SQL Move to last record?

I want to quickly move to the very last record in the database (large). Is there an sql command to do this in one step? I have the enterprise tools installed with labview 6.02

nevermind... found the VI "move to record N" if you set the number to -1 it moves to the last record.

Similar Messages

Sql query - Selecting last recorded values for each date in specified period

Hello,
Can someone please help me with my problem.
I'm trying to get last recorded balance for each day for specific box (1 or 2) in specified period of days from ms access database using ADOTool.
I'm trying to get that information with SQL query but so far unsuccessfully...
My table looks like this:
Table name: TestTable
Date Time Location Box Balance
20.10.2014. 06:00:00 1 1 345
20.10.2014. 12:00:00 1 1 7356
20.10.2014. 18:45:00 1 1 5678
20.10.2014. 23:54:00 1 1 9845
20.10.2014. 06:00:02 1 2 35
20.10.2014. 12:00:04 1 2 756
20.10.2014. 18:45:06 1 2 578
20.10.2014. 23:54:10 1 2 845
21.10.2014. 06:00:00 1 1 34
21.10.2014. 12:05:03 1 1 5789
21.10.2014. 15:00:34 1 1 1237
21.10.2014. 06:00:00 1 2 374
21.10.2014. 12:05:03 1 2 54789
21.10.2014. 15:00:34 1 2 13237
22.10.2014. 06:00:00 1 1 8562
22.10.2014. 10:00:00 1 1 1234
22.10.2014. 17:03:45 1 1 3415
22.10.2014. 22:00:00 1 1 6742
22.10.2014. 06:00:05 1 2 562
22.10.2014. 10:00:16 1 2 123
22.10.2014. 17:03:50 1 2 415
22.10.2014. 22:00:10 1 2 642
23.10.2014. 06:00:00 1 1 9876
23.10.2014. 09:13:00 1 1 223
23.10.2014. 13:50:17 1 1 7768
23.10.2014. 19:47:40 1 1 3456
23.10.2014. 21:30:00 1 1 789
23.10.2014. 23:57:12 1 1 25
23.10.2014. 06:00:07 1 2 976
23.10.2014. 09:13:45 1 2 223
23.10.2014. 13:50:40 1 2 78
23.10.2014. 19:47:55 1 2 346
23.10.2014. 21:30:03 1 2 89
23.10.2014. 23:57:18 1 2 25
24.10.2014. 06:00:55 1 1 346
24.10.2014. 12:30:22 1 1 8329
24.10.2014. 23:50:19 1 1 2225
24.10.2014. 06:01:00 1 2 3546
24.10.2014. 12:30:26 1 2 89
24.10.2014. 23:51:10 1 2 25
Let's say the period is 21.10.2014. - 23.10.2014. and I want to get last recorded balance for box 1. for each day. The result should look like this:
Date Time Location Box Balance
21.10.2014. 15:00:34 1 1 1237
22.10.2014. 22:00:00 1 1 6742
23.10.2014. 23:57:12 1 1 25
So far I've managed to write a query that gives me balance for ONLY ONE date (date with highest time in whole table), but I need balance for EVERY date in specific period.
My incorrect code (didn't manage to implement "BETWEEN" for dates...):
SELECT TestTable.[Date], TestTable.[Time], TestTable.[Location], TestTable.[Box], TestTable.[Balance]
FROM TestTable
WHERE Time=(SELECT MAX(Time)
FROM TestTable
WHERE Location=1 AND Box=1 );
Tnx!
Solved!
Go to Solution.

For loop
following query keep day (here 24 in below query) Variable from ( 1 to 28-29/30/31 as per month)
SELECT TOP 1 TestTable.[Date], TestTable.[Time], TestTable.[Location], TestTable.[Box], TestTable.[Balance]
FROM Test Table.
WHERE Time=(SELECT MAX(Time) FROM TestTable WHERE Location=1 AND Box=1 )
AND DATE = "2014-10-24";
PBP (CLAD)
Labview 6.1 - 2014
KUDOS ARE WELCOMED.
If your problem get solved then mark as solution.

