Merged gcj-eclipse branch to trunk.

git-svn-id: svn+ssh://gcc.gnu.org/svn/gcc/trunk@120621 138bc75d-0d04-0410-961f-82ee72b054a4

7 files changed, 94 insertions, 128 deletions

diff --git a/libjava/classpath/gnu/javax/swing/text/html/parser/GnuParserDelegator.java b/libjava/classpath/gnu/javax/swing/text/html/parser/GnuParserDelegator.java
index 841db667e84..273461a721e 100644
--- a/libjava/classpath/gnu/javax/swing/text/html/parser/GnuParserDelegator.java
+++ b/libjava/classpath/gnu/javax/swing/text/html/parser/GnuParserDelegator.java
@@ -43,6 +43,7 @@ import java.io.Reader;
 import java.io.Serializable;
 
 import javax.swing.text.BadLocationException;
+import javax.swing.text.SimpleAttributeSet;
 import javax.swing.text.html.HTMLEditorKit;
 import javax.swing.text.html.HTMLEditorKit.ParserCallback;
 import javax.swing.text.html.parser.DTD;
@@ -92,7 +93,7 @@ public class GnuParserDelegator extends ParserDelegator implements Serializable
 
     protected final void handleStartTag(TagElement tag)
     {
-      htmlAttributeSet attributes = gnu.getAttributes();
+      SimpleAttributeSet attributes = gnu.getAttributes();
 
       if (tag.fictional())
         attributes.addAttribute(ParserCallback.IMPLIED, Boolean.TRUE);
diff --git a/libjava/classpath/gnu/javax/swing/text/html/parser/HTML_401F.java b/libjava/classpath/gnu/javax/swing/text/html/parser/HTML_401F.java
index c3c347e36e3..1894b6a1ac6 100644
--- a/libjava/classpath/gnu/javax/swing/text/html/parser/HTML_401F.java
+++ b/libjava/classpath/gnu/javax/swing/text/html/parser/HTML_401F.java
@@ -2445,8 +2445,10 @@ public class HTML_401F
         attr(VALUE, null, null, 0, IMPLIED)
       }
     );
+      
+      // Headers in the paragraph are not allowed.
       defElement(P, 0, false, true, new ContentModel( 0,
-       new noTagModel(P), null),
+       new noTagModel(new String[] { P, H1, H2, H3, H4, H5, H6 }), null),
       NONE
       ,
       new String[] {
diff --git a/libjava/classpath/gnu/javax/swing/text/html/parser/HTML_401Swing.java b/libjava/classpath/gnu/javax/swing/text/html/parser/HTML_401Swing.java
deleted file mode 100644
index 9c934f647de..00000000000
--- a/libjava/classpath/gnu/javax/swing/text/html/parser/HTML_401Swing.java
+++ /dev/null
@@ -1,91 +0,0 @@
-/* HTML_401Swing.java -- The HTML 4.01 DTD, adapted for HTML rendering in Swing
-   Copyright (C) 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 gnu.javax.swing.text.html.parser;
-
-import javax.swing.text.html.parser.DTD;
-
-/**
- * This class is necessary because the current implementation of the GNU
- * Classpath Swing requires always enclose the text into paragraphs.
- * 
- * @author Audrius Meskauskas (AudriusA@Bioinformatics.org)
- */
-public class HTML_401Swing extends HTML_401F
-{
-  /**
-   * The singleton instance;
-   */
-  final static HTML_401Swing singleton = new HTML_401Swing();
-  
-  /**
-   * Either takes the document (by name) from DTD table, or
-   * creates a new instance and registers it in the tabe.
-   * The document is registerd under name "-//W3C//DTD HTML 4.01 Frameset//EN".
-   * @return The new or existing DTD for parsing HTML 4.01 Frameset.
-   */
-  public static DTD getInstance()
-  {
-    return singleton;
-  }  
-  
-  /**
-   * Get elements that are allowed in the document body, at the zero level.
-   * This list disallows the text at this level (the implied P tag will be
-   * generated). It also disallows A, B, I, U, CITE and other similar
-   * elements that have the plain text inside. They will also be placed
-   * inside the generated implied P tags.
-   */
-  protected String[] getBodyElements()
-  {
-    return new String[] {
-        APPLET, BASEFONT,
-        BR, BUTTON, 
-        IFRAME, IMG,
-        INPUT, LABEL, MAP, OBJECT,
-        SCRIPT, SELECT,
-        TEXTAREA, 
-        BLOCKQUOTE, CENTER, DEL, DIR,
-        DIV, DL, FIELDSET, FORM, H1,
-        H2, H3, H4, H5, H6,
-        HR, INS, ISINDEX, MENU, NOFRAMES,
-        NOSCRIPT, OL, P, PRE, TABLE,
-        UL
-      };
-  }
-}
diff --git a/libjava/classpath/gnu/javax/swing/text/html/parser/htmlValidator.java b/libjava/classpath/gnu/javax/swing/text/html/parser/htmlValidator.java
index 4d287a67763..7507850e8f9 100644
--- a/libjava/classpath/gnu/javax/swing/text/html/parser/htmlValidator.java
+++ b/libjava/classpath/gnu/javax/swing/text/html/parser/htmlValidator.java
@@ -153,7 +153,7 @@ public abstract class htmlValidator
    * Remove the given tag from the stack or (if found) from the list
    * of the forcibly closed tags.
    */
-  public void closeTag(TagElement tElement)
+  public boolean closeTag(TagElement tElement)
   {
     HTML.Tag tag = tElement.getHTMLTag();
     hTag x;
@@ -191,11 +191,12 @@ public abstract class htmlValidator
                   }
 
                 stack.remove(x);
-                return;
+                return true;
               }
           }
       }
     s_error("Closing unopened <" + tag + ">");
+    return false;
   }
 
