DSXchange

I have an Oracle table that is defined with the following SQL: create or replace type animal_array is varray(10) of varchar(30); create table mytable ( field1 animals animal_array ); I want to insert into this table from a flat file that contains: "('cat','dog','pig')" "('frog','toad'...

The solution turned out as follows: I read in the sybase timestamp from the flat file as a VARCHAR. I used a custom-made DS Routine ( 'C' program) to convert the Sybase format to the following format: %dd-%mm-%yyyy %hh:%nn%ss.3 Then I fed that result into a DS-supplied function. Here is my transform...

using Parallel job, converting timestamp from Sybase format to Oracle format. At the project level and at the job level, one can use the system-provided timestamp format, or change it to a user-defined format. I used the system-default, and I used multiple format attempts of my own in both levels. N...

Thanks, it is now working.
-- john

I am not getting anywhere with this Timestamp problem. This post will simplify the problem, in hopes of advice. Within the Seq. File Stage, I now point to a CSV file, each row with 6 fields, with each field in double quotes, and each separated by a coma (and no trailing coma). The 4th field in the r...

There is no space before AM; should there be?

I'll look into DategenericToODBCTimestep()
Thanks
-- john

There appears to be 3 different timestamp formats, viz: a) Sybase, b) Ascential, and c) Oracle. I guess my real problem is that I don't understand how we determine if the data from the input file is accepted as a valid timestamp format (in Sybase or Ascential format???), prior to exercising any deri...

I have input Seq File stage, pointing to a csv file which came from Sybase, with a Timestamp field whose data looks like: Feb 10 2007 10:30:30:450AM 1) In the columns view (of the Seq File Stage) should the length of the timestamp be specified? 2) What type of conversion (function, macro, or whateve...

I'm using Paralled Jobs (in DataStage 7.5.2) I have an input Seq File Stage (pointing to a CSV file), linked to a Transform Stage, then output to another Seq File Stage. When I run, the output log shows 4 rows that were not processed because they had too many columns (a text field occassionally had ...