A query containing a subquery and Group By all-in-one

Similar Questions

• Can someone tell me why I lost all my accurate color between iPhoto and my C8180 All-in-one?

  I was full of pictures from iPhoto printing yesterday and all of a sudden the color accuracy went in the toilet.

  Everything is green and stupid.  It's terrible.  The strange part... it allows to print very well everywhere else.

  This morning, I slipped a photo from iPhoto to the desktop and then print it from there.  Went well.

  I lost something between iPhoto and my C8180 All-in-one printer.  I've never encountered this problem

  front.  I tried all of the suggestions of the printer, but I'm sure that's not the problem.  The printer and

  iPhoto do communicate specifically.  Someone?

  Thanks for the 2 responses I received.  Unfortunately, neither contributed to my problem.  The problem

  was not with my version of OS X or the HP software that has been installed.  As I have explained, this set-up has been

  works perfectly for the last 3-4 years.  With the help of my long-time friend and Mac user many, many

  years, we determined that it was a corrupt iPhoto preferences file.  We have replaced the files preference and

  It works perfectly again.  The problem had nothing to do the printer or it's software.  I am back

  in the business.  Hope this might help someone else.

• Just upgraded to Windows 7 and my Dell All IN One Printer does not print. It is said that the pilot of the AIO 942 not found

  Just upgraded to Windows 7 and my Dell All IN One Printer does not print.  It is said that the pilot of the AIO 942 not found.

  Anyone where I can download this driver?

  See http://www.microsoft.com/windows/compatibility/Windows-7/en-us/Details.aspx?type=Hardware&p=Dell%20942%20Photo%20All-In-One%20Printer&v=Dell&uid=942STD&pf=0&pi=1&c=Printers%20%26%20Scanners&sc=Multifunction&os=32-bit.

  There are only drivers for Vista at http://www.dell.com/support/drivers/us/en/04/DriverDetails/DriverFileFormats?c=us&s=bsd&cs=04&l=en&DriverId=R151559.

  They are supposed to work if installed in compatibility mode. http://en.community.dell.com/support-forums/peripherals/f/3528/t/19319527.aspx.

• [8i] need help on query with a subquery/inline using UNION ALL view

  OK, first of all let's start with some background info:
  (1) I work in 8i
  (2) some samples for my problem (please excuse the additional columns of data not used to this problem; I already had the sample laying around tables):
  

  CREATE TABLE     vbom
  (
       part_nbr     char(25)     
  ,     bom_doc_nbr     char(25)     
  ,     bill_level     number          
  ,     comp_part_nbr     char(25)     
  ,     qty_per          number
  );
  --technically has primary and foreign keys, but whatever...
  --part_nbr is your top level (0) parent
  --comp_part_nbr is the specific child
  --bom_doc_nbr is the parent of the child (comp_part_nbr) and may be either part_nbr or a child of part_nbr that is also a parent
  
  INSERT INTO vbom
  VALUES ('SAMPLE-1','SAMPLE-1', 1,'SAMPLE-2',1);
  
  CREATE TABLE     rqmt
  (
       comp_part_nbr     char(25)     
  ,     prnt_part_nbr     char(25)
  ,     ord_nbr          char(10)     
  ,     sub_ord_nbr     char(3)
  ,     qty_reqd     number
  ,     qty_issued     number
  ,     date_reqd     date
  ,     rqmt_stat     char(2)
  ,     rqmt_type     char(2)
  );
  
  INSERT INTO rqmt
  VALUES ('SAMPLE-1','','S0000TEST1',001,30,0,TO_DATE('06/01/2010','mm/dd/yyyy'),'AL','ID');
  INSERT INTO rqmt
  VALUES ('SAMPLE-2','SAMPLE-1','0000054963',001,15,10,TO_DATE('04/01/2010','mm/dd/yyyy'),'CL','DD');
  INSERT INTO rqmt
  VALUES ('SAMPLE-2','SAMPLE-1','0000032562',001,5,5,TO_DATE('04/15/2010','mm/dd/yyyy'),'IS','DD');
  INSERT INTO rqmt
  VALUES ('SAMPLE-2','SAMPLE-1','0000022341',001,5,4,TO_DATE('04/20/2010','mm/dd/yyyy'),'SH','DD');
  INSERT INTO rqmt
  VALUES ('SAMPLE-2','SAMPLE-1','0000043469',001,10,0,TO_DATE('04/30/2010','mm/dd/yyyy'),'AL','DD');
  INSERT INTO rqmt
  VALUES ('SAMPLE-2','SAMPLE-1','0000071235',001,10,0,TO_DATE('05/01/2010','mm/dd/yyyy'),'OP','DD');
  INSERT INTO rqmt
  VALUES ('SAMPLE-2','SAMPLE-1','0000061224',001,5,0,TO_DATE('05/15/2010','mm/dd/yyyy'),'FP','DD');
  INSERT INTO rqmt
  VALUES ('SAMPLE-2','SAMPLE-1','0000033294',001,5,0,TO_DATE('05/25/2010','mm/dd/yyyy'),'PL','DD');

