Re: bk commit into 5.1 tree (davi:1.2661) BUG#32633

[Davi Arnaut]

Hi,

Alexander Nozdrin wrote:
> Hi,
> 
> thanks for working on this.
> 
> I'm generally okay with your patch, there is only one thing I haven't 
> understood yet:
> 
> On 28 January 2008 16:11:58 Davi Arnaut wrote:
>> diff -Nrup a/sql/event_db_repository.cc b/sql/event_db_repository.cc ---
>> a/sql/event_db_repository.cc	2007-11-01 18:52:50 -02:00
>> +++ b/sql/event_db_repository.cc	2008-01-28 11:11:50 -02:00
>> @@ -218,9 +218,13 @@ mysql_event_fill_row(THD *thd,
>>    */
>>    if (et->body_changed)
>>    {
>> +    LEX_STRING sql_mode;
>> +
>>      DBUG_ASSERT(sp->m_body.str);
>>
>> -    fields[ET_FIELD_SQL_MODE]->store((longlong)thd->variables.sql_mode,
>> TRUE); +    sys_var_thd_sql_mode::symbolic_mode_representation(thd,
>> thd->variables.sql_mode, +                                                 
>>      &sql_mode); +    fields[ET_FIELD_SQL_MODE]->store(sql_mode.str,
>> sql_mode.length, scs);
>>
>>      if (fields[f_num= ET_FIELD_BODY]->store(sp->m_body.str,
>>                                              sp->m_body.length,
> 
> why did you change that?
> 
> Could you please ping me when you're on IRC so that we can discuss this issue.
> 

AFAICS, nothing guarantees that sql_mode and the set field will have the
same bit pattern, so it's generally safer to use the string version
which will also do a better job of error detection (spew out a warning
or error if the bit is not found in the set).

Regards,

-- 
Davi Arnaut, Software Engineer
MySQL Inc, www.mysql.com

Are you MySQL certified?  www.mysql.com/certification