xml parsing with comments

2 files changed, 58 insertions, 42 deletions

diff --git a/sources/scala/tools/scalac/ast/parser/Parser.scala b/sources/scala/tools/scalac/ast/parser/Parser.scala
index baa8ca728e..02057a6feb 100644
--- a/sources/scala/tools/scalac/ast/parser/Parser.scala
+++ b/sources/scala/tools/scalac/ast/parser/Parser.scala
@@ -915,8 +915,10 @@ class Parser(unit: Unit) {
           val pos1 = s.skipToken();
           if (s.token == IDENTIFIER && s.name == Names.STAR) {
             s.nextToken();
+            /*
             if( s.token != RPAREN )
               syntaxError(s.pos, " escaping sequences only allowed for last argument", true);
+            */
             t = make.Typed(
               pos, t, make.Ident(pos1, TypeNames.WILDCARD_STAR));
           } else {
@@ -992,7 +994,7 @@ class Parser(unit: Unit) {
            SYMBOLLIT | TRUE | FALSE | NULL =>
         t = literal(false);
       case IDENTIFIER | THIS | SUPER =>
-        t = if( s.name == LT ) {
+        t = if(( s.name == LT )&&( unit.global.xmlMarkup )) {
           xmlExprTop();  /* top-level xml expression */
         } else {
           stableRef(true, false);
@@ -1053,7 +1055,11 @@ class Parser(unit: Unit) {
   def xmlExprTop():Tree = {
     val pos = s.pos;
     var t = xmlExpr();
+    //Console.println("old:"+token2string( s.token ) );
+    s.token = EMPTY;
     s.nextToken();
+    //Console.println("new:"+token2string( s.token ) );
+    //Console.println("line:"+s.cline);
     if(( s.token == IDENTIFIER ) && ( s.name == LT ))  {
       val ts = new myTreeList();
       ts.append( t );
@@ -1323,7 +1329,7 @@ class Parser(unit: Unit) {
     case IDENTIFIER | THIS =>
       if (s.name == BAR) {
         make.Sequence(s.pos, Tree.EMPTY_ARRAY); // ((nothing))
-      } else if (s.name == LT) {
+      } else if(( s.name == LT )&&( unit.global.xmlMarkup )) {
         xmlPatternTop()
       } else {
         var t = stableId();
@@ -1378,7 +1384,7 @@ class Parser(unit: Unit) {
       endch match {
         case '"' | '\'' => {
           s.nextch();
-          attrValue = s.xmlValue( endch );
+          attrValue = s.xmlAttribValue( endch );
           s.xmlToken( endch.asInstanceOf[char] );
           s.xmlSpaceOpt();
         }
@@ -1420,19 +1426,23 @@ class Parser(unit: Unit) {
         s.ch match {
           case '<' => {                                  /* tag */
             s.nextch();
-            if( s.ch != '/' ) { /* search end tag */
-              ts.append( xmlExpr() );
-	      s.xmlSpaceOpt();
-            } else {
-              exit = true
+            s.ch match {
+              case '/' => exit = true;
+              case '!' =>
+                s.xmlComment();
+              case _ =>
+                /* search end tag */
+                ts.append( xmlExpr() );
+	        s.xmlSpaceOpt();
             }
           }
+
           case '{' => {                                 /* Scala block */
             while( s.ch == '{' ) {
               s.nextToken();
               s.nextToken();
               val bs = new myTreeList();
-              val b = block( s.pos );
+              val b = expr(true,false); //block( s.pos );
               if( s.token != RBRACE ) {
                 s.xml_syntaxError(" expected end of Scala block");
               }
@@ -1447,7 +1457,6 @@ class Parser(unit: Unit) {
         }
       }
       xmlEndTag( elemName );
-      s.xmlSpaceOpt();
       if( attrMap.isEmpty )
 	makeXML( pos, elemName, ts.toArray() );
       else
diff --git a/sources/scala/tools/scalac/ast/parser/Scanner.scala b/sources/scala/tools/scalac/ast/parser/Scanner.scala
index b52b815e9a..174040d220 100644
--- a/sources/scala/tools/scalac/ast/parser/Scanner.scala
+++ b/sources/scala/tools/scalac/ast/parser/Scanner.scala
@@ -663,37 +663,29 @@ class Scanner(_unit: Unit) extends TokenData {
 
   def xml_syntaxError(s:String) = {
     syntaxError("in XML literal: "+s);
-    nextch();
+    xml_nextch();
   }
 
