Landing new config code. Badges are borken, the link to the relevant spec

from "This page is valid Foo" is borken, and there is a good chance the
DOCTYPE Override is borken. Otherwise this should be more or less working.

4 files changed, 517 insertions, 39 deletions

diff --git a/htdocs/base.css b/htdocs/base.css
index b231c83..12122d5 100644
--- a/htdocs/base.css
+++ b/htdocs/base.css
@@ -4,7 +4,7 @@
    Copyright 2000 W3C (MIT, INRIA, Keio). All Rights Reserved.
    See http://www.w3.org/Consortium/Legal/ipr-notice.html#Copyright
 
-   $Id: base.css,v 1.28 2002-12-02 02:52:53 link Exp $
+   $Id: base.css,v 1.29 2002-12-08 01:46:44 link Exp $
 */
 
 @import url(results.css); /* CSS for validation results. */
@@ -211,17 +211,6 @@ h6 {
   text-transform: none;
 }
 
-/* comma tools table */
-
-td.comma-desc { background: #ffeecc;  
-border : 1px solid black;
-font-family: sans-serif }
-td.comma-shortcut { background: #e7e7ff; 
-border : 1px solid black;
-font-family: monospace }
-td.comma-tool { background: #e7e7ff;  border : 1px solid black;}
-
-
 /* documentation - table of contents */
 #toc {	
   border-bottom : 2px dotted black;
@@ -257,3 +246,30 @@ input.default {
   font-family: monospace;
 }
 
