asked on

SQL: Query Question Again, calculate match with age and total

I have following table :

Table 1: PROFAIL

PROFAIL_ID | NAME | GENDER | BIRTH_DATE         | PROFAIL_TYPE | RACE    | NATIONALITY
--------------------------------------------------------------------------------------
1	   | Andy | M	   |2009-03-12 00:00:00 | A	       | Chinese | Local
2	   | Owen | M	   |2005-05-22 00:00:00 | A 	       | Others  | Local
3	   | Abby | F	   |2004-02-15 00:00:00 | A	       | Malay   | Local
4	   | Mimi | F	   |2002-03-02 00:00:00 | A	       | Indian  | Local

Open in new window

Table 2 : MATCH

MATCH_ID | PROFAIL_ID | MATCH | RESULT
--------------------------------------------
1	|  2	     |  A    | Good
2       |  4	     |  B    | Bad
3       |  3	     |  B    | Good 
4	|  2	     |  A    | Bad

Open in new window

SQL Dump as follow:

CREATE TABLE "MATCH" (
"MATCH_ID" VARCHAR2(16 BYTE) NULL ,
"PROFAIL_ID" VARCHAR2(16 BYTE) NULL ,
"MATCH" VARCHAR2(16 BYTE) NULL ,
"RESULT" VARCHAR2(16 BYTE) NULL 
)
LOGGING
NOCOMPRESS
NOCACHE

;

-- ----------------------------
-- Records of MATCH
-- ----------------------------
INSERT INTO "MATCH" VALUES ('1', '2', 'A', 'Good');
INSERT INTO "MATCH" VALUES ('2', '4', 'B', 'Bad ');
INSERT INTO "MATCH" VALUES ('3', '3', 'B', 'Good');
INSERT INTO "MATCH" VALUES ('4', '2', 'A', 'Bad');

-- ----------------------------
-- Table structure for "PROFAIL"
-- ----------------------------
DROP TABLE "PROFAIL";
CREATE TABLE "PROFAIL" (
"PROFAIL_ID" VARCHAR2(16 BYTE) NOT NULL ,
"NAME" VARCHAR2(200 BYTE) NULL ,
"NATIONALITY" VARCHAR2(200 BYTE) NULL ,
"GENDER" VARCHAR2(16 BYTE) NULL ,
"BIRTH_DATE" DATE NULL ,
"PROFAIL_TYPE" VARCHAR2(16 BYTE) NULL ,
"RACE" VARCHAR2(16 BYTE) NULL 
)
LOGGING
NOCOMPRESS
NOCACHE

;

-- ----------------------------
-- Records of PROFAIL
-- ----------------------------
INSERT INTO "PROFAIL" VALUES ('1', 'Andy', 'Local', 'M', TO_DATE('2009-03-12 00:00:00', 'YYYY-MM-DD HH24:MI:SS'), 'A', 'Chinese');
INSERT INTO "PROFAIL" VALUES ('2', 'Owen', 'Local', 'M', TO_DATE('2005-05-22 00:00:00', 'YYYY-MM-DD HH24:MI:SS'), 'A', 'Others');
INSERT INTO "PROFAIL" VALUES ('3', 'Abby', 'Local', 'F', TO_DATE('2004-02-15 00:00:00', 'YYYY-MM-DD HH24:MI:SS'), 'A', 'Malay');
INSERT INTO "PROFAIL" VALUES ('4', 'Mimi', 'Local', 'F', TO_DATE('2002-03-02 00:00:00', 'YYYY-MM-DD HH24:MI:SS'), 'A', 'Indian');

-- ----------------------------
-- Indexes structure for table PROFAIL
-- ----------------------------

-- ----------------------------
-- Primary Key structure for table "PROFAIL"
-- ----------------------------
ALTER TABLE "PROFAIL" ADD PRIMARY KEY ("PROFAIL_ID");

Open in new window

I currently have query for:

get age between 2 and 6
for Result A

SELECT
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' AND RACE = 'Malay' THEN PROFAIL END) AS GA1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' AND RACE = 'Malay' THEN PROFAIL END) AS GA2,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' AND RACE = 'Chinese' THEN PROFAIL END) AS GB1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' AND RACE = 'Chinese' THEN PROFAIL END) AS GB2,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' AND RACE = 'Indian' THEN PROFAIL END) AS GC1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' AND RACE = 'Indian' THEN PROFAIL END) AS GC2,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' AND RACE = 'Others' THEN PROFAIL END) AS GD1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' AND RACE = 'Others' THEN PROFAIL END) AS GD2,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' THEN PROFAIL END) AS J1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' THEN PROFAIL END) AS J2


FROM

(SELECT a.PROFAIL_ID as PROFAIL, GENDER , RACE, TO_CHAR (CURRENT_TIMESTAMP, 'YYYY') - TO_CHAR (BIRTH_DATE, 'YYYY') AS AGE FROM PROFAIL a, MATCH b
WHERE a.PROFAIL_ID = b.PROFAIL_ID
AND PROFAIL_TYPE ='A'
AND NATIONALITY = 'Local'
AND MATCH = 'A'
AND RESULT = 'Good'
)

Open in new window

Result B:

SELECT
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' AND RACE = 'Malay' THEN PROFAIL END) AS GA1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' AND RACE = 'Malay' THEN PROFAIL END) AS GA2,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' AND RACE = 'Chinese' THEN PROFAIL END) AS GB1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' AND RACE = 'Chinese' THEN PROFAIL END) AS GB2,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' AND RACE = 'Indian' THEN PROFAIL END) AS GC1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' AND RACE = 'Indian' THEN PROFAIL END) AS GC2,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' AND RACE = 'Others' THEN PROFAIL END) AS GD1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' AND RACE = 'Others' THEN PROFAIL END) AS GD2,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' THEN PROFAIL END) AS J1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' THEN PROFAIL END) AS J2


FROM

(SELECT a.PROFAIL_ID as PROFAIL, GENDER , RACE, TO_CHAR (CURRENT_TIMESTAMP, 'YYYY') - TO_CHAR (BIRTH_DATE, 'YYYY') AS AGE FROM PROFAIL a, MATCH b
WHERE a.PROFAIL_ID = b.PROFAIL_ID
AND PROFAIL_TYPE ='A'
AND NATIONALITY = 'Local'
AND MATCH = 'A'
AND RESULT = 'Good'
)

Open in new window

Problem : How do i combine result A and B into 1 query and able to get the result as picture below togather with total:

Thanks

Sean Stuber

what's the difference between A and B?

also note, your results are blank, do you want the number of children that fall into each category?

boon86

ASKER

sorry, result B query is:

SELECT
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' AND RACE = 'Malay' THEN PROFAIL END) AS HA1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' AND RACE = 'Malay' THEN PROFAIL END) AS HA2,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' AND RACE = 'Chinese' THEN PROFAIL END) AS HB1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' AND RACE = 'Chinese' THEN PROFAIL END) AS HB2,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' AND RACE = 'Indian' THEN PROFAIL END) AS HC1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' AND RACE = 'Indian' THEN PROFAIL END) AS HC2,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' AND RACE = 'Others' THEN PROFAIL END) AS HD1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' AND RACE = 'Others' THEN PROFAIL END) AS HD2,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' THEN PROFAIL END) AS HJ1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' THEN PROFAIL END) AS HJ2


FROM

(SELECT a.PROFAIL_ID as PROFAIL, GENDER , RACE, TO_CHAR (CURRENT_TIMESTAMP, 'YYYY') - TO_CHAR (BIRTH_DATE, 'YYYY') AS AGE FROM PROFAIL a, MATCH b
WHERE a.PROFAIL_ID = b.PROFAIL_ID
AND PROFAIL_TYPE ='A'
AND NATIONALITY = 'Local'
AND MATCH = 'A'
AND RESULT = 'Bad'
)

Open in new window

Sean Stuber

actually, I don't care what your A and B queries are,

what do you want the results to be?

boon86

ASKER

For Result A, should display the Result = Good depends on age, race and gender
by counting on number of Profail_Id involved in that condition.

For Result B, should display the Result = Bad depends on age, race and gender
by counting on number of Profail_Id involved in that condition too.

FYI : GA1 is column for Malay, Male
GA2 is column for Malay, Female
GB1 is column for Chinese, Male
GB2 is column for Chinese, Female and so on...

Sean Stuber

what are the numberic values you are expecting for each row/column?

boon86

ASKER

The numeric values should be the number of children with corresponding age, gender and race

Sean Stuber

I mean, show me the actual values you are expecting

1, 0, 3, 5, 2, 7, 1, ....
0, 2, 4, 8, 1, 3, 6, ....

boon86

ASKER

For row age between 2 and 6,

Result A should be this ,

Result B should be this,

This two results should appear in the same row

Sean Stuber

your output table has 4 rows, your queries only produce 1 row.

I don't need the queries you currently have
I won't use them anyway since the results need to be combined

can you fill in your entire table with what you are expecting?

Lowfatspread

get admin to remove ms sql server from this its an oracle(?) query

you join the two queries via a full outer join...

select coalesce(a.type,b.type) as Type
,a.GA1
,a.GA2
,a.GB1
,a.GB2
,b.GA1
,b.GA2
,b.GB1
,b.GB2
,coalesce(a.j1,0)+coalesce(b.j1,0)
from (query a) as a
full outer join (queryb) as b
on a.type=b.type
order by 1

your query a and b need to include a Type (probably the profail id?)

also you probably should set up the age range criteria in a table and use that in the a and b queries

boon86

ASKER

boon86

ASKER

Lowfat,

full outer join did not working:

[SQL] select coalesce(a.type,b.type) as Type
    ,a.GA1
    ,a.GA2 
     ,a.GB1 
     ,a.GB2
,b.GA1
    ,b.GA2 
     ,b.GB1 
     ,b.GB2
,coalesce(a.j1,0)+coalesce(b.j1,0) 
from 

(
SELECT
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' AND RACE = 'Malay' THEN PROFAIL END) AS GA1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' AND RACE = 'Malay' THEN PROFAIL END) AS GA2,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' AND RACE = 'Chinese' THEN PROFAIL END) AS GB1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' AND RACE = 'Chinese' THEN PROFAIL END) AS GB2,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' AND RACE = 'Indian' THEN PROFAIL END) AS GC1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' AND RACE = 'Indian' THEN PROFAIL END) AS GC2,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' AND RACE = 'Others' THEN PROFAIL END) AS GD1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' AND RACE = 'Others' THEN PROFAIL END) AS GD2,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' THEN PROFAIL END) AS J1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' THEN PROFAIL END) AS J2


FROM

(SELECT a.PROFAIL_ID as PROFAIL, GENDER , RACE, TO_CHAR (CURRENT_TIMESTAMP, 'YYYY') - TO_CHAR (BIRTH_DATE, 'YYYY') AS AGE FROM PROFAIL a, MATCH b
WHERE a.PROFAIL_ID = b.PROFAIL_ID
AND PROFAIL_TYPE ='A'
AND NATIONALITY = 'Local'
AND MATCH = 'A'
AND RESULT = 'Good'
)

) as a
full outer join 
(
SELECT
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' AND RACE = 'Malay' THEN PROFAIL END) AS HA1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' AND RACE = 'Malay' THEN PROFAIL END) AS HA2,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' AND RACE = 'Chinese' THEN PROFAIL END) AS HB1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' AND RACE = 'Chinese' THEN PROFAIL END) AS HB2,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' AND RACE = 'Indian' THEN PROFAIL END) AS HC1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' AND RACE = 'Indian' THEN PROFAIL END) AS HC2,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' AND RACE = 'Others' THEN PROFAIL END) AS HD1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' AND RACE = 'Others' THEN PROFAIL END) AS HD2,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='M' THEN PROFAIL END) AS HJ1,
COUNT(CASE WHEN AGE BETWEEN 2 AND 6 AND GENDER ='F' THEN PROFAIL END) AS HJ2


FROM

(SELECT a.PROFAIL_ID as PROFAIL, GENDER , RACE, TO_CHAR (CURRENT_TIMESTAMP, 'YYYY') - TO_CHAR (BIRTH_DATE, 'YYYY') AS AGE FROM PROFAIL a, MATCH b
WHERE a.PROFAIL_ID = b.PROFAIL_ID
AND PROFAIL_TYPE ='A'
AND NATIONALITY = 'Local'
AND MATCH = 'A'
AND RESULT = 'Bad'
)
) as b
on a.type=b.type
order by 1
[Err] ORA-00933: SQL command not properly ended

