Few comments for 002:

1)
@@ -210,9 +269,16 @@ ObjectsInPublicationToOids(List *pubobjspec_list,
ParseState *pstate,
  break;
  case PUBLICATIONOBJ_TABLES_IN_SCHEMA:
  schemaid = get_namespace_oid(pubobj->name, false);
+ schema_name = pubobj->name;
+ schema_repeated = list_member_oid(*schemas, schemaid);

  /* Filter out duplicates if user specifies "sch1, sch1" */
  *schemas = list_append_unique_oid(*schemas, schemaid);
+
+ ProcessSchemaExceptTables(schemaid, schema_name,
+   pubobj->except_tables, pstate,
+   schema_repeated, &schemas_with_except,
+

Should we extract schema_name inside ProcessSchemaExceptTables, as we
are passing schemaid to it. Generally one such arguement should
suffice.

2)
We can also avoid passing 'schema_repeated' to function
ProcessSchemaExceptTables(). Instead we can pass input/output arg
'schemas' similar to 'schemas_with_except'. Let it extract 'repeated'
from schemas internally and also populate it internally with unique
schema-oid.

3)
ProcessSchemaExceptTables:

can we avoid repeated dereferences of eobj->pubtable->relation. We can
work with:
RangeVar   *relation = eobj->pubtable->relation;

4)

+
+ if (except_tables != NIL)
+ *schemas_with_except = lappend_oid(*schemas_with_except, schemaid);

We can return early here:
if (except_tables == NIL)
return;

*schemas_with_except = lappend_oid(*schemas_with_except, schemaid);

thanks
Shveta


Reply via email to