Reading a fixed length flat file- no delimiter

[dsedi]

Hi All,

We started receiving some flat file inputs without any delimiter on it.
(continious stream of charecters) We need to split into columns based on the charecter position in the paricular row.
(for ex, char 1 to 10 Column1,char 11 to 15 column2..etc....)
which stage will be most appropriate to handle this?
any valuable Idea's?

Thanks in advance.
Edi

[jenkinsrob]

You should be able to handle this with the Sequential File Stage.

Set all your datatypes to CHAR and specify the length of each column in the file.

[ray.wurlod]

Exactly; the Sequential File stage is precisely the correct stage type to use.

You will need to specify the format as fixed width. The best way to do this is to import the file's "table definition" and specify there that it's fixed width format and the width of each column.

From the table definition in the Repository you can then Load the format into a Sequential File stage in your job design. The format information is largely untouched by human hands and therefore likely to be correct.