+/* comma tools table */
+
+td.comma-desc { background: #ffeecc;  
+border : 1px solid black;
+font-family: sans-serif }
+td.comma-shortcut { background: #e7e7ff; 
+border : 1px solid black;
+font-family: monospace }
+td.comma-tool { background: #e7e7ff;  border : 1px solid black;}
+
+
+/* config param table */
+
+table.config tr th {
+  text-align: left;
+  background: #e7e7ff;
+}
+table.config tr td {
+  background: #ffeecc;  
+  border : 1px solid black;
+  font-family: sans-serif;
+}
+
+table.config tr td.subtable {
+  background: white;
+  border: none;
+}
diff --git a/htdocs/config/types.conf b/htdocs/config/types.conf
new file mode 100644
index 0000000..e75183b
--- /dev/null
+++ b/htdocs/config/types.conf
@@ -0,0 +1,396 @@
+#
+# Main Document Type Database for the W3C MarkUp Validation Service.
+#
+# $Id: types.conf,v 1.1 2002-12-08 01:46:44 link Exp $
+#
+# Maintains all information for each of the document types we support.
+# See 'perldoc Config::General' for the syntax, and be aware that the
+# 'SplitPolicy' is 'equalsign', ie. keys and values are separated by '\s*=\s*'.
+#
+# The meaning of the parameters are mostly obvious and all are documented
+# in docs/devel.html. Of particular note, the names of each section is
+# arbitrary and the "datastructure" is turned inside out in the code so
+# that it is indexed by the PubID. This means you can not have multiple
+# entries with identical PubID!
+#
+
+#
+# The five different ways to refer to HTML 2.0:
+# "HTML 2.0", "HTML 2.0 Level 2", "HTML 2.0 Level 1",
+# "HTML 2.0 Strict", "HTML 2.0 Strict Level 1".
+<HTML_2_0>
+  Name       = html
+  Display    = HTML 2.0
+  Info_URL   = http://www.w3.org/MarkUp/html-spec/
+  PubID      = -//IETF//DTD HTML 2.0//EN
+  Parse_Mode = SGML
+  <Content_Types>
+    Allowed   = text/html
+    Forbidden = application/xhtml+xml
+    Preferred = text/html
+  </Content_Types>
+  <Badge>
+    URI    = http://validator.w3.org/images/vh20
+  </Badge>
+</HTML_2_0>
+
+<HTML_2_0_Level_2>
+  Name       = html
+  Display    = HTML 2.0 Level 2
+  Info_URL   = http://www.w3.org/MarkUp/html-spec/
+  PubID      = -//IETF//DTD HTML 2.0 Level 2//EN
+  Parse_Mode = SGML
+  <Content_Types>
+    Allowed   = text/html
+    Forbidden = application/xhtml+xml
+    Preferred = text/html
+  </Content_Types>
+  <Badge>
+    URI    = http://validator.w3.org/images/vh20
+  </Badge>
+</HTML_2_0_Level_2>
+
+<HTML_2_0_Level_1>
+  Name       = html
+  Display    = HTML 2.0 Level 1
+  Info_URL   = http://www.w3.org/MarkUp/html-spec/
+  PubID      = -//IETF//DTD HTML 2.0 Level 2//EN
+  Parse_Mode = SGML
+  <Content_Types>
+    Allowed   = text/html
+    Forbidden = application/xhtml+xml
+    Preferred = text/html
+  </Content_Types>
+  <Badge>
+    URI    = http://validator.w3.org/images/vh20
+  </Badge>
+</HTML_2_0_Level_1>
+
+<HTML_2_0_Strict>
+  Name       = html
+  Display    = HTML 2.0 Strict
+  Info_URL   = http://www.w3.org/MarkUp/html-spec/
+  PubID      = -//IETF//DTD HTML 2.0 Strict//EN
+  Parse_Mode = SGML
+  <Content_Types>
+    Allowed   = text/html
+    Forbidden = application/xhtml+xml
+    Preferred = text/html