   /**
diff --git a/libjava/classpath/gnu/javax/swing/text/html/parser/support/Parser.java b/libjava/classpath/gnu/javax/swing/text/html/parser/support/Parser.java
index 92f9b27c5d9..3a407310e74 100644
--- a/libjava/classpath/gnu/javax/swing/text/html/parser/support/Parser.java
+++ b/libjava/classpath/gnu/javax/swing/text/html/parser/support/Parser.java
@@ -56,6 +56,7 @@ import java.util.TreeSet;
 import java.util.Vector;
 
 import javax.swing.text.ChangedCharSetException;
+import javax.swing.text.SimpleAttributeSet;
 import javax.swing.text.html.HTML;
 import javax.swing.text.html.parser.AttributeList;
 import javax.swing.text.html.parser.DTD;
@@ -250,9 +251,9 @@ public class Parser
    * Get the attributes of the current tag.
    * @return The attribute set, representing the attributes of the current tag.
    */
-  public htmlAttributeSet getAttributes()
+  public SimpleAttributeSet getAttributes()
   {
-    return attributes;
+    return new SimpleAttributeSet(attributes);
   }
 
   /**
@@ -497,6 +498,9 @@ public class Parser
         mustBe(t.kind);
       }
     hTag = new Token(start, last);
+
+    // Consume any whitespace immediately following a comment.
+    optional(WS);
     handleComment();
   }
 
@@ -579,6 +583,8 @@ public class Parser
                  );
           }
       }
+    // Consume any whitespace that follows the Sgml insertion.
+    optional(WS);
   }
 
   /**
@@ -658,7 +664,10 @@ public class Parser
     else
       text = textProcessor.preprocess(buffer);
 
-    if (text != null && text.length > 0)
+    if (text != null && text.length > 0
+        // According to the specs we need to discard whitespace immediately
+        // before a closing tag.
+        && (text.length > 1 || text[0] != ' ' || ! TAG_CLOSE.matches(this)))
       {
         TagElement pcdata = new TagElement(dtd.getElement("#pcdata"));
         attributes = htmlAttributeSet.EMPTY_HTML_ATTRIBUTE_SET;
@@ -889,6 +898,8 @@ public class Parser
   protected void parseDocument()
                         throws ParseException
   {
+    // Read up any initial whitespace.
+    optional(WS);
     while (getTokenAhead().kind != EOF)
       {
         advanced = false;
@@ -979,13 +990,15 @@ public class Parser
                           + next.getImage() + "'");
                     attrValue = value.getImage();
                   }
-                else if (next.kind == SLASH)
-                // The slash in this context is treated as the ordinary
-                // character, not as a token. The slash may be part of
+                else if (next.kind == SLASH || next.kind == OTHER)
+                // The slash and other characters (like %) in this context is
+                // treated as the ordinary
+                // character, not as a token. The character may be part of
                 // the unquoted URL.
                   {
                     StringBuffer image = new StringBuffer(value.getImage());
-                    while (next.kind == NUMTOKEN || next.kind == SLASH)
+                    while (next.kind == NUMTOKEN || next.kind == SLASH
+                           || next.kind == OTHER)
                       {
                         image.append(getNextToken().getImage());
                         next = getTokenAhead();
@@ -1177,6 +1190,13 @@ public class Parser
       {
         validator.validateTag(tag, attributes);
         handleEmptyTag(tag);
+        HTML.Tag h = tag.getHTMLTag();
+        // When a block tag is closed, consume whitespace that follows after
+        // it.
+        // For some unknown reason a FRAME tag is not treated as block element.
+        // However in this case it should be treated as such.
+        if (isBlock(h))
+          optional(WS);
       }
     catch (ChangedCharSetException ex)
       {
@@ -1192,8 +1212,8 @@ public class Parser
    */
   private void _handleEndTag(TagElement tag)
   {
-    validator.closeTag(tag);
-    _handleEndTag_remaining(tag);
+    if (validator.closeTag(tag))
+       _handleEndTag_remaining(tag);
   }
 
   /**
@@ -1213,6 +1233,11 @@ public class Parser
     if (preformatted < 0)
       preformatted = 0;
 
+    // When a block tag is closed, consume whitespace that follows after
+    // it.
+    if (isBlock(h))
+      optional(WS);
+
     if (h == HTML.Tag.TITLE)
       {
         titleOpen = false;
@@ -1239,6 +1264,9 @@ public class Parser
 