  So, my first question was that I needed to find in the table RQMT who corresponded with everything on my VBOM have vbom.part_nbr table ' SAMPLE-1'. '. (Please note, in reality the VBOM table has thousands of rows of data, rather than just the 1 I have planned, some of them sharing the same vbom.part_nbr, others not). In addition to finding all RQMT data corresponding to the vbom.comp_part_nbr to vbom.part_nbr (where vbom.comp_part_nbr = rqmt.comp_part_nbr), I also need to find RQMT data for vbom.part_nbr itself (then, records where vbom.part_nbr = rqmt.comp_part_nbr).
  
  To resolve this problem, I started with the following query:

  SELECT     v.bill_level          AS bill_lvl
  ,     v.comp_part_nbr          AS component
  ,     v.bom_doc_nbr          AS parent
  FROM     VBOM v
  WHERE     v.part_nbr     ='SAMPLE-1'
  UNION ALL
  SELECT     0               AS bill_lvl
  ,     'SAMPLE-1'          AS component
  ,     NULL               AS parent
  FROM DUAL

  It allows me to add a line to the top of the page parent (vbom.part_nbr), which does not exist in vbom.
  My first question is, is UNION ALL of the best way to do it, or could I do a kind of instruction based on the line number box (or something) that would be better? (in this application, or in the final query)
  
  Then, I used the above query as a point of view online:

  SELECT     a.bill_lvl
  ,     a.component
  ,     a.parent
  ,     r.ord_nbr
  ,     r.sub_ord_nbr
  ,     r.qty_reqd
  FROM     (
       SELECT     v.bill_level          AS bill_lvl
       ,     v.comp_part_nbr          AS component
       ,     v.bom_doc_nbr          AS parent
       FROM     VBOM v
       WHERE     v.part_nbr     ='SAMPLE-1'
       UNION ALL
       SELECT     0               AS bill_lvl
       ,     'SAMPLE-1'          AS component
       ,     NULL               AS parent
       FROM DUAL
       ) a
  ,     RQMT r
  WHERE     a.component     = r.comp_part_nbr

  The problem here is that I have the same results (7 files) with the above query as I do with:

  SELECT     v.bill_level
  ,     v.comp_part_nbr
  ,     v.bom_doc_nbr
  ,     r.ord_nbr
  ,     r.sub_ord_nbr
  ,     r.qty_reqd
  FROM     VBOM v
  ,     RQMT r
  WHERE     v.comp_part_nbr     = r.comp_part_nbr
  AND     v.part_nbr     = 'SAMPLE-1'

  .. .it does not include RQMT recording for rqmt.comp_part_nbr = "SAMPLE-1'.
  
  To understand what was going on, I ran:

  SELECT     a.bill_lvl
  ,     a.component
  ,     a.parent
  FROM     (
       SELECT     v.bill_level          AS bill_lvl
       ,     v.comp_part_nbr          AS component
       ,     v.bom_doc_nbr          AS parent
       FROM     VBOM v
       WHERE     v.part_nbr     ='SAMPLE-1'
       UNION ALL
       SELECT     0               AS bill_lvl
       ,     'SAMPLE-1'          AS component
       ,     NULL               AS parent
       FROM DUAL
       ) a

  .. .and I get exactly this that I wait, 1 card for the SAMPLE-2 component and 1 plug for SAMPLE-1 (i.e. the recording I generated with the UNION ALL).
  
  Then I ran:

  SELECT     r.comp_part_nbr
  ,     r.ord_nbr
  ,     r.sub_ord_nbr
  ,     r.qty_reqd
  FROM     RQMT r
  WHERE     r.comp_part_nbr like 'SAMPLE-%'

  .. .and I get all 8 records, whose SAMPLE-1.
  
