I am new to PostgreSQL.
Can you please tell me why the below query is not working.
Select *
from custom_test as dbtable
inner join aLocalArray as localtable on dbtable.id = localtable.id
Here custom_test is database table in PostgreSQL. aLocalArray is an array which is prepared by me.
If you want to join the array, you probably want to use UNNEST(array):
SELECT *
FROM custom_test AS dbtable
JOIN UNNEST (aLocalArray) AS localtable (id)
ON dbtable.id = localtable.id
But as suggested in the comments, the dbtable.id = any(aLocalArray) predicate or even the ARRAY[dbtable.id] #> aLocalArray operator might also do the trick.
Related
I am trying to select from one table using the select result from another table. I can run this in two queries but would like to optimize it into just one.
First query.. Select ids where matching other id
select id from lookuptable where paid = '547'
This results in something like this
6316352
6316353
6318409
6318410
6320468
6320469
6320470
6322526
6322527
6324586
6324587
6326648
I would like to then use this result to make another selection. I can do it manually like below. Note, there could be many rows with these values so I've been using a IN statement
select * from "othertable" where id in (6316352,6316353,6318409,6318410,6320468,6320469,6320470,6322526,6322527,6324586,6324587,6326648);
select
ot.*
from
"othertable" as ot
join
lookuptable as lt
on
ot.id = lt.id
where
lt.paid = '547'
The IN operator supports not just value lists but also subqueries, so you can literally write
select * from "othertable" where id in (select id from lookuptable where paid = '547');
Hi Guys I'm doing inner join for two tables and selecting all the columns from both tables. But I'm getting three cols with same name like id, created_at and updated_at.
Query:
SELECT addresses.* , facilities.* FROM facilities
INNER JOIN addresses
ON facilities.main_address_id = addresses.id
Is there any possible way that I can mention alias for above cols having same name while selecting all cols with * ?
Help of any kind would be appreciated! Thanks!
No you can't do this other than aliasing each column separately.
But if your query will be repetitive you could create VIEW:
CREATE OR REPLACE VIEW facilities_addresses AS
SELECT
addresses.column AS "addresses_column",
facilities.column AS "facilities_column"
FROM facilities
INNER JOIN addresses ON (facilities.main_address_id = addresses.id)
and then you can query:
SELECT * FROM facilities_addresses
yes you can
SELECT
addr.id as addressesId ,
addr.created_at as addresses_created_at,
addr.updated_at as addresses_update_at,
fac.id as facilitiesId,
fac.created_at as facilities_created_at,
fac.updated_at as facilities_updated_at FROM facilities as fac
INNER JOIN addresses as addr
ON facilities.main_address_id = addresses.id
I have given a task to optimize the below sql query. Currently the query is timing out and causing a lot of blocking . I just started using t-sql, so please help me with optimizing the query.
select ExcludedID
from OfferConditions with (NoLock)
where OfferID = 27251
and ExcludedID in (210,223,409,423,447,480,633,...lots and lots of these...,
13346,13362,13380,13396,13407,1,2)
union
select CustomerGroupID as ExcludedID
from CPE_IncentiveCustomerGroups ICG with (NoLock)
inner join CPE_RewardOptions RO with (NoLock)
on RO.RewardOptionID = ICG.RewardOptionID
where RO.IncentiveID = 27251
AND ICG.Deleted = 0 and RO.Deleted = 0 and
and ExcludedUsers = 1
and CustomerGroupID in (210,223,409,423,447,480,633,...lots and lots of these...,
13346,13362,13380,13396,13407,1,2);
You can try to insert those IDs to temp table and join it instead of using IN statement.
The key to solving you problem is NOT to fix the SQL, but to fix indexes on your tables. For example, you should have a compound index on the OfferConditions table with OfferID and ExcludedID.
When you create the indexes on the other tables, remember that if the field is in the where OR the join filter, it should be part of your compound index.
Is there an equivalent to the following SQL using the Zend_Db modules?
UPDATE items,month SET items.price=month.price
WHERE items.id=month.id;
The mysql manual states
... multiple-table UPDATE statements can use any type of join permitted in
SELECT statements, such as LEFT JOIN.
You can always execute a query string.
$db->query("UPDATE items,month SET items.price=month.price WHERE items.id=month.id")
Disclaimer: I haven't tested it.
With MySQL, you can use JOIN (left, right or inner) in UPDATE the same way as you would in a SELECT statement.
Your code would look like this:
$sql = <<<END
UPDATE table1
INNER JOIN table2 ON table1.id = table2.other_id
SET table1.newvalue = 'value'
WHERE table1.id = 99
AND table2.other_value = 'value'
END;
$result = $db->query($sql);
The $result variable will contain an object related to your adapter type. I use PDO_MYSQL so I ended up with a Zend_Db_Statement_Pdo object.
I am attempting to call data after joining all of my tables in a postgreSQL query.
I am getting the following error:
Error in postgresqlExecStatement(conn, statement, ...) :
RS-DBI driver: (could not Retrieve the result : ERROR: table name "place" specified more than once
)
Failed to execute SQL chunk
After referring to similar posts and online resources, I have attempted setting aliases (which only create new errors about not referring to the other tables in the FROM clause), reordering the table names (the cleanest, reordered chunk is posted below), and experimenting with UPDATE/FROM/JOIN as an alternative to SELECT/FROM/JOIN. However, I think I ultimately need to be using the SELECT clause.
SELECT *
FROM place
INNER JOIN place ON place.key = form.key
INNER JOIN form ON form.id = items.formid
INNER JOIN items ON items.recid = rec.id
INNER JOIN rec ON rec.id = sub.recid
INNER JOIN sub ON sub.id = type.subid
INNER JOIN type ON type.name = det.typeid;
You have "place" in your query twice, which is allowed but you would have to alias the second use of the table "place". Also you reference something called "det", but it's not a table or alias anywhere in your query. If you want only one place that's fine, just remove the INNER JOIN place and move your place.key = form.key to the form join or to a where clause.
If you want to place tables in their because you are trying to join the table to itself the alias the second one, but you will want to make sure that you then have a clause to join those two tables on (could be part of an on or a where)
The issue ended up being the order of table names. The code below joined everything just fine:
SELECT *
FROM place
INNER JOIN form ON place.key = form.key
INNER JOIN items ON form.id = items.formid
INNER JOIN rec ON items.recid = rec.id
INNER JOIN sub ON rec.id = sub.recid
INNER JOIN type ON sub.id = type.subid
INNER JOIN det ON type.name = det.typeid;