+  </Content_Types>
+  <Badge>
+    URI    = http://validator.w3.org/images/vh20
+  </Badge>
+</HTML_2_0_Strict>
+
+<HTML_2_0_Strict_Level_1>
+  Name       = html
+  Display    = HTML 2.0 Strict Level 1
+  Info_URL   = http://www.w3.org/MarkUp/html-spec/
+  PubID      = -//IETF//DTD HTML 2.0 Strict Level 1//EN
+  Parse_Mode = SGML
+  <Content_Types>
+    Allowed   = text/html
+    Forbidden = application/xhtml+xml
+    Preferred = text/html
+  </Content_Types>
+  <Badge>
+    URI    = http://validator.w3.org/images/vh20
+  </Badge>
+</HTML_2_0_Strict_Level_1>
+
+#
+# HTML 3.2.
+<HTML_3_2>
+  Name       = html
+  Display    = HTML 3.2
+  Info_URL   = http://www.w3.org/TR/REC-html32
+  PubID      = -//W3C//DTD HTML 3.2 Final//EN
+  Parse_Mode = SGML
+  <Content_Types>
+    Allowed   = text/html
+    Forbidden = application/xhtml+xml
+    Preferred = text/html
+  </Content_Types>
+  <Badge>
+    URI    = http://www.w3.org/Icons/valid-html32
+    Height = 31
+    Width  = 88
+  </Badge>
+</HTML_3_2>
+
+#
+# More "current" document types:
+# HTML 4.0, HTML 4.01, XHTML 1.0, XHTML 1.1, XHTML Basic.
+# (The three first in "Strict", "Transitional", and "Frameset" variants)
+<HTML_4_0_Strict>
+  Name       = html
+  Display    = HTML 4.0 Strict
+  Info_URL   = http://www.w3.org/TR/1998/REC-html40-19980424/
+  PubID      = -//W3C//DTD HTML 4.0//EN
+  SysID      = http://www.w3.org/TR/1998/REC-html40-19980424/strict.dtd
+  Parse_Mode = SGML
+  <Content_Types>
+    Allowed   = text/html
+    Forbidden = application/xhtml+xml
+    Preferred = text/html
+  </Content_Types>
+  <Namespaces>
+    Allowed   = 
+    Required  = 0
+  </Namespaces>
+  <Badge>
+    URI    = http://www.w3.org/Icons/valid-html401
+    Height = 31
+    Width  = 88
+  </Badge>
+</HTML_4_0_Strict>
+
+<HTML_4_0_Transitional>
+  Name       = html
+  Display    = HTML 4.0 Transitional
+  Info_URL   = http://www.w3.org/TR/1998/REC-html40-19980424/
+  PubID      = -//W3C//DTD HTML 4.0 Transitional//EN
+  SysID      = http://www.w3.org/TR/1998/REC-html40-19980424/loose.dtd
+  Parse_Mode = SGML
+  <Content_Types>
+    Allowed   = text/html
+    Forbidden = application/xhtml+xml
+    Preferred = text/html
+  </Content_Types>
+  <Namespaces>
+    Allowed   = 
+    Required  = 0
+  </Namespaces>
+  <Badge>
+    URI    = http://www.w3.org/Icons/valid-html401
+    Height = 31
+    Width  = 88
+  </Badge>
+</HTML_4_0_Transitional>
+
+<HTML_4_0_Frameset>
+  Name       = html
+  Display    = HTML 4.0 Frameset
+  Info_URL   = http://www.w3.org/TR/1998/REC-html40-19980424/
+  PubID      = -//W3C//DTD HTML 4.0 Frameset//EN
+  SysID      http://www.w3.org/TR/1998/REC-html40-19980424/frameset.dtd
+  Parse_Mode = SGML
+  <Content_Types>
+    Allowed   = text/html
+    Forbidden = application/xhtml+xml
+    Preferred = text/html
+  </Content_Types>
+  <Namespaces>
+    Allowed   = 
+    Required  = 0
+  </Namespaces>
+  <Badge>
+    URI    = http://www.w3.org/Icons/valid-html401
+    Height = 31
+    Width  = 88
+  </Badge>
+</HTML_4_0_Frameset>
+
+<HTML_4_01_Strict>
+  Name       = html
+  Display    = HTML 4.01 Strict
