obtained gcc-4.6.4.tar.bz2 from upstream website;upstream

verified gcc-4.6.4.tar.bz2.sig;
imported gcc-4.6.4 source tree from verified upstream tarball.

downloading a git-generated archive based on the 'upstream' tag
should provide you with a source tree that is binary identical
to the one extracted from the above tarball.

if you have obtained the source via the command 'git clone',
however, do note that line-endings of files in your working
directory might differ from line-endings of the respective
files in the upstream repository.

1 files changed, 1010 insertions, 0 deletions

diff --git a/libjava/classpath/java/net/URL.java b/libjava/classpath/java/net/URL.java
new file mode 100644
index 000000000..1b778b40a
--- /dev/null
+++ b/libjava/classpath/java/net/URL.java
@@ -0,0 +1,1010 @@
+/* URL.java -- Uniform Resource Locator Class
+   Copyright (C) 1998, 1999, 2000, 2002, 2003, 2004, 2005, 2006
+   Free Software Foundation, Inc.
+
+This file is part of GNU Classpath.
+
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING.  If not, write to the
+Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+02110-1301 USA.
+
+Linking this library statically or dynamically with other modules is
+making a combined work based on this library.  Thus, the terms and
+conditions of the GNU General Public License cover the whole
+combination.
+
+As a special exception, the copyright holders of this library give you
+permission to link this library with independent modules to produce an
+executable, regardless of the license terms of these independent
+modules, and to copy and distribute the resulting executable under
+terms of your choice, provided that you also meet, for each linked
+independent module, the terms and conditions of the license of that
+module.  An independent module is a module which is not derived from
+or based on this library.  If you modify this library, you may extend
+this exception to your version of the library, but you are not
+obligated to do so.  If you do not wish to do so, delete this
+exception statement from your version. */
+
+package java.net;
+
+import gnu.classpath.SystemProperties;
+import gnu.java.net.URLParseError;
+
+import java.io.IOException;
+import java.io.InputStream;
+import java.io.ObjectInputStream;
+import java.io.ObjectOutputStream;
+import java.io.Serializable;
+import java.security.AccessController;
+import java.security.PrivilegedAction;
+import java.util.HashMap;
+import java.util.StringTokenizer;
+
+
+/*
+ * Written using on-line Java Platform 1.2 API Specification, as well
+ * as "The Java Class Libraries", 2nd edition (Addison-Wesley, 1998).
+ * Status:  Believed complete and correct.
+ */
+
+/**
+  * This final class represents an Internet Uniform Resource Locator (URL).
+  * For details on the syntax of URL's and what they can be used for,
+  * refer to RFC 1738, available from <a
+  * href="http://ds.internic.net/rfcs/rfc1738.txt">
+  * http://ds.internic.net/rfcs/rfc1738.txt</a>
+  * <p>
+  * There are a great many protocols supported by URL's such as "http",
+  * "ftp", and "file".  This object can handle any arbitrary URL for which
+  * a URLStreamHandler object can be written.  Default protocol handlers
+  * are provided for the "http" and "ftp" protocols.  Additional protocols
+  * handler implementations may be provided in the future.  In any case,
+  * an application or applet can install its own protocol handlers that
+  * can be "chained" with other protocol hanlders in the system to extend
+  * the base functionality provided with this class. (Note, however, that
+  * unsigned applets cannot access properties by default or install their
+  * own protocol handlers).
+  * <p>
+  * This chaining is done via the system property java.protocol.handler.pkgs
+  * If this property is set, it is assumed to be a "|" separated list of
+  * package names in which to attempt locating protocol handlers.  The
+  * protocol handler is searched for by appending the string
+  * ".&lt;protocol&gt;.Handler" to each packed in the list until a hander is
+  * found. If a protocol handler is not found in this list of packages, or if
+  * the property does not exist, then the default protocol handler of
+  * "gnu.java.net.&lt;protocol&gt;.Handler" is tried.  If this is
+  * unsuccessful, a MalformedURLException is thrown.