SQL (first and last record)

Hi All,
WITH t AS
SELECT 100 region_id, 1001 Client_Id, 'client 1' Client_Name, 1 Status_id, 'progress' Status_name, SYSDATE + 1 Start_Date, SYSDATE + 1 End_Date FROM dual UNION ALL
SELECT 100 region_id, 1001 Client_Id, 'client 1' Client_Name, 2 Status_id, 'hold' Status_name ,SYSDATE + 1 Start_Date, SYSDATE + 1 End_Datet FROM dual UNION ALL
SELECT 100 region_id, 1001 Client_Id, 'client 1' Client_Name, 3 Status_id ,'reject' Status_name, SYSDATE + 1 Start_Date, SYSDATE + 1 End_Date FROM dual UNION ALL
SELECT 200 region_id, 1002 Client_Id, 'client 2' Client_Name, 1 Status_id, 'progress' Status_name, SYSDATE + 1 Start_Date, SYSDATE + 1 End_Date FROM dual UNION ALL
SELECT 200 region_id, 1002 Client_Id, 'client 2' Client_Name, 2 Status_id, 'hold' Status_name ,SYSDATE + 1 Start_Date, SYSDATE + 1 End_Datet FROM dual UNION ALL
SELECT 200 region_id, 1002 Client_Id, 'client 2' Client_Name, 3 Status_id ,'reject' Status_name, SYSDATE + 1 Start_Date, SYSDATE + 1 End_Date FROM dual UNION ALL
SELECT 200 region_id, 1002 Client_Id, 'client 2' Client_Name, 3 Status_id ,'progress' Status_name, SYSDATE + 1 Start_Date, SYSDATE + 1 End_Date FROM dual
SELECT * FROM t;Desired output:
REGION_ID     CLIENT_ID     CLIENT_NAME     STATUS_ID     STATUS_NAME     START_DATE                    END_DATE
100     1001     client 1                      1                     progress                     9/3/2011 12:26               9/3/2011 12:26
100     1001     client 1                      3                     reject                     9/3/2011 12:26                     9/3/2011 12:26
200     1002     client 2                      1                     progress                     9/3/2011 12:26                     9/3/2011 12:26thanks,

Hi,
Like Sven, I don't understand why you only want one output row for client_id=1002. Also, I don't understand what order determine "first" and "last". It's hard to guess because so many of the columns have the same value on every row.
My best guess is that status_id plays a part in deciding who is "first" and "last", and that you don't want to include ties (multiple rows with a equal claim to being "first" or "last") in the output.
WITH     got_analytics     AS
     SELECT     t.*     -- or whatever columns you want to display
     ,     ROW_NUMBER () OVER ( PARTITION BY region_id
                         ,             client_id
                         ,             client_name
                         ORDER BY        end_date
                         ,             start_date
                         ,             status_id
                        )     AS a_num
     ,     ROW_NUMBER () OVER ( PARTITION BY region_id
                         ,             client_id
                         ,             client_name
                         ORDER BY        end_date     DESC
                         ,             start_date     DESC
                         ,             status_id     DESC
                        )     AS d_num
     ,     COUNT (*)     OVER ( PARTITION BY region_id
                         ,             client_id
                         ,             client_name
                         ,             end_date     -- NOTE: no ORDER BY
                         ,             start_date
                         ,             status_id
                        )     AS cnt
     FROM     t
--     WHERE     ...     -- if you need any filtering, this is where it goes
SELECT       region_id, client_id, client_name, status_id, start_date, end_date
FROM       got_analytics
WHERE       (     a_num     = 1
       OR     d_num     = 1
AND       cnt     = 1
ORDER BY region_id, client_id, client_name, start_date, end_date, status_id
;If client_name and region_id depend on client_id, then you don't need to include them in the PARTITION BY clauses. Depending on your requirements, you may not want them in the PARTITION BY clauses even if they are independent of client_id. Then again, you may not want client_id in th PARTITION BY clauses. It all depends on what you're trying to do.

Moving the last record to 1st position

Hi,
I have 200 records in my internal table.I want to move the last record(200 position) to 1st position(i.e Index1). How to do it?

Hi
U can try something like this
DATA: BEGIN OF ITAB OCCURS 0,
        FIELD(1) TYPE C,
      END   OF ITAB.
DO 9 TIMES.
MOVE SY-INDEX TO ITAB-FIELD.
APPEND ITAB.
ENDDO.
* Check table
DESCRIBE TABLE ITAB LINES SY-TABIX.
* Read last record
READ TABLE ITAB INDEX SY-TABIX.
* Delete last record
DELETE ITAB     INDEX SY-TABIX.
* Append the last record to the top
INSERT ITAB     INTO ITAB INDEX 1.
LOOP AT ITAB.
WRITE: / ITAB.
ENDLOOP.
Max

Cursor Jumping to last record when query mode for Multi-record block.

