On 4/30/18 07:01, Daniel Verite wrote:
>> As of v11, DO blocks can do transactions. I think this will meet your needs.
>
> They do support COMMIT and ROLLBACK in the current
> development tree, but not VACUUM as in Pierre's example.
Support for VACUUM can be added in the future. There is no big
On Mon, Apr 30, 2018 at 7:05 AM Pierre Ducroquet <
pierre.ducroq...@people-doc.com> wrote:
> On Monday, April 30, 2018 1:01:25 PM CEST Daniel Verite wrote:
> > Corey Huinker wrote:
> > > As of v11, DO blocks can do transactions. I think this will meet your
> > > needs.
> > They do support CO
On Monday, April 30, 2018 1:01:25 PM CEST Daniel Verite wrote:
> Corey Huinker wrote:
> > As of v11, DO blocks can do transactions. I think this will meet your
> > needs.
> They do support COMMIT and ROLLBACK in the current
> development tree, but not VACUUM as in Pierre's example.
>
> postg
Corey Huinker wrote:
> As of v11, DO blocks can do transactions. I think this will meet your needs.
They do support COMMIT and ROLLBACK in the current
development tree, but not VACUUM as in Pierre's example.
postgres=# \echo :SERVER_VERSION_NAME
11devel
postgres=# do ' begin vacuum; en
2018-04-28 21:25 GMT+02:00 Corey Huinker :
>
>> I did it too. It is not too hard - there was not a agreement on syntax.
>>
>> can be nice some like CTE
>>
>> WITH PROCEDURE x(a int, b int) AS ... $$ SELECT x(10);
>>
>
> I've seen "WITH function" syntax in Oracle (here's an example:
> https://o
>
>
> I did it too. It is not too hard - there was not a agreement on syntax.
>
> can be nice some like CTE
>
> WITH PROCEDURE x(a int, b int) AS ... $$ SELECT x(10);
>
I've seen "WITH function" syntax in Oracle (here's an example:
https://oracle-base.com/articles/12c/with-clause-enhancements-
2018-04-28 7:36 GMT+02:00 Corey Huinker :
>
>
>
>>> What you don't see here is that you're using your psql process's
>>> available open file handles as a stack, and when you hit that limit psql
>>> will fail. If you remove that limit, then you get a bit further before psql
>>> segfaults on you. I
>> What you don't see here is that you're using your psql process's
>> available open file handles as a stack, and when you hit that limit psql
>> will fail. If you remove that limit, then you get a bit further before psql
>> segfaults on you. I think I got ~2700 files deep before that happened. Yo
Hi
2018-04-27 21:40 GMT+02:00 Corey Huinker :
> As of v11, DO blocks can do transactions. I think this will meet your
> needs.
>
> A loop that starts at point X in the code and terminates at point Y has to
> know how to jump back in the file (if there even is a file!) to point X and
> re-interpre
As of v11, DO blocks can do transactions. I think this will meet your needs.
A loop that starts at point X in the code and terminates at point Y has to
know how to jump back in the file (if there even is a file!) to point X and
re-interpret commands as it makes it's way back through the "file" tow
Hi
2018-04-24 9:58 GMT+02:00 Pierre Ducroquet
:
> Hi
>
> When running database migrations with .sql files on a live database, it's
> not
> uncommon to have to run a migration in a loop to prevent a big lock on a
> table.
> For instance if one want to delete some old datas from a big table one
>
Hi
When running database migrations with .sql files on a live database, it's not
uncommon to have to run a migration in a loop to prevent a big lock on a
table.
For instance if one want to delete some old datas from a big table one would
write :
DELETE FROM big_table WHERE id IN (SELECT id FRO
12 matches
Mail list logo
