DSXchange

Read the XML file and parse the tags. When you've found the start tag start collecting characters until you find the end tag. You can use the Index() function to determine the character position of the tags.

When you're processing Shipping Address, use a different file dictionary that identifies the business address fields as something innocuous.

Make sure that your data are, or are transformed to, fixed width so that there can be no ambiguity about what's in what column.

Isn't 39202 indicative that the server licence has expired?

If you're in a server job you can use Iconv(thevalue, "MDn") where n is the number of decimal places. This function handles signs at either end of the value.

My team lead has decided not to go with BASIC Tfr as he needs this to be a pristine DS-EE environment. Thinking of a word that rhymes with "anchor" but isn't "banker". The DS-EE environment includes the BASIC Transformer stage. Therefore including such a stage does not in any way detract from your ...

Integer (or int8) can only contain values less than 2^31 (2147483648).

AsInteger must fail for BigInt where the value is larger than that if the target column's data type is Integer.

The manual (Parallel Job Developer's Guide) is not clear on whether AsInteger can handle BigInt (or int16).

Technically the Null Field Value must be of the correct data type for the target to receive. Luckily, in this case, the target is a text file. Text files do not actually have data types - they contain only text.

In the parallel job the inputs to the Change Capture stage need not only to be sorted identically on the comparison keys but also hash partitioned identically on the comparison keys. Have you ensured that this is the case? Change Capture stage will report whether the second input row is an edit of t...

Actually, the value $APT_CONFIG_FILE is meaningless in the UNIX environment.

Subtract the years.
Compare the dates (month and day only).
If current month and day > DOB month and day in this year, add one to year difference.

The Compare stage has a Warn on Record Count Mismatch property that will log a warning. While that's not your complete solution (nor is Abort on Mismatch), it's worth noting. A Compare stage only effects the comparison (philosophy: one stage, one task) and emits a coded value to indicate the result....

Try defining the subrecords so that the individual column names within them do not clash. The following is conceptual, may not be syntactically accurate. record schema ( result: int8; oldrecord: subrecord { oldfirstcol: int32; oldsecondcol: string[max=32]; oldthirdcol: nullable date; }...

Are you using the Enterprise stage or the API stage?

You can explicitly code Notification activity or Routine activity or Execute Command activity in Job Sequence 2 to handle either of Job3 or Job4 failing.