Bug#11765108 (Bug#58036) client utf32, utf16, ucs2 should be disallowed, they crash server

A separate fix for 5.1 (as 5.1 and 5.5 have seriously
differged in the related pieces of the code).
A patch for 5.5 was approved earlier.

Problem: ucs2 was correctly disallowed in "SET NAMES" only,
while mysql_real_connect() and mysql_change_user() still allowed
to use ucs2, which made server crash.

Fix: disallow ucs2 in mysql_real_connect() and mysql_change_user().

  @ sql/sql_priv.h
    - changing return type for thd_init_client_charset() to bool,
      to return errors to the caller

  @ sql/sql_var.cc
    - using new function

  @ sql/sql_connect.cc
    - thd_client_charset_init:
      in case of unsupported client character set send error and return true;
      in case of success return false
    - check_connection:
      Return error if character set initialization failed

  @ sql/sql_parse.cc
    - check charset in the very beginnig of the CMD_CHANGE_USER handling code

  @ tests/mysql_client_test.c
    - adding tests

4 files changed, 50 insertions, 9 deletions

diff --git a/sql/mysql_priv.h b/sql/mysql_priv.h
index 1838f0c924d..67631b265ab 100644
--- a/sql/mysql_priv.h
+++ b/sql/mysql_priv.h
@@ -1019,7 +1019,11 @@ void reset_mqh(LEX_USER *lu, bool get_them);
 bool check_mqh(THD *thd, uint check_command);
 void time_out_user_resource_limits(THD *thd, USER_CONN *uc);
 void decrease_user_connections(USER_CONN *uc);
-void thd_init_client_charset(THD *thd, uint cs_number);
+bool thd_init_client_charset(THD *thd, uint cs_number);
+inline bool is_supported_parser_charset(CHARSET_INFO *cs)
+{
+  return test(cs->mbminlen == 1);
+}
 bool setup_connection_thread_globals(THD *thd);
 
 int mysql_create_db(THD *thd, char *db, HA_CREATE_INFO *create, bool silent);
diff --git a/sql/set_var.cc b/sql/set_var.cc
index 26c9b06a912..831b68bbe14 100644
--- a/sql/set_var.cc
+++ b/sql/set_var.cc
@@ -2187,7 +2187,7 @@ bool sys_var_character_set_client::check(THD *thd, set_var *var)
   if (sys_var_character_set_sv::check(thd, var))
     return 1;
   /* Currently, UCS-2 cannot be used as a client character set */
-  if (var->save_result.charset->mbminlen > 1)
+  if (!is_supported_parser_charset(var->save_result.charset))
   {
     my_error(ER_WRONG_VALUE_FOR_VAR, MYF(0), name, 
              var->save_result.charset->csname);
diff --git a/sql/sql_connect.cc b/sql/sql_connect.cc
index 9fa6966baa2..8129324e300 100644
--- a/sql/sql_connect.cc
+++ b/sql/sql_connect.cc
@@ -582,8 +582,23 @@ void reset_mqh(LEX_USER *lu, bool get_them= 0)
 }
 
 
-void thd_init_client_charset(THD *thd, uint cs_number)
+/**
+  Set thread character set variables from the given ID
+
+  @param  thd         thread handle
+  @param  cs_number   character set and collation ID
+
+  @retval  0  OK; character_set_client, collation_connection and
+              character_set_results are set to the new value,
+              or to the default global values.
+
+  @retval  1  error, e.g. the given ID is not supported by parser.
+              Corresponding SQL error is sent.
+*/
+
+bool thd_init_client_charset(THD *thd, uint cs_number)
 {
+  CHARSET_INFO *cs;
   /*
    Use server character set and collation if
    - opt_character_set_client_handshake is not set
@@ -592,10 +607,10 @@ void thd_init_client_charset(THD *thd, uint cs_number)
    - client character set doesn't exists in server
   */
   if (!opt_character_set_client_handshake ||
-      !(thd->variables.character_set_client= get_charset(cs_number, MYF(0))) ||
+      !(cs= get_charset(cs_number, MYF(0))) ||
       !my_strcasecmp(&my_charset_latin1,
                      global_system_variables.character_set_client->name,
-                     thd->variables.character_set_client->name))
+                     cs->name))
   {
     thd->variables.character_set_client=
       global_system_variables.character_set_client;
@@ -606,10 +621,18 @@ void thd_init_client_charset(THD *thd, uint cs_number)
   }
   else
   {
+    if (!is_supported_parser_charset(cs))
+    {
+      /* Disallow non-supported parser character sets: UCS2, UTF16, UTF32 */
+      my_error(ER_WRONG_VALUE_FOR_VAR, MYF(0), "character_set_client",
+               cs->csname);
+      return true;
+    }
     thd->variables.character_set_results=
       thd->variables.collation_connection= 
-      thd->variables.character_set_client;
+      thd->variables.character_set_client= cs;
   }
+  return false;
 }
 
 
@@ -782,7 +805,8 @@ static int check_connection(THD *thd)
     thd->client_capabilities|= ((ulong) uint2korr(net->read_pos+2)) << 16;
     thd->max_client_packet_length= uint4korr(net->read_pos+4);
     DBUG_PRINT("info", ("client_character_set: %d", (uint) net->read_pos[8]));
-    thd_init_client_charset(thd, (uint) net->read_pos[8]);
+    if (thd_init_client_charset(thd, (uint) net->read_pos[8]))
+      return 1;
     thd->update_charset();
     end= (char*) net->read_pos+32;
   }
diff --git a/sql/sql_parse.cc b/sql/sql_parse.cc
index 6ce800312ff..9bf55f4dcdd 100644
--- a/sql/sql_parse.cc
+++ b/sql/sql_parse.cc
@@ -1153,13 +1153,22 @@ bool dispatch_command(enum enum_server_command command, THD *thd,
 
     if (ptr < packet_end)
     {
+      CHARSET_INFO *cs;
       if (ptr + 2 > packet_end)
       {
         my_message(ER_UNKNOWN_COM_ERROR, ER(ER_UNKNOWN_COM_ERROR), MYF(0));
         break;
       }
 
-      cs_number= uint2korr(ptr);
+      if ((cs_number= uint2korr(ptr)) &&
+          (cs= get_charset(cs_number, MYF(0))) &&
+          !is_supported_parser_charset(cs))
+      {
+        /* Disallow non-supported parser character sets: UCS2, UTF16, UTF32 */
+        my_error(ER_WRONG_VALUE_FOR_VAR, MYF(0), "character_set_client",
+                 cs->csname);
+        break;
+      }        
     }
 
     /* Convert database name to utf8 */
@@ -1205,7 +1214,11 @@ bool dispatch_command(enum enum_server_command command, THD *thd,
 
       if (cs_number)
       {
-        thd_init_client_charset(thd, cs_number);
+        /*
+          We have checked charset earlier,
+          so thd_init_client_charset cannot fail.
+        */
+        DBUG_ASSERT(!thd_init_client_charset(thd, cs_number));
         thd->update_charset();
       }
     }