* Fixed Bug#212:  is an error

swi-to-yap · Dec 15, 2004 · 78e0882 · 78e0882
1 parent cf77813
commit 78e0882
Show file tree

Hide file tree

Showing 4 changed files with 24 additions and 6 deletions.
diff --git a/ChangeLog b/ChangeLog
@@ -1,3 +1,7 @@
+Dec 15, 2004
+
+  * FIXED: Bug#212: Disallow &#0; character entities.
+
 Nov 25, 2004
 
   * FIXED: utf-8 tests, avoid conflict with UTF-8 support in Prolog

diff --git a/Test/cent-nul.xml b/Test/cent-nul.xml
@@ -0,0 +1 @@
+<test>This content holds a &#0; byte that should be skipped</test>
diff --git a/Test/ok/cent-nul.ok b/Test/ok/cent-nul.ok
@@ -0,0 +1,2 @@
+[element(test, [], ['This content holds a  byte that should be skipped'])].
+[sgml(sgml_parser(482992), 'cent-nul.xml', 1, 'Syntax error: Bad character entity, found "#0"')].
diff --git a/parser.c b/parser.c
@@ -431,9 +431,13 @@ expand_pentities(dtd_parser *p, const ichar *in, ichar *out, int len)
 
       if ( (s=isee_character_entity(dtd, in, &chr)) &&
 	   representable_char(p, chr) )
-      { *out++ = chr;
-        in = s;
-        continue;
+      { if ( chr == 0 )
+	{ gripe(ERC_SYNTAX_ERROR, "Illegal character entity", in);
+	} else
+	{ *out++ = chr;
+	  in = s;
+	  continue;
+	}
       }
     }
 
@@ -520,7 +524,11 @@ expand_entities(dtd_parser *p, const ichar *in, ochar *out, int len)
 
       if ( (s=isee_character_entity(dtd, in, &chr)) )
       { if ( chr <= 0 || chr >= OUTPUT_CHARSET_SIZE )
-	  gripe(ERC_REPRESENTATION, "character");
+	{ if ( chr == 0 )
+	    gripe(ERC_SYNTAX_ERROR, "Illegal character entity", in);
+	  else
+	    gripe(ERC_REPRESENTATION, "character");
+	}
 	if ( --len <= 0 )
 	  return gripe(ERC_REPRESENTATION, "CDATA string too long");
 	*out++ = chr;
@@ -4133,7 +4141,7 @@ process_entity(dtd_parser *p, const ichar *name)
 { if ( name[0] == '#' )			/* #charcode: character entity */
   { int v = char_entity_value(name);
 
-    if ( v < 0 )
+    if ( v <= 0 )
       return gripe(ERC_SYNTAX_ERROR, "Bad character entity", name);
 
     if ( v >= OUTPUT_CHARSET_SIZE )
@@ -4177,7 +4185,10 @@ process_entity(dtd_parser *p, const ichar *name)
     { case EC_SGML:
       case EC_CDATA:
 	if ( (s=isee_character_entity(dtd, text, &chr)) && *s == '\0' )
-	{ if ( p->blank_cdata == TRUE && !HasClass(dtd, chr, CH_BLANK) )
+	{ if ( chr == 0 )
+	    return gripe(ERC_SYNTAX_ERROR, "Illegal character entity", text);
+
+	  if ( p->blank_cdata == TRUE && !HasClass(dtd, chr, CH_BLANK) )
 	  { p->cdata_must_be_empty = !open_element(p, CDATA_ELEMENT, FALSE);
 	    p->blank_cdata = FALSE;
 	  }
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1 @@
		<test>This content holds a byte that should be skipped</test>
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1,2 @@
		[element(test, [], ['This content holds a byte that should be skipped'])].
		[sgml(sgml_parser(482992), 'cent-nul.xml', 1, 'Syntax error: Bad character entity, found "#0"')].