Hi, > > Attached is v2 with slightly improved code, nothing has changed > feature-wise. >
Here are some comments on this patch from my side: grant.sgml * Maybe we should use <replaceable class="parameter">schemaname</replaceable> in the sgml references instead of just <replaceable>schemaname</replaceable> + There is also the posibility of granting permissions to all objects of + given type inside one or multiple schemas. This functionality is supported + for tables, views, sequences and functions and can done by using + ALL TABLES IN schemanema syntax in place of object name. + </para> + + <para> typo "posibility" It should be ALL [TABLES|VIEWS|SEQUENCES|FUNCTIONS} IN schemaname (note the other typo here) syntax to be precise IMHO. aclchk.c + elog(ERROR, "unrecognized GrantStmt.objtype: %d", + (int) objtype); Kinda funny to mention the C structure name in the error. But I see that the other functions in the file do the same, so should be ok. I doubt if the syntax allows any other object type to reach upto this function anyways :) parsenodes.h GrantObjectType objtype; /* kind of object being operated on */ + bool is_schema; /* if true we want all objects + * of objtype in schema */ You forgot to make changes in _copyGrantStmt and _equalGrantStmt to account for this new field. Rest of the changes look straightforward and ok to me. make installcheck passes cleanly too. I also do not see any new warnings due to this patch. As an aside, I was just wondering the behaviour for RELKIND_INDEX? Regards, Nikhils -- http://www.enterprisedb.com -- Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-hackers