+  * <p>
+  * All of the constructor methods of URL attempt to load a protocol
+  * handler and so any needed protocol handlers must be installed when
+  * the URL is constructed.
+  * <p>
+  * Here is an example of how URL searches for protocol handlers.  Assume
+  * the value of java.protocol.handler.pkgs is "com.foo|com.bar" and the
+  * URL is "news://comp.lang.java.programmer".  URL would looking the
+  * following places for protocol handlers:
+  * <p><pre>
+  * com.foo.news.Handler
+  * com.bar.news.Handler
+  * gnu.java.net.news.Handler
+  * </pre><p>
+  * If the protocol handler is not found in any of those locations, a
+  * MalformedURLException would be thrown.
+  * <p>
+  * Please note that a protocol handler must be a subclass of
+  * URLStreamHandler.
+  * <p>
+  * Normally, this class caches protocol handlers.  Once it finds a handler
+  * for a particular protocol, it never tries to look up a new handler
+  * again.  However, if the system property
+  * gnu.java.net.nocache_protocol_handlers is set, then this
+  * caching behavior is disabled.  This property is specific to this
+  * implementation.  Sun's JDK may or may not do protocol caching, but it
+  * almost certainly does not examine this property.
+  * <p>
+  * Please also note that an application can install its own factory for
+  * loading protocol handlers (see setURLStreamHandlerFactory).  If this is
+  * done, then the above information is superseded and the behavior of this
+  * class in loading protocol handlers is dependent on that factory.
+  *
+  * @author Aaron M. Renn (arenn@urbanophile.com)
+  * @author Warren Levy (warrenl@cygnus.com)
+  *
+  * @see URLStreamHandler
+  */
+public final class URL implements Serializable
+{
+  private static final String DEFAULT_SEARCH_PATH =
+    "gnu.java.net.protocol|gnu.inet";
+
+  // Cached System ClassLoader
+  private static ClassLoader systemClassLoader;
+
+  /**
+   * The name of the protocol for this URL.
+   * The protocol is always stored in lower case.
+   */
+  private String protocol;
+
+  /**
+   * The "authority" portion of the URL.
+   */
+  private String authority;
+
+  /**
+   * The hostname or IP address of this protocol.
+   * This includes a possible user. For example <code>joe@some.host.net</code>.
+   */
+  private String host;
+
+  /**
+   * The user information necessary to establish the connection.
+   */
+  private String userInfo;
+
+  /**
+   * The port number of this protocol or -1 if the port number used is
+   * the default for this protocol.
+   */
+  private int port = -1; // Initialize for constructor using context.
+
+  /**
+   * The "file" portion of the URL. It is defined as <code>path[?query]</code>.
+   */
+  private String file;
+
+  /**
+   * The anchor portion of the URL.
+   */
+  private String ref;
+
+  /**
+   * This is the hashCode for this URL
+   */
+  private int hashCode;
+
+  /**
+   * The protocol handler in use for this URL
+   */
+  transient URLStreamHandler ph;
+
+  /**
+   * If an application installs its own protocol handler factory, this is
+   * where we keep track of it.
+   */
+  private static URLStreamHandlerFactory factory;
+  private static final long serialVersionUID = -7627629688361524110L;
+
+  /**
+   * This a table where we cache protocol handlers to avoid the overhead
+   * of looking them up each time.
+   */
+  private static HashMap<String, URLStreamHandler> ph_cache
+    = new HashMap<String, URLStreamHandler>();
+
+  /**
+   * Whether or not to cache protocol handlers.
+   */
+  private static boolean cache_handlers;
+
+  static
+    {
+      String s = SystemProperties.getProperty("gnu.java.net.nocache_protocol_handlers");
+
+      if (s == null)
+        cache_handlers = true;
+      else
+        cache_handlers = false;
+    }
+
+  /**
+   * Constructs a URL and loads a protocol handler for the values passed as
+   * arguments.
+   *
+   * @param protocol The protocol for this URL ("http", "ftp", etc)
+   * @param host The hostname or IP address to connect to
+   * @param port The port number to use, or -1 to use the protocol's
+   * default port
+   * @param file The "file" portion of the URL.
