Patch: FYI: PRs 9715, 19132

Patch: FYI: PRs 9715, 19132

[Tom Tromey]

I'm checking this in on the trunk and the 4.1 branch.

This re-merges parts of the nio charset support, adds a couple of
libgcj-style charset converters based on the nio support, and adds
missing constructors to InputStreamReader and OutputStreamWriter.
Along the way this fixes PRs 9715 and 19132.

This is based on the approach first suggested by David Daney some time
ago, and which I inexplicably missed at the time.

Tom

Index: ChangeLog
from  David Daney <ddaney@avtrex.com>
	Tom Tromey  <tromey@redhat.com>

	PR libgcj/9715, PR libgcj/19132:
	* java/nio/charset/Charset.java (charsetForName): Try default
	provider first.
	(availableCharsets): Re-merged.
	(providers2): Likewise.
	(defaultCharset): Likewise.
	* sources.am, Makefile.in: Rebuilt.
	* gnu/java/nio/charset/Provider.java: Removed.
	* java/io/OutputStreamWriter.java
	(OutputStreamWriter(OutputStream,Charset)): New constructor.
	(OutputStreamWriter(OutputStream,CharsetEncoder)): Likewise.
	* java/io/InputStreamReader.java
	(InputStreamReader(InputStream,CharsetDecoder)): New constructor.
	(InputStreamReader(InputStream,Charset)): Likewise.
	* gnu/gcj/convert/BytesToUnicode.java (getDecoder): Try a
	BytesToCharsetAdaptor.
	* gnu/gcj/convert/UnicodeToBytes.java (getEncoder): Try a
	CharsetToBytesAdaptor.
	* gnu/gcj/convert/CharsetToBytesAdaptor.java: New file.
	* gnu/gcj/convert/BytesToCharsetAdaptor.java: New file.
	* mauve-libgcj: Remove getEncoding exclusion.

Index: gnu/java/nio/charset/Provider.java
===================================================================
--- gnu/java/nio/charset/Provider.java	(revision 108955)
+++ gnu/java/nio/charset/Provider.java	(working copy)
@@ -1,154 +0,0 @@
-/* Provider.java -- 
-   Copyright (C) 2002, 2005 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 gnu.java.nio.charset;
-
-import java.nio.charset.Charset;
-import java.nio.charset.spi.CharsetProvider;
-import java.util.Collections;
-import java.util.HashMap;
-import java.util.Iterator;
-
-/**
- * Charset provider for the required charsets.  Used by
- * {@link Charset#charsetForName} and * {@link Charset#availableCharsets}.
- *
- * @author Jesse Rosenstock
- * @author Robert Schuster (thebohemian@gmx.net)
- * @see Charset
- */
-public final class Provider extends CharsetProvider
-{
-  private static Provider singleton;
-
-  static
-  {
-    synchronized (Provider.class)
-      {
-        singleton = null;
-      }
-  }
-
-  /**
-   * Map from charset name to charset canonical name. The strings
-   * are all lower-case to allow case-insensitive retrieval of
-   * Charset instances. 
-   */
-  private final HashMap canonicalNames;
-
-  /**
-   * Map from lower-case canonical name to Charset.
-   * TODO: We may want to use soft references.  We would then need to keep
-   * track of the class name to regenerate the object.
-   */
-  private final HashMap charsets;
-
-  private Provider ()
-  {
-    canonicalNames = new HashMap ();
-    charsets = new HashMap ();
-
-    // US-ASCII aka ISO646-US
-    addCharset (new US_ASCII ());
-
-    // ISO-8859-1 aka ISO-LATIN-1
-    addCharset (new ISO_8859_1 ());
-
-    // UTF-8
-    addCharset (new UTF_8 ());
-
-    // UTF-16BE
-    addCharset (new UTF_16BE ());
-
-    // UTF-16LE
-    addCharset (new UTF_16LE ());
-
-    // UTF-16
-    addCharset (new UTF_16 ());
-  }
-
-  public Iterator charsets ()
-  {
-    return Collections.unmodifiableCollection (charsets.values ())
-                      .iterator ();
-  }
-
-  /**
-   * Returns a Charset instance by converting the given
-   * name to lower-case, looking up the canonical charset
-   * name and finally looking up the Charset with that name.
-   * 
-   * <p>The lookup is therefore case-insensitive.</p>
-   * 
-   *  @returns The Charset having <code>charsetName</code>
-   *  as its alias or null if no such Charset exist.
-   */
-  public Charset charsetForName (String charsetName)
-  {
-    return (Charset) charsets.get(canonicalNames.get(charsetName.toLowerCase()));
-  }
-
-  /**
-   * Puts a Charset under its canonical name into the 'charsets' map.
-   * Then puts a mapping from all its alias names to the canonical name.
-   * 
-   * <p>All names are converted to lower-case</p>.
-   * 
-   * @param cs
-   */
-  private void addCharset (Charset cs)
-  {
-    String canonicalName = cs.name().toLowerCase();
-    charsets.put (canonicalName, cs);
-    
-    /* Adds a mapping between the canonical name
-     * itself making a lookup using that name
-     * no special case.
-     */  
-    canonicalNames.put(canonicalName, canonicalName);
-
-    for (Iterator i = cs.aliases ().iterator (); i.hasNext (); )
-      canonicalNames.put (((String) i.next()).toLowerCase(), canonicalName);
-  }
-
-  public static synchronized Provider provider ()
-  {
-    if (singleton == null)
-      singleton = new Provider ();
-    return singleton;
-  }
-}
Index: gnu/gcj/convert/BytesToCharsetAdaptor.java
===================================================================
--- gnu/gcj/convert/BytesToCharsetAdaptor.java	(revision 0)
+++ gnu/gcj/convert/BytesToCharsetAdaptor.java	(revision 0)
@@ -0,0 +1,90 @@
+/* Copyright (C) 2005  Free Software Foundation
+
+This file is part of libgcj.
+
+This software is copyrighted work licensed under the terms of the
+Libgcj License.  Please consult the file "LIBGCJ_LICENSE" for
+details.  */
+
+package gnu.gcj.convert; 
+
+import java.nio.ByteBuffer;
+import java.nio.CharBuffer;
+import java.nio.charset.Charset;
+import java.nio.charset.CharsetDecoder;
+import java.nio.charset.CodingErrorAction;
+import java.nio.charset.CoderResult;
+import gnu.java.nio.charset.EncodingHelper;
+
+/**
+ * Adaptor class that allow any {@link Charset} to be used
+ * as a BytesToUnicode converter.
+ */
+public class BytesToCharsetAdaptor extends BytesToUnicode
+{
+  /**
+   * The CharsetDecoder that does all the work.
+   */
+  private final CharsetDecoder decoder;
+
+  /**
+   * ByteBuffer wrapper for this.buf.
+   */
+  private ByteBuffer inBuf;
+
+  /**
+   * Create a new BytesToCharsetAdaptor for the given Charset.
+   *
+   * @param cs the Charset.
+   */
+  public BytesToCharsetAdaptor(Charset cs)
+  {
+    this(cs.newDecoder());
+  }
+
+  /**
+   * Create a new BytesToCharsetAdaptor for the given CharsetDecoder.
+   *
+   * @param dec the CharsetDecoder.
+   */
+  public BytesToCharsetAdaptor(CharsetDecoder dec)
+  {
+    decoder = dec;
+    // Use default replacments on bad input so that we don't have to
+    // deal with errors.
+    decoder.onMalformedInput(CodingErrorAction.REPLACE);
+    decoder.onUnmappableCharacter(CodingErrorAction.REPLACE);
+  }
+
+  /**
+   * Return the decoder's name.  The backing Charset's name is
+   * returned.
+   *
+   * @return The name.
+   */
+  public String getName()
+  {
+    return EncodingHelper.getOldCanonical(decoder.charset().name());
+  }
+
+  public int read(char[] outbuffer, int outpos, int count)
+  {
+    if (inBuf == null || ! inBuf.hasArray() || inBuf.array() != inbuffer)
+      inBuf = ByteBuffer.wrap(inbuffer);
+    inBuf.limit(inpos + inlength);
+    inBuf.position(inpos);
+
+    CharBuffer outBuf = CharBuffer.wrap(outbuffer, outpos, count);
+    decoder.decode(inBuf, outBuf, false);
+
+    // Update this.inpos to reflect the bytes consumed.
+    inpos = inBuf.position();
+    // Return the number of characters that were written to outbuffer.
+    return outBuf.position() - outpos;
+  }
+
+  // These aren't cached.
+  public void done()
+  {
+  }
+}
Index: gnu/gcj/convert/UnicodeToBytes.java
===================================================================
--- gnu/gcj/convert/UnicodeToBytes.java	(revision 108955)
+++ gnu/gcj/convert/UnicodeToBytes.java	(working copy)
@@ -1,4 +1,4 @@
-/* Copyright (C) 1999, 2000, 2001, 2003  Free Software Foundation
+/* Copyright (C) 1999, 2000, 2001, 2003, 2005  Free Software Foundation
 
