Method uri2path

basex-core/src/main/java/org/basex/util/Strings.java:355–385

Converts a URI to a directory path. See https://docs.basex.org/wiki/Repository#URI_Rewriting for details. @param uri namespace URI @return converted path

(final String uri)

   * @return converted path
   */
  public static String uri2path(final String uri) {
    String path = uri;
    try {
      final URI u = new URI(uri);
      final TokenBuilder tb = new TokenBuilder();
      if(u.isOpaque()) {
        tb.add(u.getScheme()).add('/').add(u.getSchemeSpecificPart().replace(':', '/'));
      } else {
        final String auth = u.getAuthority();
        if(auth != null) {
          // reverse authority, replace dots by slashes. example: basex.org → org/basex
          final String[] comp = split(auth, '.');
          for(int c = comp.length - 1; c >= 0; c--) tb.add('/').add(comp[c]);
        }
        // add remaining path
        final String p = u.getPath();
        tb.add(p == null || p.isEmpty() ? "/" : p.replace('.', '/'));
      }
      path = tb.toString();
    } catch(final URISyntaxException ex) {
      Util.debug(ex);
    }

    // replace special characters with dashes; remove multiple slashes
    path = path.replaceAll("[^\\w.-/]+", "-").replaceAll("//+", "/");
    // add "index" string
    if(endsWith(path, '/')) path += "index";
    // remove heading slash
    if(startsWith(path, '/')) path = path.substring(1);
    return path;
  }

  /**
   * Canonicalizes a string: heuristic CP437/UTF-8 mojibake fix followed by Unicode normalization.