I have two tables A and B which I would like to select data where both table are common in a column (user_id) and are not similar in another column (A.columnX != B.columnY).
Actually, I would like the opposite of SQL joins.
Can anybody help me?
If you want to filter a join based on further logic, just use a WHERE clause:
SELECT
table_a.something,
table_a.something_else,
table_b.another_thing
FROM table_a
INNER JOIN table_b
ON table_b.user_id = table_a.user_id
WHERE
table_a.column_x != table_b.column_y;
You can provide the join query as:
select <<Column list>> from table1 a, table2 b where a.userid=b.userid and a.columnX!=b.columnY;
As you describe the question, you seem to want NOT EXISTS:
SELECT a.something, a.something_else,
b.another_thing
FROM table_a a INNER JOIN
table_b b
ON b.user_id = a.user_id
WHERE NOT EXISTS (SELECT 1
FROM table_b b2
WHERE b2.user_id = b.user_id and
b2.column_y = a.column_x
);
It is unclear if you want the user_id comparison in the correlation clause.
Related
Good Day!
I already saw other post about avoiding subquery / using JOIN statement, but still I can't figure it out why my query is so very slow to execute the 9 result data. but when the result data is more than thousands the query execute 0.7k ms only.
My question is , how can I speed up this query execution, what will I remove/add, Or do I need to re-construct the query? how?
here's my query
SELECT a.ts, b.bc, b.rem, c.dept FROM table1 a
INNER JOIN (select doc, max(ID) from table1 Group By doc) d ON d.doc=a.doc AND d.ID=a.ID
INNER JOIN table2 c ON c.u_id=a.u_id
INNER JOIN table3 b ON b.id=a.doc
WHERE c.depart = 'deparment' AND b.end = 0
here is the Screen Shot of
EXPLAIN result
Updated SS for EXPLAIN result
I already set INDEX :
ALTER TABLE table3 ADD INDEX max_id (end,id,bc);
ALTER TABLE table1 ADD INDEX max_id (ID,doc,u_id,ts,rem);
ALTER TABLE table1 ADD INDEX m_id (doc,ID);
ALTER TABLE table2 ADD INDEX user_max (dept,u_id);
Try a correlated subquery:
SELECT a.ts, b.bc, b.rem, c.dept
FROM table1 a INNER JOIN
table2 c
ON c.u_id = a.u_id INNER JOIN
table3 b
ON b.id =a.doc
WHERE c.depart = 'deparment' AND b.end = 0 AND
d.id = (SELECT MAX(t1.id) FROM table1 t1 WHERE t1.doc = a.doc);
For the correlated subquery you want an index on table1(doc, id).
I have two tables tableA and tableB. Both have two similar columns ID and B_ID.
I want to check whether both table have similar values. My code is:
$ac = $mysql->query("(SELECT ID,B_ID FROM tableA) INTERSECT (SELECT ID,B_ID FROM tableB)");
if($ac){
while($row = $ac->fetch_assoc()){
echo "ID ".$row["ID"]." B_ID".$row["B_ID"]."<br>";
}
}
But this doesn't give any result.
ps: tableA(ID,B_ID)
1->23
2->23
3->23
4->56
5->67
tableB(ID,B_ID)
3->23
8->26
11->27
12->66
here both table has 3->23 but 1->23 2->23 is not in tableB how can i figure that sort of records. same B_ID but different ID
If you have B.ID column is present in both table then use following JOIN query
SELECT a.ID, a.B_ID
FROM tableA AS a
JOIN tableB AS b ON (a.B_ID = b.B_ID AND a.ID = b.ID)
Use a join to get the data and just iterate over your results. Run the query and if there is any record that satisfies the query or not.
select t1.ID, t1.B_ID from tableA t1, tableB t2
where t1.ID = t2.ID
and t1.B_ID =t2.B_ID
Use an INNER JOIN and COUNT
SELECT COUNT(*) as cnt
FROM tableA INNER JOIN tableB
ON tableA.B_ID = tableB.B_ID AND tableA.ID = tableB.ID
now if cnt is greater than zero than common values exist, otherwise no
Please try with this query may be help full.
SELECT
tbla.ID,
tbla.B_ID
FROM
tablea AS tbla,
tableb AS tblb
WHERE
tbla.B_ID = tblb.B_ID
I can do the following in 2 queries but want to make it simpler. Can this be combined in one query? If so how more efficient is it than doing two queries vs one?
query1: SELECT page_id, coupon_id from table_1 WHERE key = :key
query2: SELECT folder from table_2 WHERE page_id = table_1.page_id
For my final result I need to have a coupon_id from table_1, and a folder from table_2.
In query2 I need to use the page_id result from query1 to get the folder
Is there a simpler way to do this?
Use JOIN (LEFT, RIGHT or INNER is up to your needs):
SELECT
t1.page_id,
t1.coupon_id,
t2.folder
FROM
table_1 AS t1
LEFT JOIN table_2 AS t2 ON
t2.page_id = t1.page_id
WHERE
t1.key = :key
You will want to JOIN the tables on the page_id:
SELECT t1.page_id,
t1.coupon_id,
t2.folder
from table_1 t1
inner join table_2 t2
on t1.page_id = t2.page_id
WHERE key = :key
If you need help learning join syntax, here is a great visual explanation of joins.