+   *
+   * @exception MalformedURLException If a protocol handler cannot be loaded or
+   * a parse error occurs.
+   */
+  public URL(String protocol, String host, int port, String file)
+    throws MalformedURLException
+  {
+    this(protocol, host, port, file, null);
+  }
+
+  /**
+   * Constructs a URL and loads a protocol handler for the values passed in
+   * as arugments.  Uses the default port for the protocol.
+   *
+   * @param protocol The protocol for this URL ("http", "ftp", etc)
+   * @param host The hostname or IP address for this URL
+   * @param file The "file" portion of this URL.
+   *
+   * @exception MalformedURLException If a protocol handler cannot be loaded or
+   * a parse error occurs.
+   */
+  public URL(String protocol, String host, String file)
+    throws MalformedURLException
+  {
+    this(protocol, host, -1, file, null);
+  }
+
+  /**
+   * This method initializes a new instance of <code>URL</code> with the
+   * specified protocol, host, port, and file.  Additionally, this method
+   * allows the caller to specify a protocol handler to use instead of
+   * the default.  If this handler is specified, the caller must have
+   * the "specifyStreamHandler" permission (see <code>NetPermission</code>)
+   * or a <code>SecurityException</code> will be thrown.
+   *
+   * @param protocol The protocol for this URL ("http", "ftp", etc)
+   * @param host The hostname or IP address to connect to
+   * @param port The port number to use, or -1 to use the protocol's default
+   * port
+   * @param file The "file" portion of the URL.
+   * @param ph The protocol handler to use with this URL.
+   *
+   * @exception MalformedURLException If no protocol handler can be loaded
+   * for the specified protocol.
+   * @exception SecurityException If the <code>SecurityManager</code> exists
+   * and does not allow the caller to specify its own protocol handler.
+   *
+   * @since 1.2
+   */
+  public URL(String protocol, String host, int port, String file,
+             URLStreamHandler ph) throws MalformedURLException
+  {
+    if (protocol == null)
+      throw new MalformedURLException("null protocol");
+    protocol = protocol.toLowerCase();
+    this.protocol = protocol;
+
+    if (ph != null)
+      {
+        SecurityManager s = System.getSecurityManager();
+        if (s != null)
+          s.checkPermission(new NetPermission("specifyStreamHandler"));
+
+        this.ph = ph;
+      }
+    else
+      this.ph = getURLStreamHandler(protocol);
+
+    if (this.ph == null)
+      throw new MalformedURLException("Protocol handler not found: "
+                                      + protocol);
+
+    this.host = host;
+    this.port = port;
+    this.authority = (host != null) ? host : "";
+    if (port >= 0 && host != null)
+        this.authority += ":" + port;
+
+    int hashAt = file.indexOf('#');
+    if (hashAt < 0)
+      {
+        this.file = file;
+        this.ref = null;
+      }
+    else
+      {
+        this.file = file.substring(0, hashAt);
+        this.ref = file.substring(hashAt + 1);
+      }
+    hashCode = hashCode(); // Used for serialization.
+  }
+
+  /**
+   * Initializes a URL from a complete string specification such as
+   * "http://www.urbanophile.com/arenn/".  First the protocol name is parsed
+   * out of the string.  Then a handler is located for that protocol and
+   * the parseURL() method of that protocol handler is used to parse the
+   * remaining fields.
+   *
+   * @param spec The complete String representation of a URL
+   *
+   * @exception MalformedURLException If a protocol handler cannot be found
+   * or the URL cannot be parsed
+   */
+  public URL(String spec) throws MalformedURLException
+  {
+    this((URL) null, spec != null ? spec : "", (URLStreamHandler) null,
+         false);
+  }
+
+  /**
+   * This method parses a String representation of a URL within the
+   * context of an existing URL.  Principally this means that any
+   * fields not present the URL are inheritied from the context URL.
+   * This allows relative URL's to be easily constructed.  If the
+   * context argument is null, then a complete URL must be specified
+   * in the URL string.  If the protocol parsed out of the URL is
+   * different from the context URL's protocol, then then URL String
+   * is also expected to be a complete URL.
