DSXchange

Convert(Char(191),'',LinkName.FieldName)

Could you define "connect" a bit more? a) by "connect outside of my client" you mean that from your PC you can use some tool to connect to your database (e.q. DB2connect or slqplus). b) what do you mean by "in DS" with regards to connecting? c) what do you mean by "...

Ok, let us take a step back. Is your problem reading the timestamp value in your datastage job or writing it to the CFF?

Is the column nullable in datastage? What is the data type of the target DB2 column?

The join stage will output data sorted on the join keys. If the sort order is different before the join, DataStage will perform implicit sorts on the input streams.

It looks good and correct from what I see - except that you define "RtMCExtract" but call "ParesenExtract" in your job, could you change that?

Floating point numbers are represented internally as a mantissa and exponent of e; thus not all possible floating point numbers can be represented exactly (this is one reason why financial institutions don't use the type but used fixed decimal representation instead). In this case I would guess that...

Paul has been in the news a lot today, as many consider it his fault that Germany is not world champion :) They are thinking of cremating him and keeping his remains - perhaps this is going to be the birth of a new tradition, something like "The ASHES" in cricket! http://www.11freunde.de/i...

What is the actual format of the TIMESTAMP field in the file? If it is not an ASCII representation but a binary value, then you would need to know the exact format before it can be converted to another representation. Often the format, if binary, is a floating point number with some form of Julian d...

Declare your output format as a single VarChar column. In your job, detect the value group changes using either stage variables in a transform or within a sort stage. When your job detects a group change it outputs an additional record with the footer values as indicated in your example.

Field 31 is a multivalue one which is expanded correctly when I perform this query manually in my environment and all elements are returned. In your case somehow the first and last elements are skipped. Is it the same behaviour if you choose another job sequence?

There is a reason for premium memberships to DSXChange and in this case there would seem to be two possible choices, either I could post the answer again as a non-premium message or you could change to premium membership. I am sure you can understand why choice 2 is the better route. Of course anoth...

Print out the contents of the variable "Output" to see what happens, i.e. do a

You can use the convert() function to convert the characters you don't want into empty strings, i.e.
(ASCII 9, 10 and 13 are TAB, LF and CR respectively)

If you remove the "ParsenExtract" your job will compile, so you need to fix your definition of the external routine. Have you declared it correctly and compiled it?