/**
* Escapes instances of File.separator from the query. These are safe from filename overlap, but
* can't convert into extended paths and directories.
*
* @param query the query
* @return the escaped query
*/
static String escapeQuery(String query) {
if (query.indexOf(File.separator) >= 0) {
return StringUtil.replaceString(query, File.separator, ESCAPE_STR + ENCODED_SEPARATOR_CHAR);
} else {
return query;
}
}
/**
* Escapes instances of the ESCAPE_CHAR from the path. This avoids name conflicts with the
* repository files, such as '#nodestate.xml'.
*
* @param path the path
* @return the escaped path
*/
static String escapePath(String path) {
// XXX escaping disabled because of URL encoding
if (false && path.indexOf(ESCAPE_CHAR) >= 0) {
return StringUtil.replaceString(path, ESCAPE_STR, ESCAPE_STR + ESCAPE_STR);
} else {
return path;
}
}
/**
* mapUrlToFileLocation() is the method used to resolve urls into file names. This maps a given
* url to a file location, using the au top directory as the base. It creates directories which
* mirror the html string, so 'http://www.journal.org/issue1/index.html' would be cached in the
* file: <rootLocation>/www.journal.org/http/issue1/index.html
*
* @param rootLocation the top directory for ArchivalUnit this URL is in
* @param urlStr the url to translate
* @return the url file location
* @throws java.net.MalformedURLException
*/
public static String mapUrlToFileLocation(String rootLocation, String urlStr)
throws MalformedURLException {
int totalLength = rootLocation.length() + urlStr.length();
URL url = new URL(urlStr);
StringBuilder buffer = new StringBuilder(totalLength);
buffer.append(rootLocation);
if (!rootLocation.endsWith(File.separator)) {
buffer.append(File.separator);
}
buffer.append(url.getHost().toLowerCase());
int port = url.getPort();
if (port != -1) {
buffer.append(PORT_SEPARATOR);
buffer.append(port);
}
buffer.append(File.separator);
buffer.append(url.getProtocol());
if (RepositoryManager.isEnableLongComponents()) {
String escapedPath =
escapePath(
StringUtil.replaceString(url.getPath(), UrlUtil.URL_PATH_SEPARATOR, File.separator));
String query = url.getQuery();
if (query != null) {
escapedPath = escapedPath + "?" + escapeQuery(query);
}
String encodedPath = RepositoryNodeImpl.encodeUrl(escapedPath);
// encodeUrl strips leading / from path
buffer.append(File.separator);
buffer.append(encodedPath);
} else {
buffer.append(
escapePath(
StringUtil.replaceString(url.getPath(), UrlUtil.URL_PATH_SEPARATOR, File.separator)));
String query = url.getQuery();
if (query != null) {
buffer.append("?");
buffer.append(escapeQuery(query));
}
}
return buffer.toString();
}