how seq_long create

[hechmi]

HI ALL;

i want to create column with au to incrrement depend from ID key 

ACCOUNTID and ACCOUNTID+1 have same ACCOUNTNUmber

exemple :

ACCOUNTID ACCOUNTNUmber

1   002-00000001
2   002-00000002
3   002-00000003
i want to have this :

ACCOUNTID ACCOUNTNUmber

1   002-00000001
2   002-00000001
3   002-00000002
4   002-00000002
5   002-00000003

6   002-00000003

i put screen frothe job and the table i need .

Thank you .

[akumar2301]

Your requirement not very clear. why do you want to duplicate the accountnumber in output?

Anyways , if you would like to duplicate , One of the solution is as below

(Assuming your input is already sorted by AccountId)

duplicate record using tunite ( use same input two times to unite) -- tsortrow ( sort on AccountId asc)

1   002-00000001

1   002-00000001
2   002-00000002

2   002-00000002
3   002-00000003

3   002-00000003

Now, add additional column ( seq-id) using tMap/tJavaRow

1  1 002-00000001

1  2 002-00000001
2  3 002-00000002

2  4 002-00000002
3  5 002-00000003

3  6 002-00000003

Drop 1st Column

[hechmi]

Thank you ;
i duplicate AccountNumber cause it s type string in output .
*AccountId type Long
itry with tunite but i can t .can you help me hw to use tunite and tjava row .
Regards
Hechmi

[hechmi]

Thank you ;
i duplicate AccountNumber cause it s type string in output and i want to have at this format "001-00000001".
*AccountId type Long
i try with tunite but i can't .can you help me how to use tunite and tjava row .
Regards
Hechmi

[akumar2301]

Hello Another simple way to do the same

Connect your input to tMap as Main. Create a Lookup having two rows and do cross-Join. Refer below screen shot.

Result

[statistics] connecting to socket on port 3449
[statistics] connected
1|002-00000001
2|002-00000001
3|002-00000002
4|002-00000002
5|002-00000003
6|002-00000003
[statistics] disconnected

[hechmi]

Thank you for your answer @uganesh 

I try with your idea but in have different result you can see screen attach .

can you give me where the problem ?

job.JPG
result.JPG
tmap.JPG

[akumar2301]

In tMap , do not join id column to dummy. If you will join , result will not be duplicated.

You do not need two parallel flow at the starts and tJoin. One flow itself would be enough as shown in previous screenshot.

[hechmi]

thank you  @uganesh 

I try with this job but i don't have result 

you can see it ,what is the problem 

|account_number|AccountID|name

001-00000002 |1 |teste|
|003-00000004 |2 |teste|  account number different ?
|005-00000006 |3 |teste|
|007-00000008 |4 |teste| 

best regards

tmap2.JPG

[hechmi · Accepted solution]

Tkank you @uganesh 

I find a solution that i change the name of string in sequence at new sequence .

(out2.CUAccountID%2)==0 ? 
((""+(1000 +NUMERIC_S_LONG.LongSequence("s2",001L, 1))).substring(1,4) ) +"-"+

((""+(100000000 +NUMERIC_S_LONG.LongSequence("s3", 00000001L, 1) ) ).substring(1,9)):
((""+(1000 +NUMERIC_S_LONG.LongSequence("s4",001L, 1))).substring(1,4) ) +"-"+
((""+(100000000 +NUMERIC_S_LONG.LongSequence("s5", 00000001L, 1))).substring(1,9))

/*NUMERIC_S_LONG.LongSequence("s1",001L, 1)) :CUAccountID*/

result 

1    001-00000001
2    001-00000001
3    002-00000002
4   002-00000002

Best regards