+   *
+   * @param context The context on which to parse the specification
+   * @param spec The string to parse an URL
+   *
+   * @exception MalformedURLException If a protocol handler cannot be found
+   * for the URL cannot be parsed
+   */
+  public URL(URL context, String spec) throws MalformedURLException
+  {
+    this(context, spec,
+         (context == null) ? (URLStreamHandler) null : context.ph,
+         false);
+  }
+
+  /**
+   * Creates an URL from given arguments
+   * This method parses a String representation of a URL within the
+   * context of an existing URL.  Principally this means that any fields
+   * not present the URL are inheritied from the context URL.  This allows
+   * relative URL's to be easily constructed.  If the context argument is
+   * null, then a complete URL must be specified in the URL string.
+   * If the protocol parsed out of the URL is different
+   * from the context URL's protocol, then then URL String is also
+   * expected to be a complete URL.
+   * <p>
+   * Additionally, this method allows the caller to specify a protocol handler
+   * to use instead of  the default.  If this handler is specified, the caller
+   * must have the "specifyStreamHandler" permission
+   * (see <code>NetPermission</code>) or a <code>SecurityException</code>
+   * will be thrown.
+   *
+   * @param context The context in which to parse the specification
+   * @param spec The string to parse as an URL
+   * @param ph The stream handler for the URL
+   *
+   * @exception MalformedURLException If a protocol handler cannot be found
+   * or the URL cannot be parsed
+   * @exception SecurityException If the <code>SecurityManager</code> exists
+   * and does not allow the caller to specify its own protocol handler.
+   *
+   * @since 1.2
+   */
+  public URL(URL context, String spec, URLStreamHandler ph)
+    throws MalformedURLException
+  {
+    this(context, spec, ph, true);
+  }
+
+  /**
+   * Private constructor called by all other constructors taking
+   * a context and spec.
+   *
+   * @param context The context in which to parse the specification
+   * @param spec The string to parse as an URL
+   * @param ph The stream handler for the URL
+   * @param phFromUser Whether or not the user supplied the URLStreamHandler
+   *
+   */
+  private URL(URL context, String spec, URLStreamHandler ph,
+              boolean phFromUser)
+    throws MalformedURLException
+  {
+    /* A protocol is defined by the doc as the substring before a ':'
+     * as long as the ':' occurs before any '/'.
+     *
+     * If context is null, then spec must be an absolute URL.
+     *
+     * The relative URL need not specify all the components of a URL.
+     * If the protocol, host name, or port number is missing, the value
+     * is inherited from the context.  A bare file component is appended
+     * to the context's file.  The optional anchor is not inherited.
+     */
+
+    // If this is an absolute URL, then ignore context completely.
+    // An absolute URL must have chars prior to "://" but cannot have a colon
+    // right after the "://".  The second colon is for an optional port value
+    // and implies that the host from the context is used if available.
+    int colon;
+    int slash = spec.indexOf('/');
+    if ((colon = spec.indexOf("://", 1)) > 0
+        && ((colon < slash || slash < 0))
+        && ! spec.regionMatches(colon, "://:", 0, 4))
+      {
+        context = null;
+        if (! phFromUser)
+          ph = null;
+      }
+
+    boolean protocolSpecified = false;
+
+    if ((colon = spec.indexOf(':')) > 0
+        && (colon < slash || slash < 0))
+      {
+        // Protocol may have been specified in spec string.
+        protocolSpecified = true;
+        protocol = spec.substring(0, colon).toLowerCase();
+        if (context != null)
+          {
+            if (context.protocol.equals(protocol))
+              {
+                // The 1.2 doc specifically says these are copied to the new URL.
+                host = context.host;
+                port = context.port;
+                userInfo = context.userInfo;
+                authority = context.authority;
+              }
+            else
+              {
+                // There was a colon in the spec.  Check to see if
+                // what precedes it is a valid protocol.  If it was
+                // not, assume that it is relative to the context.
