From aef091a8491852470f71f5c3c2eb3c0db2fb7360 Mon Sep 17 00:00:00 2001
From: brectanus <brectanus@9017d574-64ec-4062-9424-5e00b32a252b>
Date: Fri, 15 Feb 2008 23:05:30 +0000
Subject: [PATCH] Reverted r950 which moved the periods from the message to
 after the "[offset ...]" tag.  This tag was intended to be interpreted as
 metadata. Enhanced the documentation from r951 to reflect "[offset ...]" as
 metadata and not the message.

---
 apache2/re_operators.c                | 14 +++++++-------
 doc/modsecurity2-apache-reference.xml | 24 ++++++++++++++----------
 2 files changed, 21 insertions(+), 17 deletions(-)

diff --git a/apache2/re_operators.c b/apache2/re_operators.c
index 00a22e7d..494b2449 100644
--- a/apache2/re_operators.c
+++ b/apache2/re_operators.c
@@ -1134,7 +1134,7 @@ static int msre_op_verifyCC_execute(modsec_rec *msr, msre_rule *rule, msre_var *
         /* Match. */
 
         /* This message will be logged. */
-        *error_msg = apr_psprintf(msr->mp, "CC# match \"%s\" at %s [offset \"%d\"].",
+        *error_msg = apr_psprintf(msr->mp, "CC# match \"%s\" at %s. [offset \"%d\"]",
             regex->pattern, var->name, offset);
 
         return 1;
@@ -1669,30 +1669,30 @@ static int msre_op_validateUtf8Encoding_execute(modsec_rec *msr, msre_rule *rule
             case UNICODE_ERROR_CHARACTERS_MISSING :
                 *error_msg = apr_psprintf(msr->mp, "Invalid UTF-8 encoding: "
                     "not enough bytes in character "
-                    "at %s [offset \"%d\"].", var->name, i);
+                    "at %s. [offset \"%d\"]", var->name, i);
                 return 1;
                 break;
             case UNICODE_ERROR_INVALID_ENCODING :
                 *error_msg = apr_psprintf(msr->mp, "Invalid UTF-8 encoding: "
                     "invalid byte value in character "
-                    "at %s [offset \"%d\"].", var->name, i);
+                    "at %s. [offset \"%d\"]", var->name, i);
                 return 1;
                 break;
             case UNICODE_ERROR_OVERLONG_CHARACTER :
                 *error_msg = apr_psprintf(msr->mp, "Invalid UTF-8 encoding: "
                     "overlong character detected "
-                    "at %s [offset \"%d\"].", var->name, i);
+                    "at %s. [offset \"%d\"]", var->name, i);
                 return 1;
                 break;
             case UNICODE_ERROR_RESTRICTED_CHARACTER :
                 *error_msg = apr_psprintf(msr->mp, "Invalid UTF-8 encoding: "
                     "use of restricted character "
-                    "at %s [offset \"%d\"].", var->name, i);
+                    "at %s. [offset \"%d\"]", var->name, i);
                 return 1;
                 break;
             case UNICODE_ERROR_DECODING_ERROR :
-                *error_msg = apr_psprintf(msr->mp, "Invalid UTF-8 encoding: decoding error "
-                    "at %s [offset \"%d\"].", var->name, i);
+                *error_msg = apr_psprintf(msr->mp, "Error validating UTF-8 decoding "
+                    "at %s. [offset \"%d\"]", var->name, i);
                 return 1;
                 break;
         }
diff --git a/doc/modsecurity2-apache-reference.xml b/doc/modsecurity2-apache-reference.xml
index 53f627fe..b24a67dc 100644
--- a/doc/modsecurity2-apache-reference.xml
+++ b/doc/modsecurity2-apache-reference.xml
@@ -5599,33 +5599,32 @@ against "REQUEST_PROTOCOL" required.</programlisting>
           <itemizedlist>
             <listitem>
               <para>Invalid UTF-8 encoding: not enough bytes in character at
-              %0 [offset %1].</para>
+              %0.</para>
             </listitem>
 
             <listitem>
               <para>Invalid UTF-8 encoding: invalid byte value in character at
-              %0 [offset %1].</para>
+              %0.</para>
             </listitem>
 
             <listitem>
-              <para>Invalid UTF-8 encoding: overlong character detected at %0
-              [offset %1].</para>
+              <para>Invalid UTF-8 encoding: overlong character detected at
+              %0.</para>
             </listitem>
 
             <listitem>
-              <para>Invalid UTF-8 encoding: use of restricted character at %0
-              [offset %1].</para>
+              <para>Invalid UTF-8 encoding: use of restricted character at
+              %0.</para>
             </listitem>
 
             <listitem>
-              <para>Invalid UTF-8 encoding: decoding error at %0 [offset
-              %1].</para>
+              <para>Invalid UTF-8 encoding: decoding error at %0.</para>
             </listitem>
           </itemizedlist>
         </listitem>
 
         <listitem>
-          <para>@verifyCC - CC# match %0 at %1 [offset %2].</para>
+          <para>@verifyCC - CC# match %0 at %1.</para>
         </listitem>
       </itemizedlist>
 
@@ -5654,6 +5653,11 @@ against "REQUEST_PROTOCOL" required.</programlisting>
       <para>The following metadata fields are currently used:</para>
 
       <orderedlist>
+        <listitem>
+          <para><literal>offset</literal> - The byte offset where a match
+          occured within the target data. This is not always available.</para>
+        </listitem>
+
         <listitem>
           <para><literal>id</literal> - Unique rule ID, as specified by the
           <literal>id</literal> action.</para>
@@ -5912,4 +5916,4 @@ Server: Apache/2.x.x
       </section>
     </section>
   </section>
-</article>
+</article>
\ No newline at end of file