Extraction

India2000 · Post by **India2000** » Thu Sep 23, 2010 1:58 pm

I have a series of data coming in for different column fields.How shall I extract values for each field using seq file stage?

chulett · Post by **chulett** » Thu Sep 23, 2010 2:04 pm

No idea what that means.

Can you provide an example or three? Something to show us, make us understand what "a series of data coming in for different column fields" would mean? Thanks.

chulett · Post by **chulett** » Thu Sep 23, 2010 2:19 pm

Or maybe you just need to read it as a fixed-width file. Can you post examples please of what you are looking at, before and after?

India2000 · Post by **India2000** » Thu Sep 23, 2010 2:29 pm

chulett wrote:Or maybe you just need to read it as a fixed-width file. Can you post examples please of what you are looking at, before and after?

etc,..

chulett · Post by **chulett** » Thu Sep 23, 2010 2:33 pm

Read it as fixed-width.

India2000 · Post by **India2000** » Thu Sep 23, 2010 2:39 pm

chulett wrote:Read it as fixed-width.

it has 20 fields.. if i read it as a fixed with file i get only 5 fields..

kris007 · Post by **kris007** » Thu Sep 23, 2010 2:54 pm

Not sure how you will end up with 5 fields if you read it as a fixed width file

If you know the layout of the file, it should work just fine.

Are you expecting blanks within your data where you intend to just add commas? You can read it as a single column and split them in a transformer stage as well if you want to.

India2000 · Post by **India2000** » Thu Sep 23, 2010 2:56 pm

kris007 wrote:Not sure how you will end up with 5 fields if you read it as a fixed width file :?

If you know the layout of the file, it should work just fine.

Are you expecting blanks within your data where you intend to just add commas? You can read it as a single column and split them in a transformer stage as well if you want to.

first field is like as I sent in the example..next are fields with tabs

anbu · Post by **anbu** » Thu Sep 23, 2010 3:01 pm

Read as tab delimited file. Then use substring function to split further

arunkumarmm · Post by **arunkumarmm** » Thu Sep 23, 2010 3:06 pm

Do you know the exact length of the record? If so as suggested earlier you can read it in a single column. And if you are not sure about the record length and if you are sure you have a line terminator, you can define a column like varchar 999 and read it as a delimited file with '000' in the delimiter and the quote character and split it in a transformer

kris007 · Post by **kris007** » Thu Sep 23, 2010 3:13 pm

India2000 wrote: first field is like as I sent in the example..next are fields with tabs

Again, if you can provide an example for people to you help you correctly, you will get an answer fast.

As Craig had asked, provide a sample record for all the 20 fields that are coming in so that we can chip in with our suggestions. It's hard for us to come up with an answer that works for you when you provide bits and pieces of your actual case.

India2000 · Post by **India2000** » Thu Sep 23, 2010 4:53 pm

kris007 wrote:
India2000 wrote: first field is like as I sent in the example..next are fields with tabs
Again, if you can provide an example for people to you help you correctly, you will get an answer fast.

As Craig had asked, provide a sample record for all the 20 fields that are coming in so that we can chip in with our suggestions. It's hard for us to come up with an answer that works for you when you provide bits and pieces of your actual case.

ray.wurlod · Post by **ray.wurlod** » Thu Sep 23, 2010 6:06 pm

Import the table definition. Specify fixed width. Specify column widths of 5,1,6 (or whatever is correct). Define "proper" column names. Use that table definition for the Format as well as the Columns tab. All will be well.

IBM Analytics Champion 2009 - 2020 · Post by **asorrell** » Mon Sep 27, 2010 9:15 am

India2000 - You need to do what people ask you to do. You have been asked five times by different posters to give examples of your text if you want proper help. Stop wasting our time!