  So, it seems that the problem occurred when I join my opinion RQMT one online, although both separately seem to work very well.
  
  A point to note: all parts of parent and child may not be appointed in the same way as in my example. SAMPLE-1 could have children of the SAMPLE-2, abc123, 20735, for example.
  
  Any help on this would be appreciated, thanks!

  Hello

  Concerns of problem by comparing 8 literal characters ' SAMPLE-1' for the 25-character CHAR column that contains "SAMPLE 1".»
  17 spaces including the literal or use LPAD or CAST to 25 characters.
  For example:

  SELECT     a.bill_lvl
  ,     a.component
  ,     a.parent
  ,     r.ord_nbr
  ,     r.sub_ord_nbr
  ,     r.qty_reqd
  FROM     (
       SELECT     v.bill_level          AS bill_lvl
       ,     v.comp_part_nbr          AS component
       ,     v.bom_doc_nbr          AS parent
       FROM     VBOM v
       WHERE     v.part_nbr     ='SAMPLE-1'
       UNION ALL
       SELECT     0               AS bill_lvl
       ,     CAST ('SAMPLE-1' AS CHAR (25))               -- CAST added here
                                    AS component
       ,     NULL               AS parent
       FROM DUAL
       ) a
  ,     RQMT r
  WHERE     a.component     = r.comp_part_nbr
  ;
  

  Is there a reason why you use CHAR instead of VARCHAR2? (One reason other than "I can't change it now.", otherwise said.)

• Portable printing and Scaning between an HP Pavilion dv6 PC and F310 HP all in one printer / Scanner

  I'm trying to interface a new HP laptop that has a 64-bit (HP Pavilion dv6-701) operating system and running the operating system Windows 7 on a HP All in one series F310 printer / Scanner

  HP staff helped get printing to work via a 6980 HP printer Driver software download, try to download the software 310 driver didn't work so we tried this driver 6980 credits and print function works now

  However I can't scan a document on the laptop of the 310, when I attmpt one scan, I receive an error message

  Given that the printer uses a generic driver and is not a software that is supported on Windows 7, you can scan only programs such as paint, Windows Fax and Scan and image viewer.

• Impossible to copy and scan on all-in-one printer, HP Deskjet F4180.

  I can't make a copy and scan on the surface of the glass of my printer HP deskjet F4180 all-in-one. The printer says: error in the copy menu and seeing red flashes on color and black. However, I am able to print from my laptop and a pop-up message would be in my laptop showing the ink cartridges (black and color) are placed incorrectly. Check the ink cartridges, both are in their right place and broken up. Need help... Please.

  I found this document which decodes what mean flashing lights.  There are a lot of different codes described, but I found a heading "question: one or two of the lights blink check print cartridge" which seems to be what's going on your printer.  Basically it's saying that the cartridges are out of alignment or defective.  If the document is not help clarify, let me know and I'll see what I can find.

• Windows 7 and scanning. All-in-one printer, HP Photosmart C5180

  I had some experience in the past with an upgrade of Windows 2000, XP and a printer/Scanner, I remember more that now, but I managed to get an upgrade for the printer but the scanner never worked despite the many phone calls from HP again. I have a HP Photosmart C5180 all-in-One printer and I'm about to upgrade my XP computer to Windows 7. Y at - it an upgrade that will manage the printer and the Scanner?

  Thank you. New PC comes next week. I will be presenting on the question of whether it works or not

• How the program REDOWNLOAD SHREDDER64. MSI? Windows 7 and HP 4620 all-in-One Printer

  I have to clean programs unused and deleted the Shredder4.msi, which I just found out that I must have to operate properly the functions of my

  4620 all-in-One Priner/Fax/Scanner HP.

  I looked and looked and can't find the program to reinstall.

  I would appreciate any direction that could help me in this company?

  Kind regards

  Judy

  * E-mail address is removed from the privacy * or

  E-mail address is removed from the privacy *.

  I recommend that you check with HP.

  You still have the CD that came with the printer?

  It is suitable for this topic.

  Edit: You have an ACER computer?

  They have a preinstalled by Egistec shredder program.