Open in new window

Sean Stuber

SELECT age_group,
a_malay_m,
a_malay_f,
a_chinese_m,
a_chinese_f,
a_indian_m,
a_indian_f,
a_others_m,
a_others_f,
(a_malay_m + a_chinese_m + a_indian_m + a_others_m) a_total_m,
(a_malay_f + a_chinese_f + a_indian_f + a_others_f) a_total_f,
b_malay_m,
b_malay_f,
b_chinese_m,
b_chinese_f,
b_indian_m,
b_indian_f,
b_others_m,
b_others_f,
(b_malay_m + b_chinese_m + b_indian_m + b_others_m) b_total_m,
(b_malay_f + b_chinese_f + b_indian_f + b_others_f) b_total_f
FROM (SELECT DECODE(GROUPING_ID(lo), 1, 'OVERALL TOTAL', lo) age_group,
COUNT(CASE WHEN race = 'Malay' AND gender = 'M' AND result = 'Good' THEN 1 END)
a_malay_m,
COUNT(CASE WHEN race = 'Malay' AND gender = 'F' AND result = 'Good' THEN 1 END)
a_malay_f,
COUNT(CASE WHEN race = 'Chinese' AND gender = 'M' AND result = 'Good' THEN 1 END)
a_chinese_m,
COUNT(CASE WHEN race = 'Chinese' AND gender = 'F' AND result = 'Good' THEN 1 END)
a_chinese_f,
COUNT(CASE WHEN race = 'Indian' AND gender = 'M' AND result = 'Good' THEN 1 END)
a_indian_m,
COUNT(CASE WHEN race = 'Indian' AND gender = 'F' AND result = 'Good' THEN 1 END)
a_indian_f,
COUNT(CASE WHEN race = 'Others' AND gender = 'M' AND result = 'Good' THEN 1 END)
a_others_m,
COUNT(CASE WHEN race = 'Others' AND gender = 'F' AND result = 'Good' THEN 1 END)
a_others_f,
COUNT(CASE WHEN race = 'Malay' AND gender = 'M' AND result = 'Good' THEN 1 END)
b_malay_m,
COUNT(CASE WHEN race = 'Malay' AND gender = 'F' AND result = 'Good' THEN 1 END)
b_malay_f,
COUNT(CASE WHEN race = 'Chinese' AND gender = 'M' AND result = 'Good' THEN 1 END)
b_chinese_m,
COUNT(CASE WHEN race = 'Chinese' AND gender = 'F' AND result = 'Good' THEN 1 END)
b_chinese_f,
COUNT(CASE WHEN race = 'Indian' AND gender = 'M' AND result = 'Good' THEN 1 END)
b_indian_m,
COUNT(CASE WHEN race = 'Indian' AND gender = 'F' AND result = 'Good' THEN 1 END)
b_indian_f,
COUNT(CASE WHEN race = 'Others' AND gender = 'M' AND result = 'Good' THEN 1 END)
b_others_m,
COUNT(CASE WHEN race = 'Others' AND gender = 'F' AND result = 'Good' THEN 1 END)
b_others_f
FROM profail a
JOIN match b
ON a.profail_id = b.profail_id
AND profail_type = 'A'
AND nationality = 'Local'
AND match = 'A'
RIGHT JOIN (SELECT 0 lo, 1 hi FROM DUAL
UNION ALL
SELECT 2, 6 FROM DUAL
UNION ALL
SELECT 7, 12 FROM DUAL)
ON FLOOR(MONTHS_BETWEEN(SYSDATE, birth_date) / 12) BETWEEN lo AND hi
GROUP BY ROLLUP(lo)
ORDER BY GROUPING_ID(lo), lo)

Lowfatspread

more like this

you didn't complete the statement correctly... or substitute in a value for "type"

select coalesce(a.lo,b.lo) as lo,coalesce(a.hi,b.hi) as hi
    ,a.GA1
    ,a.GA2 
     ,a.GB1 
     ,a.GB2
,b.hA1
    ,b.hA2 
     ,b.hB1 
     ,b.hB2
,coalesce(a.j1,0)+coalesce(b.hj1,0) 
from 

(
SELECT lo,hi,
COUNT(CASE WHEN AGE between lo and hi AND GENDER ='M' AND RACE = 'Malay' THEN PROFAIL END) AS GA1,
COUNT(CASE WHEN AGE between lo and hi AND GENDER ='F' AND RACE = 'Malay' THEN PROFAIL END) AS GA2,
COUNT(CASE WHEN AGE between lo and hi AND GENDER ='M' AND RACE = 'Chinese' THEN PROFAIL END) AS GB1,
COUNT(CASE WHEN AGE between lo and hi AND GENDER ='F' AND RACE = 'Chinese' THEN PROFAIL END) AS GB2,
COUNT(CASE WHEN AGE between lo and hi AND GENDER ='M' AND RACE = 'Indian' THEN PROFAIL END) AS GC1,
COUNT(CASE WHEN AGE between lo and hi AND GENDER ='F' AND RACE = 'Indian' THEN PROFAIL END) AS GC2,
COUNT(CASE WHEN AGE between lo and hi AND GENDER ='M' AND RACE = 'Others' THEN PROFAIL END) AS GD1,
COUNT(CASE WHEN AGE between lo and hi AND GENDER ='F' AND RACE = 'Others' THEN PROFAIL END) AS GD2,
COUNT(CASE WHEN AGE between lo and hi AND GENDER ='M' THEN PROFAIL END) AS J1,
COUNT(CASE WHEN AGE between lo and hi AND GENDER ='F' THEN PROFAIL END) AS J2