    This file is part of libgcj.
 
@@ -8,6 +8,8 @@
 
 package gnu.gcj.convert; 
 
+import java.nio.charset.Charset;
+
 public abstract class UnicodeToBytes extends IOConverter
 {
   /** Buffer to emit bytes to.
@@ -99,10 +101,22 @@
     catch (Throwable ex) 
       { 
 	try
-	  {
+          {
 	    // We pass the original name to iconv and let it handle
-	    // its own aliasing.
+	    // its own aliasing.  Note that we intentionally prefer
+	    // iconv over nio.
 	    return new Output_iconv (encoding);
+          }
+        catch (Throwable _)
+          {
+            // Ignore, and try the next method.
+          }
+	try
+	  {
+            // Try using finding java.nio.charset.Charset and using
+            // the adaptor.  Use the original name as Charsets have
+            // their own canonical names.
+            return new CharsetToBytesAdaptor(Charset.forName(encoding));
 	  }
 	catch (Throwable _)
 	  {
Index: gnu/gcj/convert/BytesToUnicode.java
===================================================================
--- gnu/gcj/convert/BytesToUnicode.java	(revision 108955)
+++ gnu/gcj/convert/BytesToUnicode.java	(working copy)
@@ -1,4 +1,4 @@
-/* Copyright (C) 1999, 2000, 2001  Free Software Foundation
+/* Copyright (C) 1999, 2000, 2001, 2005  Free Software Foundation
 
    This file is part of libgcj.
 
@@ -8,6 +8,8 @@
 
 package gnu.gcj.convert;
 
+import java.nio.charset.Charset;
+
 public abstract class BytesToUnicode extends IOConverter
 {
   /** Buffer to read bytes from.
@@ -104,11 +106,20 @@
 	try
 	  {
 	    // We pass the original name to iconv and let it handle
-	    // its own aliasing.
+	    // its own aliasing.  Note that we intentionally prefer
+	    // iconv over nio.
 	    return new Input_iconv (encoding);
 	  }
 	catch (Throwable _)
 	  {
+	    // Ignore, and try the next method.
+	  }
+	try
+	  {
+	    return new BytesToCharsetAdaptor(Charset.forName(encoding));
+	  }
+	catch (Throwable _)
+	  {
 	    throw new java.io.UnsupportedEncodingException(encoding
 							   + " (" + ex + ')');
 	  }
Index: gnu/gcj/convert/CharsetToBytesAdaptor.java
===================================================================
--- gnu/gcj/convert/CharsetToBytesAdaptor.java	(revision 0)
+++ gnu/gcj/convert/CharsetToBytesAdaptor.java	(revision 0)
@@ -0,0 +1,133 @@
+/* Copyright (C) 2005  Free Software Foundation
+
+This file is part of libgcj.
+
+This software is copyrighted work licensed under the terms of the
+Libgcj License.  Please consult the file "LIBGCJ_LICENSE" for
+details.  */
+
+package gnu.gcj.convert; 
+
+import java.nio.ByteBuffer;
+import java.nio.CharBuffer;
+import java.nio.charset.Charset;
+import java.nio.charset.CharsetEncoder;
+import java.nio.charset.CodingErrorAction;
+import java.nio.charset.CoderResult;
+import gnu.java.nio.charset.EncodingHelper;
+
+/**
+ * Adaptor class that allow any {@link Charset} to be used
+ * as a UnicodeToBytes converter.
+ */
+public class CharsetToBytesAdaptor extends UnicodeToBytes
+{
+  /**
+   * The CharsetEncoder that does all the work.
+   */
+  private final CharsetEncoder encoder;
+
+  /**
+   * ByteBuffer wrapper for this.buf.
+   */
+  private ByteBuffer outBuf;
+
+  /**
+   * Dummy ByteBuffer of size zero.
+   */
+  private ByteBuffer fullBuf;
+
+
+  /**
+   * Create a new CharsetToBytesAdaptor for the given Charset.
+   *
+   * @param cs The Charset.
+   */
+  public CharsetToBytesAdaptor(Charset cs)
+  {
+    this(cs.newEncoder());
+  }
+
+  /**
+   * Create a new CharsetToBytesAdaptor for the given CharsetEncoder.
+   *
+   * @param enc The CharsetEncoder.
+   */
+  public CharsetToBytesAdaptor(CharsetEncoder enc)
+  {
+    encoder = enc;
+    // Use default replacments on bad input so that we don't have to
+    // deal with errors.
+    encoder.onMalformedInput(CodingErrorAction.REPLACE);
+    encoder.onUnmappableCharacter(CodingErrorAction.REPLACE);
+  }
+
+  /**
+   * Return the encoder's name.  The backing Charset's name is
+   * returned.
+   *
+   * @return The name.
+   */
+  public String getName()
+  {
+    return EncodingHelper.getOldCanonical(encoder.charset().name());
+  }
+
+  public int write (char[] inbuffer, int inpos, int inlength)
+  {
+    // Wrap the char array so it can be used by the encoder.
+    CharBuffer b = CharBuffer.wrap(inbuffer, inpos, inlength);
+    write(b);
+    return b.position() - inpos; // Number of chars consumed.
+  }
+
+  public int write (String str, int inpos, int inlength, char work)
+  {
+    // Wrap the String so it can be used by the encoder.
+    CharBuffer b = CharBuffer.wrap(str, inpos, inlength);
+    write(b);
+    return b.position() - inpos; // Number of chars consumed.
+  }
+
+  /**
+   * Encode as much of inBuf as will fit in buf.  The number of
+   * chars consumed is reflected by the new position of inBuf.  The
+   * output is put in buf and count is incremented by the number of
+   * bytes written.
+   *
+   * @param inBuf The input.
+   */
+  private void write(CharBuffer inBuf)
+  {
+    // Reuse existing outBuf if it is still wrapping the same array
+    // it was created with.
+    if (outBuf == null || !outBuf.hasArray() || outBuf.array() != buf)
+      outBuf = ByteBuffer.wrap(buf);
+
+    // Set the current position.
+    outBuf.position(count);
+    // Do the conversion.
+    encoder.encode(inBuf, outBuf, false);
+    // Mark the new end of buf.
+    count = outBuf.position();
+  }
+
+  /**
+   * Check for cached output in the converter.  This is done by
+   * checking the return value from a flush into a full ByteBuffer
+   *
+   * @return true if there is cached output that has not been
+   * written to buf.
+   */
+  public boolean havePendingBytes()
+  {
+    if (fullBuf == null)
+      fullBuf = ByteBuffer.allocate(0);
+    return CoderResult.OVERFLOW == encoder.flush(fullBuf);
+  }
+
+  // These aren't cached.
+  public void done()
+  {
+  }
+}
Index: java/nio/charset/Charset.java
===================================================================
--- java/nio/charset/Charset.java	(revision 108955)
+++ java/nio/charset/Charset.java	(working copy)
@@ -38,6 +38,8 @@
 
 package java.nio.charset;
 
+import gnu.classpath.ServiceFactory;
+import gnu.classpath.SystemProperties;
 import gnu.java.nio.charset.Provider;
 
 import java.io.BufferedReader;
@@ -116,6 +118,53 @@
       }
   }
 
+  /**
+   * Returns the system default charset.
+   *
+   * This may be set by the user or VM with the file.encoding
+   * property.
+   *
+   * @since 1.5
+   */
+  public static Charset defaultCharset()
+  {
+    String encoding;
+    
+    try 
+      {
+	encoding = SystemProperties.getProperty("file.encoding");
+      }
+    catch(SecurityException e)
+      {
+	// Use fallback.
+	encoding = "ISO-8859-1";
+      }
+    catch(IllegalArgumentException e)
+      {
+	// Use fallback.
+	encoding = "ISO-8859-1";
+      }
+
+    try
+      {
+	return forName(encoding);
+      }
+    catch(UnsupportedCharsetException e)
+      {
+	// Ignore.
+      }
+    catch(IllegalCharsetNameException e)
+      {
+	// Ignore.
+      }
+    catch(IllegalArgumentException e)
+      {
+	// Ignore.
+      }
+    
+    throw new IllegalStateException("Can't get default charset!");
+  }
+
   public static boolean isSupported (String charsetName)
   {
     return charsetForName (charsetName) != null;
@@ -155,13 +204,19 @@
   private static Charset charsetForName(String charsetName)
   {
     checkName (charsetName);
-    Charset cs = null;
-    CharsetProvider[] providers = providers2();
-    for (int i = 0; i < providers.length; i++)
+    // Try the default provider first
+    // (so we don't need to load external providers unless really necessary)
+    // if it is an exotic charset try loading the external providers.
+    Charset cs = provider().charsetForName(charsetName);
+    if (cs == null)
       {
-        cs = providers[i].charsetForName(charsetName);
-        if (cs != null)
-	  break;
+	CharsetProvider[] providers = providers2();
+	for (int i = 0; i < providers.length; i++)
+	  {
+	    cs = providers[i].charsetForName(charsetName);
+	    if (cs != null)
+	      break;
+	  }
       }
     return cs;
   }
@@ -169,6 +224,11 @@
   public static SortedMap availableCharsets()
   {
     TreeMap charsets = new TreeMap(String.CASE_INSENSITIVE_ORDER);
+    for (Iterator i = provider().charsets(); i.hasNext(); )
+      {
+	Charset cs = (Charset) i.next();
+	charsets.put(cs.name(), cs);
+      }
 