• find emails in email group, by sending one, update contact and map Data - all in one

  Hello

  I wanted to know if I'll be able to do the following:

  1. pull the names of all emails in a given e-mail group

  2 extract contact data

  3. send an email from (1)

  4 put to update the data of contact and one or two were data cards

  All before I would need to do a program step, with custom cloud connector / API.

  Possible? Any guidance welcome.

  Thanks for your help,

  Adam

  Hi Adam,.

  (1) you can search by name, or list all emails and filter them.  You are not able to pass a name or a group id and retrieve emails within the group.

  (2) you can use the standard or DTS WSDL to extract data contact (E9 and E10).

  With a (generator program with external services) cloud connector you first list of contacts in the status stage and pull their contact ID. You can then use the Standard Service extract to extract contact data.

  (3) you can send an email to 1 contact both using electronic mail (E10 and E9) using the id contact comes from the stage of the program or send to an existing distribution list (E9 only)

  (4) you can do so via the standard WSDL (E9 and E10) or DTS WSDL (E9 only)

  Cheers, Aaron.

• Vista and Dell A920 all-in-one

  I've asked this question before and received the response on the withdrawal of the old driver and install the driver updated from the Web site.  I now have the problem of no communication between the computer and the printer/scanner.  I've done all the troubleshooting section for this problem of the printer page and looked in Device Manager stating that the printer/scanner is connected and ready for use according to the diagnostic test... firewall is configured to allow the printer to go through it, etc...   Any suggestions other than that?    Thank you

  chrysiis-chrysiisart,

  Can you please run the search for system information REMOVE YOUR PRODUCT ID and post the rest of feedback here. If the information is in another language, please do NOT translate it. It is important information that I need.

  You have a Dell A920 printer and what program anti-virus and firewall are you using?

  Rick

• And clearing memory all-in-one HP Deskjet 5440 and Officejet 7310xi

  How to erase the memory of the printer on a HP Deskjet 5440 and an Officejet 7310xi before recycling?

  dgilchri wrote: How do I clear the memory of the printer on a HP Deskjet 5440 and an Officejet 7310xi before recycling?

  Hi dgilchri, when you unplug the power cord on the printers, all memory stored in the printers are refreshed (disabled).

• updated to correct driver for windows 7 and photosmart c6280 all-in-one

  I bought the new computer with windows 7.  Tried to download recommended driver [a/o 02 Netwired_Full_Win_WW_130_140.exe.], but got the message that it is not commonly downloaded and it may harm your computer is not sure as what to do now?  Bought printer Jan 2008

  Thanks for the info.  I found myself just try the download again and it went through smoothly and works!  I used the HP website for support.

• Sorting and grouping - two months in this query

  Hi all
  
  Thanks a lot for JAC
  
  I am doing a project for the construction company, I faced this problem by grouping points according to the relationship between these points the
  Relationship is from 1 to 100. If the point between what means there is relationship between these points has come.
  
  resolve this question already, but the results does not correct when the table contains more data.
  
  SQL - sorting and grouping.
  Jac and thanks a lot to him.
  
  This example for more details
  
  for example, I have these points

  id   location         percentage   comments 
  1     loc 1,2          20%                that mean point  1 and 2 close to each other by 20%
  2     loc 1,3          40%              that mean point 1 and 3 close to each other byy 40%
  3     Loc 8,6          25%               that mean point 8 and 6 close to each other by 25% 
  4     Loc  6,10        20%
  5     LOC 11,10        10 %
  6     LOC 15,14         0%

  In addition, we can see the relationship between these points as follows
  -points 1,2,3 in a group why because close 1.2 and 1.3 a relationship which means 1.3 also hid the relationship.
  -Points 6,8,10,11 in the second group there are the relationships between them.
  - but no relationship between 1, 2 or 3 with any point of 6,8,9,10,11
  -as well as no relationship between 15, 14, which means 14 in the third group and 15 in the fourth group.
  
  
  Whati need?
  
  to group the points that growing a relationship according to the percentage value
  
  
  The most important part is to group the points. For EXAMPLE, the query below, the gropuing is not correct.
  
