good afternoon
i'm going crazy with passing parameters to a very simple query with a subquery; i wrote hundreds of complicated queries, i'm afraid my fault is just in passing parameters to the subquery or everything else seems very normally manageable
making things simpler i prepare my query this way:
Set myConn = Server.CreateObject("ADODB.Connection")
myConn.Mode = 3 '3 = adModeReadWrite
myConn.Open MM_DEFDB_STRING
Set myCmd = Server.CreateObject("ADODB.Command")
myCmd.ActiveConnection = myConn
myCmd.CommandType = 1 '1 = adCmdText
myCmd.CommandText = strQuery 'this is my query with ? placeholders
myCmd.Prepared = false 'i won't repeat it
Set myRs = Server.CreateObject("ADODB.Recordset")
myRs.CursorType = 1 '0 = adOpenForwardOnly, 1 = adOpenKeyset, ...
myRs.LockType = 3 '3 = adLockOptimistic, ...
(here i prepare query and parameters)
myRs.Open myCmd
this should be correct, it is always the same for a lot of queries
1st case:
strQuery = SELECT * FROM TableA WHERE TableA.ID IN (SELECT myIDs FROM TableB WHERE TableB.NumericField=?) AND StringField=?
then i use this command cycling on two arrays, one for parameter values, another for types:
myCmd.Parameters.Append myCmd.CreateParameter("#Par" & i, VTypes(i), 1, 0, VValues(i))
here VValues is [8975, "011M1005D"] and VTypes is [3, 8]
result is 12 records, it is correct, i see those records on my database
2nd case (i only changed order of my two AND clauses):
strQuery = SELECT * FROM TableA WHERE StringField=? AND TableA.ID IN (SELECT myIDs FROM TableB WHERE TableB.NumericField=?)
VValues is ["011M1005D", 8975] and VTypes is [8, 3]
result is ZERO records, not an error but simply no records
WHY in your opinion ??? i would expect the same 12 records as before, the AND logic operator is simmetric!
3rd case (i use values on 2nd case instead of parameters):
strQuery = SELECT * FROM TableA WHERE StringField='011M1005D' AND TableA.ID IN (SELECT myIDs FROM TableB WHERE TableB.NumericField=8975)
result is correct: 12 records
curious, with plain values the order seems not important anymore
i also tried to debug the parameters collection, printing for each parameter it's value and it's type in order to check them just a line before the Open commmand: they seems correct and in the correct order in both cases
you are last chance or i have to rearrange a lot of code using some JOIN (but i really would understand my mistake before!). thank you for patience
Related
Back Story: I am looking to do a left join based on if a condition is true with an if statement. However if it is false, I want to still join the said table in, but with the same columns or with a new set of columns. This query is being wrote for a Pervasive SQL DB.
This query is fairly large, without this particular issue it executes and returns the data sets as expected. Below is a snapshot of the issue I am currently running into..
SELECT A.ONUM, B.JN, C.SEQ, C.PN
From Z.OH
LEFT JOIN OL A ON Z.ONUM = A.ONUM
LEFT JOIN JH B ON A.ONUM = B.SNUM AND A.OLNUM = B.SLNUM AND B.CLSD <> 'Y'
LEFT JOIN JO C ON IF(A.LC <> 'ZY', B.JN = C.JN, LEFT(B.C_PO, 6) = C.JN OR B.JN = C.JN) AND C.OP_T NOT IN ('Z','C')
WHERE Z.OT <> 'T' AND C.PN NOT IN (SELECT X.PN FROM JH X WHERE B.JN = X.JN)
Again, very summarized version with lots of joins/filters/select statement removed.
I am running into issues on the join with the IF statement. Without the if statement, the query executes as expected.
The original join being: B.JN = C.JN AND C.OP_T NOT IN ('Z', 'C')
When executing the query in PCC it would give the following syntax error at the following point: "B.JN << ??? >> = C.JN"
I tried switching over to an OR statement as shown below, but the run time of the query made it an impossible choice.
LEFT JOIN JO C ON
(B.JN = C.JN) OR (A.LC = 'ZY' AND LEFT(B.C_PO, 6) = C.JN)
AND C.OP_T NOT IN ('Z','C')
Checking the documentation, it looks like the query on the if statement is following the correct syntax...