     CharsetProvider[] providers = providers2();
     for (int j = 0; j < providers.length; j++)
@@ -206,7 +266,7 @@
   /**
    * We need to support multiple providers, reading them from
    * java.nio.charset.spi.CharsetProvider in the resource directory
-   * META-INF/services.
+   * META-INF/services. This returns the "extra" charset providers.
    */
   private static CharsetProvider[] providers2()
   {
@@ -214,24 +274,10 @@
       {
         try
           {
-            Enumeration en = ClassLoader.getSystemResources
-	      ("META-INF/services/java.nio.charset.spi.CharsetProvider");
+            Iterator i = ServiceFactory.lookupProviders(CharsetProvider.class);
             LinkedHashSet set = new LinkedHashSet();
-            set.add(provider());
-            while (en.hasMoreElements())
-              {
-                BufferedReader rdr = new BufferedReader(new InputStreamReader
-                  (((URL) (en.nextElement())).openStream()));
-                while (true)
-                  {
-                    String s = rdr.readLine();
-                    if (s == null)
-		      break;
-                    CharsetProvider p =
-		      (CharsetProvider) ((Class.forName(s)).newInstance());
-                    set.add(p);
-                  }
-               }
+            while (i.hasNext())
+              set.add(i.next());
 
             providers = new CharsetProvider[set.size()];
             set.toArray(providers);
Index: java/io/OutputStreamWriter.java
===================================================================
--- java/io/OutputStreamWriter.java	(revision 108955)
+++ java/io/OutputStreamWriter.java	(working copy)
@@ -39,6 +39,9 @@
 package java.io;
 
 import gnu.gcj.convert.UnicodeToBytes;
+import gnu.gcj.convert.CharsetToBytesAdaptor;
+import java.nio.charset.Charset;
+import java.nio.charset.CharsetEncoder;
 
 /**
  * This class writes characters to an output stream that is byte oriented
@@ -129,6 +132,31 @@
   }
 
   /**
+   * This method initializes a new instance of <code>OutputStreamWriter</code>
+   * to write to the specified stream using a given <code>Charset</code>.
+   *
+   * @param out The <code>OutputStream</code> to write to
+   * @param cs The <code>Charset</code> of the encoding to use
+   */
+  public OutputStreamWriter(OutputStream out, Charset cs)
+  {
+    this(out, new CharsetToBytesAdaptor(cs));
+  }
+
+  /**
+   * This method initializes a new instance of <code>OutputStreamWriter</code>
+   * to write to the specified stream using a given
+   * <code>CharsetEncoder</code>.
+   *
+   * @param out The <code>OutputStream</code> to write to
+   * @param enc The <code>CharsetEncoder</code> to encode the output with
+   */
+  public OutputStreamWriter(OutputStream out, CharsetEncoder enc)
+  {
+    this(out, new CharsetToBytesAdaptor(enc));
+  }
+
+  /**
    * This method closes this stream, and the underlying 
    * <code>OutputStream</code>
    *
Index: java/io/InputStreamReader.java
===================================================================
--- java/io/InputStreamReader.java	(revision 108955)
+++ java/io/InputStreamReader.java	(working copy)
@@ -39,6 +39,8 @@
 package java.io;
 
 import gnu.gcj.convert.*;
+import java.nio.charset.Charset;
+import java.nio.charset.CharsetDecoder;
 
 /**
  * This class reads characters from a byte input stream.   The characters
@@ -131,6 +133,25 @@
     this(in, BytesToUnicode.getDecoder(encoding_name));
   }
 
+  /**
+   * Creates an InputStreamReader that uses a decoder of the given
+   * charset to decode the bytes in the InputStream into
+   * characters.
+   */
+  public InputStreamReader(InputStream in, Charset charset)
+  {
+    this(in, new BytesToCharsetAdaptor(charset));
+  }
+
+  /**
+   * Creates an InputStreamReader that uses the given charset decoder
+   * to decode the bytes in the InputStream into characters.
+   */
+  public InputStreamReader(InputStream in, CharsetDecoder decoder)
+  {
+    this(in, new BytesToCharsetAdaptor(decoder));
+  }
+
   private InputStreamReader(InputStream in, BytesToUnicode decoder)
   {
     // FIXME: someone could pass in a BufferedInputStream whose buffer
Index: mauve-libgcj
===================================================================
--- mauve-libgcj	(revision 108955)
+++ mauve-libgcj	(working copy)
@@ -23,4 +23,3 @@
 !java.awt.event.MouseEvent.modifiersEx
 !org.omg.
 !javax.rmi
-!java.io.InputStreamReader.getEncoding

Re: Patch: FYI: PRs 9715, 19132

[Tom Tromey]

Tom> Along the way this fixes PRs 9715 and 19132.

Oops, I missed a couple mauve fails related to 9715.
I'm going to hold off checking this in until I investigate those.

Tom

Re: Patch: FYI: PRs 9715, 19132

[Tom Tromey]

>>>>> "Tom" == Tom Tromey <tromey@redhat.com> writes:

Tom> Oops, I missed a couple mauve fails related to 9715.
Tom> I'm going to hold off checking this in until I investigate those.