     HTML.Tag h = tag.getHTMLTag();
 
+    if (isBlock(h))
+      optional(WS);
+
     if (h.isPreformatted())
       preformatted++;
 
@@ -1418,8 +1446,6 @@ public class Parser
 
     hTag = new Token(start, next);
 
-    attributes.setResolveParent(defaulter.getDefaultParameters(name.getImage()));
-
     if (!end)
       {
         // The tag body contains errors. If additionally the tag
@@ -1457,7 +1483,12 @@ public class Parser
         if (te.getElement().type == DTDConstants.EMPTY)
           _handleEmptyTag(te);
         else
-          _handleStartTag(te);
+          {
+            // According to the specs we need to consume whitespace following
+            // immediately after a opening tag.
+            optional(WS);
+            _handleStartTag(te);
+          }
       }
   }
 
@@ -1483,4 +1514,19 @@ public class Parser
   {
     error("Whitespace here is not permitted");
   }
+
+  /**
+   * Returns true when the specified tag should be considered a block tag
+   * wrt whitespace handling. We need this special handling, since there
+   * are a couple of tags that we must treat as block tags but which aren't
+   * officially block tags.
+   *
+   * @param tag the tag to check
+   * @return true when the specified tag should be considered a block tag
+   *         wrt whitespace handling
+   */
+  private boolean isBlock(HTML.Tag tag)
+  {
+    return tag.isBlock() || tag == HTML.Tag.STYLE || tag == HTML.Tag.FRAME;
+  }
 }
diff --git a/libjava/classpath/gnu/javax/swing/text/html/parser/support/low/Constants.java b/libjava/classpath/gnu/javax/swing/text/html/parser/support/low/Constants.java
index 283d32385ef..5416582adca 100644
--- a/libjava/classpath/gnu/javax/swing/text/html/parser/support/low/Constants.java
+++ b/libjava/classpath/gnu/javax/swing/text/html/parser/support/low/Constants.java
@@ -209,6 +209,17 @@ public class Constants
                 }
                );
 
+  /**
+   * Ordinary HTML tag closing pattern.
+   */
+  public static final pattern TAG_CLOSE =
+    new pattern(new node[]
+                {
+                  new node(BEGIN), new node(WS, true), new node(SLASH),
+                  new node(WS, true), new node(NUMTOKEN)
+                }
+               );
+
   /* Special tokens */
 
   /**
diff --git a/libjava/classpath/gnu/javax/swing/text/html/parser/support/textPreProcessor.java b/libjava/classpath/gnu/javax/swing/text/html/parser/support/textPreProcessor.java
index cc1610585a6..6fd79e2589e 100644
--- a/libjava/classpath/gnu/javax/swing/text/html/parser/support/textPreProcessor.java
+++ b/libjava/classpath/gnu/javax/swing/text/html/parser/support/textPreProcessor.java
@@ -42,17 +42,17 @@ import gnu.javax.swing.text.html.parser.support.low.Constants;
 
 /**
  * Pre - processes text in text parts of the html document.
- * Not thread - safe.
+ *
  * @author Audrius Meskauskas, Lithuania (AudriusA@Bioinformatics.org)
  */
 public class textPreProcessor
 {
   /**
-   * Pre - process non-preformatted text.
-   * \t, \r and \n mutate into spaces, then multiple spaces mutate
-   * into single one, all whitespace around tags is consumed.
-   * The content of the passed buffer is destroyed.
-   * @param text A text to pre-process.
+   * Pre - process non-preformatted text. \t, \r and \n mutate into spaces, then
+   * multiple spaces mutate into single one, all whitespace around tags is
+   * consumed. The content of the passed buffer is destroyed.
+   * 
+   * @param a_text A text to pre-process.
    */
   public char[] preprocess(StringBuffer a_text)
   {
@@ -64,18 +64,15 @@ public class textPreProcessor
     int a = 0;
     int b = text.length - 1;
 
-    try
-      {
-        while (Constants.bWHITESPACE.get(text [ a ]))
-          a++;
-        while (Constants.bWHITESPACE.get(text [ b ]))
-          b--;
-      }
-    catch (ArrayIndexOutOfBoundsException sx)
-      {
-        // A text fragment, consisting from line breaks only.
-        return null;
-      }
+    // Remove leading/trailing whitespace, leaving at most one character
+    int len = text.length;
+    while (a + 1 < len && Constants.bWHITESPACE.get(text[a])
+           && Constants.bWHITESPACE.get(text[a + 1]))
+      a++;
+
+    while (b > a && Constants.bWHITESPACE.get(text[b])
+               && Constants.bWHITESPACE.get(text[b - 1]))
+      b--;
 
     a_text.setLength(0);
 
@@ -83,10 +80,9 @@ public class textPreProcessor
     boolean spaceNow;
     char c;
 
-    chars: 
-    for (int i = a; i <= b; i++)
+    chars: for (int i = a; i <= b; i++)
       {
-        c = text [ i ];
+        c = text[i];
         spaceNow = Constants.bWHITESPACE.get(c);
         if (spacesWere && spaceNow)
           continue chars;