DSXchange

Hi ,

I have a file created by a server job with data type integer(10) and while I'm designing a parallel job I mentioned it as varchar(255), my question is what is the maximum varchar length I can give in datastage. Because when ever I import a meta data it gives me varchar(255) , is varchar(255) maximum or I can define varchar(1023) for integer(10) field.



Thanks,
Raju

Because text files don't have data types (other than "text") VarChar(255) is used as a default. DataStage has a really high limit on length of a VarChar (64K ?); however some databases limit it (for example to 1024) making you use Long VarChar for anything larger.

Thanks ray for the reply. If I don't provide any length to the varchar field in datastage, what is the maximum it takes is it ( 255 ) or is it going to read entire data from the column as you said it is a text field with out trimming or dropping any data from the column.


Thanks,
Raju

it will not drop any data i.e. it will read entire column data. Best way is to try it out

Thanks Kandy/sandy and Ray , I tested the job and if I don't provide any length to varchar field it reads the entire column data.



Thanks,
Raju

This is what's called an "unbounded VarChar".

These have a specific storage strategy in Data Sets, which ordinarily store data (even bounded VarChar) in fixed width format, because it's far more efficient to pack and unpack.