-  /* this helper functions updates ccol and cline, only necessary in whitespace
-  *  production
+  /* this helper functions updates ccol and cline
   */
   def xml_nextch() = {
-    try {
-      nextch();
-    } catch {
-      case e:ArrayIndexOutOfBoundsException =>  {
-        token = EOF;
-        syntaxError(lastpos, "unclosed XML literal");
-        throw new ApplicationError("unclosed XML literal");
-      }
-    }
+    nextch();
     ch match {
-      case '\r' => {
+      case SU => syntaxError(lastpos, "unclosed XML literal"); token = EOF;
+      case CR =>
         cline = cline + 1;
         ccol = 0;
         nextch(); /* in compliance with XML spec */
-        if( ch == '\n' ) {
+        if( ch == LF ) {
           ccol = 0;
         }
-      }
-      case '\n' => {
+      case LF => {
         cline = cline + 1;
         ccol = 0;
       }
       case _ =>
     }
+    pos = Position.encode(cline, ccol);
   }
 
   def xml_isSpace() = ch match {
@@ -705,7 +697,6 @@ class Scanner(_unit: Unit) extends TokenData {
     while( xml_isSpace() ) {
       xml_nextch();
     }
-    pos = Position.encode(cline, ccol);
   }
 
   /** [3] S ::= (#x20 | #x9 | #xD | #xA)+
@@ -715,7 +706,6 @@ class Scanner(_unit: Unit) extends TokenData {
       xml_nextch();
       xmlSpaceOpt()
     } else {
-      pos = Position.encode(cline, ccol);
       xml_syntaxError("whitespace expected");
     }
   }
@@ -752,9 +742,8 @@ class Scanner(_unit: Unit) extends TokenData {
     if( xml_isNameStart() ) {
       val index = bp;
       while( xml_isNameChar() ) {
-        nextch();
+        xml_nextch();
       }
-      pos = Position.encode(cline, ccol);
       Name.fromAscii(buf, index, bp - index);
     } else {
       xml_syntaxError("name expected");
@@ -762,28 +751,46 @@ class Scanner(_unit: Unit) extends TokenData {
     }
   }
 
-  def xmlValue(endch:char):String = {
+  /* consuming everything up to the next endch */
+  def xmlValue(endch:char):String = xmlValue(endch, true);
+
+  def xmlValue(endch:char, keep:boolean):String = {
     lastpos = pos;
     val index = bp;
-    while ( ch != endch ) {
-      if(( ch == '<' )||( ch == '&' )) {
-	pos = Position.encode(cline, ccol);
-        xml_syntaxError(ch.asInstanceOf[char]+" not allowed here");
-      }
-      xml_nextch();
-    };
-    pos = Position.encode(cline, ccol);
-    new String(buf, index, bp-index);
+    while ( ch != endch ) { xml_nextch();};
+    pos = Position.encode( cline, ccol );
+    if( keep )
+      new String(buf, index, bp-index);
+    else
+      null
+  }
+
+  def xmlAttribValue(endch:char):String = {
+    val s = xmlValue(endch, true);
+    if( s.indexOf('<') != -1 ) {
+        xml_syntaxError("'<' not allowed in attrib value");
+        "--syntax error--"
+    } else {
+      s
+    }
   }
 
   def xmlText():String = xmlValue('<');
 
+  def xmlComment() = {
+    xmlToken('!');
+    xmlToken('-');
+    xmlToken('-');
+    xmlValue('-', false);
+    xmlToken('-');
+    xmlToken('-');
+    xmlToken('>');
+  };
+
   def xmlToken(that:char):unit = {
     if( ch == that ) {
       xml_nextch();
-      pos = Position.encode(cline, ccol);
     } else {
-      pos = Position.encode(cline, ccol);
       xml_syntaxError("'"+that+"' expected instead of '"+ch.asInstanceOf[char]+"'");
     }
   }