#86 cubrid_affected_rows() requires a different parameter than mysql_affected_rows()

[Jira Trac]

cubrid_affected_rows() has the following definition:

{code}
int cubrid_affected_rows ([ resource $result ] )
{code}

While mysql_affected_rows():

{code}
int mysql_affected_rows ([ resource $link_identifier ] )
{code}

Thus, they are incompatible as MySQL requires a connection resource, while CUBRID requires result resource. Is there any specific reason we require result resource instead of connection resource like in MySQL?