/**
* A method to remove any non-canonical '..' or '.' elements in the path, as well as protecting
* against illegal path traversal.
*
* @param url the raw url
* @return String the canonicalized url
* @throws MalformedURLException
*/
public String canonicalizePath(String url) throws MalformedURLException {
String canonUrl = UrlUtil.normalizeUrl(url, UrlUtil.PATH_TRAVERSAL_ACTION_THROW);
// canonicalize "dir" and "dir/"
// XXX if these are ever two separate nodes, this is wrong
if (canonUrl.endsWith(UrlUtil.URL_PATH_SEPARATOR)) {
canonUrl = canonUrl.substring(0, canonUrl.length() - 1);
}
return canonUrl;
}
private void buildUrlSets(String url) {
try {
outputMessage("\nFetching " + url, TEST_SUMMARY_MESSAGE);
URL srcUrl = new URL(url);
// URLConnection conn = srcUrl.openConnection();
// String type = conn.getContentType();
// type = conn.getHeaderField("content-type");
// InputStream istr = conn.getInputStream();
LockssUrlConnection conn = UrlUtil.openConnection(url, connectionPool);
if (proxyHost != null) {
conn.setProxy(proxyHost, proxyPort);
}
if (userAgent != null) {
conn.setRequestProperty("user-agent", userAgent);
}
try {
conn.execute();
int resp = conn.getResponseCode();
if (resp != 200) {
outputMessage("Resp: " + resp + ": " + conn.getResponseMessage(), TEST_SUMMARY_MESSAGE);
return;
}
depth_fetched[m_curDepth - 1]++;
String cookies = conn.getResponseHeaderValue("Set-Cookie");
if (cookies != null) {
outputMessage("Cookies: " + cookies, PLAIN_MESSAGE);
}
String type = conn.getResponseContentType();
if (type == null || !type.toLowerCase().startsWith("text/html")) {
outputMessage("Type: " + type + ", not parsing", URL_SUMMARY_MESSAGE);
return;
}
outputMessage("Type: " + type + ", extracting Urls", URL_SUMMARY_MESSAGE);
InputStream istr = conn.getResponseInputStream();
InputStreamReader reader = new InputStreamReader(istr);
// MyMockCachedUrl mcu = new MyMockCachedUrl(srcUrl.toString(), reader);
GoslingHtmlLinkExtractor extractor = new GoslingHtmlLinkExtractor();
extractor.extractUrls(null, istr, null, srcUrl.toString(), new MyLinkExtractorCallback());
istr.close();
depth_parsed[m_curDepth - 1]++;
} finally {
conn.release();
}
} catch (MalformedURLException murle) {
murle.printStackTrace();
outputErrResults(url, "Malformed URL:" + murle.getMessage());
} catch (IOException ex) {
ex.printStackTrace();
outputErrResults(url, "IOException: " + ex.getMessage());
}
}
public void foundLink(String url) {
m_extracted.add(url);
try {
String normUrl = UrlUtil.normalizeUrl(url);
if (BaseCrawler.isSupportedUrlProtocol(normUrl) && m_au.shouldBeCached(normUrl)) {
m_incls.add(normUrl);
} else {
m_excls.add(normUrl);
}
} catch (MalformedURLException e) {
m_excls.add(url);
}
}