+  Info_URL   = http://www.w3.org/TR/1999/REC-html401-19991224/
+  PubID      = -//W3C//DTD HTML 4.01//EN
+  SysID      = http://www.w3.org/TR/1999/REC-html401-19991224/strict.dtd
+  Parse_Mode = SGML
+  <Content_Types>
+    Allowed   = text/html
+    Forbidden = application/xhtml+xml
+    Preferred = text/html
+  </Content_Types>
+  <Namespaces>
+    Allowed   = 
+    Required  = 0
+  </Namespaces>
+  <Badge>
+    URI    = http://www.w3.org/Icons/valid-html401
+    Height = 31
+    Width  = 88
+  </Badge>
+</HTML_4_01_Strict>
+
+<HTML_4_01_Transitional>
+  Name       = html
+  Display    = HTML 4.01 Transitional
+  Info_URL   = http://www.w3.org/TR/1999/REC-html401-19991224/
+  PubID      = -//W3C//DTD HTML 4.01 Transitional//EN
+  SysID      = http://www.w3.org/TR/1999/REC-html401-19991224/loose.dtd
+  Parse_Mode = SGML
+  <Content_Types>
+    Allowed   = text/html
+    Forbidden = application/xhtml+xml
+    Preferred = text/html
+  </Content_Types>
+  <Namespaces>
+    Allowed   = 
+    Required  = 0
+  </Namespaces>
+  <Badge>
+    URI    = http://www.w3.org/Icons/valid-html401
+    Height = 31
+    Width  = 88
+  </Badge>
+</HTML_4_01_Transitional>
+
+<HTML_4_01_Frameset>
+  Name       = html
+  Display    = HTML 4.01 Frameset
+  Info_URL   = http://www.w3.org/TR/1999/REC-html401-19991224/
+  PubID      = -//W3C//DTD HTML 4.01 Transitional//EN
+  SysID      = http://www.w3.org/TR/1999/REC-html401-19991224/frameset.dtd
+  Parse_Mode = SGML
+  <Content_Types>
+    Allowed   = text/html
+    Forbidden = application/xhtml+xml
+    Preferred = text/html
+  </Content_Types>
+  <Namespaces>
+    Allowed   = 
+    Required  = 0
+  </Namespaces>
+  <Badge>
+    URI    = http://www.w3.org/Icons/valid-html401
+    Height = 31
+    Width  = 88
+  </Badge>
+</HTML_4_01_Frameset>
+
+<XHTML_1_0_Strict>
+  Name       = html
+  Display    = XHTML 1.0 Strict
+  Info_URL   = http://www.w3.org/TR/2000/REC-xhtml1-20000126/
+  PubID      = -//W3C//DTD XHTML 1.0 Strict//EN
+  SysID      = http://www.w3.org/TR/2002/REC-xhtml1-20020801/DTD/xhtml1-strict.dtd
+  Parse_Mode = XML
+  <Content_Types>
+    Allowed   = text/html
+    Allowed   = application/xhtml+xml
+    Preferred = application/xhtml+xml
+  </Content_Types>
+  <Namespaces>
+    Allowed   = http://www.w3.org/1999/xhtml
+    Required  = 1
+  </Namespaces>
+  <Badge>
+    URI    = http://www.w3.org/Icons/valid-xhtml10
+    Height = 31
+    Width  = 88
+  </Badge>
+</XHTML_1_0_Strict>
+
+<XHTML_1_0_Transitional>
+  Name       = html
+  Display    = XHTML 1.0 Transitional
+  Info_URL   = http://www.w3.org/TR/2000/REC-xhtml1-20000126/
+  PubID      = -//W3C//DTD XHTML 1.0 Transitional//EN
+  SysID      = http://www.w3.org/TR/2002/REC-xhtml1-20020801/DTD/xhtml1-transitional.dtd
+  Parse_Mode = XML
+  <Content_Types>
+    Allowed   = text/html
+    Allowed   = application/xhtml+xml
+    Preferred = application/xhtml+xml
+  </Content_Types>
+  <Namespaces>
+    Allowed   = http://www.w3.org/1999/xhtml
+    Required  = 1
+  </Namespaces>
+  <Badge>
+    URI    = http://www.w3.org/Icons/valid-xhtml10
+    Height = 31
+    Width  = 88
+  </Badge>
