Whats wrong with this nested query? - sql

I am trying to write a query to return the id of the latest version of a market index stored in a database.
SELECT miv.market_index_id market_index_id from ref_market_index_version miv
INNER JOIN ref_market_index mi ON miv.market_index_id = mi.id
WHERE mi.short_name='dow30'
AND miv.version_num = (SELECT MAX(m1.version_num) FROM ref_market_index_version m1 INNER JOIN ref_market_index m2 ON m1.market_index_id = m2.id )
The above SQL statement can be (roughly) translated into the form:
SELECT some columns FROM SOME CRITERIA MATCHED TABLES
WHERE mi.short_name='some name'
AND miv.version_num = SOME NUMBER
What I don't understand is that when I supply an actual number (instead of a sub query), the SQL statement works - also, when I test the SUB query used to determine the latest version number, that also works - however, when I attempt to use the result returned by sub query in the outer (parent?) query, it returns 0 rows - what am I doing wrong here?
Incidentally, I also tried an IN CLAUSE instead of the strict equality match i.e.
... AND miv.version_num IN (SUB QUERY)
That also resulted in 0 rows, although as before, when running the parent query with a hard coded version number, I get 1 row returned (as expected).
BTW I am using postgeresql, but I prefer the solution to be db agnostic.

The problem is probably that the max(version_num) doesn't exist for 'dow30'.
Try the following correlated subquery:
SELECT miv.market_index_id market_index_id
from ref_market_index_version miv INNER JOIN
ref_market_index mi
ON miv.market_index_id = mi.id
WHERE mi.short_name='dow30' AND
miv.version_num = (SELECT MAX(m1.version_num)
FROM ref_market_index_version m1 INNER JOIN
ref_market_index m2
ON m1.market_index_id = m2.id
where m1.short_name = 'dow30'
)
I added the where clause in the subquery.

Related

MS Access SQL sub Query using outer query result

I'm attempting to run a sub query based on the result of an outer query. The issue that I am having is that instead of using the outer query, I get a prompt for a value from the subquery.
SELECT Facilities.CustomerName, Facilities.FacilityName,
Facilities.AnnualPlan, Facilities.AppCo1,
(SELECT YeildDB.CornYield
FROM YeildDB
WHERE Facilites.AppCo1 = YeildDB.FIPS) AS Expr1
FROM Facilities
The goal is that the sub query should use the value from Facilities.AppCo1 to match with the value in YeildDB.FIPS and then return the corresponding value in YeildDB.CornYeild.
Currently I get a prompt asking for the YeildDB.FIPS value instead of the sub query using the outer query value.
Your code should work. But you can also express this using a LEFT JOIN:
SELECT Facilities.CustomerName, Facilities.FacilityName,
Facilities.AnnualPlan, Facilities.AppCo1,
YeildDB.CornYield
FROM Facilities LEFT JOIN
YeildDB
ON Facilties.AppCo1 = YeildDB.FIPS;
I noticed that you misspelled Facilities -- and that is probably why your version doesn't work. This is one reason to use table aliases:
SELECT f.CustomerName, f.FacilityName,
f.AnnualPlan, f.AppCo1,
y.CornYield
FROM Facilities as f LEFT JOIN
YeildDB as y
ON f.AppCo1 = y.FIPS;
Your sub query may be returning multiple values, hence the prompt asking you to specify which one you want. You can fix this (or at least hide this issue) by specifying top 1:
SELECT Facilities.CustomerName, Facilities.FacilityName,
Facilities.AnnualPlan, Facilities.AppCo1,
(SELECT TOP 1 YeildDB.CornYield
FROM YeildDB
WHERE Facilites.AppCo1 = YeildDB.FIPS) AS Expr1
FROM Facilities

The "where" condition worked not as expected ("or" issue)

