
// Get node value
Node childKey = list.item(0);
String key = childKey.getNodeValue();
// Check if we have to create a new bean
if (SongBean.NAME_TRACK_ID.equals(key)) {
isTrackID = true;
SongBean previousSong = song;
song = new SongBean();
if (previousSong != null
&& !("AAC audio file".equals(previousSong.getKind())
|| "MPEG audio file".equals(previousSong.getKind()))) {
songMap.remove(previousSong.getTrack_ID());
} else {
// Add an album bean
addOrUpdateAlbum(albumMap, previousSong, false);
}
}
// The first parameter is the key
String prop = childKey.getNodeValue().replace(' ', '_');
// The second parameter is the value
i++;
// Get element and child nodes
elem = (Element) nodelist.item(i);
// Check for boolean properties
Object value = null;
// Get node value
list = elem.getChildNodes();
childKey = list.item(0);
value = (childKey == null) ? elem.getNodeName() : childKey.getNodeValue();
if (isTrackID) {
isTrackID = false;
}
// Set the property of the song bean
Statement stmt = new Statement(song, "set" + prop, new Object[] {value});
try {
stmt.execute();
} catch (Exception e) {
// Ignore that field, we do not have it in our bean
}
// If the property is the track ID, add the song to the hash
// map
if (SongBean.NAME_TRACK_ID.equals(key)) {
int trackID = Integer.valueOf((String) value);
songMap.put(trackID, song);
}
}
// Update album for last song
addOrUpdateAlbum(albumMap, song, false);
// Check the album map for inconsistencies
Iterator<AlbumBean> albums = albumMap.values().iterator();
while (albums.hasNext()) {
AlbumBean album = albums.next();
if (album.checkConsistency()) {
libraryList.add(album);
album.setHashCode();
} else {
// Add an inconsistent album only using the album title
SongBean[] songs = album.getSongs();
for (int i = 0; i < songs.length; i++) {
addOrUpdateAlbum(secondaryAlbumMap, songs[i], true);
}
}
}
// Check secondary album map for consistency
albums = secondaryAlbumMap.values().iterator();
while (albums.hasNext()) {
AlbumBean album = albums.next();
if (album.checkConsistency()) {
libraryList.add(album);
album.setHashCode();
} else {
// This album cannot be matched
// TODO: Add to warning message
}
}
setChanged();
}
return document;
} catch (IOException ioe) {
// Log an expected connect exception
throw new LibraryException(ioe);
} catch (SAXException se) {
// Catch all other exceptions
throw new LibraryException(se);
} catch (ParserConfigurationException pce) {
// Catch all other exceptions
Utils.logSevere(pce);
throw new LibraryException(pce);
} catch (XPathExpressionException xpe) {
// Catch all other exceptions
Utils.logSevere(xpe);
throw new LibraryException(xpe);
} catch (NumberFormatException nfe) {
// Catch all other exceptions
throw new LibraryException(nfe);
}
}
@Override
public HashSet<ScoredAnnotation> solveSa2W(String text) throws AnnotationException {
HashSet<ScoredAnnotation> res;
try {
res = new HashSet<ScoredAnnotation>();
lastTime = Calendar.getInstance().getTimeInMillis();
URL wikiApi = new URL(url);
String parameters =
"references=true&repeatMode=all&minProbability=0.0&source="
+ URLEncoder.encode(text, "UTF-8");
HttpURLConnection slConnection = (HttpURLConnection) wikiApi.openConnection();
slConnection.setRequestProperty("accept", "text/xml");
slConnection.setDoOutput(true);
slConnection.setDoInput(true);
slConnection.setRequestMethod("POST");
slConnection.setRequestProperty("Content-Type", "application/x-www-form-urlencoded");
slConnection.setRequestProperty("charset", "utf-8");
slConnection.setRequestProperty(
"Content-Length", "" + Integer.toString(parameters.getBytes().length));
slConnection.setUseCaches(false);
DataOutputStream wr = new DataOutputStream(slConnection.getOutputStream());
wr.writeBytes(parameters);
wr.flush();
wr.close();
DocumentBuilderFactory factory = DocumentBuilderFactory.newInstance();
DocumentBuilder builder = factory.newDocumentBuilder();
Document doc = builder.parse(slConnection.getInputStream());
/* URL wikiApi = new URL(url+"?references=true&repeatMode=all&minProbability=0.0&source="+URLEncoder.encode(text, "UTF-8"));
URLConnection wikiConnection = wikiApi.openConnection();
DocumentBuilderFactory factory = DocumentBuilderFactory.newInstance();
DocumentBuilder builder = factory.newDocumentBuilder();
Document doc = builder.parse(wikiConnection.getInputStream());
*/
lastTime = Calendar.getInstance().getTimeInMillis() - lastTime;
XPathFactory xPathfactory = XPathFactory.newInstance();
XPath xpath = xPathfactory.newXPath();
XPathExpression idExpr = xpath.compile("//detectedTopic/@id");
XPathExpression weightExpr = xpath.compile("//detectedTopic/@weight");
XPathExpression referenceExpr = xpath.compile("//detectedTopic/references");
NodeList ids = (NodeList) idExpr.evaluate(doc, XPathConstants.NODESET);
NodeList weights = (NodeList) weightExpr.evaluate(doc, XPathConstants.NODESET);
NodeList references = (NodeList) referenceExpr.evaluate(doc, XPathConstants.NODESET);
for (int i = 0; i < weights.getLength(); i++) {
if (weights.item(i).getNodeType() != Node.TEXT_NODE) {
int id = Integer.parseInt(ids.item(i).getNodeValue());
float weight = Float.parseFloat(weights.item(i).getNodeValue());
// System.out.println("ID="+ids.item(i).getNodeValue()+" weight="+weight);
XPathExpression startExpr =
xpath.compile("//detectedTopic[@id=" + id + "]/references/reference/@start");
XPathExpression endExpr =
xpath.compile("//detectedTopic[@id=" + id + "]/references/reference/@end");
NodeList starts =
(NodeList) startExpr.evaluate(references.item(i), XPathConstants.NODESET);
NodeList ends = (NodeList) endExpr.evaluate(references.item(i), XPathConstants.NODESET);
for (int j = 0; j < starts.getLength(); j++) {
int start = Integer.parseInt(starts.item(j).getNodeValue());
int end = Integer.parseInt(ends.item(j).getNodeValue());
int len = end - start;
res.add(new ScoredAnnotation(start, len, id, weight));
}
}
}
} catch (Exception e) {
e.printStackTrace();
throw new AnnotationException(
"An error occurred while querying Wikipedia Miner API. Message: " + e.getMessage());
}
return res;
}