  I have the following table with data

  drop table temp_value; 
  create table temp_value(id number(10),location varchar2(20), percentage number(9)); 
   
   
  insert into temp_value values  (1,'LOC 1,2',10); 
  insert into  temp_value values (2,'LOC 1,3',0); 
  insert into  temp_value values (3,'LOC 1,4',0); 
  insert into  temp_value values (4,'LOC 1,5',0); 
  insert into  temp_value values (5,'LOC 1,6',0); 
  insert into  temp_value values (6,'LOC 2,3',0); 
  insert into  temp_value  values(7,'LOC 2,4',0); 
  insert into  temp_value values (8,'LOC 2,5',30); 
  insert into  temp_value values (9,'LOC 2,6',0); 
  insert into  temp_value values (10,'LOC 3,4',0); 
  insert into  temp_value values (11,'LOC 3,5',0); 
  insert into  temp_value values (12,'LOC 4,5',40); 
  insert into  temp_value values (13,'LOC 4,6',0); 
  insert into  temp_value values (14,'LOC 6,7',40);
  insert into  temp_value values (15,'LOC 7,2',0);
  insert into  temp_value values (16,'LOC 8,2',60);
  insert into  temp_value values (17,'LOC 8,3',0);
  insert into  temp_value values (18,'LOC 3,1',0);
  insert into  temp_value values (19,'LOC 9,6',30);
  insert into  temp_value values (20,'LOC 11,2',0);
  insert into  temp_value values (22,'LOC 12,3',10);
  insert into  temp_value values (23,'LOC 19,3',0);
  insert into  temp_value values (24,'LOC 17,3',0);
  insert into  temp_value values (24,'LOC 20,3',0);

  When I used this query, the results is not correct
  
  

   with t as
      (select percentage,loc1,loc2,sum(case when percentage = 0 then 1
                         when loc1 in (l1,l2) then 0
                     when loc2 in (l1,l2) then 0
                     when l1 is null and l2 is null then 0
                     else 1
                end) over(order by rn) sm
      from (     select id,location,percentage,
                         regexp_substr(location,'\d+',1,1) LOC1,
                        regexp_substr(location,'\d+',1,2)  LOC2,
                       lag(regexp_substr(location,'\d+',1,1))
                        over(order by percentage desc) l1,
                        lag(regexp_substr(location,'\d+',1,2))
                        over(order by percentage desc) l2,
                row_number() over(order by percentage desc) rn
        from temp_value
        order by percentage desc
          )
    )
     select loc,min(sm)+1 grp
       from(
         select loc,rownum rn,sm
         from(
         select percentage,decode(rn,1,loc1,loc2) loc,sm
         from t a,
              (select 1 rn from dual union all
               select 2 from dual ) b
         order by percentage desc,decode(rn,1,loc1,loc2) asc
        )
     )
      group by loc
     order by min(sm),min(rn);

  results
  
  
  SQL > /.
  

  LOC                         GRP
  -------------------- ----------
  2                             1
  8                             1
  6                             2
  7                             2
  4                             3
  5                             3
  9                             4
  1                             5
  12                            6
  3                             6
  11                           13
  
  LOC                         GRP
  -------------------- ----------
  19                           14
  17                           15
  20                           22
  
  14 rows selected.

  SQL >
  
  
  but the just is

  Location        group No
  2                  1
  8                  1
  4                  1
  5                  1
  1                  1
  6                  2
  7                  2
  9                  2
  12                 3
  3                  3
  19                 4
  17                 5
  20                 6

  Thanks in advance.
  
  Published by: Isabelle on November 30, 2012 03:07

  OK, I thought an it once again and found out how to include with any such percentage points.
  In your example expected output you missed the 11 that's why we got 7 groups now.
  The order of the Group 2 and 3 is ambiguous, because the highest percentage of these groups is the same.

  with connects as (
  select distinct
   loc1
  ,loc2
  ,connect_by_root(loc1) grp
  ,percentage per
  from
  temp_value
  --start with
  --percentage != 0
  connect by nocycle
  (prior loc2 = loc1
  or
  prior loc1 = loc2
  or
  prior loc1 = loc1
  or
  prior loc2 = loc2)
  and
  percentage != 0
  and
  prior percentage != 0
  )
  
  select
   loc
  ,dense_rank() over (order by decode(per,0,1,0), grp) grp
  from (
  select
   loc
  ,max(grp) keep (dense_rank first order by per desc, grp) grp
  ,max(per) keep (dense_rank last order by per nulls first) per
  from (
  select
   loc1 loc
  ,grp
  ,per
  from connects
  union
  select
   loc2
  ,grp
  ,per
  from connects
  )
  group by
  loc )
  order by 2,per desc,1
  