I looked into this.  Some of the failures were bugs in my patch, I've
fixed these.  The new patch is appended.

the UTF-16, UnicodeBig, and UnicodeLittle test failures are caused by
BOM-related problems... in one case we choose a different ordering
than what the test expects (I don't see this one as a bug really, that
is what the BOM is for), and in the latter two we do not write a BOM
at all.  I suspect that comes from preferring iconv over the nio
charset converters.  This bears more investigation -- but this patch
does not hurt.

Tom


Index: ChangeLog
from  David Daney <ddaney@avtrex.com>
	Tom Tromey  <tromey@redhat.com>

	PR libgcj/9715, PR libgcj/19132:
	* java/nio/charset/Charset.java (charsetForName): Try default
	provider first.
	(availableCharsets): Re-merged.
	(providers2): Likewise.
	(defaultCharset): Likewise.
	* sources.am, Makefile.in: Rebuilt.
	* gnu/java/nio/charset/Provider.java: Removed.
	* java/io/OutputStreamWriter.java
	(OutputStreamWriter(OutputStream,Charset)): New constructor.
	(OutputStreamWriter(OutputStream,CharsetEncoder)): Likewise.
	* java/io/InputStreamReader.java
	(InputStreamReader(InputStream,CharsetDecoder)): New constructor.
	(InputStreamReader(InputStream,Charset)): Likewise.
	* gnu/gcj/convert/BytesToUnicode.java (getDecoder): Try a
	BytesToCharsetAdaptor.
	* gnu/gcj/convert/UnicodeToBytes.java (getEncoder): Try a
	CharsetToBytesAdaptor.
	* gnu/gcj/convert/CharsetToBytesAdaptor.java: New file.
	* gnu/gcj/convert/BytesToCharsetAdaptor.java: New file.
	* mauve-libgcj: Remove getEncoding exclusion.

Index: gnu/java/nio/charset/Provider.java
===================================================================
--- gnu/java/nio/charset/Provider.java	(revision 109042)
+++ gnu/java/nio/charset/Provider.java	(working copy)
@@ -1,154 +0,0 @@
-/* Provider.java -- 
-   Copyright (C) 2002, 2005 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 gnu.java.nio.charset;
-
-import java.nio.charset.Charset;
-import java.nio.charset.spi.CharsetProvider;
-import java.util.Collections;
-import java.util.HashMap;
-import java.util.Iterator;
-
-/**
- * Charset provider for the required charsets.  Used by
- * {@link Charset#charsetForName} and * {@link Charset#availableCharsets}.
- *
- * @author Jesse Rosenstock
- * @author Robert Schuster (thebohemian@gmx.net)
- * @see Charset
- */
-public final class Provider extends CharsetProvider
-{
-  private static Provider singleton;
-
-  static
-  {
-    synchronized (Provider.class)
-      {
-        singleton = null;
-      }
-  }
-
-  /**
-   * Map from charset name to charset canonical name. The strings
-   * are all lower-case to allow case-insensitive retrieval of
-   * Charset instances. 
-   */
-  private final HashMap canonicalNames;
-
-  /**
-   * Map from lower-case canonical name to Charset.
-   * TODO: We may want to use soft references.  We would then need to keep
-   * track of the class name to regenerate the object.
-   */
-  private final HashMap charsets;
-
-  private Provider ()
-  {
-    canonicalNames = new HashMap ();
-    charsets = new HashMap ();
-
-    // US-ASCII aka ISO646-US
-    addCharset (new US_ASCII ());
-
-    // ISO-8859-1 aka ISO-LATIN-1
-    addCharset (new ISO_8859_1 ());
-
-    // UTF-8
-    addCharset (new UTF_8 ());
-
-    // UTF-16BE
-    addCharset (new UTF_16BE ());
-
-    // UTF-16LE
-    addCharset (new UTF_16LE ());
-
-    // UTF-16
-    addCharset (new UTF_16 ());
-  }
-
-  public Iterator charsets ()
-  {
-    return Collections.unmodifiableCollection (charsets.values ())
-                      .iterator ();
-  }
-
-  /**
-   * Returns a Charset instance by converting the given
-   * name to lower-case, looking up the canonical charset
-   * name and finally looking up the Charset with that name.
-   * 
-   * <p>The lookup is therefore case-insensitive.</p>
-   * 
-   *  @returns The Charset having <code>charsetName</code>
-   *  as its alias or null if no such Charset exist.
-   */
-  public Charset charsetForName (String charsetName)
-  {
-    return (Charset) charsets.get(canonicalNames.get(charsetName.toLowerCase()));
-  }
-
-  /**
-   * Puts a Charset under its canonical name into the 'charsets' map.
-   * Then puts a mapping from all its alias names to the canonical name.
-   * 
-   * <p>All names are converted to lower-case</p>.
-   * 
-   * @param cs
-   */
-  private void addCharset (Charset cs)
-  {
-    String canonicalName = cs.name().toLowerCase();
-    charsets.put (canonicalName, cs);
-    
-    /* Adds a mapping between the canonical name
-     * itself making a lookup using that name
-     * no special case.
-     */  
-    canonicalNames.put(canonicalName, canonicalName);
-
-    for (Iterator i = cs.aliases ().iterator (); i.hasNext (); )
-      canonicalNames.put (((String) i.next()).toLowerCase(), canonicalName);
-  }
-
-  public static synchronized Provider provider ()
-  {
-    if (singleton == null)
-      singleton = new Provider ();
-    return singleton;
-  }
-}
Index: gnu/gcj/convert/BytesToCharsetAdaptor.java
===================================================================
--- gnu/gcj/convert/BytesToCharsetAdaptor.java	(revision 0)
+++ gnu/gcj/convert/BytesToCharsetAdaptor.java	(revision 0)
@@ -0,0 +1,90 @@
+/* Copyright (C) 2005  Free Software Foundation
+
+This file is part of libgcj.
+
+This software is copyrighted work licensed under the terms of the
+Libgcj License.  Please consult the file "LIBGCJ_LICENSE" for
+details.  */
+
+package gnu.gcj.convert; 
+
+import java.nio.ByteBuffer;
+import java.nio.CharBuffer;
+import java.nio.charset.Charset;
+import java.nio.charset.CharsetDecoder;
+import java.nio.charset.CodingErrorAction;
+import java.nio.charset.CoderResult;
+import gnu.java.nio.charset.EncodingHelper;
+
+/**
+ * Adaptor class that allow any {@link Charset} to be used
+ * as a BytesToUnicode converter.
+ */
+public class BytesToCharsetAdaptor extends BytesToUnicode
+{
+  /**
+   * The CharsetDecoder that does all the work.
+   */
+  private final CharsetDecoder decoder;
+
+  /**
+   * ByteBuffer wrapper for this.buf.
+   */
+  private ByteBuffer inBuf;
+
+  /**
+   * Create a new BytesToCharsetAdaptor for the given Charset.
+   *
+   * @param cs the Charset.
+   */
+  public BytesToCharsetAdaptor(Charset cs)
+  {
+    this(cs.newDecoder());
+  }
+
+  /**
+   * Create a new BytesToCharsetAdaptor for the given CharsetDecoder.
+   *
+   * @param dec the CharsetDecoder.
+   */
+  public BytesToCharsetAdaptor(CharsetDecoder dec)
+  {
+    decoder = dec;
+    // Use default replacments on bad input so that we don't have to
+    // deal with errors.
+    decoder.onMalformedInput(CodingErrorAction.REPLACE);
+    decoder.onUnmappableCharacter(CodingErrorAction.REPLACE);
+  }
+
+  /**
+   * Return the decoder's name.  The backing Charset's name is
+   * returned.
+   *
+   * @return The name.
+   */
+  public String getName()
+  {
+    return EncodingHelper.getOldCanonical(decoder.charset().name());
+  }
+
+  public int read(char[] outbuffer, int outpos, int count)
+  {
+    if (inBuf == null || ! inBuf.hasArray() || inBuf.array() != inbuffer)
+      inBuf = ByteBuffer.wrap(inbuffer);
+    inBuf.limit(inpos + inlength);
+    inBuf.position(inpos);
+
+    CharBuffer outBuf = CharBuffer.wrap(outbuffer, outpos, count);
+    decoder.decode(inBuf, outBuf, false);
+
+    // Update this.inpos to reflect the bytes consumed.
+    inpos = inBuf.position();
+    // Return the number of characters that were written to outbuffer.
+    return outBuf.position() - outpos;
+  }
+
+  // These aren't cached.
+  public void done()
+  {
+  }
+}
Index: gnu/gcj/convert/BytesToUnicode.java
===================================================================
--- gnu/gcj/convert/BytesToUnicode.java	(revision 109042)
+++ gnu/gcj/convert/BytesToUnicode.java	(working copy)
@@ -1,4 +1,4 @@
-/* Copyright (C) 1999, 2000, 2001  Free Software Foundation
+/* Copyright (C) 1999, 2000, 2001, 2005  Free Software Foundation
 
