* Attribute qualification in XML mode

ChangeLog

@@ -1,3 +1,6 @@
+* MODIFIED/ADDED: qualify_attributes option for xmlns mode.  Default is
+  now *not* to qualify attributes.
+
 * FIXED: Handle elements inside shorttag values.  Thanks Richard O'Keefe
   for providing a clear description how to handle this.
 

Test/netc.sgml

@@ -1,10 +1,12 @@
 <!DOCTYPE shref [
-  <!ELEMENT shref - - (netc|twit)*>
+  <!ELEMENT shref - - (netc|twit|twat)*>
   <!ELEMENT netc  - - (frob+)>
   <!ELEMENT frob  O O (twit)>
   <!ELEMENT twit  - O (#PCDATA)>
+  <!ELEMENT twat  - - CDATA>
 ]>    
 <shref>
 <netc/<frob><twit>This slash should end netc:/
+<twat/This is a twat/
 <twit/This is a simple shortref/
 </shref>

Test/ok/netc.ok

@@ -1,2 +1,2 @@
-[element(shref, [], [element(netc, [], [element(frob, [], [element(twit, [], ['This slash should end netc:'])])]), element(twit, [], ['This is a simple shortref'])])].
+[element(shref, [], [element(netc, [], [element(frob, [], [element(twit, [], ['This slash should end netc:'])])]), element(twat, [], ['This is a twat']), element(twit, [], ['This is a simple shortref'])])].
 [].

doc/sgml2pl.html

@@ -10,7 +10,7 @@
 <center>
 
 <h1>SWI-Prolog SGML/XML parser<br>
-<font size="+0"><em>Version 1.0.14, March 2001</em></font></h1>
+<font size="+0"><em>Version 1.0.15, September 2002</em></font></h1>
 
 <br>
 <a href="mailto:[email protected]"><em>Jan Wielemaker</em></a><br>
@@ -823,7 +823,7 @@ <h2><a name="sec:sec-3.4">DTD-Handling</a></h2>
 
 <dt>&nbsp;<br>
 <b>notation(<var>Name, Decl</var>)</b><dd>
-Unify <var>Decl</var> with a list if <code>system(+File)</code> and/or
+Unify <var>Decl</var> with a list if <code>system(+File)</code> andor
 <code>public(+PublicId)</code>.
 </dl>
 </dl>
@@ -962,10 +962,20 @@ <h2><a name="sec:sec-3.6">Parsing Primitives</a></h2>
 <dt>&nbsp;<br>
 <b><a name="const:xmlns%5"><b><tt>xmlns</tt></b></a></b><dd>
 Process file as XML file with namespace support. See <a href="#sec:xmlns">xmlns</a> for
-details. 
+details.  See also the <code>qualify_attributes</code> option below.
 </dl>
 
 
+<dt>&nbsp;<br>
+<b>qualify_attributes(<var>Boolean</var>)</b><dd>
+How to handle unqualified attribute (i.e. without an explicit namespace)
+in XML namespace (<a name="const:xmlns%6"><b><tt>xmlns</tt></b></a>) mode.  Default and standard compliant
+is not to qualify such elements.  If <a name="const:true%2"><b><tt>true</tt></b></a>, such attributes are qualified
+with the namespace of the element they appear in.  This option is for
+backward compatibility as this is the behaviour of older versions.  In
+addition, the namespace document suggests unqualified attributes are
+often interpreted in the namespace of their element.
+
 <dt>&nbsp;<br>
 <b>space(<var>SpaceMode</var>)</b><dd>
 Define the initial handling of white-space in PCDATA.  This attribute is
@@ -1024,7 +1034,7 @@ <h2><a name="sec:sec-3.6">Parsing Primitives</a></h2>
 
 <dt>&nbsp;<br>
 <b>dialect(<var>-Dialect</var>)</b><dd>
-Return the current dialect used by the parser (<a name="const:sgml%5"><b><tt>sgml</tt></b></a>, <a name="const:xml%4"><b><tt>xml</tt></b></a> or <a name="const:xmlns%6"><b><tt>xmlns</tt></b></a>).
+Return the current dialect used by the parser (<a name="const:sgml%5"><b><tt>sgml</tt></b></a>, <a name="const:xml%4"><b><tt>xml</tt></b></a> or <a name="const:xmlns%7"><b><tt>xmlns</tt></b></a>).
 
 
 <dt>&nbsp;<br>
@@ -1278,14 +1288,14 @@ <h2><a name="sec:sec-3.6">Parsing Primitives</a></h2>
 
 <dt>&nbsp;<br>
 <b>xmlns</b><dd>
-When parsing an in <a name="const:xmlns%7"><b><tt>xmlns</tt></b></a> mode, a new namespace declaraction is 
+When parsing an in <a name="const:xmlns%8"><b><tt>xmlns</tt></b></a> mode, a new namespace declaraction is 
 pushed on the environment. The named handler is called with three 
 arguments: <code><var>Handler</var>(+NameSpace, +URL, +Parser)</code>.
 See <a href="#sec:xmlns">xmlns</a> for details. 
 