+</XHTML_1_0_Transitional>
+
+<XHTML_1_0_Frameset>
+  Name       = html
+  Display    = XHTML 1.0 Frameset
+  Info_URL   = http://www.w3.org/TR/2000/REC-xhtml1-20000126/
+  PubID      = -//W3C//DTD XHTML 1.0 Frameset//EN
+  SysID      = http://www.w3.org/TR/2002/REC-xhtml1-20020801/DTD/xhtml1-frameset.dtd
+  Parse_Mode = XML
+  <Content_Types>
+    Allowed   = text/html
+    Allowed   = application/xhtml+xml
+    Preferred = application/xhtml+xml
+  </Content_Types>
+  <Namespaces>
+    Allowed   = http://www.w3.org/1999/xhtml
+    Required  = 1
+  </Namespaces>
+  <Badge>
+    URI    = http://www.w3.org/Icons/valid-xhtml10
+    Height = 31
+    Width  = 88
+  </Badge>
+</XHTML_1_0_Frameset>
+
+<XHTML_Basic_1_0>
+  Name       = html
+  Display    = XHTML Basic 1.0
+  Info_URL   = http://www.w3.org/TR/xhtml-basic/
+  PubID      = -//W3C//DTD XHTML Basic 1.0//EN
+  SysID      = http://www.w3.org/TR/2000/REC-xhtml-basic-20001219/xhtml-basic10.dtd
+  Parse_Mode = XML
+  <Content_Types>
+    Allowed   = text/html
+    Allowed   = application/xhtml+xml
+    Preferred = application/xhtml+xml
+  </Content_Types>
+  <Namespaces>
+    Allowed   = http://www.w3.org/1999/xhtml
+    Required  = 1
+  </Namespaces>
+  <Badge>
+    URI    = http://validator.w3.org/images/vxhtml-basic10
+    Height = 31
+    Width  = 88
+  </Badge>
+</XHTML_Basic_1_0>
+
+<XHTML_1_1>
+  Name       = html
+  Display    = XHTML 1.1
+  Info_URL   = http://www.w3.org/TR/xhtml11/
+  PubID      = -//W3C//DTD XHTML 1.1//EN
+  SysID      = http://www.w3.org/TR/2001/REC-xhtml11-20010531/DTD/xhtml11-flat.dtd
+  Parse_Mode = XML
+  <Content_Types>
+    Allowed   = application/xhtml+xml
+    Forbidden = text/html
+    Preferred = application/xhtml+xml
+  </Content_Types>
+  <Namespaces>
+    Allowed   = http://www.w3.org/1999/xhtml
+    Required  = 1
+  </Namespaces>
+  <Badge>
+    URI    = http://www.w3.org/Icons/valid-xhtml11
+    Height = 31
+    Width  = 88
+  </Badge>
+</XHTML_1_1>
+
+<ISO_IEC_15445_2000>
+  Name       = html
+  Display    = ISO/IEC 15445:2000 ("ISO HTML")
+  Info_URL   = http://purl.org/NET/ISO+IEC.15445/15445.html
+  PubID      = ISO/IEC 15445:2000//DTD HTML//EN
+  Parse_Mode = SGML
+  <Content_Types>
+    Allowed   = text/html
+    Forbidden = application/xhtml+xml
+    Preferred = text/html
+  </Content_Types>
+  <Badge>
+    URI    = http://validator.w3.org/images/v15445
+  </Badge>
+</ISO_IEC_15445_2000>
+
+
+#
+# @@@FIXME: Need to add in SVG, SMIL, etc.
diff --git a/htdocs/config/validator.conf b/htdocs/config/validator.conf
index 5a4466c..6e1543e 100644
--- a/htdocs/config/validator.conf
+++ b/htdocs/config/validator.conf
@@ -1,7 +1,7 @@
 #
 # Main Configuration File for the W3C MarkUp Validation Service.
 #
-# $Id: validator.conf,v 1.4 2002-12-07 04:00:18 link Exp $
+# $Id: validator.conf,v 1.5 2002-12-08 01:46:44 link Exp $
 #
 # See 'perldoc Config::General' for the syntax, and be aware that the
 # 'SplitPolicy' is 'equalsign', ie. keys and values are separated by '\s*=\s*'.
@@ -40,10 +40,6 @@ SGML_Library        = /usr/local/validator/htdocs/sgml-lib
   Include eref.cfg
 </Element_Map>
 