+                URLStreamHandler specPh = getURLStreamHandler(protocol.trim());
+                if (null == specPh)
+                    protocolSpecified = false;
+              }
+          }
+      }
+
+    if (!protocolSpecified)
+      {
+        if (context != null)
+          {
+            // Protocol NOT specified in spec string.
+            // Use context fields (except ref) as a foundation for relative URLs.
+            colon = -1;
+            protocol = context.protocol;
+            host = context.host;
+            port = context.port;
+            userInfo = context.userInfo;
+            if (spec.indexOf(":/", 1) < 0)
+              {
+                file = context.file;
+                if (file == null || file.length() == 0)
+                  file = "/";
+              }
+            authority = context.authority;
+          }
+        else // Protocol NOT specified in spec. and no context available.
+          throw new MalformedURLException("Absolute URL required with null"
+                                          + " context: " + spec);
+      }
+
+    protocol = protocol.trim();
+
+    if (ph != null)
+      {
+        SecurityManager s = System.getSecurityManager();
+        if (s != null && phFromUser)
+          s.checkPermission(new NetPermission("specifyStreamHandler"));
+
+        this.ph = ph;
+      }
+    else
+      this.ph = getURLStreamHandler(protocol);
+
+    if (this.ph == null)
+      throw new MalformedURLException("Protocol handler not found: "
+                                      + protocol);
+
+    // JDK 1.2 doc for parseURL specifically states that any '#' ref
+    // is to be excluded by passing the 'limit' as the indexOf the '#'
+    // if one exists, otherwise pass the end of the string.
+    int hashAt = spec.indexOf('#', colon + 1);
+
+    try
+      {
+        this.ph.parseURL(this, spec, colon + 1,
+                         hashAt < 0 ? spec.length() : hashAt);
+      }
+    catch (URLParseError e)
+      {
+        MalformedURLException mue = new MalformedURLException(e.getMessage());
+        mue.initCause(e);
+        throw mue;
+      }
+    catch (RuntimeException e)
+      {
+        // This isn't documented, but the JDK also catches
+        // RuntimeExceptions here.
+        MalformedURLException mue = new MalformedURLException(e.getMessage());
+        mue.initCause(e);
+        throw mue;
+      }
+
+    if (hashAt >= 0)
+      ref = spec.substring(hashAt + 1);
+
+    hashCode = hashCode(); // Used for serialization.
+  }
+
+  /**
+   * Test another URL for equality with this one.  This will be true only if
+   * the argument is non-null and all of the fields in the URL's match
+   * exactly (ie, protocol, host, port, file, and ref).  Overrides
+   * Object.equals(), implemented by calling the equals method of the handler.
+   *
+   * @param obj The URL to compare with
+   *
+   * @return true if the URL is equal, false otherwise
+   */
+  public boolean equals(Object obj)
+  {
+    if (! (obj instanceof URL))
+      return false;
+
+    return ph.equals(this, (URL) obj);
+  }
+
+  /**
+   * Returns the contents of this URL as an object by first opening a
+   * connection, then calling the getContent() method against the connection
+   *
+   * @return A content object for this URL
+   * @exception IOException If opening the connection or getting the
+   * content fails.
+   *
+   * @since 1.3
+   */
+  public Object getContent() throws IOException
+  {
+    return openConnection().getContent();
+  }
+
+  /**
+   * Gets the contents of this URL
+   *
+   * @param classes The allow classes for the content object.
+   *
+   * @return a context object for this URL.
+   *
+   * @exception IOException If an error occurs
+   */
+  public Object getContent(Class[] classes) throws IOException
+  {
+    return openConnection().getContent(classes);
+  }
+
+  /**
+   * Returns the file portion of the URL.
+   * Defined as <code>path[?query]</code>.
+   * Returns the empty string if there is no file portion.
+   *
+   * @return The filename specified in this URL, or an empty string if empty.
+   */
+  public String getFile()
+  {
+    return file == null ? "" : file;
+  }
+
+  /**
+   * Returns the path of the URL. This is the part of the file before any '?'
+   * character.
+   *
+   * @return The path specified in this URL, or null if empty.
