private byte[] downloadByteArray(URL url) throws IOException {
HttpURLConnection conn = (HttpURLConnection) url.openConnection();
conn.setRequestProperty("Cookie", cookies);
if (conn.getResponseCode() == HttpURLConnection.HTTP_NOT_FOUND) return (null);
InputStream in = conn.getInputStream();
byte[] buf = new byte[conn.getContentLength()];
int read, offset = 0;
while ((read = in.read(buf, offset, buf.length - offset)) != -1) {
offset += read;
}
return (buf);
}
private boolean handshake() throws Exception {
URL homePage = new URL("http://mangaonweb.com/viewer.do?ctsn=" + ctsn);
HttpURLConnection urlConn = (HttpURLConnection) homePage.openConnection();
urlConn.connect();
if (urlConn.getResponseCode() == HttpURLConnection.HTTP_NOT_FOUND) return (false);
// save the cookie
String headerName = null;
for (int i = 1; (headerName = urlConn.getHeaderFieldKey(i)) != null; i++) {
if (headerName.equals("Set-Cookie")) {
cookies = urlConn.getHeaderField(i);
}
}
// save cdn and crcod
String page = "", line;
BufferedReader stream =
new BufferedReader(new InputStreamReader(urlConn.getInputStream(), "UTF-8"));
while ((line = stream.readLine()) != null) page += line;
cdn = param(page, "cdn");
crcod = param(page, "crcod");
return (true);
}
@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;
}