I have a multi-record block. the seq field has this validation to check that there should not be a gap
in sequence for S_type and cer_dl field.The program unit is as below.
PROCEDURE seq_validation IS
--Validation to check that there is no gaps in sequence for S_TYPE and cer_dl fields
l_value_to_check varchar2(100);
l_seq_found number;
l_curr_sequence number;
l_new_value varchar2(100);
l_found boolean:=FALSE;
l_new_set boolean := FALSE; --s_type and cer_dl are different from previous set.
begin
if trim(:b1.s_type) is not null
or trim(:b1.cer_dl) is not null
then
-- Program continues here only if all the items are not null
-- Get information from record that needs to be validated
l_value_to_check := trim(:b1.s_type) || ':'|| trim(:b1.cer_dl) ;
l_curr_sequence := :sequence;
if :SYSTEM.CURSOR_RECORD = '1' then
---to check sequence is entered as 1 in the first record
message('Error:Sequence should start with 1');
raise form_trigger_failure;
end if;
go_item('b1.XYZ');
else
FIRST_RECORD;
while :SYSTEM.LAST_RECORD != 'TRUE'
loop
l_found := FALSE;
l_new_value := trim(:b1.S_type) || ':'|| trim(:cer_dl) ;
if l_new_value = l_value_to_check then --
l_new_set:= FALSE;
l_seq_found := :seq;
if l_seq_found >= l_curr_sequence then
go_item('b1.seq');
l_new_set := FALSE;
elsif l_seq_found < l_curr_sequence - 1 then
go_item('b1.sequence');
l_found := FALSE;
elsif l_seq_found = l_curr_sequence - 1 then
l_found:= TRUE;
go_item('b1.xyz); --go to next item
end if;
else
l_new_set := TRUE;
end if;
NEXT_RECORD;
end loop;
if l_new_set then
go_item('b1.xyz'); -- go to next item
else
if not l_found then
message('Error:Sequence should be in order.');
go_item('b1.seq');
raise form_trigger_failure;
end if;
end if;
go_item('b1.xyz'); ---go to next item
end if;
end if;
end;
In the insert mode its working fine(unless you gurus feel to change to make it more perfect.
When I say enter-query and put 'IAS' in the s_type field and then do a execute query it brings all the records of 'IAS'.
Now as I want to update the records, and when I navigate across field(s_type,cer_dl,Seq,xyz) using tab and move the cursor down on second
record(IAS N 2 N), what happens is when I navigate to sequence field the cursor jumps to LAST RECORD( IAS Y 3 N ) of LAST field XYZ.
I am putting the sample record
s_type cer_dl seq xyz
IAS N 1 N
IAS N 2 N
IAS N 3 N
IAS N 4 N
IAS Y 1 N
IAS Y 2 N
IAS Y 3 N
The cursor moves to last record --IAS -Y -3-N when I tab across the second record.

When I say enter-query and put 'IAS' in the s_type field...For your validation to work you must not allow users to enter records in the middle of a block, clear records from the block or query specific records. If you allow the users to enter a query then you are allowing 2 of the rules to be broken.

How to get the first and the last record of every month within a time range in sql

I am trying to get the first record and the last record of each and every month in a given date range using sql. I have a sample code
where i have just selected everything within the date range now i have to extract the first and the last records of each and every month.
SELECT PurOrderNum,
OrderDate
FROM Purchasing.PurOrder
WHERE OrderDate >= '2013-02-28'
AND OrderDate <= '2014-12-29'

SELECT PurOrderNum,
OrderDate
FROM
SELECT PurOrderNum,
OrderDate,
MAX(OrderDate) OVER (PARTITION BY DATEDIFF(mm,0,OrderDate)) AS MaxDate,
MIN(OrderDate) OVER (PARTITION BY DATEDIFF(mm,0,OrderDate)) AS MinDate
FROM Purchasing.PurOrder
WHERE OrderDate >= '2013-02-28'
AND OrderDate <= '2014-12-29'
)t
WHERE OrderDate = MaxDate
OR OrderDate = MinDate
Please Mark This As Answer if it solved your issue
Please Vote This As Helpful if it helps to solve your issue
Visakh
My MSDN Page
My Personal Blog
My Facebook Page

When I send mi presentation to a Quick Movie, the program record only the last frame

Why when I send mi presentation to a Quick Movie, the program record only the last frame, how I do to record the whole presentation of 52 frames

I mean when I try to burn a CD.

Can I use SQL to move a device record?

It appears all/most of the powershell scripts out there to move a device record into a collection utilize WMI.
Is it possible to use SQL to add a device record into a collection.
I also mean to execute this via C#.
Any assistance with this wether direct or indirect would be very much appreciated.

It appears all/most of the powershell scripts out there to move a device record into a collection utilize WMI.
Is it possible to use SQL to add a device record into a collection.
I also mean to execute this via C#.
Any assistance with this wether direct or indirect would be very much appreciated.
No, you can't use SQL to directly add a record to CM12. (Short of dropping to DOS and executing a batch,powershell, etc.)
Have you look at the SDK, it should allow you to do what you want in C#.
Garth Jones | My blogs: Enhansoft and
Old Blog site | Twitter:
@GarthMJ

See my problem no master record move to next record

hi master
sir
i have master detail block master table have all data but detail no data i need feed data in detail then i goto in detail block write value and press F10 for save
for new record in detail
in detail blick post-block or post_record or post_insert event i write this code
go_block('masterblock');
next_record;
go_item('detailblock.item');
but not responses no master block move to next record still and my curser sitll in last postion
sir how i move master record after save the detail record and jump to detail block for new record
please give me idea or code
thanking you
aamir

thank for reply
sir your help very usefull for my but when i press F10 then system show massage do you want to save change the record
sir how i avoid that window and directly move to next record
please give me idea
thanking you
aamir

Select first and last records in grouped results - Oracle 11g

Say I have the following information in an Oracle 11g table:
Qty
Production order
Date and time
20
00000000000000001
12-JAN-14 00:02
20
00000000000000001
12-JAN-14 00:05
20
00000000000000001
12-JAN-14 00:07
20
00000000000000001
13-JAN-14 00:09
30
00000000000000002
12-JAN-14 00:11
30
00000000000000002
12-JAN-14 00:15
30
00000000000000002
12-JAN-14 00:20
30
00000000000000002
14-JAN-14 00:29
I would like to write a query that would return the following:
Qty
Production order
First
Last
80
00000000000000001
12-JAN-14 00:02
13-JAN-14 00:09
120
00000000000000002
12-JAN-14 00:11
14-JAN-14 00:29
That is, the sum of the Qty column grouped by Production order, and the date/time of the first and last records for each Production order.
I came up with a query that yielded this result:
Qty
Production order
First
Last
80
00000000000000001
12-JAN-14 00:02
14-JAN-14 00:29
120
00000000000000002
12-JAN-14 00:02
14-JAN-14 00:29
Which means that the First and Last columns show the overall first and last date / time of the whole table. Please note that this is a dummy table. Sorry I am now allowed to write the actual query
I came up with since work policies do not allow me to share it. Also, I tried with windowing functions such as rank()and row_number() but my user does not have enough privileges to do so.
Any help or hints will be greatly appreciated.

Due to the fact that Oracle does not record the rows in any particular order, it would be wrong that the "first date" would be the first row processed by the query.
Therefore you would have to supply some other column if you do not want to consider the table as ordered by date.
Also, any analytical functions will need you to supply the "order by" and if its the date, then just a simple query will do:
SQL>WITH Tab1 (Qty, Production_Order, Pdate)
2       AS (SELECT 20, '00000000000000001', TO_DATE ( '12-JAN-14 00:02', 'DD-MON-YY HH24:MI') FROM DUAL UNION ALL
3           SELECT 20, '00000000000000001', TO_DATE ( '12-JAN-14 00:05', 'DD-MON-YY HH24:MI') FROM DUAL UNION ALL
4           SELECT 20, '00000000000000001', TO_DATE ( '12-JAN-14 00:07', 'DD-MON-YY HH24:MI') FROM DUAL UNION ALL
5           SELECT 20, '00000000000000001', TO_DATE ( '13-JAN-14 00:09', 'DD-MON-YY HH24:MI') FROM DUAL UNION ALL
6           SELECT 30, '00000000000000002', TO_DATE ( '12-JAN-14 00:11', 'DD-MON-YY HH24:MI') FROM DUAL UNION ALL
7           SELECT 30, '00000000000000002', TO_DATE ( '12-JAN-14 00:15', 'DD-MON-YY HH24:MI') FROM DUAL UNION ALL
8           SELECT 30, '00000000000000002', TO_DATE ( '12-JAN-14 00:20', 'DD-MON-YY HH24:MI') FROM DUAL UNION ALL
9           SELECT 30, '00000000000000002', TO_DATE ( '14-JAN-14 00:29', 'DD-MON-YY HH24:MI') FROM DUAL)
10 SELECT   SUM ( Qty), Production_Order, MIN ( Pdate), MAX ( Pdate)
11      FROM Tab1
12 GROUP BY Production_Order
13* ORDER BY Production_Order
SQL> /
SUM(QTY) PRODUCTION_ORDER     MIN(PDATE)                    MAX(PDATE)
        80 00000000000000001    12-Jan-2014 00:02:00          13-Jan-2014 00:09:00
       120 00000000000000002    12-Jan-2014 00:11:00          14-Jan-2014 00:29:00

Query takes too much time in fetching last records.

Hi,
I am using oracle 8.1 and trying to execute a SQL statement, it takes few minutes and display records.
When trying to fetch all the records, it is fast up to some level and takes much time to fetch last record.
Ex: If total records = 16336 , then it fetches records faster up to 16300 and takes app.500 sec to fetch last 36 records.
Could you kindly let me know the reason.
I have copied the explain plan below for your reference.Please let me know if anything is required.
SELECT STATEMENT, GOAL = RULE               4046     8     4048
NESTED LOOPS OUTER               4046     8     4048
NESTED LOOPS OUTER               4030     8     2952
   FILTER
    NESTED LOOPS OUTER
     NESTED LOOPS OUTER               4014     8     1728
      NESTED LOOPS               3998     8     936
       TABLE ACCESS BY INDEX ROWID     IFSAPP     CUSTOMER_ORDER_TAB     3966     8     440
        INDEX RANGE SCAN     IFSAPP     CUSTOMER_ORDER_1_IX     108     8
       TABLE ACCESS BY INDEX ROWID     IFSAPP     CUSTOMER_ORDER_LINE_TAB     4     30667     1901354
        INDEX RANGE SCAN     IFSAPP     CUSTOMER_ORDER_LINE_PK     3     30667
      TABLE ACCESS BY INDEX ROWID     IFSAPP     PWR_CONS_PARCEL_CONTENT_TAB     2     2000     198000
       INDEX RANGE SCAN     IFSAPP     PWR_CONS_PARCEL_CONTENT_1_IDX     1     2000
     TABLE ACCESS BY INDEX ROWID     IFSAPP     PWR_CONS_PARCEL_TAB     1     2000     222000
      INDEX UNIQUE SCAN     IFSAPP     PWR_CONS_PARCEL_PK          2000
   TABLE ACCESS BY INDEX ROWID     IFSAPP     CONSIGNMENT_PARCEL_TAB     1     2000     84000
    INDEX UNIQUE SCAN     IFSAPP     CONSIGNMENT_PARCEL_PK          2000
TABLE ACCESS BY INDEX ROWID     IFSAPP     PWR_OBJECT_CONNECTION_TAB     2     20     2740
   INDEX RANGE SCAN     IFSAPP     PWR_OBJECT_CONNECTION_IX1     1     20     Thanks.

We are using PL/SQL Developer tool. The time what we have mentioned in the post is approximated time.
Apologies for not mentioning these details in previous thread.Let it be approximate time but how did you arrive at that time? When a query fetches records how did you derived that one portion is fetched in x and the remaining in y time limit?
I would suggest this could be some issue with PL/SQL Developer (Never used this tool by myself) But for performance testing i would suggest you to use SQL Plus. Thats the best tool to test performance.

How to get last Record ior Total rows in For Loop Cursor ?

Hi Friends
I would like to know , the last record in for loop cursor, i have the code in following format
cursor c1 is
select * from emp;
begin
for r1 in c1 loop
v_total_rec := ? ( i would like to know total rows in the cursor , say for example if cursor has 10 rows, i want10 into this variable )
v_count := v_count +1;
dbms_output.put_line(r1.emp_name);
end loop;
end;
Hope i am clear
Any suggestions?
Thanks
Ravi

Even though cursor loops are generally a Bad Idea ^tm^ as Dan says, here's an example of how you can get the information you wanted within the query itself...
SQL> ed
Wrote file afiedt.buf
1 declare
2    cursor c1 is
3      select emp.*
4            ,count(*) over (order by empno) as cnt
5            ,count(*) over () as total_cnt
6      from emp
7      order by empno;
8 begin
9    for r1 in c1 loop
10      dbms_output.put_line(r1.ename||' - row: '||r1.cnt||' of '||r1.total_cnt);
11    end loop;
12* end;
SQL> /
SMITH - row: 1 of 14
ALLEN - row: 2 of 14
WARD - row: 3 of 14
JONES - row: 4 of 14
MARTIN - row: 5 of 14
BLAKE - row: 6 of 14
CLARK - row: 7 of 14
SCOTT - row: 8 of 14
KING - row: 9 of 14
TURNER - row: 10 of 14
ADAMS - row: 11 of 14
JAMES - row: 12 of 14
FORD - row: 13 of 14
MILLER - row: 14 of 14
PL/SQL procedure successfully completed.
SQL>

Puzzle why query returned wrong value for last record in update.

Hi all,
10.2.0.4, Windows 32 bit.
Apply 15% dis.count to related 5 items ( total amount 522 ) sorted by lowest amount first (amount = 1).
Last record ( amount = 200 ) will be the total dis.count amount ( 78.3 ) less applied accumulated dis.count amount ( 48.50 ).
Query runs fine without update clause but wrong result for last record with update.
CREATE TABLE "T1"
    "ID"       NUMBER NOT NULL ENABLE,
    "ITEM_ID" VARCHAR2(20 BYTE) NOT NULL ENABLE,
    "AMOUNT"   NUMBER(10,2) NOT NULL ENABLE,
    "dizcount" NUMBER(10,2)
INSERT INTO T1 ( ID, ITEM_ID, AMOUNT, dizcount ) VALUES ( 65, '101', 1, NULL ) ;
INSERT INTO T1 ( ID, ITEM_ID, AMOUNT, dizcount ) VALUES ( 65, '102', 1, NULL ) ;
INSERT INTO T1 ( ID, ITEM_ID, AMOUNT, dizcount ) VALUES ( 65, '201', 200, NULL ) ;
INSERT INTO T1 ( ID, ITEM_ID, AMOUNT, dizcount ) VALUES ( 65, '215', 155, NULL ) ;
INSERT INTO T1 ( ID, ITEM_ID, AMOUNT, dizcount ) VALUES ( 65, '111', 165, NULL ) ;
UPDATE t1 a
SET a.dizcount =
(SELECT
    CASE
      WHEN rec_count = row_count
      THEN (78.3 - NVL(lag(temp_total,1) over ( order by rec_count) ,0))-- 78.3 is total dizcount amount from 522 * .15
      ELSE disc_amt
    END amt
FROM
    (SELECT id,
      item_id,
      disc_amt,
      rec_count,
      row_count,
      CASE
        WHEN rec_count != row_count -- accumulate dizcount amount except for last record
        THEN SUM(disc_amt) over (order by rec_count)
        ELSE 0
      END temp_total
    FROM
      (SELECT ID ,
        item_id,
        amount amt,
        ROUND(amount * .15,1) disc_amt, -- dizcount is 15%
        row_number () over (order by amount) rec_count,
        COUNT ( *) over () row_count
      FROM t1
      WHERE ID = 65
    GROUP BY id,
      item_id,
      disc_amt,
      rec_count,
      row_count
    )b
WHERE a.item_id = b.item_id
);Regards
Zack
Edited by: Zack.L on Jul 26, 2010 1:26 AM

Zack.L wrote:
Query runs fine without update clause but wrong result for last record with update.Not sure why but looks like another case in favour of MERGE.
MERGE INTO T1
using (SELECT id, item_id,
    CASE
      WHEN rec_count = row_count
      THEN (78.3 - NVL(lag(temp_total,1) over ( order by rec_count) ,0))-- 78.3 is total dizcount amount from 522 * .15
      ELSE disc_amt
    END amt
FROM
    (SELECT id,
      item_id,
      disc_amt,
      rec_count,
      row_count,
      CASE
        WHEN rec_count != row_count -- accumulate dizcount amount except for last record
        THEN SUM(disc_amt) over (order by rec_count)
        ELSE 0
      END temp_total
    FROM
      (SELECT ID ,
        item_id,
        amount amt,
        ROUND(amount * .15,1) disc_amt, -- dizcount is 15%
        row_number () over (order by amount) rec_count,
        COUNT ( *) over () row_count
      FROM t1
      WHERE ID = 65
    GROUP BY id,
      item_id,
      disc_amt,
      rec_count,
      row_count
    ))b
on (t1.id = b.id and t1.item_id = b.item_id)
when matched then update set dizcount = b.amt ;This worked for me. I tested on 11.2 but should work on 10.2.0.4
SQL> select * from t1 order by amount ;
        ID ITEM_ID                  AMOUNT   DIZCOUNT
        65 101                           1         .2
        65 102                           1         .2
        65 215                         155       23.3
        65 111                         165       24.8
        65 201                         200       29.8p.s. BTW, not sure if you want the column names in different cases, but your script, as it is, gave me an error (on 11.2)

File import skips last record without blank line

The following code works great for importing a csv file and inserting the records into tables..
but the last record is not read unless there is a blank line after it..
what needs to be changed so that the last line works with or without a carriage return after it..
create or replace PACKAGE BODY "HTMLDB_TOOLS"
AS
TYPE varchar2_t IS TABLE OF VARCHAR2(32767) INDEX BY binary_integer;
-- Private functions --{{{
PROCEDURE delete_collection ( --{{{
-- Delete the collection if it exists
p_collection_name IN VARCHAR2
IS
BEGIN
IF (htmldb_collection.collection_exists(p_collection_name))
THEN
htmldb_collection.delete_collection(p_collection_name);
END IF;
END delete_collection; --}}}
PROCEDURE csv_to_array ( --{{{
-- Utility to take a CSV string, parse it into a PL/SQL table
-- Note that it takes care of some elements optionally enclosed
-- by double-quotes.
p_csv_string IN VARCHAR2,
p_array OUT wwv_flow_global.vc_arr2,
p_separator IN VARCHAR2 := ','
IS
l_start_separator PLS_INTEGER := 0;
l_stop_separator PLS_INTEGER := 0;
l_length PLS_INTEGER := 0;
l_idx BINARY_INTEGER := 0;
l_quote_enclosed BOOLEAN := FALSE;
l_offset PLS_INTEGER := 1;
BEGIN
l_length := NVL(LENGTH(p_csv_string),0);
IF (l_length <= 0)
THEN
RETURN;
END IF;
LOOP
l_idx := l_idx + 1;
l_quote_enclosed := FALSE;
IF SUBSTR(p_csv_string, l_start_separator + 1, 1) = '"'
THEN
l_quote_enclosed := TRUE;
l_offset := 2;
l_stop_separator := INSTR(p_csv_string, '"', l_start_separator + l_offset, 1);
ELSE
l_offset := 1;
l_stop_separator := INSTR(p_csv_string, p_separator, l_start_separator + l_offset, 1);
END IF;
IF l_stop_separator = 0
THEN
l_stop_separator := l_length + 1;
END IF;
p_array(l_idx) := (SUBSTR(p_csv_string, l_start_separator + l_offset,(l_stop_separator - l_start_separator - l_offset)));
EXIT WHEN l_stop_separator >= l_length;
IF l_quote_enclosed
THEN
l_stop_separator := l_stop_separator + 1;
END IF;
l_start_separator := l_stop_separator;
END LOOP;
END csv_to_array; --}}}
PROCEDURE get_records(p_blob IN blob,p_records OUT varchar2_t) --{{{
IS
l_record_separator VARCHAR2(2) := chr(13)||chr(10);
l_last INTEGER;
l_current INTEGER;
BEGIN
-- Sigh, stupid DOS/Unix newline stuff. If HTMLDB has generated the file,
-- it will be a Unix text file. If user has manually created the file, it
-- will have DOS newlines.
-- If the file has a DOS newline (cr+lf), use that
-- If the file does not have a DOS newline, use a Unix newline (lf)
IF (NVL(dbms_lob.instr(p_blob,utl_raw.cast_to_raw(l_record_separator),1,1),0)=0)
THEN
l_record_separator := chr(10);
END IF;
l_last := 1;
LOOP
l_current := dbms_lob.instr( p_blob, utl_raw.cast_to_raw(l_record_separator), l_last, 1 );
EXIT WHEN (nvl(l_current,0) = 0);
p_records(p_records.count+1) := utl_raw.cast_to_varchar2(dbms_lob.substr(p_blob,l_current-l_last,l_last));
l_last := l_current+length(l_record_separator);
END LOOP;
END get_records; --}}}
-- Utility functions --{{{
PROCEDURE parse_textarea ( --{{{
p_textarea IN VARCHAR2,
p_collection_name IN VARCHAR2
IS
l_index INTEGER;
l_string VARCHAR2(32767) := TRANSLATE(p_textarea,chr(10)||chr(13)||' ,','@@@@');
l_element VARCHAR2(100);
BEGIN
l_string := l_string||'@';
htmldb_collection.create_or_truncate_collection(p_collection_name);
LOOP
l_index := instr(l_string,'@');
EXIT WHEN NVL(l_index,0)=0;
l_element := substr(l_string,1,l_index-1);
IF (trim(l_element) IS NOT NULL)
THEN
htmldb_collection.add_member(p_collection_name,l_element);
END IF;
l_string := substr(l_string,l_index+1);
END LOOP;
END parse_textarea; --}}}
PROCEDURE parse_file( --{{{
p_file_name IN VARCHAR2,
p_collection_name IN VARCHAR2,
p_headings_item IN VARCHAR2,
p_columns_item IN VARCHAR2,
p_ddl_item IN VARCHAR2,
p_table_name IN VARCHAR2 DEFAULT NULL
IS
l_blob blob;
l_records varchar2_t;
l_record wwv_flow_global.vc_arr2;
l_datatypes wwv_flow_global.vc_arr2;
l_headings VARCHAR2(4000);
l_columns VARCHAR2(4000);
l_seq_id NUMBER;
l_num_columns INTEGER;
l_ddl VARCHAR2(4000);
BEGIN
IF (p_table_name is not null)
THEN
l_ddl := 'insert into '||p_table_name||' '||
'select '||v(p_columns_item)||' '||
'from htmldb_collections '||
'where seq_id > 0 and collection_name='''||p_collection_name||'''';
execute immediate l_ddl;
RETURN;
END IF;
BEGIN
select blob_content into l_blob from wwv_flow_files
where name=p_file_name;
EXCEPTION
WHEN NO_DATA_FOUND THEN
raise_application_error(-20000,'File not found, id='||p_file_name);
END;
get_records(l_blob,l_records);
IF (l_records.count < 2)
THEN
raise_application_error(-20000,'File must have at least 2 ROWS, ONE HEADER and ONE OR MORE DATA, id='||p_file_name);
END IF;
-- Initialize collection
htmldb_collection.create_or_truncate_collection(p_collection_name);
-- Get column headings
csv_to_array(l_records(1),l_record);
l_num_columns := l_record.count;
if (l_num_columns > 50) then
raise_application_error(-20000,'Max. of 50 columns allowed, id='||p_file_name);
end if;
-- Get column headings and names
FOR i IN 1..l_record.count
LOOP
l_headings := l_headings||':'||l_record(i);
l_columns := l_columns||',c'||lpad(i,3,'0');
END LOOP;
l_headings := ltrim(l_headings,':');
l_columns := ltrim(l_columns,',');
htmldb_util.set_session_state(p_headings_item,l_headings);
htmldb_util.set_session_state(p_columns_item,l_columns);
-- Save data into specified collection
FOR i IN 2..l_records.count
LOOP
csv_to_array(l_records(i),l_record);
l_seq_id := htmldb_collection.add_member(p_collection_name,'dummy');
FOR i IN 1..l_record.count
LOOP
htmldb_collection.update_member_attribute(
p_collection_name=> p_collection_name,
p_seq => l_seq_id,
p_attr_number => i,
p_attr_value => l_record(i)
END LOOP;
END LOOP;
DELETE FROM wwv_flow_files WHERE name=p_file_name;
END;
BEGIN
NULL;
END;

found someone that helped me.. here was the fix:
l_current := dbms_lob.instr( p_blob, utl_raw.cast_to_raw(l_record_separator), l_last, 1 );
-- START of new code, lines above as per existing
IF nvl(l_current,0) = 0 AND LENGTH(TRIM(utl_raw.cast_to_varchar2(dbms_lob.substr(p_blob,32767,l_last)))) > 0 THEN
p_records(p_records.count+1) := utl_raw.cast_to_varchar2(dbms_lob.substr(p_blob,32767,l_last));
END IF;
-- END of new code,, lines below as per existing
EXIT WHEN (nvl(l_current,0) = 0);

Compare 2 data sets and display the last record

I have one record in first table and 2 records in second table.
Ex:Table 1:
A     100     200   0
B     200     300   0
Table 2:
A      0        0     100
A      0        0     500
B      0        0     700
In one final template I want to display like below with only 2 records like below.
A      100    200    500
B      200    300    700
Anybody suggest me how to pick only last record if the count is greater than 1 in second table.

not sure what are you trying to do
if you want to join two table then sql is more useful IMHO
Anybody suggest me how to pick only last record if the count is greater than 1 in second table.
you can get last record by last() function
<?for-each:SOMETAGorGROUP[position()=last()]?>

SQL Move to last record?

Similar Messages

Maybe you are looking for