Most simple solution would be to avoid the IF in the WHERE-clause, and do:
SELECT A.ONUM, B.JN, C.SEQ, C.PN
From Z.OH
LEFT JOIN OL A ON Z.ONUM = A.ONUM
LEFT JOIN JH B ON A.ONUM = B.SNUM AND A.OLNUM = B.SLNUM AND B.CLSD <> 'Y'
LEFT JOIN JO C ON (A.LC <> 'ZY' AND B.JN = C.JN) OR (A.LC = 'ZY' AND (LEFT(B.C_PO, 6) = C.JN OR B.JN = C.JN))
AND C.OP_T NOT IN ('Z','C')
WHERE Z.OT <> 'T' AND C.PN NOT IN (SELECT X.PN FROM JH X WHERE B.JN = X.JN)
I have update query which returns updated rows ID. Execution time of query is about 90 seconds. When i remove Returning clause, then execution time is 1ms.
Table update_table has 39000 rows.
Query updates 0 rows in this case. When updates 3 rows- execution time is same.
DECLARE
type intTable IS TABLE OF INTEGER;
idCol intTable;
BEGIN
UPDATE
update_table
SET
prop1 = 3, prop2 = NULL
WHERE EXISTS (
SELECT null FROM update_table f
INNER JOIN rel_table1 u ON f.ID= u.ID
INNER JOIN rel_table2 VP ON f.another_ID = VP.another_ID
WHERE (u.prop1 = 3)
AND VP.prop1 = 1
AND (u.prop2 = 75)
AND f.ID = update_table.ID
)
ReTURNING ID BULK COLLECT INTO idCol;
.
.
.
END;
Why returning clause slows down query?
A good part of using Oracle is knowing what is "supposed" to happen and what isn't.
Adding a RETURNING INTO clause is not "supposed" to make your update run more slowly. When something happens that isn't supposed to happen, check Oracle's support site to see whether it is a known bug.
In your case, it looks like you are encountering:
Bug 27131648 - SUB OPTIMAL PLAN ON UPDATE STATEMENT WITH RETURNING INTO
I am not sure if there is a patch, but there is a simple workaround: use the UNNEST hint. In your case, that would be:
UPDATE
update_table
SET
prop1 = 3, prop2 = NULL
WHERE EXISTS (
SELECT /*+ UNNEST */ null FROM update_table f
INNER JOIN rel_table1 u ON f.ID= u.ID
INNER JOIN rel_table2 VP ON f.another_ID = VP.another_ID
WHERE (u.prop1 = 3)
AND VP.prop1 = 1
AND (u.prop2 = 75)
AND f.ID = update_table.ID
)
ReTURNING ID BULK COLLECT INTO idCol;
I would recommend splitting it into two parts, first BULK COLLECT and next FORALL collected ID's, both extremely fast and you'll keep being able to further reference updated ID's from idCol.
DECLARE
type intTable IS TABLE OF INTEGER;
idCol intTable;
BEGIN
SELECT f.id
BULK COLLECT INTO idCol
FROM update_table f
INNER JOIN rel_table1 u ON f.ID= u.ID
INNER JOIN rel_table2 VP ON f.another_ID = VP.another_ID
WHERE (u.prop1 = 3)
AND VP.prop1 = 1
AND (u.prop2 = 75);
FORALL indx IN 1..idCol.COUNT
UPDATE update_table
SET prop1 = 3, prop2 = NULL
WHERE id = idCol(indx);
.
.
.
END;
I hope I helped!
I am developing in ASP VBScript at work and need to run a SELECT DISTINCT query but I am having some troubles.
I have other queries in my code that work perfectly fine, that do not use SELECT DISTINCT.
