Rank - no null

Declare @test table
calldate date,
calllogid int,
id int
Insert into @test select '2013-12-02', 52713, 48287
Insert into @test select '2013-12-02', 52713, 271
Insert into @test select '2014-05-01', 71817,NULL
Insert into @test select '2014-05-01', 71817, 62805
Insert into @test select '2014-02-28', 63995, NULL
Select * from @test
order by calllogid
Above is temp table with sample data and below is the output I'm looking for..
calldate calllogid id
2013-12-02 52713 48287
2013-12-02 52713 271
2014-02-28 63995 NULL
2014-05-01 71817 62805
Say suppose, if a callogid is having multiple id's I'm interested to display both. And which if one is null I just want to show not null record for that calllogid, in above for calllogid 71817 I'm only interested in seeing not null record. and If the calllogid
has only one record with null id, I would like to show that as well..
I can achieve with rank function but this dealing with NULL giving me issue.Help is appreciated.
- please mark correct answers

Please do not post those proprietary table variables. We need to know keys, constraints, DRI, etc. to give you answers.
What little you did post is a mess. What math do you do on the call_log_id? That is the only reason to make it a numeric data type. Why do you use a vague, generic “id”? Let's call it foobar, so we will know it is not defined and cannot be a key. But means
that this is not a table!
CREATE TABLE Tests
(call_date DATE DEFAULT CURRENT_TIMESTAMP NOT NULL,
call_log_id CHAR(5) NOT NULL
CHECK(call_log_id LIKE '[0-9][0-9][0-9][0-9][0-9]'),
foobar INTEGER,
PRIMARY KEY (??????????) );
You also got the syntax for insertion wrong.
INSERT INTO Tests
VALUES
('2013-12-02', '52713', 48287),
('2013-12-02', '52713', 271),
('2014-05-01', '71817', NULL),
('2014-05-01', '71817', 62805),
('2014-02-28', '63995', NULL);
What is foobar? Why is it NULL? What about multiple NULLs?
Do you want to try again and give us valid DDL?
--CELKO-- Books in Celko Series for Morgan-Kaufmann Publishing: Analytics and OLAP in SQL / Data and Databases: Concepts in Practice Data / Measurements and Standards in SQL SQL for Smarties / SQL Programming Style / SQL Puzzles and Answers / Thinking
in Sets / Trees and Hierarchies in SQL

Similar Messages

Need help with RANK() on NULL data

Hi All
I am using Oracle 10g and running a query with RANK(), but it is not returning a desired output. Pleas HELP!!
I have a STATUS table that shows the history of order status.. I have a requirement to display the order and the last status date (max). If there is any NULL date for an order then show NULL.
STATUS
ORD_NO | STAT | DT
1 | Open |
1 | Pending |
2 | Open |
2 | Pending |
3 | Open |1/1/2009
3 | Pending |1/6/2009
3 | Close |
4 | Open |3/2/2009
4 | Close |3/4/2009
Result should be (max date for each ORD_NO otherwise NULL):
ORD_NO |DT
1 |
2 |
3 |
4 |3/4/2009
CREATE TABLE Status (ORD_NO NUMBER, STAT VARCHAR2(10), DT DATE);
INSERT INTO Status VALUES(1, 'Open', NULL);
INSERT INTO Status VALUES(1, 'Pending', NULL);
INSERT INTO Status VALUES(2, 'Open', NULL);
INSERT INTO Status VALUES(2, 'Pending',NULL);
INSERT INTO Status VALUES(3, 'Open', '1 JAN 2009');
INSERT INTO Status VALUES(3,'Pending', '6 JAN 2009');
INSERT INTO Status VALUES(3, 'Close', NULL);
INSERT INTO Status VALUES(4, 'Open', '2 MAR 2009');
INSERT INTO Status VALUES(4, 'Close', '4 MAR 2009');
COMMIT;
I tried using RANK function to rank all the orders by date. So used ORDER BY cluse on date in descending order thinking that the null dates would be on top and will be grouped together by each ORD_NO.
SELECT ORD_NO, DT, RANK() OVER (PARTITION BY ORD_NO ORDER BY DT DESC)
FROM Status;
...but the result was something..
ORD_NO |DT |RANKING
*1 | | 1*
*1 | | 1*
*2 | | 1*
*2 | | 1*3 | | 1
3 |1/6/2009 | 2
3 |1/1/2009 | 3
4 |3/4/2009 | 1
4 |3/2/2009 | 2
I am not sure why didn't the first two ORD_NOs didn't group together and why ranking of 1 was assigned to them. I was assuming something like:
ORD_NO |DT |RANKING
*1 | | 1*
*1 | | 2*
*2 | | 1*
*2 | | 1*
3 | | 1
3 |1/6/2009 | 2
3 |1/1/2009 | 3
4 |3/4/2009 | 1
4 |3/2/2009 | 2
Please guide me if I am missing something here?
Regards
Sri

Hi,
If i well understood, you don't need rank
SELECT   ord_no, MAX (dt)KEEP (DENSE_RANK LAST ORDER BY dt) dt
    FROM status
GROUP BY ord_no
SQL> select * from status;
    ORD_NO STAT       DT
         1 Open
         1 Pending
         2 Open
         2 Pending
         3 Open       2009-01-01
         3 Pending    2009-01-06
         3 Close
         4 Open       2009-03-02
         4 Close      2009-03-04
9 ligne(s) sélectionnée(s).
SQL> SELECT   ord_no, MAX (dt)KEEP (DENSE_RANK LAST ORDER BY dt) dt
2      FROM status
3 GROUP BY ord_no;
    ORD_NO DT
         1
         2
         3
         4 2009-03-04
SQL>

How do I run query per report page?

I have a report with two subreports which both need queries to run on the fly per supplier(per report page). All the other subreports on the page as well as the main report queries are loaded in advance before the report and includes all records needed for all suppliers. However, these two querys that I'm trying to have run per report though can only be loaded one at a time because of the way these queries are formated (it allows a bar graph to show a label for only the current supplier on the x-axis and all other suppliers will display a "" for x-axis label, effectively only showing one legend label per report). I already have the necessary parameter needed for the report(Pm-Supplier) so how can I run an oracle query on the fly per report page?
Eric
Query 1:
select vendor,supplierloc,commodity,ip from (select c.Vendor, q.supplierloc, c.commodity,SUM (q.indexpoints+q.lateindexpoints) AS ip
FROM qa_occ q, glovia_prod.c_vencom@GL7TEST c
where q.occdate BETWEEN TO_DATE ('4/1/2006', 'mm/dd/yyyy')
AND TO_DATE ('3/9/2007', 'mm/dd/yyyy')
and c.vendor=q.supplier
and (upper(trim(foundby)) not like '%MARKET%' or foundby is null)
and (upper(trim(foundby)) not like '%TRIAL%' or foundby is null)
and (upper(trim(rank)) not like '%MARKET%' or rank is null)
and (upper(trim(rank)) not like '%TRIAL%' or rank is null)
and nvl(void,'N') = 'N'
and q.supplier= @SUPPLIER
GROUP BY C.vendor,q.supplierloc,c.commodity) qa
union
select '' as vendor,supplierloc,commodity,ip from (select c.Vendor, q.supplierloc, c.commodity,SUM (q.indexpoints+q.lateindexpoints) AS ip
FROM qa_occ q, glovia_prod.c_vencom@GL7TEST c
where q.occdate BETWEEN TO_DATE (@BegDate, 'mm/dd/yyyy')
AND TO_DATE (@EndDate, 'mm/dd/yyyy')
and c.vendor=q.supplier
and (upper(trim(foundby)) not like '%MARKET%' or foundby is null)
and (upper(trim(foundby)) not like '%TRIAL%' or foundby is null)
and (upper(trim(rank)) not like '%MARKET%' or rank is null)
and (upper(trim(rank)) not like '%TRIAL%' or rank is null)
and nvl(void,'N') = 'N'
and q.supplier<> @SUPPLIER
GROUP BY C.vendor,q.supplierloc,c.commodity) qa
order by commodity,ip desc
Query 2:
select c.name as name,c.vendor,c.ven_loc, SUM(q.indexpoints+q.lateindexpoints) as ip
from qa_occ q,glovia_prod.ven_loc c
WHERE q.occdate BETWEEN TO_DATE (@BegDate, 'mm/dd/yyyy')
AND TO_DATE (@EndDate, 'mm/dd/yyyy')
and (upper(trim(q.foundby)) not like '%MARKET%' or q.foundby is null)
and (upper(trim(q.foundby)) not like '%TRIAL%' or q.foundby is null)
and (upper(trim(q.rank)) not like '%MARKET%' or q.rank is null)
and (upper(trim(q.rank)) not like '%TRIAL%' or q.rank is null)
and q.supplier is not null
and nvl(q.void,'N') = 'N'
and q.supplier=@SUPPLIER
and q.supplier=c.vendor
and q.supplierloc=c.ven_loc
GROUP BY c.name,c.vendor,c.ven_loc
union
select '' as name,c.vendor,c.ven_loc, SUM(q.indexpoints+q.lateindexpoints) as ip
from qa_occ q,glovia_prod.ven_loc c
WHERE q.occdate BETWEEN TO_DATE (@BegDate, 'mm/dd/yyyy')
AND TO_DATE (@EndDate, 'mm/dd/yyyy')
and (upper(trim(q.foundby)) not like '%MARKET%' or q.foundby is null)
and (upper(trim(q.foundby)) not like '%TRIAL%' or q.foundby is null)
and (upper(trim(q.rank)) not like '%MARKET%' or q.rank is null)
and (upper(trim(q.rank)) not like '%TRIAL%' or q.rank is null)
--and q.supplier is not null
and nvl(q.void,'N') = 'N'
and q.supplier=c.vendor
and q.supplierloc=c.ven_loc
and q.supplier <> @SUPPLIER
GROUP BY c.name,c.vendor,c.ven_loc
ORDER BY ip DESC

You can't, CR considers each subreport as a separate report and therefore it makes a new connection and runs the SQL for the subreport.

Rollup/Cube Operation in Oracle 10g

I'm using Oracle 10g as my database.
Suppose I have a table that has data:
ID SEMESTER SUBJECT MARKS
9 4 Maths 50
9 4 Science 45
9 4 English 42
10 5 Maths 56
10 5 History 23
Now the output should look like this
ID SEMESTER SUBJECT MARKS RANK
9 4 Maths 50
Science 45
English 42
Total 137 1
10 5 Maths 56
History 23
Total 79 2
Can anybody please help me out.
Thanx in advance

Select * from tmp_sp_marks
SEMESTER     SUBJECT     MARKS          STUDID
3          Maths          40          8
3          English     52          8
3          ujarati          40          8
4          Science     45          9
4          English     42          9
5          Maths          43          10
5          English     44          10
4          Maths          50          9
select case when rn = 1 then studid else null end studid,
case when rn = 1 then semester else null end semester,
subject,
marks,
case when subject = 'Total' then rank else null end rank
from
select studid, semester, subject, marks, max(rank) over(partition by studid, semester) rank,
row_number() over(partition by studid, semester order by marks) rn
from
select studid, semester, subject, marks, 0 rank from tmp_sp_marks a
union
select studid, semester, 'Total', sm, dense_rank() over(order by sm desc) rank
from (
select studid, semester, sum(marks) sm
from tmp_sp_marks
group by rollup(studid, semester) )
where studid is not null and semester is not null
order by max(rank) over(partition by studid, semester) desc , marks asc
OUTPUT:
STUDID     SEMESTER     SUBJECT     MARKS          RANK
8          3     Maths     40
               Gujarati     40
          English     52
          Total     132          2
9          4     English     42
          Science     45
          Maths     50
          Total     137          1
10          5     Maths     43
          English     44
          Total     87          3

Inconsistent datatypes: expected NUMBER got CHAR error

Hi,
I have the following table
create GLOBAL TEMPORARY TABLE br_total_rtn_data_tmp
code varchar(50)     NOT NULL,
name                         varchar(255),
cum_ytd_rtn_amt          varchar(255),
cum_one_mon_rtn_amt          varchar(255)     ,
cum_thr_mon_rtn_amt          varchar(255)     ,
cum_six_mon_rtn_amt          varchar(255),
cum_nine_mon_rtn_amt     varchar(255),
cum_one_yr_rtn_amt          varchar(255),
cum_thr_yr_rtn_amt          varchar(255),
cum_five_yr_rtn_amt          varchar(255),
cum_ten_yr_rtn_amt          varchar(255),
cum_lof_rtn_amt               varchar(255),
avg_anl_one_yr_rtn_amt     varchar(255),
avg_anl_thr_yr_rtn_amt     varchar(255),
avg_anl_five_yr_rtn_amt     varchar(255),
avg_anl_ten_yr_rtn_amt     varchar(255),
avg_anl_lof_rtn_amt          varchar(255),
cum_prev_1m_month_end     varchar(255),
cum_prev_2m_month_end     varchar(255)
)ON COMMIT PRESERVE ROWS;
I have a case statement
CASE
             WHEN code = 'MDN' THEN
                       max(case when p.m_date = v_prev2_yr_mon and p.period_type = '1M' then p.mdn /100 else null end)
             WHEN code = 'QRT' THEN
                  max(case when p.m_date = v_prev2_yr_mon and p.period_type = '1M' then p.quartile else null end)
             WHEN code = 'PCT' THEN
                  max(case when p.m_date = v_prev2_yr_mon and p.period_type = '1M' then p.pct_beaten / 100 else null end)
             WHEN code = 'RNK' THEN
                       case when (p.m_date = v_prev2_yr_mon and p.period_type = '1M' and p.rank is null and p.cnt is null)
                    THEN
                                   P.RANK
                    else
                                    p.rank||'/'||p.cnt
                    end
             ELSE NULL
             END CASE The output for code = RNK should be somewhat like 3/5 which is rank/count
but i get the error "Inconsistent datatypes: expected NUMBER got CHAR error" when i put p.rank||'/'||p.cnt
How can that be solved.
ORacle version is 10g.

Taken from the documentation of the CASE expression:
"For a simple CASE expression, the expr and all comparison_expr values must either have the same datatype (CHAR, VARCHAR2, NCHAR, or NVARCHAR2, NUMBER, BINARY_FLOAT, or BINARY_DOUBLE) or must all have a numeric datatype. If all expressions have a numeric datatype, then Oracle determines the argument with the highest numeric precedence, implicitly converts the remaining arguments to that datatype, and returns that datatype.
For both simple and searched CASE expressions, all of the return_exprs must either have the same datatype (CHAR, VARCHAR2, NCHAR, or NVARCHAR2, NUMBER, BINARY_FLOAT, or BINARY_DOUBLE) or must all have a numeric datatype. If all return expressions have a numeric datatype, then Oracle determines the argument with the highest numeric precedence, implicitly converts the remaining arguments to that datatype, and returns that datatype."
You need to use the same data type for all your expressions. If you want to return a string, then you need to convert the remaining numbers explicitly to strings. E.g. you could try something like this:
CASE
             WHEN code = 'MDN' THEN
                       to_char(max(case when p.m_date = v_prev2_yr_mon and p.period_type = '1M' then p.mdn /100 else null end), 'TM')
             WHEN code = 'QRT' THEN
                  to_char(max(case when p.m_date = v_prev2_yr_mon and p.period_type = '1M' then p.quartile else null end), 'TM')
             WHEN code = 'PCT' THEN
                  to_char(max(case when p.m_date = v_prev2_yr_mon and p.period_type = '1M' then p.pct_beaten / 100 else null end), 'TM')
             WHEN code = 'RNK' THEN
                       case when (p.m_date = v_prev2_yr_mon and p.period_type = '1M' and p.rank is null and p.cnt is null)
                    THEN
                                   to_char(P.RANK, 'TM')
                    else
                                    p.rank||'/'||p.cnt
                    end
             ELSE NULL
             END CASE I see another potential issue, you're mixing aggregate functions with non-aggregate expressions, this can only work if these non-aggregate expressions are part of the group by clause, but you haven't posted the complete statement so I can only guess.
Regards,
Randolf
Oracle related stuff blog:
http://oracle-randolf.blogspot.com/
SQLTools++ for Oracle:
http://www.sqltools-plusplus.org:7676/
http://sourceforge.net/projects/sqlt-pp/

To Update a table

HI all,
I have to move values from a column of data type varchar2 to a column of data type number.
How could I do that?
It throws an error
WHERE range_low = (SELECT status_short_message
ERROR at line 3:
ORA-01722: invalid number
UPDATE edr_rpt_by_ranges_output
     SET range_label = range_low
   WHERE range_low = (SELECT status_short_message
                        FROM status_message
                       WHERE edr_rpt_by_ranges_output.range_low = status_message.status_code);
                    Edited by: Indhu Ram on Dec 17, 2009 9:41 AM
Edited by: Indhu Ram on Dec 17, 2009 9:55 AM
Edited by: Indhu Ram on Dec 17, 2009 10:00 AM

edr_rpt_by_ranges_output table structure
RANK                           NOT NULL NUMBER(3)
ROW_TYPE                                 NOT NULL VARCHAR2 (30)
LANE_ID                                      NOT NULL NUMBER(3)
DIRECTION_ID                              NOT NULL NUMBER(3)
INTERVAL_START_DATE_TIME       NOT NULL DATE
INTERVAL_END_DATE_TIME         NOT NULL DATE
RANGE_LABEL                             VARCHAR2(20)
RANGE_LOW                      NOT NULL NUMBER(12)
RANGE_HIGH                     NOT NULL NUMBER(12)   desc status_message
Name Null Type
STATUS_CODE NOT NULL NUMBER(12)
STATUS_ABBREVIATION NOT NULL VARCHAR2(20)
STATUS_SHORT_MESSAGE NOT NULL VARCHAR2(10)
STATUS_MESSAGE NOT NULL VARCHAR2(50)
Edited by: Indhu Ram on Dec 17, 2009 9:50 AM

Regarding case statement

Hi guys,
i am getting a very strange problem. i have score column. basis of this column i am sorting the dataand after it i am giving rank to it.
but for one employee it is not giving rank. giving null.so please any suggestion.
i am applying case statement for it.

It can be because of case statement...Whats the purpose of case statement here ? plz elaborate..
Try removing case and check if you are getting same error

Calculate serial no according to groups

Name ID
Prime    12
Prime    12
Prime    12
Prime    12
Prime    12
Prime    16
Prime 18
Prime    18
Prime    18
I have 2 colums say Name and iD. I want result of sql as:
Name ID      Order
Prime    12    1
Prime    12    2
Prime    12    3
Prime2    12    1
Prime2    12   2
Prime    16    1
Prime 18    1
Prime    18    2
Prime    18    3
Means the count of product rows based on name.

>>Prime2
How you are getting Prime2, I am not seeing Prime2 in your input table
in-case its a typo then try this
declare @tab table (Name varchar(50),id int, [Rank] int NULL)
Insert into @tab (name,ID) Values
('Prime',12),
('Prime',12),
('Prime',12),
('Prime2',12),
('Prime2',12),
('Prime',16),
('Prime',18),
('Prime',18),
('Prime',18);
with cte
as(
Select name,id,rank, row_number()Over(partition by Name,Id Order by Id asc)RowNum From @tab
)update cte set [rank]= RowNum
select * from @tab
Satheesh
My Blog

Sql query slowness due to rank and columns with null values:


