Quick comment:
Alfranio Correia wrote:
[snip]
> === modified file 'sql/mysql_priv.h'
> --- a/sql/mysql_priv.h 2009-06-05 11:23:58 +0000
> +++ b/sql/mysql_priv.h 2009-07-13 04:01:35 +0000
> @@ -1025,8 +1025,8 @@ check_and_unset_inject_value(int value)
>
> #endif
>
> -void write_bin_log(THD *thd, bool clear_error,
> - char const *query, ulong query_length);
> +int write_bin_log(THD *thd, bool clear_error,
> + char const *query, ulong query_length);
>
> /* sql_connect.cc */
> int check_user(THD *thd, enum enum_server_command command,
>
>
> Here we have a funny function. It is defined in the current file,
> used by several modules, implemented in the sql/sql_table.cc and
> calls functions in the log.cc :)
>
> We need to do a refactory here. I don't know if in the context
> of this bug. What do you think?
Yes, it is a funny function with a funny placement, but don't refactor it in the
context of this bug. I am working on splitting up mysql_priv.h, and after that
we can start to move functions to proper components. If you do the refactor now,
it will cause merge problems.
You have worse examples in the code than this, but more about that later.
Best wishes,
Mats Kindahl
--
Mats Kindahl
Senior Software Engineer
Database Technology Group
Sun Microsystems