FROM

(SELECT a.PROFAIL_ID as PROFAIL, GENDER , RACE, TO_CHAR (CURRENT_TIMESTAMP, 'YYYY') - TO_CHAR (BIRTH_DATE, 'YYYY') AS AGE 
  FROM PROFAIL a, MATCH b,
  (SELECT 0 lo, 1 hi FROM DUAL
                            UNION ALL
                            SELECT 2, 6 FROM DUAL
                            UNION ALL
                            SELECT 7, 12 FROM DUAL) as c
WHERE a.PROFAIL_ID = b.PROFAIL_ID
AND PROFAIL_TYPE ='A'
AND NATIONALITY = 'Local'
AND MATCH = 'A'
AND RESULT = 'Good'
) as x
group by lo,hi
) as a
full outer join 
(
SELECT lo,hi
COUNT(CASE WHEN AGE between lo and hi AND GENDER ='M' AND RACE = 'Malay' THEN PROFAIL END) AS HA1,
COUNT(CASE WHEN AGE between lo and hi AND GENDER ='F' AND RACE = 'Malay' THEN PROFAIL END) AS HA2,
COUNT(CASE WHEN AGE between lo and hi AND GENDER ='M' AND RACE = 'Chinese' THEN PROFAIL END) AS HB1,
COUNT(CASE WHEN AGE between lo and hi AND GENDER ='F' AND RACE = 'Chinese' THEN PROFAIL END) AS HB2,
COUNT(CASE WHEN AGE between lo and hi AND GENDER ='M' AND RACE = 'Indian' THEN PROFAIL END) AS HC1,
COUNT(CASE WHEN AGE between lo and hi AND GENDER ='F' AND RACE = 'Indian' THEN PROFAIL END) AS HC2,
COUNT(CASE WHEN AGE between lo and hi AND GENDER ='M' AND RACE = 'Others' THEN PROFAIL END) AS HD1,
COUNT(CASE WHEN AGE between lo and hi AND GENDER ='F' AND RACE = 'Others' THEN PROFAIL END) AS HD2,
COUNT(CASE WHEN AGE between lo and hi AND GENDER ='M' THEN PROFAIL END) AS HJ1,
COUNT(CASE WHEN AGE between lo and hi AND GENDER ='F' THEN PROFAIL END) AS HJ2


FROM

(SELECT a.PROFAIL_ID as PROFAIL, GENDER , RACE, TO_CHAR (CURRENT_TIMESTAMP, 'YYYY') - TO_CHAR (BIRTH_DATE, 'YYYY') AS AGE
 FROM PROFAIL a, MATCH b, (SELECT 0 lo, 1 hi FROM DUAL
                            UNION ALL
                            SELECT 2, 6 FROM DUAL
                            UNION ALL
                            SELECT 7, 12 FROM DUAL) as c
WHERE a.PROFAIL_ID = b.PROFAIL_ID
AND PROFAIL_TYPE ='A'
AND NATIONALITY = 'Local'
AND MATCH = 'A'
AND RESULT = 'Bad'
) as x
group by lo,hi
) as b
on a.lo=b.lo
and a.hi=b.hi
order by 1

Open in new window

Sean Stuber

as noted above this question is zoned for Oracle and MS SQL Server
your previous question was zoned for Oracle and MySQL

the previous question is closed, but this one can still be changed.

my answers have been oracle specific, what are the appropriate zones here?

boon86

ASKER

sdstuber, thanks for the outstanding query, it work
perfect. but i dont understand where did you get:

lo
hi
DUAL

Sean Stuber

dual is a built in table of oracle
it has one row of one column but that's not important.
the utility of it is for including functions or constants where you don't really care about the contents of table or view

in this case I wanted hi and lo values for the age ranges so, this

SELECT 0 lo, 1 hi FROM DUAL
UNION ALL
SELECT 2, 6 FROM DUAL
UNION ALL
SELECT 7, 12 FROM DUAL

lets me create data from "nothing"

ASKER CERTIFIED SOLUTION

Sean Stuber

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

boon86

ASKER

If my start age range is nt start from zero,

FROM (SELECT DECODE(GROUPING_ID(lo), 1, 'OVERALL TOTAL', lo) age_group,

the above "lo" can replace with integer '0' ?