Re: A few questions about ltree
| От | Alban Hertroys |
|---|---|
| Тема | Re: A few questions about ltree |
| Дата | |
| Msg-id | 4448EBF1.4040708@magproductions.nl обсуждение исходный текст |
| Ответ на | Re: A few questions about ltree (Teodor Sigaev <teodor@sigaev.ru>) |
| Ответы |
Re: A few questions about ltree
Re: A few questions about ltree |
| Список | pgsql-general |
Teodor Sigaev wrote:
>> Maybe something along the lines of the following is possible?:
>
> Exact, it's for what ltree was developed.
Cool, looks like it is what I need then.
> contrib_regression=# select 'a.b.c' <@ 'a.b'::ltree;
> ?column?
> ----------
> t
> (1 row)
How would you use this to constrain a foreign key?
We've been experimenting with a table containing a branch 'a', 'a.b' and
'a.b.c', but deleting 'a.b' didn't cause a constraint violation.
SQL> CREATE TABLE ltree_test (path ltree PRIMARY KEY REFERENCES
ltree_test(path));
NOTICE: CREATE TABLE / PRIMARY KEY will create implicit index
"ltree_test_pkey" for table "ltree_test"
CREATE TABLE
SQL> INSERT INTO ltree_test VALUES ('a'::ltree);
INSERT 84117368 1
SQL> INSERT INTO ltree_test VALUES ('a.b'::ltree);
INSERT 84117369 1
SQL> INSERT INTO ltree_test VALUES ('a.b.c'::ltree);
INSERT 84117370 1
SQL> DELETE FROM ltree_test WHERE path = 'a.b'::ltree;
DELETE 1
SQL> select * from ltree_test;
path
-------
a
a.b.c
(2 rows)
Is there some obvious/easy way to prevent this?
Regards,
--
Alban Hertroys
alban@magproductions.nl
magproductions b.v.
T: ++31(0)534346874
F: ++31(0)534346876
M:
I: www.magproductions.nl
A: Postbus 416
7500 AK Enschede
// Integrate Your World //
В списке pgsql-general по дате отправления: