Re: Add LINE: hint when schemaname.typename is a non-existent schema

Поиск
Список
Период
Сортировка
От Jeevan Chalke
Тема Re: Add LINE: hint when schemaname.typename is a non-existent schema
Дата
Msg-id 20150304073121.2540.19086.pgcf@coridan.postgresql.org
обсуждение исходный текст
Ответ на Re: Add LINE: hint when schemaname.typename is a non-existent schema  (Tom Lane <tgl@sss.pgh.pa.us>)
Список pgsql-hackers
The following review has been posted through the commitfest application:
make installcheck-world:  not tested
Implements feature:       tested, failed
Spec compliant:           not tested
Documentation:            not tested

Tom suggested few changes already which I too think author needs to address.
So marking it "Waiting on Author".

However, I see following, example does not work well.

postgres=# create or replace function
f1(a abc.test.id%type) returns int as
$$ select 1; $$
language sql;
ERROR:  schema "abc" does not exist

Is that expected?
I guess we need it at all places in parse_*.c where we will look for namespace.
Please fix.


Also, like Tom's suggestion on make_oper_cache_key, can we push down this
inside func_get_detail() as well, just to limit it for namespace lookup?

However, patch is not getting applied cleanly on latest sources. Need rebase.

> On Tom comments on parse_utilcmd.c:
I guess the block is moved after the pstate and CreateStmtContext are setup
properly.  I guess, we can move just after pstate setup, so that it will
result into minimal changes?

Can we have small test-case? Or will it be too much for this feature?

The new status of this patch is: Waiting on Author



В списке pgsql-hackers по дате отправления:

Предыдущее
От: Jim Nasby
Дата:
Сообщение: Re: How about to have relnamespace and relrole?
Следующее
От: Oskari Saarenmaa
Дата:
Сообщение: Re: Configurable location for extension .control files