removing the list of supported character encodings,

as we have a better and more reliable technical way to do this.

Replacing with a mechanism to:
* suggest a better alias if a "bad" encoding alias is used
* refuse to transcode if encoding used is forbidden by policy
(I don't know any, but the mechanism is here...)

More details:
http://lists.w3.org/Archives/Public/public-qa-dev/2007Jul/0009.html
and surrounding thread.

5 files changed, 61 insertions, 69 deletions

diff --git a/htdocs/config/charset.cfg b/htdocs/config/charset.cfg
index 69ef847..aed81a6 100644
--- a/htdocs/config/charset.cfg
+++ b/htdocs/config/charset.cfg
@@ -1,60 +1,23 @@
 #
-# List of accepted encodings.
+# List of encodings aliases and forbidden encodings
 #
-# $Id: charset.cfg,v 1.13 2006-10-12 01:15:50 ot Exp $
-#
-# The Validator will refuse to decode documents in an encoding
-# other than those listed here. The list is independent of what
+# $Id: charset.cfg,v 1.14 2007-07-19 03:59:23 ot Exp $
+
+# This list indicates character encoding aliases that are 
+# not recommended, along with a recommended equivalent, e.g:
+# encoding-obscure = encoding-well-known
+
+# It also lists encoding names that the validator will refuse to treat:
+# bogus_encoding = Encoding Forbidden (Reason why)
+
+# The list is independent of what
 # is supported on a specific system but subject to the Validator
 # policy for acceptable encodings.
-#
 
-utf-8                          = 1
-utf-16                         = 1
-utf-16be                       = 1
-utf-16le                       = 1
-iso-8859-1                     = 1
-iso-8859-2                     = 1
-iso-8859-3                     = 1
-iso-8859-4                     = 1
-iso-8859-5                     = 1
-iso-8859-6                     = 1
-iso-8859-6-i                   = 1
-iso-8859-7                     = 1
-iso-8859-8                     = 1
-iso-8859-8-i                   = 1
-iso-8859-9                     = 1
-iso-8859-10                    = 1
-iso-8859-11                    = 1
-iso-8859-13                    = 1
-iso-8859-14                    = 1
-iso-8859-15                    = 1
-iso-8859-16                    = 1
-us-ascii                       = 1
-iso-2022-jp                    = 1
-shift_jis                      = 1
-euc-jp                         = 1
-gb2312                         = 1
-big5                           = 1
-iso-2022-kr                    = 1
-euc-kr                         = 1
-gb18030                        = 1
-tis-620                        = 1
-koi8-r                         = 1
-koi8-u                         = 1
-windows-1250                   = 1
-windows-1251                   = 1
-windows-1252                   = 1
-windows-1253                   = 1
-windows-1254                   = 1
-windows-1255                   = 1
-windows-1256                   = 1
-windows-1257                   = 1
-macintosh                      = 1
-x-mac-roman                    = 1
-x-sjis                         = 1
-iso8859-1                      = 1
-ascii                          = 1
-iso-8859-1-Windows-3.1-Latin-1 = 1
-ks_c_5601-1987                 = 1
-ksc_5601                       = 1
+
+x-mac-roman		= macintosh
+x-sjis			= shift_jis
+iso8859-1		= iso-8859-1
+ascii			= us-ascii
+# this one is in IANA, but better use only windows-1252
+iso-8859-1-Windows-3.1-Latin-1	= windows-1252
diff --git a/httpd/cgi-bin/check b/httpd/cgi-bin/check
index 97063af..2a4a79d 100755
--- a/httpd/cgi-bin/check
+++ b/httpd/cgi-bin/check
@@ -14,7 +14,7 @@
 # This source code is available under the license at:
 #     http://www.w3.org/Consortium/Legal/copyright-software
 #
-# $Id: check,v 1.540 2007-07-17 02:32:36 ot Exp $
+# $Id: check,v 1.541 2007-07-19 03:59:24 ot Exp $
 
 #
 # Disable buffering on STDOUT!
@@ -183,7 +183,7 @@ Directory not readable (permission denied): @_r
 
   #
   # Strings
-  $VERSION =  q$Revision: 1.540 $;
+  $VERSION =  q$Revision: 1.541 $;
   $VERSION =~ s/Revision: ([\d\.]+) /$1/;
 
   #
@@ -2083,19 +2083,25 @@ sub transcode {
 
   my $cs = $exact_charset;
 