    This file is part of libgcj.
 
@@ -8,6 +8,8 @@
 
 package gnu.gcj.convert;
 
+import java.nio.charset.Charset;
+
 public abstract class BytesToUnicode extends IOConverter
 {
   /** Buffer to read bytes from.
@@ -104,11 +106,20 @@
 	try
 	  {
 	    // We pass the original name to iconv and let it handle
-	    // its own aliasing.
+	    // its own aliasing.  Note that we intentionally prefer
+	    // iconv over nio.
 	    return new Input_iconv (encoding);
 	  }
 	catch (Throwable _)
 	  {
+	    // Ignore, and try the next method.
+	  }
+	try
+	  {
+	    return new BytesToCharsetAdaptor(Charset.forName(encoding));
+	  }
+	catch (Throwable _)
+	  {
 	    throw new java.io.UnsupportedEncodingException(encoding
 							   + " (" + ex + ')');
 	  }
Index: gnu/gcj/convert/UnicodeToBytes.java
===================================================================
--- gnu/gcj/convert/UnicodeToBytes.java	(revision 109042)
+++ gnu/gcj/convert/UnicodeToBytes.java	(working copy)
@@ -1,4 +1,4 @@
-/* Copyright (C) 1999, 2000, 2001, 2003  Free Software Foundation
+/* Copyright (C) 1999, 2000, 2001, 2003, 2005  Free Software Foundation
 
    This file is part of libgcj.
 
@@ -8,6 +8,8 @@
 
 package gnu.gcj.convert; 
 