 <dt>&nbsp;<br>
 <b>urlns</b><dd>
-When parsing an in <a name="const:xmlns%8"><b><tt>xmlns</tt></b></a> mode, this predicate can be used to map a 
+When parsing an in <a name="const:xmlns%9"><b><tt>xmlns</tt></b></a> mode, this predicate can be used to map a 
 url into either a canonical URL for this namespace or another internal 
 identifier. See <a href="#sec:xmlns">xmlns</a> for details. 
 </dl>
@@ -1761,13 +1771,13 @@ <h1>Index</h1>
 <li><a href="#const:shorttag"><b><tt>shorttag</tt></b></a>
 <li><a href="#const:start"><b><tt>start</tt></b></a>
 <li><a href="#const:token"><b><tt>token</tt></b></a> [ <a href="#const:token%2">2</a>]
-<li><a href="#const:true"><b><tt>true</tt></b></a>
+<li><a href="#const:true"><b><tt>true</tt></b></a> [ <a href="#const:true%2">2</a>]
 <li><a href="#const:urlns"><b><tt>urlns</tt></b></a>
 <li><a href="#const:warning"><b><tt>warning</tt></b></a>
 <li><a href="#const:xml"><b><tt>xml</tt></b></a> [ <a href="#const:xml%2">2</a> <a href="#const:xml%3">3</a> <a href="#const:xml%4">4</a>]
 <li><a href="#xml_quote_attribute/2"><b>xml_quote_attribute/2</b></a>
 <li><a href="#xml_quote_cdata/2"><b>xml_quote_cdata/2</b></a>
-<li><a href="#const:xmlns"><b><tt>xmlns</tt></b></a> [ <a href="#const:xmlns%2">2</a> <a href="#const:xmlns%3">3</a> <a href="#const:xmlns%4">4</a> <a href="#const:xmlns%5">5</a> <a href="#const:xmlns%6">6</a> <a href="#const:xmlns%7">7</a> <a href="#const:xmlns%8">8</a>]
+<li><a href="#const:xmlns"><b><tt>xmlns</tt></b></a> [ <a href="#const:xmlns%2">2</a> <a href="#const:xmlns%3">3</a> <a href="#const:xmlns%4">4</a> <a href="#const:xmlns%5">5</a> <a href="#const:xmlns%6">6</a> <a href="#const:xmlns%7">7</a> <a href="#const:xmlns%8">8</a> <a href="#const:xmlns%9">9</a>]
 </ul>
 </body>
 </html>

doc/sgml2pl.sgml

@@ -2,8 +2,8 @@
 
 <pldoc>
 <head>
-<version>1.0.14
-<release-date>March 2001
+<version>1.0.15
+<release-date>September 2002
 <package>sgml2pl
 </head>
 
@@ -706,8 +706,16 @@ This dialect is selected automatically if the processing instruction
 <verb><?xml ...></verb> is encountered. See <ref sec/xml/ for details. 
     <termitem><c/xmlns/
 Process file as XML file with namespace support. See <ref sec/xmlns/ for
-details. 
+details.  See also the <verb>qualify_attributes</verb> option below.
 </argdef>
+    <termitem>qualify_attributes(Boolean)
+How to handle unqualified attribute (i.e. without an explicit namespace)
+in XML namespace (<c/xmlns/) mode.  Default and standard compliant
+is not to qualify such elements.  If <c/true/, such attributes are qualified
+with the namespace of the element they appear in.  This option is for
+backward compatibility as this is the behaviour of older versions.  In
+addition, the namespace document suggests unqualified attributes are
+often interpreted in the namespace of their element.
     <termitem>space(SpaceMode)
 Define the initial handling of white-space in PCDATA.  This attribute is
 described in <ref sec/space/.

parser.c

@@ -3366,7 +3366,9 @@ process_net(dtd_parser *p)
       p->first = FALSE;
 
       if ( p->on_end_element )
-	(*p->on_end_element)(p, env->element);
+      { WITH_CLASS(p, EV_SHORTTAG,
+		   (*p->on_end_element)(p, env->element));
+      }
 
       free_environment(env);
       p->environments = parent;
@@ -4459,7 +4461,8 @@ putchar_dtd_parser(dtd_parser *p, int chr)
       }
 
       if ( p->waiting_for_net && f[CF_ETAGO2] == chr ) /* shorttag */
-      { WITH_PARSER(p,
+      { setlocation(&p->startloc, &p->location, line, lpos);
+	WITH_PARSER(p,
 		    process_net(p));
 	return;
       }
@@ -4523,7 +4526,23 @@ putchar_dtd_parser(dtd_parser *p, int chr)
       if ( f[CF_MDO1] == chr )		/* < */
       { setlocation(&p->startloc, &p->location, line, lpos);
 	p->state = S_ECDATA1;
-      } 
+      }
+
+					/* / in CDATA shorttag element */
+      if ( p->waiting_for_net && f[CF_ETAGO2] == chr )
+      { setlocation(&p->startloc, &p->location, line, lpos);
+	p->cdata->size--;
+	terminate_ocharbuf(p->cdata);
+	terminate_icharbuf(p->buffer);
+	if ( p->mark_state == MS_INCLUDE )
+	{ WITH_PARSER(p,
+		      process_cdata(p, TRUE);
+		      process_net(p));
+	  empty_cdata(p);
+	}
+	empty_icharbuf(p->buffer);
+	p->cdata_state = p->state = S_PCDATA;
+      }
 
       return;
     }