-  if (!$CFG->{Charsets}->{$cs}) {
-    # The encoding is not supported due to policy
-    # and possibly other reasons
-
-    $File->{'Error Flagged'} = TRUE;
-    $File->{Templates}->{Error}->param(fatal_transcode_error   => TRUE);
-    $File->{Templates}->{Error}->param(fatal_transcode_charset => $cs);
-
-    # @@FIXME might need better text
-    $File->{Templates}->{Error}->param(fatal_transcode_errmsg  =>
-                                       "Encoding not supported.");
+  if ($CFG->{Charsets}->{$cs}) {
+    if ($CFG->{Charsets}->{$cs} =~ /Encoding Forbidden/) {
+      # The encoding is not supported due to policy
+      
+      $File->{'Error Flagged'} = TRUE;
+      $File->{Templates}->{Error}->param(fatal_transcode_error   => TRUE);
+      $File->{Templates}->{Error}->param(fatal_transcode_charset => $cs);
 
-    return $File;
+      # @@FIXME might need better text
+      $File->{Templates}->{Error}->param(fatal_transcode_errmsg  =>
+                                         "This encoding is not supported by the validator.");
+      return $File;
+    }
+    else {
+      &add_warning('W22', {
+        W22_declared => $cs,
+        W22_suggested => $CFG->{Charsets}->{$cs},
+      });      
+    }
   }
 
   # Does the system support decoding this encoding?
diff --git a/share/templates/en_US/soap_warnings.tmpl b/share/templates/en_US/soap_warnings.tmpl
index 4275972..463480d 100644
--- a/share/templates/en_US/soap_warnings.tmpl
+++ b/share/templates/en_US/soap_warnings.tmpl
@@ -73,6 +73,10 @@
 <TMPL_IF NAME="W21">
   <m:warning><m:message>Byte-Order Mark found in UTF-8 File.</m:message></m:warning>
 </TMPL_IF>
+<TMPL_IF NAME="W22">
+<m:warning><m:message>Character Encoding suggestion: use 
+<TMPL_VAR NAME="W22_suggested" ESCAPE="HTML"> instead of <TMPL_VAR NAME="W22_declared" ESCAPE="HTML"></m:warning></m:message>
+</TMPL_IF>
 
 <TMPL_IF NAME="W@@">
   <m:warning><m:message></m:message></m:warning>
diff --git a/share/templates/en_US/ucn_warnings.tmpl b/share/templates/en_US/ucn_warnings.tmpl
index ca33c71..5222cba 100644
--- a/share/templates/en_US/ucn_warnings.tmpl
+++ b/share/templates/en_US/ucn_warnings.tmpl
@@ -73,6 +73,10 @@
 <TMPL_IF NAME="W21">
   <warning><message>Byte-Order Mark found in UTF-8 File.</message></warning>
 </TMPL_IF>
+<TMPL_IF NAME="W22">
+<warning><message>Character Encoding suggestion: use 
+<TMPL_VAR NAME="W22_suggested" ESCAPE="HTML"> instead of <TMPL_VAR NAME="W22_declared" ESCAPE="HTML"></warning></message>
+</TMPL_IF>
 
 <TMPL_IF NAME="W@@">
   <warning><message></message></warning>
diff --git a/share/templates/en_US/warnings.tmpl b/share/templates/en_US/warnings.tmpl
index 6393aca..d99dcec 100644
--- a/share/templates/en_US/warnings.tmpl
+++ b/share/templates/en_US/warnings.tmpl
@@ -428,6 +428,21 @@
     </p>
   </li>
 </TMPL_IF>
+<TMPL_IF NAME="W22">
+  <li class="msg_warn" id="W22"><span class="err_type"><img src="images/info_icons/warning.png" alt="Warning" title="Warning" /></span> <span class="msg">Character Encoding suggestion: use 
+  <code><TMPL_VAR NAME="W22_suggested" ESCAPE="HTML"></code> instead of <code><TMPL_VAR NAME="W22_declared" ESCAPE="HTML"></code></span>
+  
+    <p>
+      The character encoding declared for this document 
+      (<code><TMPL_VAR NAME="W22_declared" ESCAPE="HTML"></code>) may not be widely supported.
+      This encoding is equivalent to <code><TMPL_VAR NAME="W22_suggested" ESCAPE="HTML"></code>
+      which may be better supported across platforms.
+    </p> 
+    <p>
+      More information on <a href="http://www.w3.org/International/O-charset.html">declaring a character encoding on your Web server or in your document</a> can be found on the W3C Internationalization site.
+    </p>
+  </li>
+</TMPL_IF>