+import java.nio.charset.Charset;
+
 public abstract class UnicodeToBytes extends IOConverter
 {
   /** Buffer to emit bytes to.
@@ -99,10 +101,22 @@
     catch (Throwable ex) 
       { 
 	try
-	  {
+          {
 	    // We pass the original name to iconv and let it handle
-	    // its own aliasing.
+	    // its own aliasing.  Note that we intentionally prefer
+	    // iconv over nio.
 	    return new Output_iconv (encoding);
+          }
+        catch (Throwable _)
+          {
+            // Ignore, and try the next method.
+          }
+	try
+	  {
+            // Try using finding java.nio.charset.Charset and using
+            // the adaptor.  Use the original name as Charsets have
+            // their own canonical names.
+            return new CharsetToBytesAdaptor(Charset.forName(encoding));
 	  }
 	catch (Throwable _)
 	  {
Index: gnu/gcj/convert/CharsetToBytesAdaptor.java
===================================================================
--- gnu/gcj/convert/CharsetToBytesAdaptor.java	(revision 0)
+++ gnu/gcj/convert/CharsetToBytesAdaptor.java	(revision 0)
@@ -0,0 +1,150 @@
+/* Copyright (C) 2005, 2006  Free Software Foundation
+
+This file is part of libgcj.
+
+This software is copyrighted work licensed under the terms of the
+Libgcj License.  Please consult the file "LIBGCJ_LICENSE" for
+details.  */
+
+package gnu.gcj.convert; 
+
+import java.nio.ByteBuffer;
+import java.nio.CharBuffer;
+import java.nio.charset.Charset;
+import java.nio.charset.CharsetEncoder;
+import java.nio.charset.CodingErrorAction;
+import java.nio.charset.CoderResult;
+import gnu.java.nio.charset.EncodingHelper;
+
+/**
+ * Adaptor class that allow any {@link Charset} to be used
+ * as a UnicodeToBytes converter.
+ */
+public class CharsetToBytesAdaptor extends UnicodeToBytes
+{
+  /**
+   * The CharsetEncoder that does all the work.
+   */
+  private final CharsetEncoder encoder;
+
+  /**
+   * ByteBuffer wrapper for this.buf.
+   */
+  private ByteBuffer outBuf;
+
+  /**
+   * True if we've told the CharsetEncoder that there are no more
+   * characters available.
+   */
+  private boolean closedEncoder;
+
+  /**
+   * True if we're finished.
+   */
+  private boolean finished;
+
+  /**
+   * Create a new CharsetToBytesAdaptor for the given Charset.
+   *
+   * @param cs The Charset.
+   */
+  public CharsetToBytesAdaptor(Charset cs)
+  {
+    this(cs.newEncoder());
+  }
+
+  /**
+   * Create a new CharsetToBytesAdaptor for the given CharsetEncoder.
+   *
+   * @param enc The CharsetEncoder.
+   */
+  public CharsetToBytesAdaptor(CharsetEncoder enc)
+  {
+    encoder = enc;
+    // Use default replacments on bad input so that we don't have to
+    // deal with errors.
+    encoder.onMalformedInput(CodingErrorAction.REPLACE);
+    encoder.onUnmappableCharacter(CodingErrorAction.REPLACE);
+  }
+
+  /**
+   * Return the encoder's name.  The backing Charset's name is
+   * returned.
+   *
+   * @return The name.
+   */
+  public String getName()
+  {
+    return EncodingHelper.getOldCanonical(encoder.charset().name());
+  }
+
+  public int write (char[] inbuffer, int inpos, int inlength)
+  {
+    // Wrap the char array so it can be used by the encoder.
+    CharBuffer b = CharBuffer.wrap(inbuffer, inpos, inlength);
+    write(b);
+    return b.position() - inpos; // Number of chars consumed.
+  }
+
+  public int write (String str, int inpos, int inlength, char work)
+  {
+    // Wrap the String so it can be used by the encoder.
+    CharBuffer b = CharBuffer.wrap(str, inpos, inlength);
+    write(b);
+    return b.position() - inpos; // Number of chars consumed.
+  }
+
+  /**
+   * Encode as much of inBuf as will fit in buf.  The number of
+   * chars consumed is reflected by the new position of inBuf.  The
+   * output is put in buf and count is incremented by the number of
+   * bytes written.
+   *
+   * @param inBuf The input.
+   */
+  private void write(CharBuffer inBuf)
+  {
+    // Reuse existing outBuf if it is still wrapping the same array
+    // it was created with.
+    if (outBuf == null || !outBuf.hasArray() || outBuf.array() != buf)
+      outBuf = ByteBuffer.wrap(buf);
+
+    // Set the current position.
+    outBuf.position(count);
+
+    // If we've already said that there is no more input available,
+    // then we simply try to flush again.
+    if (closedEncoder)
+      {
+	CoderResult result = encoder.flush(outBuf);
+	if (result == CoderResult.UNDERFLOW)
+	  finished = true;
+       }
+    else
+      {
+	// Do the conversion.  If there are no characters to write,
+	// then we are finished.
+	closedEncoder = ! inBuf.hasRemaining();
+	encoder.encode(inBuf, outBuf, closedEncoder);
+      }
+
+    // Mark the new end of buf.
+    count = outBuf.position();
+  }
+
+  /**
+   * Check for cached output in the converter.
+   *
+   * @return true if there is cached output that has not been
+   * written to buf.
+   */
+  public boolean havePendingBytes()
+  {
+    return ! finished;
+  }
+
+  // These aren't cached.
+  public void done()
+  {
+  }
+}
Index: java/nio/charset/Charset.java
===================================================================
--- java/nio/charset/Charset.java	(revision 109042)
+++ java/nio/charset/Charset.java	(working copy)
@@ -38,6 +38,8 @@
 
 package java.nio.charset;
 
+import gnu.classpath.ServiceFactory;
+import gnu.classpath.SystemProperties;
 import gnu.java.nio.charset.Provider;
 
 import java.io.BufferedReader;
@@ -116,6 +118,53 @@
       }
   }
 
+  /**
+   * Returns the system default charset.
+   *
+   * This may be set by the user or VM with the file.encoding
+   * property.
+   *
+   * @since 1.5
+   */
+  public static Charset defaultCharset()
+  {
+    String encoding;
+    
+    try 
+      {
+	encoding = SystemProperties.getProperty("file.encoding");
+      }
+    catch(SecurityException e)
+      {
+	// Use fallback.
+	encoding = "ISO-8859-1";
+      }
+    catch(IllegalArgumentException e)
+      {
+	// Use fallback.
+	encoding = "ISO-8859-1";
+      }
+
+    try
+      {
+	return forName(encoding);
+      }
+    catch(UnsupportedCharsetException e)
+      {
+	// Ignore.
+      }
+    catch(IllegalCharsetNameException e)
+      {
+	// Ignore.
+      }
+    catch(IllegalArgumentException e)
+      {
+	// Ignore.
+      }
+    
+    throw new IllegalStateException("Can't get default charset!");
+  }
+
   public static boolean isSupported (String charsetName)
   {
     return charsetForName (charsetName) != null;
@@ -155,13 +204,19 @@
   private static Charset charsetForName(String charsetName)
   {
     checkName (charsetName);
-    Charset cs = null;
-    CharsetProvider[] providers = providers2();
-    for (int i = 0; i < providers.length; i++)
+    // Try the default provider first
+    // (so we don't need to load external providers unless really necessary)
+    // if it is an exotic charset try loading the external providers.
+    Charset cs = provider().charsetForName(charsetName);
+    if (cs == null)
       {
-        cs = providers[i].charsetForName(charsetName);
-        if (cs != null)
-	  break;
+	CharsetProvider[] providers = providers2();
+	for (int i = 0; i < providers.length; i++)
+	  {
+	    cs = providers[i].charsetForName(charsetName);
+	    if (cs != null)
+	      break;
+	  }
       }
     return cs;
   }
@@ -169,6 +224,11 @@
   public static SortedMap availableCharsets()
   {
     TreeMap charsets = new TreeMap(String.CASE_INSENSITIVE_ORDER);
+    for (Iterator i = provider().charsets(); i.hasNext(); )
+      {
+	Charset cs = (Charset) i.next();
+	charsets.put(cs.name(), cs);
+      }
 
     CharsetProvider[] providers = providers2();
     for (int j = 0; j < providers.length; j++)
@@ -206,7 +266,7 @@
   /**
    * We need to support multiple providers, reading them from
    * java.nio.charset.spi.CharsetProvider in the resource directory
-   * META-INF/services.
+   * META-INF/services. This returns the "extra" charset providers.
    */
   private static CharsetProvider[] providers2()
   {
@@ -214,24 +274,10 @@
       {
         try
           {
-            Enumeration en = ClassLoader.getSystemResources
-	      ("META-INF/services/java.nio.charset.spi.CharsetProvider");
+            Iterator i = ServiceFactory.lookupProviders(CharsetProvider.class);
             LinkedHashSet set = new LinkedHashSet();
-            set.add(provider());
-            while (en.hasMoreElements())
-              {
-                BufferedReader rdr = new BufferedReader(new InputStreamReader
-                  (((URL) (en.nextElement())).openStream()));
-                while (true)
-                  {
-                    String s = rdr.readLine();
-                    if (s == null)
-		      break;
-                    CharsetProvider p =
-		      (CharsetProvider) ((Class.forName(s)).newInstance());
-                    set.add(p);
-                  }
-               }
+            while (i.hasNext())
+              set.add(i.next());
 
             providers = new CharsetProvider[set.size()];
             set.toArray(providers);
Index: java/io/OutputStreamWriter.java
===================================================================
--- java/io/OutputStreamWriter.java	(revision 109042)
+++ java/io/OutputStreamWriter.java	(working copy)
@@ -39,6 +39,9 @@
 package java.io;
 
 import gnu.gcj.convert.UnicodeToBytes;
+import gnu.gcj.convert.CharsetToBytesAdaptor;
+import java.nio.charset.Charset;
+import java.nio.charset.CharsetEncoder;
 
 /**
  * This class writes characters to an output stream that is byte oriented
@@ -129,6 +132,31 @@
   }
 
   /**
+   * This method initializes a new instance of <code>OutputStreamWriter</code>
+   * to write to the specified stream using a given <code>Charset</code>.
+   *
+   * @param out The <code>OutputStream</code> to write to
+   * @param cs The <code>Charset</code> of the encoding to use
+   */
+  public OutputStreamWriter(OutputStream out, Charset cs)
+  {
+    this(out, new CharsetToBytesAdaptor(cs));
+  }
+
+  /**
+   * This method initializes a new instance of <code>OutputStreamWriter</code>
+   * to write to the specified stream using a given
+   * <code>CharsetEncoder</code>.
+   *
+   * @param out The <code>OutputStream</code> to write to
+   * @param enc The <code>CharsetEncoder</code> to encode the output with
+   */
+  public OutputStreamWriter(OutputStream out, CharsetEncoder enc)
+  {
+    this(out, new CharsetToBytesAdaptor(enc));
+  }
+
+  /**
    * This method closes this stream, and the underlying 
    * <code>OutputStream</code>
    *
Index: java/io/InputStreamReader.java
===================================================================
--- java/io/InputStreamReader.java	(revision 109042)
+++ java/io/InputStreamReader.java	(working copy)
@@ -39,6 +39,8 @@
 package java.io;
 
 import gnu.gcj.convert.*;
+import java.nio.charset.Charset;
+import java.nio.charset.CharsetDecoder;
 
 /**
  * This class reads characters from a byte input stream.   The characters
@@ -131,6 +133,25 @@
     this(in, BytesToUnicode.getDecoder(encoding_name));
   }
 
+  /**
+   * Creates an InputStreamReader that uses a decoder of the given
+   * charset to decode the bytes in the InputStream into
+   * characters.
+   */
+  public InputStreamReader(InputStream in, Charset charset)
+  {
+    this(in, new BytesToCharsetAdaptor(charset));
+  }
+
+  /**
+   * Creates an InputStreamReader that uses the given charset decoder
+   * to decode the bytes in the InputStream into characters.
+   */
+  public InputStreamReader(InputStream in, CharsetDecoder decoder)
+  {
+    this(in, new BytesToCharsetAdaptor(decoder));
+  }
+
   private InputStreamReader(InputStream in, BytesToUnicode decoder)
   {
     // FIXME: someone could pass in a BufferedInputStream whose buffer
Index: mauve-libgcj
===================================================================
--- mauve-libgcj	(revision 109042)
+++ mauve-libgcj	(working copy)
@@ -23,4 +23,3 @@
 !java.awt.event.MouseEvent.modifiersEx
 !org.omg.
 !javax.rmi
-!java.io.InputStreamReader.getEncoding

Re: Patch: FYI: PRs 9715, 19132

[David Daney]

Tom Tromey wrote:
>>>>>>"Tom" == Tom Tromey <tromey@redhat.com> writes:
> Index: ChangeLog
> from  David Daney <ddaney@avtrex.com>
> 	Tom Tromey  <tromey@redhat.com>
> 
Thanks Tom!.  I'm famous :-)