I have a problem to join thoses 4 tables
Model of my database
I want to count the number of reservations with different sorts (user [mrbs_users.id], room [mrbs_room.room_id], area [mrbs_area.area_id]).
Howewer when I execute this query (for the user (id=1) )
SELECT count(*)
FROM mrbs_users JOIN mrbs_entry ON mrbs_users.name=mrbs_entry.create_by
JOIN mrbs_room ON mrbs_entry.room_id = mrbs_room.id
JOIN mrbs_area ON mrbs_room.area_id = mrbs_area.id
WHERE mrbs_entry.start_time BETWEEN "145811700" and "1463985000"
or
mrbs_entry.end_time BETWEEN "1458120600" and "1463992200" and mrbs_users.id = 1
The result is the total number of reservations of every user, not just the user who has the id = 1.
So if anyone could help me.. Thanks in advance.
Use parentheses in the where clause whenever you have more than one condition. Your where is parsed as:
WHERE (mrbs_entry.start_time BETWEEN "145811700" and "1463985000" ) or
(mrbs_entry.end_time BETWEEN "1458120600" and "1463992200" and
mrbs_users.id = 1
)
Presumably, you intend:
WHERE (mrbs_entry.start_time BETWEEN 145811700 and 1463985000 or
mrbs_entry.end_time BETWEEN 1458120600 and 1463992200
) and
mrbs_users.id = 1
Also, I removed the quotes around the string constants. It is bad practice to mix data types, and in some databases, the conversion between types can make the query less efficient.
The problem you've faced caused by the incorrect condition WHERE.
So, should be:
WHERE (mrbs_entry.start_time BETWEEN 145811700 AND 1463985000 )
OR
(mrbs_entry.end_time BETWEEN 1458120600 AND 1463992200 AND mrbs_users.id = 1)
Moreover, when you use only INNER JOIN (JOIN) then it be better to avoid WHERE clause, because the ON clause is executed before the WHERE clause, so criteria there would perform faster.
Your query in this case should be like this:
SELECT COUNT(*)
FROM mrbs_users
JOIN mrbs_entry ON mrbs_users.name=mrbs_entry.create_by
JOIN mrbs_room ON mrbs_entry.room_id = mrbs_room.id
AND
(mrbs_entry.start_time BETWEEN 145811700 AND 1463985000
OR ( mrbs_entry.end_time BETWEEN 1458120600 AND 1463992200 AND mrbs_users.id = 1)
)
JOIN mrbs_area ON mrbs_room.area_id = mrbs_area.id

SQL - updating Oracle 10g database from a complex query

trying to update a single column based on criteria. The rows that need to be updated are returned by the query
SELECT
it.objectid,
it.versionnumber,
it.itemnumber "Item Number",
it.itemtype,
itcovprem.basisofsettlement,
itcovprem.coverage "Coverage"
FROM
itemcoveragesandpremiums itcovprem,
items it,
policies pl
WHERE
pl.objectid = it.parentobjectid AND
pl.policytype in ('H','F') AND
it.objectid = itcovprem.itemobjectid AND
it.itemtype in ('SHOA','SHOB','SHOC','SHOD','SHOK','SHOL') and
it.versionnumber = itcovprem.itemversionnumber AND
((itcovprem.coverage like ('A - Dwg Bldg%')) or
(itcovprem.coverage like ('#42 - Repl Cost Plus%')) or
(itcovprem.coverage like ('#42 - Limited GRC%'))) and
it.effectivedate >= TO_DATE('01-JAN-2006', 'DD-MON-YYYY') and
exists (select * from itemcoveragesandpremiums icp where icp.itemobjectid = it.objectid and icp.itemversionnumber = it.versionnumber and icp.coverage like ('#42%')) and
itcovprem.basisofsettlement not in ('LGRC')
I've tried a few options to convert this to an update query, but no joy.
I get Error - line 3 - SQL command not properly ended when using
update itemcoveragesandpremiums
set basisofsettlement = 'LGRC'
from itemcoveragesandpremiums as itcovprem
inner join items as it
on it.objectid = itcovprem.itemobjectid and it.versionnumber = itcovprem.itemversionnumber
inner join policies as pl
on pl.objectid = it.parentobjectid
where [cut, rest of query was below]
I get Error - line 6 - missing right parenthesis when trying use an inline query
update
(
SELECT
itcovprem.basisofsettlement as OLD
FROM
itemcoveragesandpremiums as itcovprem inner join items as it on it.objectid = itcovprem.itemobjectid and it.versionnumber = itcovprem.itemversionnumber inner join policies AS pl on pl.objectid = it.parentobjectid
WHERE [query snipped]
) t
set t.old = 'LGRC'
Seems that SQL*Plus just wants to stop looking at the update before it gets to the meat of my select query. I'm not sure if I'm formatting it incorrectly or going at it from the wrong direction. Any advice?
In your first attempt, the error at line 3 is because update doesn't support a from or join clause. In your second attempt the immediate error at line 6 is because you're trying to alias a table with as itcovprem; but you can only use as for column aliases, not for table aliases. (The first attempt is trying to do that too, but it isn't getting as far as encountering that problem). But you can't generally update an inline-view with joins anyway, so it would still error with that - something like an ORA-01779.
You would need to do something like:
update itemcoveragesandpremiums
set basisofsettlement = 'LGRC'
where (itemobjectid, itemversionnumber, basisofsettlement, coverage) in (
select it.objectid,
it.versionnumber,
itcovprem.basisofsettlement,
itcovprem.coverage
from ...
)
Assuming that itemobjectid, itemversionnumber, basisofsettlement, coverage identifies a row sufficiently such that you don't risk updating anything you shouldn't. It might be safer to add a rowid to the select and use that for the update instead to avoid ambiguity:
update itemcoveragesandpremiums
set basisofsettlement = 'LGRC'
where rowid in (
select itcovprem.rowid as target_rowid
from ...
)

