Home Page for the TeradataForum
 

Archives of the TeradataForum

Message Posted: Mon, 30 Dec 2002 @ 18:48:05 GMT


     
  <Prev Next>   <<First <Prev Next> Last>>  


Subj:   Re: Inmod and Mload VARTEXT
 
From:   Calvert, Jim

Dara:

Multiload usually expects to receive the rows from the INMOD in FASTLOAD format with a twist.

1) The fields must be in datatype format, so in your case, the VARCHARs must be preceded with a valid length in 2 bytes.

2) There must NOT be a row length field preceding the actual data row. In the INMOD, you must supply a valid row length as one of the return parameters.

3) There must be NO separator character ('|' for example ).

4) In a UNIX network environment, the last character in the row must be \n (a newline character).

5) You must take care of the NULLS somehow. Some methods:

- ensure every field is not NULL, not very practical.

- test known values with the Multiload NULLIF clause; very cpu expensive in the host.

- add NULL test fields to the data row and test with CASE statements in the Values clause; somewhat better cpu economy.

- calculate INDICATORS and preface the data row with these; best cpu economy and my preference.

6) In a VARTEXT scenario, look for adjacent separator characters.

I think that should solve your issues.

Jim Calvert - Teradata



     
  <Prev Next>   <<First <Prev Next> Last>>  
 
 
 
 
 
 
 
 
  
  Top Home Privacy Feedback  
 
 
Copyright for the TeradataForum (TDATA-L), Manta BlueSky    
Copyright 2016 - All Rights Reserved    
Last Modified: 15 Jun 2023