I used an INNER JOIN which will return all rows that match between the two tables. If you want to return all rows from table_1
even if it doesn't have a matching row in table_2, then you would use a LEFT JOIN
SELECT
table_1.coupon_id AS coupon_id,
table_2.folder AS folder
FROM
table_1
INNER JOIN table_2 ON table_2.page_id = table_1.page_id
WHERE
table_1.key = :key
SELECT t1.page_id, t1.coupon_id, t2.folder
FROM table_1 t1 LEFT JOIN table_2 t2 ON (t1.page_id = t2.page_id)
WHERE t1.key = :key
This will be faster than two queries, how much depends on your data.
Try this please:
SELECT a.page_id, a.coupon_id, b.folder_id
from table_1 a
join table_2 b
ON a.page_id = b.page_id
WHERE a.key = :key
group by a.page_id
;
I need to get both table and here is the table structure
Table A
UserID
Username
Status
IntroCode
Table B
IntroCode
UserID
I want to get the table a data and join with table b on tblA.IntroCode = tblB.IntroCode, then get the username of tblB.userID. How can i do such join ?
I tried half way and stuck in the middle, please help. Thanks for reply
This is just a simple join.
SELECT a.*, b.* -- select your desired columns here
FROM tableA a
INNER JOIN tableB b
ON a.IntroCode = b.IntroCode
WHERE b.userid = valueHere
UPDATE 1
SELECT a.UserID,
a.`Username` OrigUserName,
a.`Status`,
c.`Username` IntroUserName
FROM tableA a
INNER JOIN tableB b
ON a.IntroCode = b.IntroCode
INNER JOIN tableA c
ON b.userID = c.userID
-- WHERE b.UserID = valueHere -- extra condition here
SELECT column_name(s)
FROM TableA
LEFT JOIN TableB
ON TableA.UserID=TableB.UserID
SELECT B.userID from TableA A
LEFT JOIN TableB B on A.IntroCode=B.IntroCode
select a.*,b.IntroCode from TableA a left join TableB b
on a.IntroCode = b.IntroCode
you have to give the columns with same name an unique value:
SELECT a.UserID as uid_a, b.UserID as uid_b
FROM tableA a
INNER JOIN tableB b ON a.IntroCode = b.IntroCode
WHERE b.UserID = 1
Use this query.
SELECT TableA.Username FROM TableA JOIN TableB ON (TableA.IntroCode = TableB.IntroCode);
use this query
SELECT * FROM tblA INNER JOIN tblB ON tblA.IntroCode = tblB.IntroCode where tblB.userid = value
I am trying to unify a pair of queries with a LEFT JOIN, so that I can perform an ORDER BY on a desired column.
Table A has an exclusive one-to-many relationship with Table B, and table B has an exclusive one-to-many relationship with Table C.
I want to get the maximum value of a column in Table C, for each row in Table A.
what I used to have was:
$tableA = SELECT * FROM tableA;
for each row in $tableA {
$maxValue = SELECT MAX(value) FROM tableC WHERE tableB_id IN (SELECT tableB_id FROM tableB WHERE tableA_id={$row['tableA_id']}) GROUP BY tableB_id;
}
and now I'm thinking along the lines of:
SELECT * FROM tableA LEFT JOIN (SELECT tableA_id, MAX(max_c_value) FROM (SELECT tableB_id, MAX(value) max_c_value FROM tableC GROUP BY tableB_id) t GROUP BY tableA_id) USING(tableA_id)
but I know that's gibberish. I am not sure where to go from here.
I'm finding it hard to explain the problem, sorry.
SELECT A.ID, MAX(C.MyField) as CField
FROM A
LEFT OUTER JOIN B
ON A.ID = B.A_ID
LEFT OUTER JOIN C
ON B.ID = C.B_ID
GROUP BY A.ID
You will get null value for CField if there is no rows corresponding.
select max(table3.field),table1.field from table1
join table2 on table1.id=table2.table1_id
join table3 on table2.id = table3.table2_id
group by table1.field
You were closer on the first try I think. You want something like this:
SELECT MAX(tableC.cValue) FROM tableA
LEFT JOIN tableB
ON tableA.tableA_id = tableB.tableA_id
LEFT JOIN tableC
ON tableB.tableB_id = tableC.tableB_id
http://www.w3schools.com/sql/sql_join.asp
I have managed to solve it on my own, only to come back here and find I was making things far too complicated for myself, getting lost in a sea of tables!
This code did actually work, but I have already replaced it with the simpler code suggested above:
SELECT * FROM A LEFT JOIN (
SELECT A_ID, MAX(val) FROM (
SELECT * FROM B LEFT JOIN (
SELECT B_ID, MAX(val) val FROM C GROUP BY B_ID
) x USING(B_ID)
) y GROUP BY A_ID
) z USING(A_ID);