+   *
+   * @since 1.3
+   */
+  public String getPath()
+  {
+    // The spec says we need to return an empty string, but some
+    // applications depends on receiving null when the path is empty.
+    if (file == null)
+      return null;
+    int quest = file.indexOf('?');
+    return quest < 0 ? getFile() : file.substring(0, quest);
+  }
+
+  /**
+   * Returns the authority of the URL
+   *
+   * @return The authority specified in this URL.
+   *
+   * @since 1.3
+   */
+  public String getAuthority()
+  {
+    return authority;
+  }
+
+  /**
+   * Returns the host of the URL
+   *
+   * @return The host specified in this URL.
+   */
+  public String getHost()
+  {
+    int at = (host == null) ? -1 : host.indexOf('@');
+    return at < 0 ? host : host.substring(at + 1, host.length());
+  }
+
+  /**
+   * Returns the port number of this URL or -1 if the default port number is
+   * being used.
+   *
+   * @return The port number
+   *
+   * @see #getDefaultPort()
+   */
+  public int getPort()
+  {
+    return port;
+  }
+
+  /**
+   * Returns the default port of the URL. If the StreamHandler for the URL
+   * protocol does not define a default port it returns -1.
+   *
+   * @return The default port of the current protocol.
+   */
+  public int getDefaultPort()
+  {
+    return ph.getDefaultPort();
+  }
+
+  /**
+   * Returns the protocol of the URL
+   *
+   * @return The specified protocol.
+   */
+  public String getProtocol()
+  {
+    return protocol;
+  }
+
+  /**
+   * Returns the ref (sometimes called the "# reference" or "anchor") portion
+   * of the URL.
+   *
+   * @return The ref
+   */
+  public String getRef()
+  {
+    return ref;
+  }
+
+  /**
+   * Returns the user information of the URL. This is the part of the host
+   * name before the '@'.
+   *
+   * @return the user at a particular host or null when no user defined.
+   */
+  public String getUserInfo()
+  {
+    if (userInfo != null)
+      return userInfo;
+    int at = (host == null) ? -1 : host.indexOf('@');
+    return at < 0 ? null : host.substring(0, at);
+  }
+
+  /**
+   * Returns the query of the URL. This is the part of the file before the
+   * '?'.
+   *
+   * @return the query part of the file, or null when there is no query part.
+   */
+  public String getQuery()
+  {
+    int quest = (file == null) ? -1 : file.indexOf('?');
+    return quest < 0 ? null : file.substring(quest + 1, file.length());
+  }
+
+  /**
+   * Returns a hashcode computed by the URLStreamHandler of this URL
+   *
+   * @return The hashcode for this URL.
+   */
+  public int hashCode()
+  {
+    if (hashCode != 0)
+      return hashCode; // Use cached value if available.
+    else
+      return ph.hashCode(this);
+  }
+
+  /**
+   * Returns a URLConnection object that represents a connection to the remote
+   * object referred to by the URL. The URLConnection is created by calling the
+   * openConnection() method of the protocol handler
+   *
+   * @return A URLConnection for this URL
+   *
+   * @exception IOException If an error occurs
+   */
+  public URLConnection openConnection() throws IOException
+  {
+    return ph.openConnection(this);
+  }
+
+  /**
+   * Opens a connection to this URL and returns an InputStream for reading
+   * from that connection
+   *
+   * @return An <code>InputStream</code> for this URL.
+   *
+   * @exception IOException If an error occurs
+   */
+  public InputStream openStream() throws IOException
+  {
+    return openConnection().getInputStream();
+  }
+
+  /**
+   * Tests whether or not another URL refers to the same "file" as this one.
+   * This will be true if and only if the passed object is not null, is a
+   * URL, and matches all fields but the ref (ie, protocol, host, port,
+   * and file);
+   *
+   * @param url The URL object to test with
+   *
+   * @return true if URL matches this URL's file, false otherwise
+   */
+  public boolean sameFile(URL url)
+  {
+    return ph.sameFile(this, url);
+  }
+
+  /**
+   * Sets the specified fields of the URL. This is not a public method so
+   * that only URLStreamHandlers can modify URL fields. This might be called
+   * by the <code>parseURL()</code> method in that class. URLs are otherwise
+   * constant. If the given protocol does not exist, it will keep the previously
+   * set protocol.
+   *
+   * @param protocol The protocol name for this URL
+   * @param host The hostname or IP address for this URL
+   * @param port The port number of this URL
+   * @param file The "file" portion of this URL.
+   * @param ref The anchor portion of this URL.
+   */
+  protected void set(String protocol, String host, int port, String file,
+                     String ref)
+  {
+    URLStreamHandler protocolHandler = null;
+    protocol = protocol.toLowerCase();
+    if (! this.protocol.equals(protocol))
+      protocolHandler = getURLStreamHandler(protocol);
+
+    // It is an hidden feature of the JDK. If the protocol does not exist,
+    // we keep the previously initialized protocol.
+    if (protocolHandler != null)
+      {
+        this.ph = protocolHandler;
+        this.protocol = protocol;
+      }
+    this.authority = "";
+    this.port = port;
+    this.host = host;
+    this.file = file;
+    this.ref = ref;
+
+    if (host != null)
+      this.authority += host;
+    if (port >= 0)
+      this.authority += ":" + port;
+
+    hashCode = hashCode(); // Used for serialization.
+  }
+
+  /**
+   * Sets the specified fields of the URL. This is not a public method so
+   * that only URLStreamHandlers can modify URL fields. URLs are otherwise
+   * constant. If the given protocol does not exist, it will keep the previously
+   * set protocol.
+   *
+   * @param protocol The protocol name for this URL.
+   * @param host The hostname or IP address for this URL.
+   * @param port The port number of this URL.
+   * @param authority The authority of this URL.
+   * @param userInfo The user and password (if needed) of this URL.
+   * @param path The "path" portion of this URL.
+   * @param query The query of this URL.
+   * @param ref The anchor portion of this URL.
+   *
+   * @since 1.3
+   */
+  protected void set(String protocol, String host, int port, String authority,
+                     String userInfo, String path, String query, String ref)
+  {
+    URLStreamHandler protocolHandler = null;
+    protocol = protocol.toLowerCase();
+    if (! this.protocol.equals(protocol))
+      protocolHandler = getURLStreamHandler(protocol);
+
+    // It is an hidden feature of the JDK. If the protocol does not exist,
+    // we keep the previously initialized protocol.
+    if (protocolHandler != null)
+      {
+        this.ph = protocolHandler;
+        this.protocol = protocol;
+      }
+    this.host = host;
+    this.userInfo = userInfo;
+    this.port = port;
+    this.authority = authority;
+    if (query == null)
+      this.file = path;
+    else
+      this.file = path + "?" + query;
+    this.ref = ref;
+    hashCode = hashCode(); // Used for serialization.
+  }
+
+  /**
+   * Sets the URLStreamHandlerFactory for this class.  This factory is
+   * responsible for returning the appropriate protocol handler for
+   * a given URL.
+   *
+   * @param fac The URLStreamHandlerFactory class to use
+   *
+   * @exception Error If the factory is alread set.
+   * @exception SecurityException If a security manager exists and its
+   * checkSetFactory method doesn't allow the operation
+   */
+  public static synchronized void setURLStreamHandlerFactory(URLStreamHandlerFactory fac)
+  {
+    if (factory != null)
+      throw new Error("URLStreamHandlerFactory already set");
+
+    // Throw an exception if an extant security mgr precludes
+    // setting the factory.
+    SecurityManager s = System.getSecurityManager();
+    if (s != null)
+      s.checkSetFactory();
+    factory = fac;
+  }
+
+  /**
+   * Returns a String representing this URL.  The String returned is
+   * created by calling the protocol handler's toExternalForm() method.
+   *
+   * @return A string for this URL
+   */
+  public String toExternalForm()
+  {
+    // Identical to toString().
+    return ph.toExternalForm(this);
+  }
+
+  /**
+   * Returns a String representing this URL.  Identical to toExternalForm().