SQL LEFT JOIN not working in MS Access

The following SQL query returns no data for the LEFT JOIN in MS Access.
SELECT * FROM
(
SELECT Operation_Part.PPC,
Operation_Part.TargetOperationsPerHour as JPH,
Operation_Part.Misc1 as [JPh Alt 1],
STR(Operation_Part.SeqNr) as Sequence,
Operation_Part.idPart,
Operation_Part.idOperationPart,
Operation.OperationType as Operation,
tblOperationType.OperationType as [Operation Type]
FROM tblOperationType
RIGHT JOIN (Operation INNER JOIN Operation_Part ON Operation.idOperation = Operation_Part.idOperation)
ON tblOperationType.idOpType = Operation.OperationTID
WHERE Operation_Part.VsbLDq = 0
AND Operation_Part.idPart <> 0 AND Operation_Part.idPart = 1271)
AS [AA]
LEFT JOIN (SELECT Sum([Cptotal]) AS DownTime,
TransactionDetail.idPart,
STR(TransactionDetail.seq_number) as Sequence
FROM ([Transaction] INNER JOIN TransactionDetail ON [Transaction].idTransaction = TransactionDetail.idTransaction)
WHERE [Transaction].idTransactionType=29
AND TransactionDetail.WorkOrderNumber = 'PR23144'
GROUP BY TransactionDetail.idPart, STR(TransactionDetail.seq_number))
AS [EE]
ON AA.idPart = EE.idPart AND EE.Sequence=AA.Sequence
In SQL Server the query does return the downtime value of 1.08 as required (see pics below).
First select returns:
Second select returns:
MS Access result:
SQL server result:
How do I make it work in MS Access?
This is only a guess, but it may well have something to do with the nulls in the applicable columns of the rows you dont really want.
Suggest you change
SELECT Sum([Cptotal]) AS DownTime,
to
SELECT Sum(IIf(IsNull([CpTotal]), 0, [CpTotal])) AS DownTime
In Access I always use CStr(...) instead of Str(...)
Aside from this, painful though it may be, I'd suggest turning the left-joined component into a separate query, or if you dont use queries, building a temporary table with this data which is then left joined into the original query.

Subselect a Summed col in Oracle

this is an attempted fix to a crystal reports use of 2 sub reports!
I have a query that joins 3 tables, and I wanted to use a pair of sub selects that bring in the same new table.
Here is the first of the two columns in script:
SELECT ea."LOC_ID", lo."DESCR", ea."PEGSTRIP", ea."ENTITY_OWNER"
, ea."PCT_OWNERSHIP", ea."BEG_BAL", ea."ADDITIONS", ea."DISPOSITIONS"
, ea."EXPLANATION", ea."END_BAL", ea."NUM_SHARES", ea."PAR_VALUE"
, ag."DESCR", ea."EOY", ea."FAKEPEGSTRIP",
(select sum(htb.END_FNC_CUR_US_GAAP)
from EQUITY_ACCOUNTS ea , HYPERION_TRIAL_BALANCE htb
where
htb.PEGSTRIP = ea.PEGSTRIP and
htb.PRD_NBR = 0 and
htb.LOC_ID = ea.LOC_ID and
htb.PRD_YY = ea.EOY
) firstHyp
FROM ("TAXPALL"."ACCOUNT_GROUPING" ag
INNER JOIN "TAXP"."EQUITY_ACCOUNTS" ea
ON (ag."ACCT_ID"=ea."PEGSTRIP") AND (ag."EOY"=ea."EOY"))
INNER JOIN "TAXP"."LOCATION" lo ON ea."LOC_ID"=lo."LOC_ID"
WHERE ea."EOY"=2009
ORDER BY ea."LOC_ID", ea."PEGSTRIP"
When this delivers the dataset the value of "firstHyp" fails to change by pegstrip value. It returns a single total for the join and fails to put the proper by value by pegstrip.
I thought that the where clause would have picked up the joins line by line.
I don't do Oracle syntax often so what am I missing here?
TIA
Your SQL is equivilent to the following:
SELECT ea."LOC_ID", lo."DESCR", ea."PEGSTRIP",
ea."ENTITY_OWNER" , ea."PCT_OWNERSHIP",
ea."BEG_BAL", ea."ADDITIONS", ea."DISPOSITIONS" ,
ea."EXPLANATION", ea."END_BAL", ea."NUM_SHARES",
ea."PAR_VALUE" , ag."DESCR", ea."EOY", ea."FAKEPEGSTRIP",
(select sum(htb.END_FNC_CUR_US_GAAP)
from EQUITY_ACCOUNTS iea
Join HYPERION_TRIAL_BALANCE htb
On htb.PEGSTRIP = iea.PEGSTRIP
and htb.LOC_ID = iea.LOC_ID
and htb.PRD_YY = iea.EOY
where htb.PRD_NBR = 0 ) firstHyp
FROM "TAXPALL"."ACCOUNT_GROUPING" ag
JOIN "TAXP"."EQUITY_ACCOUNTS" ea
ON ag."ACCT_ID"=ea."PEGSTRIP"
AND ag."EOY"=ea."EOY"
JOIN "TAXP"."LOCATION" lo
ON ea."LOC_ID"=lo."LOC_ID"
WHERE ea."EOY"=2009
ORDER BY ea."LOC_ID", ea."PEGSTRIP"
Notice that the subquery that generates firstHyp is not in any way dependant on the tables in the outer query... It is therefore not a Correllated SubQuery... meaning that the value it generates will NOT be different for each row in the outer query's resultset, it will be the same for every row. You need to somehow put something in the subquery that makes it dependant on the value of some row in the outer query so that it will become a correllated subquery and run over and over once for each outer row....
Also, you mention a pair of subselects, but I only see one. Where is the other one ?
Use:
SELECT ea.LOC_ID,
lo.DESCR,
ea.PEGSTRIP,
ea.ENTITY_OWNER,
ea.PCT_OWNERSHIP,
ea.BEG_BAL,
ea.ADDITIONS,
ea.DISPOSITIONS,
ea.EXPLANATION,
ea.END_BAL,
ea.NUM_SHARES,
ea.PAR_VALUE,
ag.DESCR,
ea.EOY,
ea.FAKEPEGSTRIP,
NVL(SUM(htb.END_FNC_CUR_US_GAAP), 0) AS firstHyp
FROM TAXPALL.ACCOUNT_GROUPING ag
JOIN TAXP.EQUITY_ACCOUNTS ea ON ea.PEGSTRIP = ag.ACCT_ID
AND ea.EOY = ag.EOY
AND ea.EOY = 2009
JOIN TAXP.LOCATION lo ON lo.LOC_ID = ea.LOC_ID
LEFT JOIN HYPERION_TRIAL_BALANCE htb ON htb.PEGSTRIP = ea.PEGSTRIP
AND htb.LOC_ID = ea.LOC_ID
AND htb.PRD_YY = ea.EOY
AND htb.PRD_NBR = 0
GROUP BY ea.LOC_ID,
lo.DESCR,
ea.PEGSTRIP,
ea.ENTITY_OWNER,
ea.PCT_OWNERSHIP,
ea.BEG_BAL,
ea.ADDITIONS,
ea.DISPOSITIONS,
ea.EXPLANATION,
ea.END_BAL,
ea.NUM_SHARES,
ea.PAR_VALUE,
ag.DESCR,
ea.EOY,
ea.FAKEPEGSTRIP,
ORDER BY ea.LOC_ID, ea.PEGSTRIP
I agree with Charles Bretana's assessment that the original SELECT in the SELECT clause was not correlated, which is why the value never changed per row. But the sub SELECT used the EQUITY_ACCOUNTS table, which is the basis for the main query. So I removed the join, and incorporated the HYPERION_TRIAL_BALANCE table into the main query, using a LEFT JOIN. I wrapped the SUM in an NVL rather than COALESCE because I didn't catch what version of Oracle this is for.