-<FPI_to_Text>
-  Include fpis.cfg
-</FPI_to_Text>
-
 <Error_to_URI>
   Include frag.cfg
 </Error_to_URI>
@@ -52,14 +48,14 @@ SGML_Library        = /usr/local/validator/htdocs/sgml-lib
   Include type.cfg
 </File_Type>
 
-<Doctypes>
-  Include doctypes.cfg
-</Doctypes>
-
 <Charsets>
   Include charset.cfg
 </Charsets>
 
+<Types>
+  Include types.cfg
+</Types>
+
 
 #
 # DanC territory...
diff --git a/htdocs/docs/devel.html b/htdocs/docs/devel.html
index 5b19d43..2619fd0 100755
--- a/htdocs/docs/devel.html
+++ b/htdocs/docs/devel.html
@@ -1,5 +1,5 @@
-<!--#set var="revision" value="\$Id: devel.html,v 1.11 2002-12-01 13:40:33 link Exp $"
---><!--#set var="date" value="\$Date: 2002-12-01 13:40:33 $"
+<!--#set var="revision" value="\$Id: devel.html,v 1.12 2002-12-08 01:46:44 link Exp $"
+--><!--#set var="date" value="\$Date: 2002-12-08 01:46:44 $"
 --><!--#set var="title" value="Developer Documentation for The W3C MarkUp Validation Service"
 --><!--#set var="relroot" value="../"
 --><!--#include virtual="../header.html" -->
@@ -36,7 +36,7 @@
         can be found on <a href="http://www.cpan.org/">CPAN</a> (minimum
         versions in parenthesis after the name). The following
         list was complete when CVS spit out:
-        <code>$Date: 2002-12-01 13:40:33 $</code>. <tt>:-)</tt>
+        <code>$Date: 2002-12-08 01:46:44 $</code>. <tt>:-)</tt>
       </p>
       <dl>
         <dt><code>CGI (2.81)</code></dt>
@@ -82,32 +82,102 @@
         here is a short description to get you started.
       </p>
       <dl>
+        <dt>validator.conf</dt>
+        <dd>
+          Main configuration file. Gives various parameters (such as the
+          address of the maintainer and the URL for the "Home Page") and
+          the locations of the other configuration files and mapping tables.
+        </dd>
+        <dt>types.conf</dt>
+        <dd>
+          <p>
+	    The main document type database for the Validator. This file
+            contains information on all the document types we know of. It
+            lets us map from a Public Identifier to a plain text version
+            string, lookup an URL for more information on a DOCTYPE, and
+            check which Content-Types and Namespaces are legal for this
+            particular DOCTYPE.
+          </p>
+          <p>And entry in this file looks like this:</p>
+<pre>
+&lt;XHTML_1_1&gt;
+  Name       = html
+  Display    = XHTML 1.1
+  Info_URL   = http://www.w3.org/TR/xhtml11/
+  PubID      = -//W3C//DTD XHTML 1.1//EN
+  SysID      = http://www.w3.org/TR/2001/REC-xhtml11-20010531/DTD/xhtml11-flat.dtd
+  Parse_Mode = XML
+  &lt;Content_Types&gt;
+    Allowed   = application/xhtml+xml
+    Forbidden = text/html
+    Preferred = application/xhtml+xml
+  &lt;/Content_Types&gt;
+  &lt;Namespaces&gt;
+    Allowed   = http://www.w3.org/1999/xhtml
+    Required  = 1
+  &lt;/Namespaces&gt;
+  &lt;Badge&gt;
+    URI    = http://www.w3.org/Icons/valid-xhtml11
+    Height = 31
+    Width  = 88
+  &lt;/Badge&gt;
+&lt;/XHTML_1_1&gt;
+</pre>
+          <p>
+            The name used for each section (e.g. "XHTML_1_1") is arbitrary.
+            The file will be turned inside out and will end up indexed by
+            the "PubID". This means that you cannot have two entries with
+            the same PubID. The rest of the parameters are:
+          </p>
+          <table class="config">
+            <tr><th>Name</th><td>The "Document Type Name" for this document type.</td></tr>
+	    <tr><th>Display</th><td>The pretty text version for the PubID.</td></tr>
+	    <tr><th>Info_URL</th><td>URL for more information on the PubID.</td></tr>
+            <tr><th>PubID</th><td>The Formal Public Identifier for this document type.</td></tr>
+	    <tr><th>SysID</th><td>A System Identifier for the DTD.</td></tr>
+	    <tr><th>Parse_Mode</th><td>Boolean describing whether to treat this as XML or SGML.</td></tr>
+	    <tr>
+	      <th>Content_Types</th>
+              <td class="subtable">
+	        <table>
+	          <tr><th>Allowed</th><td>Allowed Content-Types</td></tr>
+	          <tr><th>Forbidden</th><td>Forbidden Content-Types</td></tr>
+	          <tr><th>Preferred</th><td>Preferred Content-Types</td></tr>
+	        </table>
+	      </td>
+	    </tr>
+	    <tr>
+	      <th>Namespaces</th>
+	      <td class="subtable">
+	        <table>
+	          <tr><th>Allowed</th><td>Allowed Namespaces</td></tr>
+	          <tr><th>Required</th><td>Boolean describing whether a Namespace is required in this document type.</td></tr>
+	        </table>
+	      </td>
+	    </tr>
+	    <tr>
+	      <th>Badge</th>
+	      <td class="subtable">
+	        <table>
+	          <tr><th>URI</th><td>URI for a "Valid Foo" badge.</td></tr>
+	          <tr><th>Height</th><td>Height of this image.</td></tr>
+	          <tr><th>Width</th><td>Width of this image.</td></tr>
+	        </table>
+	      </td>
+	    </tr>
+          </table>
+        </dd>
         <dt>eref.cfg</dt>
         <dd>
           Contains the mappings from element names to an URI fragment
           (relative to a configurable URI) for their definitions. Used
           in output when the "Show Source Input" option is enabled.
         </dd>
-        <dt>fpis.cfg</dt>
-        <dd>
-          Maps FPIs to plain text version strings.
-        </dd>
         <dt>frag.cfg</dt>
         <dd>
           Maps error messages to an URI fragment identifier where an
           explanation of that error can be found.
         </dd>
-        <dt>type.cfg</dt>
-        <dd>
-          Maps MIME/HTTP Content-Types to an internal "document type" which
-          is used for treating HTML, XML, and XHTML in different ways.
-        </dd>
-        <dt>check.cfg</dt>
-        <dd>
-          Main configuration file. Gives various parameters (such as the
-          address of the maintainer and the URL for the "Home Page") and
-          the locations of the other configuration files and mapping tables.
-        </dd>
        </dl>
     </div>
     <div id="todo" class="stb">