regarding spliting the records

sujatha dasa · Post by **sujatha dasa** » Tue Feb 17, 2009 4:40 am

hi,

In my project there is one requirement that is
our souce is db2 table.For one column the data is comming like below

ws
ts/fe
rs
rs/js/lk/md

the requirement is i need to split the all the records which is having '/'.like below.

ws
ts
fe
rs
rs
js
lk
md

Can any one help on this.

Thanks
sujatha

Sainath.Srinivasan · Post by **Sainath.Srinivasan** » Tue Feb 17, 2009 5:22 am

Convert '/' to newline.

Search for "urgent" and see response from Ray.

sujatha dasa · Post by **sujatha dasa** » Tue Feb 17, 2009 6:18 am

Thnks for ur help.
how to conver '/' to newline .Is there any function for this

Mike · Post by **Mike** » Tue Feb 17, 2009 6:27 am

Did you bother to search the product documentation? Or search the forum? The function you need has a very obvious name.

Mike

sujatha dasa · Post by **sujatha dasa** » Tue Feb 17, 2009 6:49 am

Thnks for ur help.
how to conver '/' to newline .Is there any function for this

Sainath.Srinivasan · Post by **Sainath.Srinivasan** » Tue Feb 17, 2009 7:41 am

Search can be found right below the banner.

DSguru2B · Post by **DSguru2B** » Tue Feb 17, 2009 8:59 am

Read about the CONVERT() function from datastage help.

mandyli · Post by **mandyli** » Tue Feb 17, 2009 10:00 am

Hi

Try to use Pivot Stage or try to write Unix script to do that.

I hope convert function it will not do this function.

Thanks
Man

chulett · Post by **chulett** » Tue Feb 17, 2009 10:08 am

mandyli wrote:I hope convert function it will not do this function.

Sure it will... write out one record, read back many. There's a FAQ posting on this technique as well.

girija · Post by **girija** » Tue Feb 17, 2009 10:22 am

Could you please tell us detail about your requirement? If you are selecting only one column from DB2 and try to get the result as you shown then CONVERT can do this. Use Convert function and write into another seqfile, and read the same file in the next job. But if you are selecting multiple column from your source along with the specified column and need the output as you mentioned along with other columns then CONVERT is not the right solution for that.

ray.wurlod · Post by **ray.wurlod** » Tue Feb 17, 2009 2:47 pm

Does "very argent" mean you're prepared to pay?

sujatha dasa · Post by **sujatha dasa** » Wed Feb 18, 2009 3:24 am

Thnks for ur response.

Convert is used to Replaces every instance of specified characters in a string with substitute characters.

but my requirement is i have to convert the data in to the records whereever '/' will come in incomming data.
Iam using only one column in db2.

Eg;

if the data contains like

we
ws/ms/ds
rs/fd/ff/ss/ll
re

i want the output like below
we
ws
ms
ds
rs
fd
ff
ss
ll
re

sujatha dasa · Post by **sujatha dasa** » Wed Feb 18, 2009 3:26 am

Thnks for ur response.

Convert is used to Replaces every instance of specified characters in a string with substitute characters.

but my requirement is i have to convert the data in to the records whereever '/' will come in incomming data.
Iam using only one column in db2.

Eg;

if the data contains like

we
ws/ms/ds
rs/fd/ff/ss/ll
re

i want the output like below
we
ws
ms
ds
rs
fd
ff
ss
ll
re

Sainath.Srinivasan · Post by **Sainath.Srinivasan** » Wed Feb 18, 2009 5:17 am

Convert is used to Replaces every instance of specified characters in a string with substitute characters.

interpret as

Convert is used to Replaces every instance of '/' in a string with 'newline' characters.

Manually think through the function and expected outcome

chulett · Post by **chulett** » Wed Feb 18, 2009 8:37 am

Your question has already been answered and Convert() can do what you need. As noted, this FAQ discusses the same technique.