Here is what I am using:
Dim sections()
c = 1
set conn=Server.CreateObject("ADODB.Connection")
set rs=Server.CreateObject("ADODB.Recordset")
conn.Open "PROVIDER=Microsoft.Jet.OLEDB.4.0;Jet OLEDB:Database Password=******;User ID=Admin;" & _
"DATA SOURCE=" & Server.MapPath("modules.mdb")
rs.open "SELECT DISTINCT section FROM modules WHERE area='First' ORDER BY lvl ASC",conn
ReDim sections(10)
do while not rs.EOF
sections(c) = rs("section")
c = c + 1
rs.MoveNext
loop
rs.Close
conn.Close
set rs = nothing
set conn = nothing
Which gives me this error:
error '80004005'
on the line of the SQL query
The only way to fix this is to use "GROUP BY" instead of "DISTINCT"
SELECT DISTINCT section FROM modules WHERE area='First' ORDER BY lvl ASC
SELECT section FROM modules WHERE area='First' GROUP BY section ORDER BY lvl ASC
ADO recordset PageCount property returns "-1"
CONST PAGESIZE = 10
Set RS = Server.CreateObject("ADODB.Recordset")
RS.Open xSQL, Con1, 3, 3
RS.PageSize = PAGESIZE
If i try
response.write RS.PageCount
it returns -1
This was working a second ago, with a SQL
SELECT * FROM table
after i changed it to
SELECT DISTINCT field1, field2, field3 FROM table
the PageCount stopped working.
I've tried to change my query to
SELECT * FROM (SELECT DISTINCT field1, field2, field3 FROM table) aa
with no success.
I can't figure out why this is happening, please give me a hand.
This has somethign to do with my recordset (cursor) type. In the RS.Open line has 3,3 (two times) i remove one, and is now working.
CONST PAGESIZE = 10
Set RS = Server.CreateObject("ADODB.Recordset")
RS.Open xSQL, Con1, 3
RS.PageSize = PAGESIZE
I'm using an Access over Oracle database system (Basically using Access for the forms and getting into the tables using ADO code) and am trying to update a field in the product table with the value of the same named field in a load table.
The code I am using is:
.CommandText = "UPDATE " & strSchema & ".TBL_CAPITAL_MGMT_PRODUCT a INNER JOIN " & strSchema & ".TBL_CAPITAL_MGMT_TEMP_LOAD b ON a.AR_ID = b.AR_ID SET a.TOT_RWA_AMT = b.TOT_RWA_AMT;"
Which returns an error about missing SET keyword.. So I changed it to:
.CommandText = "UPDATE (SELECT a.TOT_RWA_AMT, b.TOT_RWA_AMT As New_RWA_AMT FROM " & strSchema & ".TBL_CAPITAL_MGMT_TEMP_LOAD a INNER JOIN " & strSchema & ".TBL_CAPITAL_MGMT_PRODUCT b ON b.AR_ID = a.AR_ID Where a.New_Rec <> '-1' AND a.IP_ID Is Not Null) c SET c.New_RWA_AMT = c.TOT_RWA_AMT;"
Which returns an error about non key-preserved table. the b table has a pk of AR_ID but the a table has no primary key and it probably won't be getting one, I can't update the structure of any of the tables.
I tried using the /*+ BYPASS_UJVC */ which lets the code run, but doesn't actually seem to do anything.
Anyone got any ideas where I should go from here?
Thanks
Alex
Ignoring the irrelevant ADO code, the update you are trying to do is:
UPDATE TBL_CAPITAL_MGMT_PRODUCT a
INNER JOIN
SET a.TOT_RWA_AMT = b.TOT_RWA_AMT;
This isn't supported by Oracle (though maybe this undocumented BYPASS_UJVC hint is supposed to overcome that, but I wasn't aware of it till now).
Given that your inline view version fails due to lack of constraints you may have to fall back on the traditional Oracle approach using correlated subqueries:
UPDATE TBL_CAPITAL_MGMT_PRODUCT a
SET a.TOT_RWA_AMT = (SELECT b.TOT_RWA_AMT
FROM TBL_CAPITAL_MGMT_TEMP_LOAD b
WHERE a.AR_ID = b.AR_ID
)
WHERE EXISTS (SELECT NULL
FROM TBL_CAPITAL_MGMT_TEMP_LOAD b
WHERE a.AR_ID = b.AR_ID
);
The final WHERE clause is to prevent TOT_RWA_AMT being set to NULL on any "a" rows that don't have a matching "b" row. If you know that can never happen you can remove the WHERE clause.
If you're using Oracle 10g or higher, an alternative to Tony's solution would be to use a MERGE statement with only a MATCHED clause.
MERGE INTO TBL_CAPITAL_MGMT_PRODUCT a
USING TBL_CAPITAL_MGMT_TEMP_LOAD b
ON (a.AR_ID = b.AR_ID)
WHEN MATCHED THEN
UPDATE SET a.TOT_RWA_AMT = b.TOT_RWA_AMT;