From: | Guillaume Lelarge <guillaume(at)lelarge(dot)info> |
---|---|
To: | Vladimir Kokovic <vladimir(dot)kokovic(at)gmail(dot)com> |
Cc: | pgadmin-hackers(at)postgresql(dot)org |
Subject: | Re: Copy/Paste table(s) functions - git context patch |
Date: | 2011-07-19 20:19:48 |
Message-ID: | 1311106789.2036.80.camel@laptop |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgadmin-hackers |
On Sun, 2011-07-17 at 23:17 +0200, Vladimir Kokovic wrote:
> OK
>
> > Now that we can copy on the same schema, it shouldn't ask for an
> > extension, but for the complete name.
>
> I do not know how to do when the table has objects that each has a name!
>
> CREATE TABLE gk_vrsta_naloga_vk --NAME1
> (
> id bigint NOT NULL DEFAULT
> nextval('"''id_gk_vrsta_naloga''"'::regclass), --NAME2
> sifra character varying NOT NULL DEFAULT ''::character varying,
> CONSTRAINT gk_vrsta_naloga_vk_pkey PRIMARY KEY (id), --NAME3
> CONSTRAINT unique_gk_vrsta_naloga_vk1 UNIQUE (sifra, id), --NAME4
> CONSTRAINT unique_gk_vrsta_naloga_vk2 UNIQUE (id, sifra), --NAME5
> ... --NAMEn
> )
> WITH (
> OIDS=FALSE
>
>
> Suffix is much better solution ...
>
Good point.
I tried a few other things and it seems pretty solid. One of my tests
got a segfault: I copy all tables (4) in a schema, create a schema,
paste the tables in the schema. Segfault directly.
So I tried with a really small schema: only one object, a table,
declared this way: "CREATE TABLE t1 (c1 bigint);". With the previous
steps, I got a crash.
The threading is a great idea but it makes things a lot harder.
--
Guillaume
http://blog.guillaume.lelarge.info
http://www.dalibo.com
From | Date | Subject | |
---|---|---|---|
Next Message | Guillaume Lelarge | 2011-07-19 20:20:34 | Re: Altering column collation |
Previous Message | Dave Page | 2011-07-19 20:12:04 | Re: Handling connection loss |