  LOC     GRP
  2     1
  8     1
  4     1
  5     1
  1     1
  12     2
  3     2
  6     3
  7     3
  9     3
  11     4
  17     5
  19     6
  20     7
  

  Think we are done now ;-)
  Edited by: chris227 at 30.11.2012 16:46

  Edited by: chris227 at 30.11.2012 17:12
  order corrected

  Edited by: chris227 at 30.11.2012 17:15
  simplification, no need to rank in linking subquery

  Edited by: chris227 at 30.11.2012 17:26

• Rewrite the query with joins, and group by

  Hello
  
  It's an interview question.
  
  Table names: bookshelf_checkout
  virtual library
  
  And the join condition between these two tables is title
  
  We need to rewrite under request without using the join condition and group by clause?
  
  

  SELECT b.title,max(bc.returned_date - bc.checkout_date) "Most Days Out"
             FROM bookshelf_checkout bc,bookshelf b
             WHERE bc.title(+)=b.title
             GROUP BY b.title;

  When I was in College, I read most of SELECT statements can be replaced by operations base SQL (DEFINE the OPERATORS). Now, I am rewriting the query with SET operators, but not able to get the exact result.
  
  Kindly help me on this.
  
  Thank you
  Suri

  Something like that?

    1  WITH books AS (
    2  SELECT 'title 1' title FROM dual UNION ALL
    3  SELECT 'title 2' FROM dual UNION ALL
    4  SELECT 'title 3' FROM dual ),
    5  bookshelf AS (
    6  SELECT 'title 1' title, DATE '2012-05-01' checkout_date, DATE '2012-05-15' returned_date FROM dual UNION ALL
    7  SELECT 'title 1' title, DATE '2012-05-16' checkout_date, DATE '2012-05-20' returned_date FROM dual UNION ALL
    8  SELECT 'title 2' title, DATE '2012-04-01' checkout_date, DATE '2012-05-15' returned_date FROM dual )
    9  SELECT bs.title, MAX(bs.returned_date - bs.checkout_date) OVER (PARTITION BY title) FROM bookshelf bs
   10  UNION
   11  (SELECT b.title, NULL FROM books b
   12  MINUS
   13* SELECT bs.title, NULL FROM bookshelf bs)
  SQL> /
  
  TITLE   MAX(BS.RETURNED_DATE-BS.CHECKOUT_DATE)OVER(PARTITIONBYTITLE)
  ------- ------------------------------------------------------------
  title 1                                                           14
  title 2                                                           44
  title 3
  

  Lukasz

• Cases and Group by

  I tried to understand what is the problem with a large sql query that does not change what I do.
  Have managed to isolate the part which is a failure, but I get the error message:
  
  ORA-00979: not a GROUP BY expression
  
  is not very useful.
  
  I suspect there is something fundamentally wrong in what I'm trying to do.
  
  I had to anonimise the names of the tables and fields that the data that I use is quite sensitive, but it should not forget the problem.
  
  Here is the query:
  

  select
  case when a='100' and cost in (select cost from lookup) then '100' else b end as main,
  count(*)
  from 
  data_table
  group by 
  case when a='100' and cost in (select cost from lookup) then '100' else b end

  DATA_TABLE has (among others) of fields:
  a, b, and the cost
  
  search contains the field:
  cost
  
  All fields are VARCHAR2 (255)
  
  
  If I remove the count (*) and any group of the query statement runs it is to say:

  select
  case when a='100' and cost in (select cost from lookup) then '100' else b end as main
  from 
  data_table

  This shows that the stated case is valid - then why I can not combine the count and group of?
  If this is a reason of syntax is there another solution that I can possibly use subqueries to work around the problem - I prefer to avoid this and can't really make the case in a decoding without him make mess.
  
  Oracle version:
  Oracle Database 10g Enterprise Edition Release 10.2.0.3.
  
  TIA
  
  Published by: user8378443 on May 11, 2011 10:26

  Try like this,

  SELECT MAIN, COUNT (*)
    FROM (SELECT CASE
                    WHEN A = '100'
                     AND COST IN (SELECT COST FROM lookup)
                    THEN '100' ELSE b
                 END
                    AS MAIN
            FROM data_table)
  GROUP BY MAIN
  

  G.