Re: Ticket 269: Add support for 9.1 ALTER TYPE new syntax for enum
От | Dave Page |
---|---|
Тема | Re: Ticket 269: Add support for 9.1 ALTER TYPE new syntax for enum |
Дата | |
Msg-id | AANLkTinG3_mDJBeaYVgrQCEe_NwFj1K574B62-QHPiDO@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: Ticket 269: Add support for 9.1 ALTER TYPE new syntax for enum (Guillaume Lelarge <guillaume@lelarge.info>) |
Ответы |
Re: Ticket 269: Add support for 9.1 ALTER TYPE
new syntax for enum
|
Список | pgadmin-hackers |
On Sat, Oct 30, 2010 at 4:07 PM, Guillaume Lelarge <guillaume@lelarge.info> wrote: > Le 29/10/2010 21:56, Guillaume Lelarge a écrit : >> Le 29/10/2010 21:11, Guillaume Lelarge a écrit : >>> [...] >>> This patch adds support to the new ALTER TYPE syntax in 9.1 for enums. >>> >>> It's working great except one thing. If a user wants to add two labels, >>> we're screwed because we can't do two "ALTER TYPE ... ADD" statements in >>> the same query execution. Any idea how to solve this? the only way I >>> found would be to disallow adding two labels at once but it results on a >>> less interesting feature. >>> >> >> So I was wrong. The issue is that we can't issue this statement in a >> explicit transaction. Any idea how to solve the "don't send begin/end >> statements"? >> > > The only idea I have is to make dlgType a two-SQL-boxes dialog and > modify the dlgProperty::apply() method so that if there is "ALTER TYPE > ... ADD {BEFORE | AFTER}" statements, they get splitted and fired > individualy. I didn't yet write the code to split the statement, but it > will surely be ugly. > > Any objection on doing this? or better idea to fix this issue? I don't understand why we can't do all this at once - what's the problem exactly? Normally we only split statements when they're not atomic and cannot be rolled back. -- Dave Page Blog: http://pgsnake.blogspot.com Twitter: @pgsnake EnterpriseDB UK: http://www.enterprisedb.com The Enterprise PostgreSQL Company
В списке pgadmin-hackers по дате отправления: