help with sql
with t as(select *)
Of
(select *)
Of
(select rownum rn
of the double
connect by rownum < 100) a
where mod(a.rn,9) <>0) b
where mod(b.rn,6) <>0).
r as (select t.rn
t)
Select r.*
r
My version of oracle is: 10 G/9i
Hi all
I have the above query, now I have a scenario where I have to fill in the missing (between the smallest and largest number) numbers.
Example:
In the series of 10 consecutive numbers (with 1 interval between 2 numbers).
1
2
3
4
5
9
10
I need produce the following output:
1
2
3
4
5
6
7
8
9
10
Please help me with the sql.
Respect of
Rahul
HII Mac Freak Rahul,
Here's a solution:
SQL> select * from t1;
COL1
----------
1
10
SQL> with a as (select level lev
2 from t1
3 connect by level <= (select max(col1) from t1))
4 select a.lev
5 from a
6 minus
7 select col1
8 from t1 ;
LEV
----------
2
3
4
5
6
7
8
9
8 rows selected.
SQL>
Tags: Database
Similar Questions
-
Under Update taking 2 h complete 3000 sets.
UPDATE ARR_TRANSACTIONS ARR
SET = NUMBER99
(SELECT ARR_1.NUMBER99
OF ARR_TRANSACTIONS ARR_1, ARR_HEADER BH
WHERE ARR_1.ARR_ID = BH. ARR_ID
AND BH. FLAG = "Y".
AND ARR_1.LINE_ID = ARR. ORIG_INV_LINE_ID
AND ARR_1.NUMBER9 IS NOT NULL
AND ROWNUM = 1)
WHERE ARR BATCH_ID = 26219
AND ARR. NUMBER99 IS NULL
I added more some condition where to reduce cost and its working fine. But I need to rise more than performance on SQL.
Please help with SQL.
I do not see the number of 3000 lines in the plan - so I guess that the optimizer does not compute with the correct numbers. Could generate the plan with dbms_xplan.display_cursor? This could tell us something about filter predicates and access.
In the given situation, I would check:
- is the access to the index in step 1 (filtered by the relevant conditions: this could be BATCH_ID = 26219) return actually 248 entries (or something similar)?
- Access table in step 2 returning 204 lines (or something similar) after applying the filters given? If the number of lines is much bigger then a Nested Loops is perhaps not a good idea.
To make this much simpler check, you could create a plan with statistics of content (RowSource) given that the plan containing the real and the number of estimated rows.
If the estimated cardinalities are not plausible the next question would be if the statistics are strong and up to date.
-
Need help with SQL/PL/SQL for Dates
Hi Experts - need help with a SQL query.
I need to insert some date fields in a table called CALENDAR_PERIOD.
Current data in CALENDAR_PERIOD table with their data types:
STARTPERIOD (DATE) YEAR (NUMBER) PERIOD_LABEL (Varchar2 255)
02/11/2014 2014 2014/02/11 SUN
03/11/2014 2014 14/03/11 MON
04/11/2014 2014 11/04/14 MAR
I have to increment above values up to the year 2025. I don't know how to write SQL and increment of these values.
Ex: My next value should insert: 05/11/2015 2014 11/05/14 WED like that I need to insert data until 12 31, 2025.
Can you please help me with PL/SQL block?
Really appreciate your help!
DB version:
Oracle Database 11 g Enterprise Edition Release 11.2.0.3.0 - 64 bit Production
PL/SQL Release 11.2.0.3.0 - Production
CORE Production 11.2.0.3.0
AMT for IBM/AIX RISC System/6000: Version 11.2.0.3.0 - Production
NLSRTL Version 11.2.0.3.0 - Production
Thank you
SandyHello Sandy,
Maybe something like
INSERT INTO calendar_period (startperiod, year, period_label)
SELECT DATE '' 2014-11-04 + LEVEL
, TO_NUMBER (TO_CHAR (DATE '' 2014-11-04 + LEVEL, "YYYY"))
, TO_CHAR (DATE '' 2014-11-04 + LEVEL, "MM/DD/YY DY")
OF the double
CONNECT BY LEVEL<= date="" '2025-12-314="" -="" date="">
;
((mais je ne comprends pas pourquoi nous créons une telle table "année" et "period_label" peuvent être calculé à partir de startperiod))
Best regards
Bruno Vroman.
-
Help with SQL Query (Subselects)
Hello community,
IAM a new Member in this forum. The first excuse my English, my native language is German.
In my workplace, we have a great Orcle Database 11 g with 30 different tables for production control issues.
I try to get a couple of different information from the database, so I started with SQL of the query, but for this problem, I wasn't able to write a query to work.
In this case, I have 2 tables:
Table 1:
ID; ORDER_NR; DESCRIPTION; CREATE_DATE
1; A500236; CLEAN HOUSE; 20/02/2012
2; A623555; REPAIR CAR; 10/01/2012
3; A866944; MAINTAIN EQUIPMENT; 11/02/2012
Table 2:
ID; ORDER_NR; WO_STEP; STEP_DATE; EMPLOYEE
1; A500236; A; 21/02/2012; W0010
2; A500239; F; 21/02/2012; W0010
3; A500239; S; 22/02/2012; W0027
4; A500239; R; 23/02/2012; H0034
5; A500239; U; 25/02/2012; L0099
6; A263555; A; 15/01/2012; G0009
7; A263555; C; 17/01/2012; S0039
5 V A263555; R; 18/01/2012; K0059
9; A263555; U; 19/01/2012; A0048
10; A866944; A; 13/02/2012; H0034
11; A866944; B; 13/02/2012; L0035
12; A866944; G; 17/02/2012; D0084
13; A866944; U; 23/02/2012; S0039
And the result of my query should look like this:
ORDER_NR; DESCRIPTION; CREATE_DATE; A_STAT_AGE; R_STAT_AGE; U_STAT_AGE
A500236; OWN HOME; 20/02/2012; 5; 3; 1
A623555; REPAIR CAR; 01/10/2012; 42; 39; 38
A866944; MAINTAIN EQUIPMENT; 11/02/2012; 15; 4; 3
The age of my query result should be calculated from the date of the creation of the order.
I would like to know 2 things, one is how old was the order when they reached this status, R and U.
The second, that is, how long did you order remaining on A stat, R and U (and if possible all the others too)
It can happen that not every order reaches every State, so he ca go directly from A you in this case I want to display a generic character in this row/column
I hope you all know what I mean and what result to expect.
Thanks for your help.
Reinhard W.Hi, Reinhard,
990524 wrote:
Hi Frank,.I thank you for your professional response and excuse my non-professional way to clarify my question.
I have now already read and understand how to ask good questions, but is there an easy an inexpensive way to run a database on my computer at home?You can download the Oracle database from this site. The Express edition is easier to install. It lacks a few features that the Enterprise edition (for example, partitioning table and safety of level line), but most of the things work in all editions of Oracle.
All editions are free if you do not use them for Production applications. Of course, at this price (or lack thereof), you get Oracle support.I have already noticed that the syntax SQL for Oracle differ from other DBMSS.
Yes and some features are different in different products.
Your query works perfectly, thank you.
Now I want to refine my (your) query, is there a way to display a range of day within 1 day? As 0.5 day to half a day between two State timestamp.
Of course, you can do it. After the exact output desired of the sample data that I posted. If the sample data do not show what you need to do, then after a few different examples of data that makes.
My table contains the Date and time in a row, I forgot to mention that.
After a CREATE TABLE statement. I was guessing a lot of things that may be important. For example, I assumed you were using a DATE for date and time column. So it is only reasonable to do so, but maybe it's not what looks like at your table.
I tried something like this:
SYSDATE - MIN ( CASE WHEN t2.wo_step = 'A' THEN t2.step_date END ) AS Age_ABut it gives me an error of arithmetic overflow and not die difference in the Date and time or days with decimal friction.
There are only about 5.4 million possible days with DATEs of Oracle. Is only 7 digits and Oracle can treat approximately 39 numbers without overflow errors, so I don't see how this can happen if t2.step_date is really a DATE. This is an example of why you need to publish a CREATE TABLE statement.
-
Where clause ' where 1 = 1 ' help with SQL tuning
Hello Tuning specialists,.
It is useful to use ' where 1 = 1 ' and then type all joins and conditions in the statements 'AND' of the SQL when writing SQL queries. I would like to know if it helps with the writing queirs this way SQL query performance.
Thanks in advance.
Published by: oracle_developer on October 8, 2012 10:41oracle_developer wrote:
Hello Tuning specialists,.
It is useful to use ' where 1 = 1 ' and then type all joins and joins in the statements "AND" when writing SQL queries.
I would like to know if it helps with the writing queirs this way SQL query performance.He has No. performance impact.
It is done when building dynamic SQL, so each sentence can start with AND -
Hi all
I need assistance with SQL FULL OUTER JOIN.
How to write with FULL OUTER JOIN, the query.
I tried like
I have 3 different queries.
Query q: SELECT emp_id LN.emp_id.
Sum (LN_amount) loan_amt
MY_LON_TAB LN
WHERE LN_amount > 20000
LN.emp_id GROUP;
Query b: SELECT PLN. EMP_ID emp_id,
Sum (PLAN_AMOUNT) plan_amt
FROM PLN MY_PLAN_TAB
where PLAN_AMOUNT <>0
GROUP BY PLN. EMP_ID;
Query C:
SELECT FLX. EMP_ID emp_id,
SUM (PRORATE_AMOUNT) PRORATE_AMT
OF MY_FLX_TAB FLX
WHERE PRORATE_AMOUNT <>0
FLX GROUP. EMP_ID;
Suppose that the different subquery above 3 a, b, c respectively. each subquery with emp_id and respective amount.
like a.emp_id, a.loan_amt
b.emp_id, b.plan_amt
c.emp_id, c.prorate_amt
I show all the empid with their amount respective.
schenario: If an account made, but not vice-versa b
and B have record, but not vice versa c
and a credit record but not vice versa c
first: I have external is associated with the A and B
Second: join external then complete the total outcome of (a & b) with c
but empid c just as empty.
My output is:
emp_id a.loan_amt, b.plan_amt c.prorate_amt
101 5000 null null
102 4500 null 2000
103 6700 null null
How to solve the foregoing to the ANSI (FULL OUTER JOIN) method.
Please suggest me.
Thanks in advance...
PKMHello
It is very difficult for anyone to say what you're doing wrong if they do not know what you are doing. Post your code FULL OUTER JOIN.
My best guess is:WITH a AS ( SELECT ... -- What you posted as query a ) , b AS ( SELECT ... -- What you posted as query b ) , c AS ( SELECT ... -- What you posted as query c ) SELECT COALESCE ( a.emp_id , b.emp_id , c.emp_id ) AS emp_id , ... -- whatever other columns you want FROM a FULL OUTER JOIN b ON b.emp_id = a.emp_id FULL OUTER JOIN c ON c.emp_id = COALESCE (a.emp_id, b.emp_id) ;I hope that answers your question.
If not, post a small example (CREATE TABLE and INSERT statements) data for all tables and the results expected from these data (if not what you have already posted). -
Hi guys,.
My main problem/situation is when sqlplus returns no data (no message selected lines)
Bash shell script condition
echo $open_items
15 if [[$open_items == "0"]]
16. can
echo 17 0
18 else
19 1 echo
20 fi
and real sql is as follows:
elect the count (*) a table with a condition.
I get the output as follows:
+ open_items ='
COUNT (*)
----------
0'
+ echo 'COUNT (*)'---0
COUNT ()) - 0
+ [[
COUNT (*)
----------
0 == \0 ]]
+ echo 1
1
What to do so that I can get a number in the output of sql is my question because the condition of shell script is based entirely on sqlplus out. I tried set feedback off, but it does not help.
Any help would be much appreciated!
SQL lover wrote:
OK I try it and give you feedback
In fact, I take out execution of sql into a variable
MyVar = $(sqlplus-s/as sysdba)<>
Set echo off feedback off verify off trimsp head off pages 0
Select count (*) from dba_users;
output
EXPRESSIONS OF FOLKLORE)
echo myvar is $myvar
-
Dear all,
I have a sql like query below
SELECT min (G.TRANSACTION_DATE), D.REQUEST_NUMBER
MTL_TXN_REQUEST_HEADERS D,.
MTL_TXN_REQUEST_LINES, E.
MTL_MATERIAL_TRANSACTIONS G,.
Mtl_Transaction_Types I have
WHERE D.HEADER_ID = E.HEADER_ID
AND G.TRANSACTION_TYPE_ID = I.TRANSACTION_TYPE_ID
AND Upper (I.Transaction_Type_Name) = Upper ('TEC outcome')
AND E.LINE_ID = G.MOVE_ORDER_LINE_ID
AND D.MOVE_ORDER_TYPE = 5
TO_DATE (G.TRANSACTION_DATE) BETWEEN TO_DATE('01-JAN-2014') AND TO_DATE('31-DEC-2014')
D.REQUEST_NUMBER GROUP
I need to get the first number of the application and finally ask for number based on the date of the transaction, how can I get the number of name application and based on the date of the transaction for the same query, please help me.
above query is back under results
If the query should return a line with the number of application like 2383 based on minimum transaction date and another column based on the maximum transaction date, please help me.
so in the example above, it must return
FIRST REQUEST NUMBER LAST NUMBER
2383 1886
Thank you
select min(request_number) keep(dense_rank first order by transaction_date asc ) request_number_min , min(request_number) keep(dense_rank first order by transaction_date desc) request_number_max from ( select min(g.transaction_date) transaction_date , d.request_number from mtl_txn_request_headers d , mtl_txn_request_lines e , mtl_material_transactions g , mtl_transaction_types i where d.header_id = e.header_id and g.transaction_type_id = i.transaction_type_id and upper(i.transaction_type_name) = upper('WIP Issue') and e.line_id = g.move_order_line_id and d.move_order_type = 5 and to_date(g.transaction_date) between to_date('01-jan-2014') and to_date('31-dec-2014') group by d.request_number ) -
Need help with sql query performance
Dear all,
I have a sql like query below, I need to give the following query please help me identify which statement I should tune to have better performanece.
Select rownum LINE_NUM,
A.LINE_ID,
TO_CHAR (A.INVITMID),
TO_NUMBER (A.PICKQTY),
UNLOADINGPNT NULL,
RRNUM NULL,
WORKORDNUM NULL,
WORKORDDESC NULL,
A.PONUM,
DTR_DUMB NULL,
A.DESCRIPTION,
FROM_SUB NULL,
TO_SUB NULL,
NO SOURCE,
ASSET_NUMBER NULL,
A.RECEIPTNUM,
MOVEORD NULL,
FROM_LOC NULL,
TO_LOC NULL,
MSD_NUM NULL,
CONTAIN_LINE NULL,
A.UOM,
A.PO_RELEASE
de)
Select headerid Po.Po_Header_Id,
rcv1. Po_Line_Id LINE_ID,
rcv1.item_id INVITMID,
(NVL(Rcv1.Transact_Qty,0)-NVL(rcv2.transact_qty,0)) PICKQTY,
Po.Segment1 PONUM,
Rcv1.Receipt_Num RECEIPTNUM,
Rcv1.Item_Desc DESCRIPTION,
Rcv1.Transact_Uom GLU,
Rcv1.Po_release
Po_Headers_All in.,.
(Select rcv3. Po_Header_Id, RCV3.receipt_num, rcv3. Po_Line_Id, rcv3. Destination_Type_Code, rcv3. Item_Id, rcv3. Item_Desc, rcv3. Transact_Uom, SUM (rcv3. Transact_Qty) Transact_Qty, rcv3. PO_RELEASE OF)
SELECT A.Po_Header_Id,
C.RECEIPT_NUM receipt_num,
A.Po_Line_Id,
A.Destination_Type_Code,
B.Item_Id,
B.item_description Item_Desc,
A.UNIT_OF_MEASURE Transact_Uom,
A.QUANTITY Transact_Qty,
D.RELEASE_NUM PO_RELEASE
OF RCV_TRANSACTIONS,.
RCV_SHIPMENT_HEADERS C.
B RCV_SHIPMENT_LINES,
PO_RELEASES_ALL D
WHERE C.SHIPMENT_HEADER_ID = A.SHIPMENT_HEADER_ID
AND B.SHIPMENT_LINE_ID = A.SHIPMENT_LINE_ID
AND UPPER (A.Transaction_Type) = "to DELIVER".
AND higher (A.Destination_Type_Code) = "EXPENSES".
AND D.PO_RELEASE_ID = A.PO_RELEASE_ID
UNION ALL
SELECT A.Po_Header_Id,
C.RECEIPT_NUM receipt_num,
A.Po_Line_Id,
A.Destination_Type_Code,
B.Item_Id,
B.item_description Item_Desc,
A.UNIT_OF_MEASURE Transact_Uom,
A.QUANTITY Transact_Qty,
D.RELEASE_NUM PO_RELEASE
OF RCV_TRANSACTIONS,.
RCV_SHIPMENT_HEADERS C.
B RCV_SHIPMENT_LINES,
PO_RELEASES_ALL D
WHERE C.SHIPMENT_HEADER_ID = A.SHIPMENT_HEADER_ID
AND B.SHIPMENT_LINE_ID = A.SHIPMENT_LINE_ID
AND B.ITEM_ID IS NULL
AND UPPER (A.Transaction_Type) = "to DELIVER".
AND higher (A.Destination_Type_Code) = "WORKSHOP".
D.PO_RELEASE_ID AND = A.PO_RELEASE_ID) rcv3
GROUP BY rcv3. Po_Header_Id, RCV3.receipt_num, rcv3. Po_Line_Id, rcv3. Destination_Type_Code, rcv3. Item_Id, rcv3. Item_Desc, rcv3. Transact_Uom, rcv3. Rcv1 PO_RELEASE),
(SELECT A.PO_LINE_ID,
Sum (A.Quantity) transact_qty,
A.PO_HEADER_ID,
C.RECEIPT_NUM
OF RCV_TRANSACTIONS,.
RCV_SHIPMENT_HEADERS C.
B RCV_SHIPMENT_LINES,
PO_RELEASES_ALL D
WHERE C.SHIPMENT_HEADER_ID = A.SHIPMENT_HEADER_ID
AND B.SHIPMENT_LINE_ID = A.SHIPMENT_LINE_ID
AND UPPER (A.Transaction_Type) = "RETURN to the RECEPTION"
AND D.PO_RELEASE_ID = A.PO_RELEASE_ID
A.PO_LINE_ID, A.PO_HEADER_ID, C.RECEIPT_NUM GROUP) Rcv2
Where Po.Po_Header_Id = Rcv1.Po_Header_Id (+)
And Rcv1.Po_Line_Id = Rcv2.Po_Line_Id (+)
And Rcv1.Receipt_Num = Rcv2.Receipt_Num (+)
And Rcv1.Transact_Qty <>Nvl(Rcv2.Transact_Qty,999999999)
Group of po.po_header_id, rcv1.po_line_id, po.segment1, rcv1.receipt_num, rcv1.item_id, Rcv1.Item_Desc, rcv1. TRANSACT_UOM, rcv1. PO_RELEASE, (NVL(Rcv1.Transact_Qty,0)-NVL(RCV2.transact_qty,0))) has
Is my version of the database: Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - 64 bit Production
Please find the attached PLAN to EXPLAIN.
Thank you
May be essentially the same join (between a, b, c and d) twice instead of three times
sounds like it could be done with a single join (between a, b, c, and d), but you do not some columns than the columns of rcv1 rcv2
We can work on what we can see only
Select rownum line_num,
rcv1.po_line_id line_id,
TO_CHAR (rcv1.item_id) invitmid,
NVL(rcv1.transact_qty,0) - nvl(rcv2.transact_qty,0) pickqty,
unloadingpnt null,
rrnum null,
workordnum null,
workorddesc null,
Po. Ponum Segment1,
dtr_dumb null,
description of the rcv1.item_desc,
from_sub null,
to_sub null,
No source,
asset_number null,
rcv1.receipt_num receiptnum,
moveord null,
from_loc null,
to_loc null,
msd_num null,
contain_line null,
Glu rcv1.transact_uom,
rcv1.po_release
of po_headers_all in.
left outer join
(select a.po_header_id,
c.receipt_num,
a.po_line_id,
a.destination_type_code,
b.item_id,
b.item_description item_desc,
a.unit_of_measure transact_uom,
Sum (a.Quantity) transact_qty,
d.release_num po_release
from (select shipment_header_id,
shipment_line_id,
po_release_id,
po_header_id,
po_line_id,
destination_type_code,
unit_of_measure,
quantity
of rcv_transactions
where upper (a.transaction_type) = "to DELIVER".
and upper (a.destination_type_code) ('charge', 'WORKSHOP')
) a
inner join
rcv_shipment_lines b
On a.shipment_line_id = b.shipment_line_id
inner join
c rcv_shipment_headers
On a.shipment_header_id = c.shipment_header_id
left outer join
po_releases_all d
On a.po_release_id = d.po_release_id
where upper (a.destination_type_code) = "EXPENSES".
or (upper (a.destination_type_code) = 'WORKSHOP'
and b.item_id is null
)
A.po_header_id group,
c.receipt_num,
a.po_line_id,
a.destination_type_code,
b.item_id,
b.item_description,
a.unit_of_measure,
d.release_num
) rcv1
On po.po_header_id = rcv1.po_header_id
left outer join
(select a.po_line_id,
Sum (a.Quantity) transact_qty,
a.po_header_id,
c.receipt_num
from (select shipment_header_id,
shipment_line_id,
po_release_id,
po_header_id,
quantity
of rcv_transactions
where upper (a.transaction_type) = "RETURN to THE RECIPIENTS.
) a
inner join
rcv_shipment_lines b
On a.shipment_line_id = b.shipment_line_id
inner join
c rcv_shipment_headers
On a.shipment_header_id = c.shipment_header_id
left outer join
po_releases_all d
On a.po_release_id = d.po_release_id
A.po_line_id group,
a.po_header_id,
c.receipt_num
) rcv2
On rcv1.po_line_id = rcv2.po_line_id
and rcv1.receipt_num = rcv2.receipt_num
where rcv1.transact_qty! = nvl(rcv2.transact_qty,999999999)
Concerning
Etbin
-
Help with SQL query definition
Hello
Could someone please help me with the sub condition.
On the example of the EMP table, I want a report that looks like:
Please see the application for the relevant columns and the definitions of the watch below:Manager Tot_Joiniees_Q1 Tot_Joiniees_Q2 Tot_Joiniees_Q3 Tot_Joiniees_Q4
Thank youSELECT mgr, CASE WHEN TO_CHAR (hiredate, 'MON') IN ('JUN', 'JUL', 'AUG') THEN 'Q1' WHEN TO_CHAR (hiredate, 'MON') IN ('SEP', 'OCT', 'NOV') THEN 'Q2' WHEN TO_CHAR (hiredate, 'MON') IN ('DEC', 'JAN', 'FEB') THEN 'Q3' WHEN TO_CHAR (hiredate, 'MON') IN ('MAR', 'APR', 'MAY') THEN 'Q4' END qtd, COUNT (DISTINCT ename) FROM emp GROUP BY mgr, CASE WHEN TO_CHAR (hiredate, 'MON') IN ('JUN', 'JUL', 'AUG') THEN 'Q1' WHEN TO_CHAR (hiredate, 'MON') IN ('SEP', 'OCT', 'NOV') THEN 'Q2' WHEN TO_CHAR (hiredate, 'MON') IN ('DEC', 'JAN', 'FEB') THEN 'Q3' WHEN TO_CHAR (hiredate, 'MON') IN ('MAR', 'APR', 'MAY') THEN 'Q4' END ORDER BY 1, 2
CJMSELECT mgr Manager, COUNT(CASE WHEN TO_CHAR(hiredate,'Q') = '1' then 1 end) Tot_Joiniees_Q1, COUNT(CASE WHEN TO_CHAR(hiredate,'Q') = '2' then 1 end) Tot_Joiniees_Q2, COUNT(CASE WHEN TO_CHAR(hiredate,'Q') = '3' then 1 end) Tot_Joiniees_Q3, COUNT(CASE WHEN TO_CHAR(hiredate,'Q') = '4' then 1 end) Tot_Joiniees_Q4 FROM emp GROUP BY mgr, TO_CHAR(hiredate,'Q') / MANAGER TOT_JOINIEES_Q1 TOT_JOINIEES_Q2 TOT_JOINIEES_Q3 TOT_JOINIEES_Q4 ---------- --------------- --------------- --------------- --------------- 7566 0 0 0 1 7839 0 3 0 0 7698 0 0 2 0 7788 0 1 0 0 7782 1 0 0 0 7566 0 1 0 0 7902 0 0 0 1 7698 2 0 0 0 0 0 0 1 7698 0 0 0 1 10 rows selected. SQL>SY.
-
Need help with SQL Query - change of name history of audit table.
I need your help to find the result in the following way...
Emp No New_name Old_Name ----------------------------------------------- 1 Name3 Name2 1 Name2 Name1
Thank you...create table emp(emp_id number(10), emp_name varchar(50), constraints emp_pk primary key(emp_id) ); create table emp_audit( audit_id number(10), emp_id number(10), emp_name varchar(50), audit_date date, constraints emp_audit_pk primary key (audit_id), constraints emp_audit_emp_fk foreign key(emp_id) references emp(emp_id)); insert into emp values(1,'Name3'); insert into EMP_AUDIT (audit_id, emp_id, emp_name, audit_date) values (1, 1, 'Name1', to_date('14-08-2011', 'dd-mm-yyyy')); insert into EMP_AUDIT (audit_id, emp_id, emp_name, audit_date) values (2, 1, 'Name2', to_date('15-08-2011', 'dd-mm-yyyy')); commit;Dipabkar Bédard (DB) wrote:
We write is the query without using "partition by" in oracle...?
with t as ( select audit_id, emp_id, emp_name, row_number() over(order by audit_id) rn from emp_audit order by emp_id, audit_id ) select a.audit_id, a.emp_id, a.emp_name old_name, nvl(b.emp_name,(select c.emp_name from emp c where c.emp_id = a.emp_id)) new_name from t a left join t b on ( b.emp_id = a.emp_id and b.rn = a.rn + 1 ) / AUDIT_ID EMP_ID OLD_NAME NEW_NAME ---------- ---------- -------------------------------------------------- ------------ 1 1 Name1 Name2 2 1 Name2 Name3 SQL>And without analytical functions:
with t1 as ( select audit_id, emp_id, emp_name from emp_audit order by emp_id, audit_id ), t2 as ( select audit_id, emp_id, emp_name, rownum rn from t1 ) select a.audit_id, a.emp_id, a.emp_name old_name, nvl(b.emp_name,(select c.emp_name from emp c where c.emp_id = a.emp_id)) new_name from t2 a left join t2 b on ( b.emp_id = a.emp_id and b.rn = a.rn + 1 ) / AUDIT_ID EMP_ID OLD_NAME NEW_NAME ---------- ---------- -------------------------------------------------- ---------- 1 1 Name1 Name2 2 1 Name2 Name3 SQL>SY.
-
Hello
could you please help write select sql to retrieve the maximum number of concurrent users, something to check?
user_id, rec_start, rec_end
1, 01.10.2009 12:01, 01.10.2009 13:11
2, 01.10.2009 13:12, 01.10.2009 14:04
3, 01.10.2009 13:35, 01.10.2009 13:55
1, 01.10.2009 14:35, 01.10.2009 14:50
1, 01.10.2009 14:45, 01.10.2009 14:50
I would like:
hour, max_of_conc_users
01.10.2009 12: 1 - no more then 1 user at the same time
01.10.2009 13: 2 - user_id 2and 3 record at the same time
01.10.2009 14: 2 - user_id 1 to 2 sessions
Thank you
D.Hello
Interesting problem!
We will change a bit, your sample data for the (total connections per hour) much simpler problem solved by Sven and Hoek has not coincidentally the same results as your problem (the high concurrent users per hour).
I changed the data of Hoek, adding a line, which has been lgged in for the entire period covered by others:CREATE TABLE t as select 9 user_id, to_date('01.10.2009 12:01', 'dd.mm.yyyy hh24:mi') rec_start, to_date('01.10.2009 14:49', 'dd.mm.yyyy hh24:mi') rec_end from dual union all select 1 user_id, to_date('01.10.2009 12:01', 'dd.mm.yyyy hh24:mi') rec_start, to_date('01.10.2009 13:11', 'dd.mm.yyyy hh24:mi') rec_end from dual union all select 2, to_date('01.10.2009 13:12', 'dd.mm.yyyy hh24:mi'), to_date('01.10.2009 14:04', 'dd.mm.yyyy hh24:mi') from dual union all select 3, to_date('01.10.2009 13:35', 'dd.mm.yyyy hh24:mi'), to_date('01.10.2009 13:55', 'dd.mm.yyyy hh24:mi') from dual union all select 1, to_date('01.10.2009 14:35', 'dd.mm.yyyy hh24:mi'), to_date('01.10.2009 14:50', 'dd.mm.yyyy hh24:mi') from dual union all select 1, to_date('01.10.2009 14:45', 'dd.mm.yyyy hh24:mi'), to_date('01.10.2009 14:50', 'dd.mm.yyyy hh24:mi') from dual ;The query below unpivots these data in separate lines for rec_start and rec_end. We use the analytical SUM function to count each + 1 rec_start and each rec_end as -1 to get the number of concurrent users at any time. As soon as we have these data, we can use GROUP BY TRUNC (dt, 'HH')as Sven Hoek has suggested, to get the maximum by houir:
WITH cntr AS ( SELECT 1 AS cnt FROM dual UNION ALL SELECT -1 AS cnt FROM dual ) , events AS ( SELECT CASE cntr.cnt WHEN 1 THEN t.rec_start WHEN -1 THEN t.rec_end END AS dt , cntr.cnt FROM t CROSS JOIN cntr ) , concurrent_users AS ( SELECT TRUNC (dt, 'HH') AS hour , SUM (cnt) OVER (ORDER BY dt) AS concurrent_users FROM events ) SELECT hour , MAX (concurrent_users) AS max_of_conc_users FROM concurrent_users GROUP BY hour ORDER BY hour;Output:
HOUR MAX_OF_CONC_USERS ---------------- ----------------- 01.10.2009 12:00 2 01.10.2009 13:00 3 01.10.2009 14:00 3 -
Need help with sql query involving distinct and County
I have 2 tables and I want to get the number of specific names. Find the details below. It's hard to explain but I will try to provide as much detail as I can.
Table A:
ID of the SR
1001 1
1002 2
2 1003
1004 3
Table B:
Name of the key SrNew
1 David 1001
2 James 1002
3 James 1002
4 James 1003
5 James 1004
6 Mike 1004
Result: I'm looking:
Count names such as if the name appears for the same ID of Table A two times, and then only count 1.
Name of County
David 1
James 2 (1002 and 1003 for the same ID (ID #2) so count as 1 for this and then 1 more for 1004 and ID 3)
1 Mike
I have following question:
SELECT distinct (b.Name), a.ID
FROM TableA, TableB b
Where a.Sr = b.SrNew Group By b.Name, a.ID
and as a result I get:
ID name
David 1
James 2
3 James
3 Mike
Now, I want to just the number of each name with the result, but don't know how I can do this with a sql?
Thanks in advance.Who help me?
SELECT b.Name, count(distinct a.ID) FROM TableA a, TableB b Where a.Sr = b.SrNew Group By b.NameNicolas.
-
Hello
I have a requirement to extract data from a table.
The data looks like this:
ENAME, DEPTNO
----------------------------------------------
ABC 10
20 Sadas
GVK 20
SCOTT 10
BLAKE 10
HARRY 30
The output will look like this:
10, ABC, SCOTT, BLAKE
20 sadas, GVK
30, HARRY
I know that this can be done using PL/SQL.
Y at - it another way to display the data using SQL Analytic functions.
Please send me an answer if you know the answer to [email protected]. **********ASAP*******
Thanks in advance.Try this
This can help u
DeptNo, MAX (LTRIM (SYS_CONNECT_BY_PATH(ename,','), ',')) SELECT ename
from (select deptno, ename,)
ROW_NUMBER() over (partition by deptno arrested by ename) rn
from emp)
START WITH rn = 1
CONNECT BY PRIOR rn = rn - 1
AND PRIOR deptno = deptno
GROUP BY deptno;Thank you
-
I have a third-party site that transforms the information into a csv summary. I can't change the format, and they are not ready to give directly from db and db, so I try and then import it. I'm trying to use SQL Loader, but I can't seem to make it work. I looked in the documentation (VERY rare examples) as well as several websites and I can't find anything that can help out me.
Control file:
Note that the line WHEN is commented out, because I can't make it work.LOAD DATA INFILE 'Daily_Diagnostics_test.csv' BADFILE 'Diag_failed.csv' APPEND INTO TABLE ROAD_DIAGS FIELDS TERMINATED BY "," --WHEN VEHICLE != 'Vehicle' AND VEHICLE != 'Total' (VEHICLE, DRIVER, START_DATE, START_TIME, START_TZ, END_DATE, END_TIME, END_TZ, TM_DRV_MINS, TM_IDLE_MINS, TM_PTO_MINS, TRIP_TM_MINS, FUEL_DRV_GALS, FUEL_IDLE_GALS, FUEL_PTO_GALS, FUEL_TRIP_GALS, FUEL_START_GALS, MPG, DRV_MPG, START_ODOM, END_ODOM, DIFF_ODOM, AVG_MPH, MAX_MPH, MAX_RPM, DATE1, TM1, TZ1, ADDRESS_NUM1, ADDRESS_STREET1, ADDRESS_ST1, ADDRESS_ZIP1, ADDRESS_COUNTY1, DATE2, TM2, TZ2, ADDRESS_NUM2, ADDRESS_STREET2, ADDRESS_ST2, ADDRESS_ZIP2, ADDRESS_COUNTY2, STOP_CT, FAULT_CODE, HARD_BRAKE)
and here is a sample of the data:
As you can see, I need to have the SQL Loader to ignore the header and summary lines that appear for each grouping. Your suggestions are welcome as I need to implement this in an automatic process, I can call with a batch script.Vehicle,Driver,Start Date,Start Time,Timezone,End Date,End Time,Timezone,Driving (hh:mm),Idle (hh:mm),PTO (hh:mm),Trip (hh:mm),Driving (gals),Idle (gals),PTO (gals),Trip (gals),Before Trip (gals), MPG,Driving MPG,Start,End,Trip Distance (miles),Average(MPH),Maximum(MPH),Maximum RPM(rpm),Date,Time,Timezone,Address,Street,City,State,Zip,County,Date,Time,Timezone,Address,Street,City,State,Zip,County,Stop Count,Fault Code,Hard Brake, 26033, ,01/23/08,12:40 AM,MST,01/23/08,12:42 AM,MST,0,3,0,3,0,0.12,0,0.12, ,0,0,295829.22,295829.22,0,0,6.84, ,01/23/08,12:23 AM,MST,0,COUNTY ROAD 108,SAVAGE,MT,59262,RICHLAND,01/23/08,12:23 AM,MST,0,COUNTY ROAD 108,SAVAGE,MT,59262,RICHLAND,0,4, , 26033, ,01/23/08,12:43 AM,MST,01/23/08,04:03 AM,MST,126,74,0,200,0,0,0,0, ,0,0,295829.22,295914.97,85.75,25.72,67.11, ,01/23/08,12:23 AM,MST,0,COUNTY ROAD 108,SAVAGE,MT,59262,RICHLAND,01/23/08,03:56 AM,MST,0,COUNTY ROAD 125,SIDNEY,MT,59270,RICHLAND,8,3, , 26033, ,01/23/08,04:03 AM,MST,01/23/08,12:29 PM,MST,317,190,0,507,0,0,0,0, ,0,0,295914.97,296139.9,224.94,26.64,65.87, ,01/23/08,03:56 AM,MST,0,COUNTY ROAD 125,SIDNEY,MT,59270,RICHLAND,01/23/08,12:29 PM,MST,756,BELL RD,SIDNEY,MT,59270,RICHLAND,14,5, , Total, , , ,443,267,0,710,0,0.12,0,0.12, ,0,0,295829.22,296139.91,310.69,26.38,67.11, , , , , , , , , , , , , , , , , , , ,22,12,0, Vehicle,Driver,Start Date,Start Time,Timezone,End Date,End Time,Timezone,Driving (hh:mm),Idle (hh:mm),PTO (hh:mm),Trip (hh:mm),Driving (gals),Idle (gals),PTO (gals),Trip (gals),Before Trip (gals), MPG,Driving MPG,Start,End,Trip Distance (miles),Average(MPH),Maximum(MPH),Maximum RPM(rpm),Date,Time,Timezone,Address,Street,City,State,Zip,County,Date,Time,Timezone,Address,Street,City,State,Zip,County,Stop Count,Fault Code,Hard Brake, 26035, ,01/23/08,03:59 AM,MST,01/23/08,09:05 AM,MST,222,85,0,306,50.87,1.5,0,52.37, ,3.32,3.42,285056.53,285230.5,173.98,34.07,68.97, ,01/23/08,03:24 AM,MST,756,BELL RD,SIDNEY,MT,59270,RICHLAND,01/23/08,09:00 AM,MST,0,COUNTY ROAD 338,SAVAGE,MT,59262,RICHLAND,7,3, , 26035, ,01/23/08,09:06 AM,MST,01/23/08,03:37 PM,MST,273,118,0,391,68.5,1.5,0,70, ,3.24,3.31,285230.5,285457.3,226.8,34.8,69.59, ,01/23/08,09:00 AM,MST,0,COUNTY ROAD 338,SAVAGE,MT,59262,RICHLAND,01/23/08,03:37 PM,MST,756,BELL RD,SIDNEY,MT,59270,RICHLAND,11,3, , 26035, ,01/23/08,03:45 PM,MST,01/23/08,04:38 PM,MST,32,21,0,53,7.37,0.25,0,7.62, ,3.42,3.54,285457.3,285483.4,26.1,29.73,67.73, ,01/23/08,03:37 PM,MST,756,BELL RD,SIDNEY,MT,59270,RICHLAND,01/23/08,04:33 PM,MST,0,COUNTY ROAD 108,SAVAGE,MT,59262,RICHLAND,2,3, , 26035, ,01/23/08,04:49 PM,MST,01/23/08,10:22 PM,MST,256,76,0,332,52,1.25,0,53.25, ,3.71,3.8,285483.4,285681.0,197.6,35.67,64, ,01/23/08,04:33 PM,MST,0,COUNTY ROAD 108,SAVAGE,MT,59262,RICHLAND,01/23/08,10:17 PM,MST,0,COUNTY ROAD 108,SAVAGE,MT,59262,RICHLAND,9,3, , 26035, ,01/23/08,10:25 PM,MST,01/23/08,10:35 PM,MST,4,5,0,10,0.37,0,0,0.37, ,1.68,1.68,285681.0,285681.62,0.62,3.9,24.85, ,01/23/08,10:17 PM,MST,0,COUNTY ROAD 108,SAVAGE,MT,59262,RICHLAND,01/23/08,10:35 PM,MST,0,COUNTY ROAD 338,SAVAGE,MT,59262,RICHLAND,1,5, , Total, , , ,787,305,0,1092,179.11,4.5,0,183.61, ,3.4,3.49,285056.53,285681.62,625.1,34.35,69.59, , , , , , , , , , , , , , , , , , , ,30,17,0,blarman74 wrote:
Can anyone suggest how to fix this problem? The log file shows this error:Record 8: Rejected - Error on table ROAD_DIAGS, column DATE2. ORA-01858: a non-numeric character was found where a numeric was expectedThe error message is clear - you try to load a value "non-date" in a DATE column and, therefore, sql * loader berms.
That points to a discrepancy between the list of columns in the control file and, in the data file.
Note the position of the table of the "DATE2" column in the table "ROAD_DIAGS":test@XE> test@XE> l 1 select table_name, 2 column_name, 3 column_id 4 from user_tab_columns 5 where table_name='ROAD_DIAGS' 6* and column_name='DATE2' test@XE> / TABLE_NAME COLUMN_NAME COLUMN_ID --------------- --------------- ---------- ROAD_DIAGS DATE2 34 test@XE>If your control file specifies that the 34th column in the data file must be "DATE2". However-
test@XE> test@XE> -- display the line number and the 34th column of the data file test@XE> test@XE> !perl -ne '{@a=split/,/; print $.,"\t",$a[33],"\n"}' Daily_Diagnostics_test.csv 1 County 2 RICHLAND 3 RICHLAND 4 RICHLAND 5 6 County 7 RICHLAND 8 RICHLAND 9 RICHLAND 10 RICHLAND 11 RICHLAND 12 test@XE>the 34th column in your data file is 'COUNTY', which is counted in your control file.
Here is the list of the columns for 29 to 37 positions in the csv file:
test@XE> test@XE> !perl -ne '{if ($. == 1) {@a=split/,/; foreach $i (28..36) {print $i+1,"\t",$a[$i],"\n"}}}' Daily_Diagnostics_test.csv 29 Address 30 Street 31 City 32 State 33 Zip 34 County 35 Date 36 Time 37 Timezone test@XE> test@XE>And here are the columns specified in the control file:
... ... ADDRESS_NUM1, ADDRESS_STREET1, ADDRESS_ST1, ADDRESS_ZIP1, ADDRESS_COUNTY1, DATE2 DATE "MM/DD/YYYY", TM2 DATE "HH:MI AM", TZ2, ... ...(You can open the csv file in MS Excel to get a better picture of the data in there).
As seen above, due to the incompatibility, sql * loader loads 'City' in ADDRESS_ST1, 'State' in ADDRESS_ZIP1, "Zip" in ADDRESS_COUNTY1 and DATE2 "County" - where it error.
The solution would be-
(a) If you want to load the 'City' information from the data file, and then add the "City" column in the table (if nonexistent) and specify in the control file.
(b) If you do not want to load the 'City' information from the data file, you can specify a column to 'FILL' in the file of control as follows:
... ... ADDRESS_NUM1, ADDRESS_STREET1, city1 FILLER, ADDRESS_ST1, ADDRESS_ZIP1, ADDRESS_COUNTY1, DATE2 DATE "MM/DD/YYYY", TM2 DATE "HH:MI AM", TZ2, ADDRESS_NUM2, ADDRESS_STREET2, city2 FILLER, ADDRESS_ST2, ADDRESS_ZIP2, ADDRESS_COUNTY2, ... ...Hope that helps,
isotope
Maybe you are looking for
-
Is it possible to kern (widen the spacing) between letters in the Pages?
Is it possible to kern (widen the spacing) between letters in the Pages?
-
new update won't let me get my iphote
Hey guys having a problem with the new update. Theres a circle with a line through my iphoto app and can't access my photos. also not connect to wifi for 2 days
-
Problems with the Guide of TV Media Centre in the United Kingdom for the new year?
Since midnight January 1, no data are available for any channel in the guide, even though it says that it has successfully downloaded the latest announcements of guide. Does anyone else have this problem? I tried a new analysis for the services and s
-
My PC runs frequently 'Power save mode', my screen is inactivated, and the keyboard or the mouse to activate the display. My only recourse is to power off and power on. I didn't intentionally activate Power save mode, it is useless and do not know h
-
What's the legal system I bought without operating system?
Hello I was looking to buy this system on ebay http://www.ebay.com.au/itm/9-7inch-Windows-7-Intel-Dual-core-Tablet-PC-2GB-RAM-32G-SSD-/221157069780?pt=AU_comp_laptop&hash=item337dfd13d4 and I noticed that the OS is supposed to be «not authentic but w