+   * The value returned is created by the protocol handler's
+   * toExternalForm method.  Overrides Object.toString()
+   *
+   * @return A string for this URL
+   */
+  public String toString()
+  {
+    // Identical to toExternalForm().
+    return ph.toExternalForm(this);
+  }
+
+  /**
+   * This internal method is used in two different constructors to load
+   * a protocol handler for this URL.
+   *
+   * @param protocol The protocol to load a handler for
+   *
+   * @return A URLStreamHandler for this protocol, or null when not found.
+   */
+  private static synchronized URLStreamHandler getURLStreamHandler(String protocol)
+  {
+    URLStreamHandler ph = null;
+
+    // First, see if a protocol handler is in our cache.
+    if (cache_handlers)
+      {
+        if ((ph = ph_cache.get(protocol)) != null)
+          return ph;
+      }
+
+    // If a non-default factory has been set, use it to find the protocol.
+    if (factory != null)
+      {
+        ph = factory.createURLStreamHandler(protocol);
+      }
+
+    // Non-default factory may have returned null or a factory wasn't set.
+    // Use the default search algorithm to find a handler for this protocol.
+    if (ph == null)
+      {
+        // Get the list of packages to check and append our default handler
+        // to it, along with the JDK specified default as a last resort.
+        // Except in very unusual environments the JDK specified one shouldn't
+        // ever be needed (or available).
+        String ph_search_path =
+          SystemProperties.getProperty("java.protocol.handler.pkgs");
+
+        // Tack our default package on at the ends.
+        if (ph_search_path != null)
+          ph_search_path += "|" + DEFAULT_SEARCH_PATH;
+        else
+          ph_search_path = DEFAULT_SEARCH_PATH;
+
+        // Finally loop through our search path looking for a match.
+        StringTokenizer pkgPrefix = new StringTokenizer(ph_search_path, "|");
+
+        // Cache the systemClassLoader
+        if (systemClassLoader == null)
+          {
+            systemClassLoader = AccessController.doPrivileged
+              (new PrivilegedAction<ClassLoader>() {
+                  public ClassLoader run()
+                  {
+                    return ClassLoader.getSystemClassLoader();
+                  }
+                });
+          }
+
+        do
+          {
+            try
+              {
+                // Try to get a class from the system/application
+                // classloader, initialize it, make an instance
+                // and try to cast it to a URLStreamHandler.
+                String clsName =
+                  (pkgPrefix.nextToken() + "." + protocol + ".Handler");
+                Class c = Class.forName(clsName, true, systemClassLoader);
+                ph = (URLStreamHandler) c.newInstance();
+              }
+            catch (ThreadDeath death)
+              {
+                throw death;
+              }
+            catch (Throwable t)
+              {
+                // Ignored.
+              }
+          }
+         while (ph == null && pkgPrefix.hasMoreTokens());
+      }
+
+    // Update the hashtable with the new protocol handler.
+    if (ph != null && cache_handlers)
+      ph_cache.put(protocol, ph);
+    else
+      ph = null;
+
+    return ph;
+  }
+
+  private void readObject(ObjectInputStream ois)
+    throws IOException, ClassNotFoundException
+  {
+    ois.defaultReadObject();
+    this.ph = getURLStreamHandler(protocol);
+    if (this.ph == null)
+      throw new IOException("Handler for protocol " + protocol + " not found");
+  }
+
+  private void writeObject(ObjectOutputStream oos) throws IOException
+  {
+    oos.defaultWriteObject();
+  }
+
+  /**
+   * Returns the equivalent <code>URI</code> object for this <code>URL</code>.
+   * This is the same as calling <code>new URI(this.toString())</code>.
+   * RFC2396-compliant URLs are guaranteed a successful conversion to
+   * a <code>URI</code> instance.  However, there are some values which
+   * form valid URLs, but which do not also form RFC2396-compliant URIs.
+   *
+   * @throws URISyntaxException if this URL is not RFC2396-compliant,
+   *         and thus can not be successfully converted to a URI.
+   */
+  public URI toURI()
+    throws URISyntaxException
+  {
+    return new URI(toString());
+  }
+
+}