Sql query slowness due to rank and columns with null values:
I have the following table in database with around 10 millions records:
Declaration:
create table PropertyOwners (
[Key] int not null primary key,
PropertyKey int not null,
BoughtDate DateTime,
OwnerKey int null,
GroupKey int null
go
[Key] is primary key and combination of PropertyKey, BoughtDate, OwnerKey and GroupKey is unique.
With the following index:
CREATE NONCLUSTERED INDEX [IX_PropertyOwners] ON [dbo].[PropertyOwners]
[PropertyKey] ASC,
[BoughtDate] DESC,
[OwnerKey] DESC,
[GroupKey] DESC
go
Description of the case:
For single BoughtDate one property can belong to multiple owners or single group, for single record there can either be OwnerKey or GroupKey but not both so one of them will be null for each record. I am trying to retrieve the data from the table using
following query for the OwnerKey. If there are same property rows for owners and group at the same time than the rows having OwnerKey with be preferred, that is why I am using "OwnerKey desc" in Rank function.
declare @ownerKey int = 40000
select PropertyKey, BoughtDate, OwnerKey, GroupKey
from (
select PropertyKey, BoughtDate, OwnerKey, GroupKey,
RANK() over (partition by PropertyKey order by BoughtDate desc, OwnerKey desc, GroupKey desc) as [Rank]
from PropertyOwners
) as result
where result.[Rank]=1 and result.[OwnerKey]=@ownerKey
It is taking 2-3 seconds to get the records which is too slow, similar time it is taking as I try to get the records using the GroupKey. But when I tried to get the records for the PropertyKey with the same query, it is executing in 10 milliseconds.
May be the slowness is due to as OwnerKey/GroupKey in the table can be null and sql server in unable to index it. I have also tried to use the Indexed view to pre ranked them but I can't use it in my query as Rank function is not supported in indexed
view.
Please note this table is updated once a day and using Sql Server 2008 R2. Any help will be greatly appreciated.

create table #result (PropertyKey int not null, BoughtDate datetime, OwnerKey int null, GroupKey int null, [Rank] int not null)Create index idx ON #result(OwnerKey ,rnk)
insert into #result(PropertyKey, BoughtDate, OwnerKey, GroupKey, [Rank])
select PropertyKey, BoughtDate, OwnerKey, GroupKey,
RANK() over (partition by PropertyKey order by BoughtDate desc, OwnerKey desc, GroupKey desc) as [Rank]
from PropertyOwners
go
declare @ownerKey int = 1
select PropertyKey, BoughtDate, OwnerKey, GroupKey
from #result as result
where result.[Rank]=1
and result.[OwnerKey]=@ownerKey
go
Best Regards,Uri Dimant SQL Server MVP,
http://sqlblog.com/blogs/uri_dimant/
MS SQL optimization: MS SQL Development and Optimization
MS SQL Consulting:
Large scale of database and data cleansing
Remote DBA Services:
Improves MS SQL Database Performance
SQL Server Integration Services:
Business Intelligence

How to use "Rank" function in Oracle?

I need to display Top 15 records by using rank function.
Here is my query...I need to pull top 15 FAQ's using the below query.. How can I use RANK function to display the Top 15 FAQ"s in the list.
Select distinct SUb1.FAQ,Sub1.FAQ_Hits,GU.display_Name_FMLS as displayname,ev.ParentLinkrecordid,ev.userid from User GU
Join Event ev
ON LOWER (ev.userid) IN (LOWER (GU.lanid), LOWER (Gu.racfid))
Join (Select distinct sm.stem as FAQ,Sum(ev.Eventresults) as FAQ_Hits,ev.ParentLinkrecordid as Topic_ID from Event ev
Join SubjectMatter sm
ON (TO_CHAR (sm.smrecordid) = ev.eventdetail1) AND ev.eventdetail1 IS NOT NULL AND sm.smtype = 1
Where (Upper(ev.eventsubtype) in (Upper('FAQ'),Upper('OPENFAQ')))
AND TO_DATE (eventdatetime, 'yyyy-mm-dd hh24:mi:ss') >= TO_DATE ('20100601', 'yyyymmdd')
and TO_DATE (eventdatetime, 'yyyy-mm-dd hh24:mi:ss') <= TO_DATE ('20100831', 'yyyymmdd')
Group by sm.stem,ev.Parentlinkrecordid
order by FAQ )sub1
ON Sub1.Topic_ID = ev.ParentLinkrecordid)

A few bits that I noticed in the query ...
in (Upper('FAQ'),Upper('OPENFAQ'))1) Do you really a upper for a string which is already in upper case.
Select distinct sm.stem as FAQ,Sum(ev.Eventresults) as FAQ_Hits,ev.ParentLinkrecordid as Topic_ID2) Do you need a distinct when you are using a GROUP function viz. SUM ?
You rank query is as follows, I am not very good at the ANSI style JOIN so changed it slightly ... :-)
Also notice the usage rank function in the "sub1" query.
select distinct sub1.faq,
                sub1.faq_hits,
                gu.display_name_fmls as displayname,
                ev.parentlinkrecordid,
                ev.userid
from user gu, event ev,
(select rank() over (order by sum(ev.eventresults) desc) rnk,
          sum(ev.eventresults) as faq_hits,
          sm.stem as faq,
          ev.parentlinkrecordid as topic_id
     from event ev, subjectmatter sm
    where (to_char(sm.smrecordid) = ev.eventdetail1)
      and ev.eventdetail1 is not null
      and sm.smtype = 1
      AND upper(ev.eventsubtype) in ('FAQ', 'OPENFAQ')
      and to_date(eventdatetime, 'yyyy-mm-dd hh24:mi:ss') >= to_date('20100601', 'yyyymmdd')
      and to_date(eventdatetime, 'yyyy-mm-dd hh24:mi:ss') <= to_date('20100831', 'yyyymmdd')
    group by sm.stem, ev.parentlinkrecordid
    order by faq) sub1
where lower(ev.userid) in (lower(gu.lanid), lower(gu.racfid))
and sub1.topic_id = ev.parentlinkrecordid)
and sub1.rnk <= 15;Like mentioned above, some sample data would have helped.

Rank function taking too long

I am running the below query. This query runs fine. However, if I uncomment the
"rank() over(partition by CONCAT_DATE,VARIABLE_ID order by VARIABLE_VALUE) RANK" and
"B.rank=1" , the query takes a very long time to execute...takes about 6-7 minutes
instead of 20 seconds(when the rank part is commented out). Is there any other way to speed
this up as I needed the one with the lowest rank.
Thanks
SELECT
EXAMCODE,
STARTDATE,
REVDATE,
ENDDATE,
VARIATION ,
GROUPNAME,
PLAN,
STDPLAN,
CORPPLAN,
PRODUCT,
PES,
CONCAT_DATE,
NOTE_ID,
MAJ_HDG_ID,
MAJ_HDG_TXT,
MIN_HDG_ID,
MIN_HDG_TXT,
VARIABLE_ID,
VARIABLE_DESC,
PROVIDERCODE,
VARFORMAT,
NOTE_NAME,
MAJHEADNGNOTE,
MINHEADNGNOTE,
VARIABLENOTE,
VARIABLE_VALUE
FROM(
SELECT
EXAMCODE,
STARTDATE,
REVDATE,
ENDDATE,
VARIATION ,
GROUPNAME,
PLAN,
STDPLAN,
CORPPLAN,
PRODUCT,
PES,
CONCAT_DATE,
NOTE_ID,
MAJ_HDG_ID,
MAJ_HDG_TXT,
MIN_HDG_ID,
MIN_HDG_TXT,
VARIABLE_ID,
VARIABLE_DESC,
PROVIDERCODE,
VARFORMAT,
NOTE_NAME,
MAJHEADNGNOTE,
MINHEADNGNOTE,
VARIABLENOTE,
VARIABLE_VALUE --,
-- rank() over(partition by CONCAT_DATE,VARIABLE_ID order by VARIABLE_VALUE) RANK
FROM
SELECT
EXAM_DIM2.EXAM_CODE EXAMCODE,
to_char(START_DATE_DIM2.FULL_DATE,'MM/DD/YYYY') STARTDATE,
to_char(REV_DATE_DIM2.FULL_DATE,'MM/DD/YYYY') REVDATE,
to_char(END_DATE_DIM2.FULL_DATE,'MM/DD/YYYY') ENDDATE,
VARIATION_DIM2.VARIATION_ID VARIATION ,
EXAM_DIM2.GROUP_NAME GROUPNAME,
EXAM_DIM2.BENEFIT_PLAN_NAME PLAN,
EXAM_DIM2.STANDARD_PLAN_NAME STDPLAN,
EXAM_DIM2.CORPORATE_PLAN_NAME CORPPLAN,
EXAM_DIM2.PRODUCT_NAME PRODUCT,
STRUCTURE_DIM2.STRUCTURE_NAME PES,
EXAM_DIM2.EXAM_CODE || ' - ' || to_char(START_DATE_DIM2.FULL_DATE,'MM/DD/YYYY') || ' - ' || nvl(to_char(REV_DATE_DIM2.FULL_DATE,'MM/DD/YYYY'),'N/A')|| ' - ' || to_char(END_DATE_DIM2.FULL_DATE ,'MM/DD/YYYY') CONCAT_DATE,
NOTES_DIM2.NOTE_ID NOTE_ID,
DECODE (MAJOR_HEADING_DIM2.HEADING_ID ,null,HEADING_DIM2.HEADING_ID,MAJOR_HEADING_DIM2.HEADING_ID) MAJ_HDG_ID ,
DECODE (MAJOR_HEADING_DIM2.HEADING_ID ,null,HEADING_DIM2.HEADING_TEXT,MAJOR_HEADING_DIM2.HEADING_TEXT) MAJ_HDG_TXT ,
DECODE(HEADING_DIM2.PARENT_HEADING_ID,null,'',HEADING_DIM2.HEADING_ID) MIN_HDG_ID,
DECODE(HEADING_DIM2.PARENT_HEADING_ID,null,'',HEADING_DIM2.HEADING_TEXT) MIN_HDG_TXT,
VARIABLE_DIM2.VARIABLE_ID VARIABLE_ID,
VARIABLE_DIM2.VARIABLE_SHORT_DESCRIPTION VARIABLE_DESC,
VARIABLE_DIM2.PROVIDER_ARRANGEMENT_CODE PROVIDERCODE,
VARIABLE_DIM2.VARIABLE_FORMAT_CODE VARFORMAT,
NOTES_DIM2.NOTE_NAME NOTE_NAME,
'' as MAJHEADNGNOTE,
'' as MINHEADNGNOTE,
DBMS_LOB.SUBSTR(NOTES_DIM2.NOTE_TEXT,DBMS_LOB.GETLENGTH(NOTES_DIM2.NOTE_TEXT) ,1) VARIABLENOTE,
EXAM_INFO_FACT2.VARIABLE_VALUE VARIABLE_VALUE
FROM
MED_DM.DATE_DIM START_DATE_DIM2,
MED_DM.DATE_DIM END_DATE_DIM2,
MED_DM.DATE_DIM REV_DATE_DIM2,
MED_DM.EXAM_DIM EXAM_DIM2,
MED_DM.STRUCTURE_DIM STRUCTURE_DIM2,
MED_DM.NOTES_FACT NOTES_FACT2,
MED_DM.HEADING_DIM MAJOR_HEADING_DIM2,
MED_DM.HEADING_DIM HEADING_DIM2,
MED_DM.VARIABLE_DIM VARIABLE_DIM2,
MED_DM.NOTES_DIM NOTES_DIM2,
MED_DM.EXAM_INFO_FACT EXAM_INFO_FACT2,
MED_DM.VARIATION_DIM VARIATION_DIM2
WHERE
( EXAM_INFO_FACT2.EXAM_DIM_ID = EXAM_DIM2.EXAM_DIM_ID )
AND ( VARIATION_DIM2.VARIATION_DIM_ID (+)= EXAM_INFO_FACT2.VARIATION_DIM_ID )
AND ( EXAM_INFO_FACT2.STRUCTURE_DIM_ID = STRUCTURE_DIM2.STRUCTURE_DIM_ID)
AND ( EXAM_DIM2.END_DATE_DIM_ID=END_DATE_DIM2.DATE_DIM_ID )
AND ( EXAM_DIM2.REVISION_DATE_DIM_ID=REV_DATE_DIM2.DATE_DIM_ID )
AND ( EXAM_DIM2.START_DATE_DIM_ID=START_DATE_DIM2.DATE_DIM_ID )
AND ( HEADING_DIM2.HEADING_DIM_ID= EXAM_INFO_FACT2.HEADING_DIM_ID )
AND ( MAJOR_HEADING_DIM2.HEADING_ID(+)=HEADING_DIM2.PARENT_HEADING_ID )
AND ( EXAM_INFO_FACT2.VARIABLE_DIM_ID = VARIABLE_DIM2.VARIABLE_DIM_ID )
AND ( EXAM_INFO_FACT2.EXAM_DIM_ID = NOTES_FACT2.EXAM_DIM_ID (+) )
AND ( EXAM_INFO_FACT2.VARIABLE_DIM_ID = NOTES_FACT2.VARIABLE_DIM_ID (+))
AND ( NOTES_FACT2.NOTE_DIM_ID = NOTES_DIM2.NOTE_DIM_ID (+))
UNION ALL
SELECT
EXAM_DIM2.EXAM_CODE EXAMCODE,
to_char(START_DATE_DIM2.FULL_DATE,'MM/DD/YYYY') STARTDATE,
to_char(REV_DATE_DIM2.FULL_DATE,'MM/DD/YYYY') REVDATE,
to_char(END_DATE_DIM2.FULL_DATE,'MM/DD/YYYY') ENDDATE,
'' as VARIATION,
EXAM_DIM2.GROUP_NAME GROUPNAME,
EXAM_DIM2.BENEFIT_PLAN_NAME PLAN,
EXAM_DIM2.STANDARD_PLAN_NAME ,
EXAM_DIM2.CORPORATE_PLAN_NAME CORPPLAN,
EXAM_DIM2.PRODUCT_NAME PRODUCT,
'' as PES,
EXAM_DIM2.EXAM_CODE || ' - ' || to_char(START_DATE_DIM2.FULL_DATE,'MM/DD/YYYY') || ' - ' || nvl(to_char(REV_DATE_DIM2.FULL_DATE,'MM/DD/YYYY'),'N/A')|| ' - ' || to_char(END_DATE_DIM2.FULL_DATE ,'MM/DD/YYYY') CONCAT_DATE,
MED_DM.NOTES_DIM.NOTE_ID NOTE_ID,
DECODE (MAJOR_HEADING_DIM2.HEADING_ID ,null,HEADING_DIM2.HEADING_ID,MAJOR_HEADING_DIM2.HEADING_ID) MAJ_HDG_ID ,
DECODE (MAJOR_HEADING_DIM2.HEADING_ID ,null,HEADING_DIM2.HEADING_TEXT,MAJOR_HEADING_DIM2.HEADING_TEXT) MAJ_HDG_TXT ,
DECODE(HEADING_DIM2.PARENT_HEADING_ID,null,'',HEADING_DIM2.HEADING_ID) MIN_HDG_ID,
DECODE(HEADING_DIM2.PARENT_HEADING_ID,null,'',HEADING_DIM2.HEADING_TEXT) MIN_HDG_TXT,
VARIABLE_DIM2.VARIABLE_ID VARIABLE_ID,
VARIABLE_DIM2.VARIABLE_SHORT_DESCRIPTION VARIABLE_DESC,
VARIABLE_DIM2.PROVIDER_ARRANGEMENT_CODE PROVIDERCODE,
VARIABLE_DIM2.VARIABLE_FORMAT_CODE VARFORMAT,
MED_DM.NOTES_DIM.NOTE_NAME NOTE_NAME,
(CASE WHEN ((HEADING_DIM2.PARENT_HEADING_ID is null) AND (NOTES_FACT.VARIABLE_DIM_ID is null))
THEN DBMS_LOB.SUBSTR(MED_DM.NOTES_DIM.NOTE_TEXT,DBMS_LOB.GETLENGTH(MED_DM.NOTES_DIM.NOTE_TEXT),1)
ELSE
END) as MAJHEADNGNOTE,
-- DECODE(HEADING_DIM2.PARENT_HEADING_ID,null,DBMS_LOB.SUBSTR(MED_DM.NOTES_DIM.NOTE_TEXT,DBMS_LOB.GETLENGTH(MED_DM.NOTES_DIM.NOTE_TEXT),1),'') MAJHEADNGNOTE,
-- DECODE(HEADING_DIM2.PARENT_HEADING_ID,null,'',DBMS_LOB.SUBSTR(MED_DM.NOTES_DIM.NOTE_TEXT,DBMS_LOB.GETLENGTH(MED_DM.NOTES_DIM.NOTE_TEXT),1)) MINHEADNGNOTE,
(CASE WHEN ((HEADING_DIM2.PARENT_HEADING_ID is not null) AND (NOTES_FACT.VARIABLE_DIM_ID is null))
THEN DBMS_LOB.SUBSTR(MED_DM.NOTES_DIM.NOTE_TEXT,DBMS_LOB.GETLENGTH(MED_DM.NOTES_DIM.NOTE_TEXT),1)
ELSE
END) as MINHEADNGNOTE,
(CASE WHEN (NOTES_FACT.VARIABLE_DIM_ID is not null)
THEN DBMS_LOB.SUBSTR(MED_DM.NOTES_DIM.NOTE_TEXT,DBMS_LOB.GETLENGTH(MED_DM.NOTES_DIM.NOTE_TEXT),1)
ELSE
END) as VARIABLENOTE,
--DECODE(NOTES_FACT.VARIABLE_DIM_ID,null,DBMS_LOB.SUBSTR(MED_DM.NOTES_DIM.NOTE_TEXT,DBMS_LOB.GETLENGTH(MED_DM.NOTES_DIM.NOTE_TEXT),1),'') VARIABLENOTE,
'' as VARIABLE_VALUE
FROM
MED_DM.DATE_DIM START_DATE_DIM2,
MED_DM.DATE_DIM END_DATE_DIM2,
MED_DM.DATE_DIM REV_DATE_DIM2,
MED_DM.EXAM_DIM EXAM_DIM2,
MED_DM.NOTES_FACT,
MED_DM.HEADING_DIM MAJOR_HEADING_DIM2,
MED_DM.HEADING_DIM HEADING_DIM2,
MED_DM.VARIABLE_DIM VARIABLE_DIM2,
MED_DM.NOTES_DIM
WHERE
( MED_DM.NOTES_DIM.NOTE_DIM_ID=MED_DM.NOTES_FACT.NOTE_DIM_ID )
AND ( MED_DM.NOTES_FACT.VARIABLE_DIM_ID=VARIABLE_DIM2.VARIABLE_DIM_ID (+) )
AND ( MED_DM.NOTES_FACT.EXAM_DIM_ID=EXAM_DIM2.EXAM_DIM_ID )
AND ( HEADING_DIM2.HEADING_DIM_ID=MED_DM.NOTES_FACT.HEADING_DIM_ID )
AND ( EXAM_DIM2.END_DATE_DIM_ID=END_DATE_DIM2.DATE_DIM_ID )
AND ( EXAM_DIM2.REVISION_DATE_DIM_ID=REV_DATE_DIM2.DATE_DIM_ID )
AND ( EXAM_DIM2.START_DATE_DIM_ID=START_DATE_DIM2.DATE_DIM_ID )
AND ( MAJOR_HEADING_DIM2.HEADING_ID(+)=HEADING_DIM2.PARENT_HEADING_ID )
AND ( MED_DM.NOTES_FACT.HEADING_DIM_ID is not null)
)B
WHERE B.EXAMCODE ='G971' and B.STARTDATE = '10/01/2002'
--and B.RANK =1

There are probably lots of things you could do to improve the performance of this query. The first thing I would try is applying the filter criteria to each query in the "union all" rather than to the final set of results. The way it is working now is that the inner queries are ranking and returning all rows for all exam codes and all start dates (likely thousands if not millions of rows). Only after all rows have been ranked are the results filtered for the desired examcode.
Try something like:
select * from (
select ...rank() over(...) rnk from (
    select ... from ...
      where examcode='G971' and startdate=to_date('10/02/2002','MM/DD/YYYY') ...
    union all
    select ... from ...
      where examcode='G971' and startdate=to_date('10/02/2002','MM/DD/YYYY') ...
) where rnk = 1Also, indexes on examcode and/or startdate might help, if they don't already exist.

Most efficient way to strip nulls from a Double[] array?

I'm trying to optimize performance of some code that needs to accept a Double[] array (or a List<Double>) and fetch the median of the non-null values. I'm using org.apache.commons.math.stat.descriptive.rank.Median to fetch the median after converting to double[]. My question is how I can most efficiently make this conversion?
public class MathStatics {
     private static Median median = new Median();
     public static Double getMedian(Double[] doubles) {
          int numNonNull = 0;
          for (int i = 0; i < doubles.length; i++) {
               if (doubles[i] != null) numNonNull++;
          double[] ds = new double[numNonNull];
          for (int i = 0; i < doubles.length; i++) {
               if (doubles[i] != null) ds[i] = doubles;
               System.out.println(ds[i]);
          return median.evaluate(ds);
     public static void main(String[] args) {
          Double[] test = new Double[] {null,null,-1.1,2.2,5.8,null};
          System.out.println(MathStatics.getMedian(test));
I'm sure that the code I wrote above is clunky and amateurish, so I'd really appreciate some insight into how to make improvements. FWIW, the arrays will typically range in size from ~1-15,000 doubles.
Thanks!

There's no need to loop over the array twice
          int numNonNull = 0;
          double[] ds = new double[numNonNull];
          for (int i = 0; i < doubles.length; i++) {
               if (doubles[i] != null) {
numNonNull++;
ds[i] = doubles;
               System.out.println(ds[i]);
Except that ds will have length zero, so you'll get a OutOfBoundsException every time you use it.
As you're using Doubles rather than doubles, you can add the non-null values to an ArrayList and then convert it to an array at the end.

Implementing Rank / Partitioned Index

Has anyone implemented a partitioned RANK in Power Query? I understand I can Sort and then create an Index, but how can I define partitions for the index? Is there some way I can make it restart with each new partition value?
For instance, I want to rank company performance in each sport:
let
Source = Table.FromRecords(
{[Company = "Company A", Event = "Swimming", Points = 8],
[Company = "Company A", Event = "Running", Points = 5],
[Company = "Company B", Event = "Swimming", Points = 10],
[Company = "Company B", Event = "Running", Points = 2],
[Company = "Company C", Event = "Swimming", Points = 2],
[Company = "Company C", Event = "Running", Points = 4] })
in
#"Source"
Is there a way to do this, besides manually filtering to a single sport, doing the Sort + Index, repeating this step for each sport, and then appending these steps?

Since there is a built-in rank function in Power Pivot and you have more control over filtering etc., if one has the choice, then Power Pivot is the better option. On the other hand, Curt provides a fine solution if you need to do the equivalent in Power
Query.
It appears though, that you'd rather not have an index column and the lot. An alternative solution involves using a couple of portable custom functions.
ListRankEqual:
(inputValue as any, inputSeries as list, optional orderDescending as nullable logical) as number =>
let
    SortedSeries = if not orderDescending or orderDescending = null then
                    List.Sort(inputSeries,Order.Descending)
                 else
                    List.Sort(inputSeries),
    RankEqual = List.PositionOf(SortedSeries,inputValue)+1
in
    RankEqual
The function has two mandatory parameters and one optional parameter. In your example, inputValue would be the current value in the Points column, specified as [Points] when passed to the function. inputSeries would be the entire Points column, or in this
case, a subset of the Points column. In the body of the function, we sort inputSeries in
descending order to get the rank in ascending order and vice versa.
The second function I call GroupList. I tend to use this function when I want to group a column, but retain the original columns in a table. In your example, we want to rank Swimming and Running independently. Therefore, we need
to create a Swimming points list and a Running points list for the ListRankEqual function.
The code for GroupList:
(keyColumnList as list, keyColumnValue as any, associatedColumnList as list)=>
let
    FilteredKeyColumnPositions = List.PositionOf(keyColumnList, keyColumnValue, Occurrence.All),
    FilteredGroupList = List.Transform(FilteredKeyColumnPositions, each associatedColumnList{_})
in
   FilteredGroupList
keyColumnList is the list of values in the grouping column. In this case, it would be specified as
Source[Event]. keyColumnValue is the current value in the Event column, specified as [Event]. associatedColumnList is the column with the values you want to rank (Source[Points]).
GroupList first finds all the positions of the current value in the Event column. It then uses List.Transform to produce a corresponding Point list. The Swimming event will produce a list containing all of the points for Swimming, and the Running event
will produce a list containing all of the points for Running.
The complete custom formula would be:
ListRankEqual([Points],GroupList(Source[Event],[Event],Source[Points]))
You just toss the formula into a new custom column.

How to use RANK function ?

Hello everyone,
here is the query I run in sql developer using RANKFunction.
SELECT Empno, Ename, Job, Mgr, Hiredate, Sal
FROM
(SELECT Empno, Ename, Job, Mgr, Hiredate, Sal,
*RANK*() OVER
(ORDER BY SAL Desc NULLS LAST) AS Emp_Rank
FROM Emp
ORDER BY SAL Desc NULLS LAST)
WHERE Emp_Rank < 6;How I can use this query in my report in obiee or is there any replacement of RANK() function in obiee so that I can use that to get my same above result.
Thanks

Kuldip wrote:
Thanks, you are absolutely correct. However, By doing this I am getting my output as whic I was not expecting.
Students Marks Rank
student1 95 1
student2 95 1
student3 93 3
student4 93 3
student5 91 5
The output should be as instead
Students Marks Rank
student1 95 1
student2 95 1
student3 93 2
student4 93 2
student5 91 3
Can It be done like this ?
Thanks.
Edited by: Kuldip on Mar 15, 2012 11:51 PMHi Boss,
I think you copied the above scenario from this site..
http://oracle-bi.siebelunleashed.com/articles/rank-and-dense-rank-functionsobiee/
Then why asking how to do this? Are you testing us? Doesn't that site say how to achieve this?

One-to-many selfjoin removing records with the same ranking or with a substitute

Sorry for my bad choice of discussion title, feel free to suggest me a more pertinent one
I've rewritten post for clarity and following the FAQ.
DB Version
I'm using Oracle Enterprise 10g 10.2.0.1.0 64bit
Tables involved
CREATE TABLE wrhwr (
wr_id INTEGER PRIMARY KEY,
eq_id VARCHAR2(50) NULL,
date_completed DATE NULL,
status VARCHAR2(20) NOT NULL,
pmp_id VARCHAR2(20) NOT NULL,
description VARCHAR2(20) NULL);
Sample data
INSERT into wrhwr VALUES (1,'MI-EXT-0001',date'2013-07-16','Com','VER-EXC','Revisione')
INSERT into wrhwr VALUES (2,'MI-EXT-0001',date'2013-07-01','Com','VER-EXC','Verifica')
INSERT into wrhwr VALUES (3,'MI-EXT-0001',date'2013-06-15','Com','VER-EXC','Revisione')
INSERT into wrhwr VALUES (4,'MI-EXT-0001',date'2013-06-25','Com','VER-EXC','Verifica')
INSERT into wrhwr VALUES (5,'MI-EXT-0001',date'2013-04-14','Com','VER-EXC','Revisione')
INSERT into wrhwr VALUES (6,'MI-EXT-0001',date'2013-04-30','Com','VER-EXC','Verifica')
INSERT into wrhwr VALUES (7,'MI-EXT-0001',date'2013-03-14','Com','VER-EXC','Collaudo')
Query used
SELECT *
FROM (SELECT eq_id,
               date_completed,
               RANK ()
               OVER (PARTITION BY eq_id
                     ORDER BY date_completed DESC NULLS LAST)
                  rn
          FROM wrhwr
         WHERE     status != 'S'
               AND pmp_id LIKE 'VER-EX%'
               AND description LIKE '%Verifica%') table1,
       (SELECT eq_id,
               date_completed,
               RANK ()
               OVER (PARTITION BY eq_id
                     ORDER BY date_completed DESC NULLS LAST)
                  rn
          FROM wrhwr
         WHERE     status != 'S'
               AND pmp_id LIKE 'VER-EX%'
               AND description LIKE '%Revisione%') table2,
       (SELECT eq_id,
               date_completed,
               RANK ()
               OVER (PARTITION BY eq_id
                     ORDER BY date_completed DESC NULLS LAST)
                  rn
          FROM wrhwr
         WHERE     status != 'S'
               AND pmp_id LIKE 'VER-EX%'
               AND description LIKE '%Collaudo%') table3
WHERE     table1.eq_id = table3.eq_id
       AND table2.eq_id = table3.eq_id
       AND table1.eq_id = table2.eq_id
Purpose of the above query is to selfjoin wrhwr table 3 times in order to have for every row:
eq_id;
completition date of a work request of type Verifica for this eq_id (table1 alias);
completition date of a wr of type Revisione (table2 alias) for this eq_id;
completition date of a wr of type Collaudo (table3 alias) for this eq_id;
A distinct eq_id:
can have many work requests (wrhwr records) with different completition dates or without completition date (date_completed column NULL);
in a date range can have all the types of wrhwr ('Verifica', 'Revisione', 'Collaudo') or some of them (ex. Verifica, Revisione but not Collaudo, Collaudo but not Verifica and Revisione, etc.);
substrings in description shouldn't repeat;
(eq_id,date_completed) aren't unique but (eq_id,date_completed,description,pmp_id) should be unique;
Expected output
Using sample data above I expect this output:
eq_id,table1.date_completed,table2.date_completed,table3.date_completed
MI-EXT-001,2013-07-01,2013-07-16,2013-03-14 <--- for this eq_id table3 doesn't have 3 rows but only 1. I want to repeat the most ranked value of table3 for every result row
MI-EXT-001,2013-07-01,2013-06-15,2013-03-14 <-- I don't wanna this row because table1 and table2 have both 3 rows so the match must be in rank terms (1st,1st) (2nd,2nd) (3rd,3rd)
MI-EXT-001,2013-06-25,2013-06-15,2013-03-14 <-- 2nd rank of table1 joins 2nd rank of table2
MI-EXT-001,2013-04-30,2013-04-14,2013-03-14 <-- 1st rank table1, 1st rank table2, 1st rank table3
In vector style syntax, expected tuple output must be:
ix = i-th ranking of tableX
(i1, i2, i3) IF EXISTS an i-th ranking row in every table
ELSE
(i1, b, b)
where b is the first available lower ranking of table2 OR NULL if there isn't any row of lower ranking.
Any clues?
With the query I'm unable to remove "spurius" rows.
I'm thinking at a solution based on analytic functions like LAG() and LEAD(), using ROLLUP() or CUBE(), using nested query but I would find a solution elegant, easy, fast and easy to maintain.
Thanks

FrankKulash ha scritto:
About duplicate dates: I was most interested in what you wanted when 2 (or more) rows with the same eq_id and row type (e.g. 'Collaudo') had exactly the same completed_date.
In the new results, did you get the columns mixed up? It looks like the row with eq_id='MI-EXT-0002' has 'Collaudo' in the desciption, but the date appears in the verifica column of the output, not the collaudo column.
Why don't you want 'MI-EXT-0001' in the results? Is it realted to the non-unique date?
For all optimization questions, see the forum FAQ:https://forums.oracle.com/message/9362003
If you can explain what you need to do in the view (and post some sample data and output as examples) then someone might help you find a better way to do it.
It looks like there's a lot of repetition in the code. Whatever you're trying to do, I suspect there's a simpler, more efficient way to do it.
About Duplicate dates: query must show ONLY one date_completed and ignore duplicated. Those records are "bad data". You can't have 2 collaudos with the same date completed.
Collaudo stands for equipment check. A craftperson does an equipment check once a day and, with a mobile app, update the work request related to equipment and procedure of preventive maintenance, so is impossibile that complete more than one check (Collaudo) in a day, by design.
In the new results, it's my fault: during digitation I've swapped columns
With "I don't want 'MI-EXT-0001'" I mean: "I don't want to show AGAIN MI-EXT-0001. In the previous post was correct the output including MI-EXT-0001.
Regarding optimisation...
repetition of
LAST_VALUE (
                        MIN (CASE WHEN r_type = THEN column_name END) IGNORE NULLS)
                     OVER (PARTITION BY eq_id ORDER BY r_num) AS alias_column_name
is because I don't know another feasible way to have all columns needed of table wrhwr in main query, maintaining the correct order. So i get them in got_r_type and propagate them in all the subquery.
In main query I join eq table (which contains all the information about a specific equipment) with "correct" dates and columns of wrhwr table.
I filter eq table for the specific equipment standard (eq_std column).
efm_eq_tablet table and where clause
AND e.eq_id = e2.eq_id
          AND e2.is_active = 'S';
means: show only rows in eq table that have an equal row in efm_eq_tablet table AND are active (represented by 'S' value in is_active column).
About the tables v2, r2 and c2
          (SELECT doc_data, doc_data_rinnovo, eq_id
             FROM efm_doc_rep edr
            WHERE edr.csi_id = '1011503' AND edr.doc_validita_temp = 'LIM') v2,
          (SELECT doc_data, doc_data_rinnovo, eq_id
             FROM efm_doc_rep edr
            WHERE     eq_id = edr.eq_id
                  AND edr.csi_id = '1011504'
                  AND edr.doc_validita_temp = 'LIM') r2,
          (SELECT doc_data, doc_data_rinnovo, eq_id
             FROM efm_doc_rep edr
            WHERE     edr.csi_id IN ('1011505', '1011507')
                  AND edr.doc_validita_temp = 'LIM'
                  AND adempimento_ok = 'SI') c2,
Those tables contains "alternate" dates of completition to be used when there isn't any wrhwr row for an eq_id OR when all date_completed are NULL.
NVL() and NVL2() functions are used in main query in order to impletement this.
The CASE/WHEN blocks inside main query implements the behavior of selecting the correct date based of the above conditions.

Rank - no null

Similar Messages

Maybe you are looking for